IFIP Advances in Information and Communication Technology
347
Editor-in-Chief A. Joe Turner, Seneca, SC, USA
Editorial Board Foundations of Computer Science Mike Hinchey, Lero, Limerick, Ireland Software: Theory and Practice Bertrand Meyer, ETH Zurich, Switzerland Education Arthur Tatnall, Victoria University, Melbourne, Australia Information Technology Applications Ronald Waxman, EDA Standards Consulting, Beachwood, OH, USA Communication Systems Guy Leduc, Université de Liège, Belgium System Modeling and Optimization Jacques Henry, Université de Bordeaux, France Information Systems Jan Pries-Heje, Roskilde University, Denmark Relationship between Computers and Society Jackie Phahlamohlaka, CSIR, Pretoria, South Africa Computer Systems Technology Paolo Prinetto, Politecnico di Torino, Italy Security and Privacy Protection in Information Processing Systems Kai Rannenberg, Goethe University Frankfurt, Germany Artificial Intelligence Tharam Dillon, Curtin University, Bentley, Australia Human-Computer Interaction Annelise Mark Pejtersen, Center of Cognitive Systems Engineering, Denmark Entertainment Computing Ryohei Nakatsu, National University of Singapore
IFIP – The International Federation for Information Processing IFIP was founded in 1960 under the auspices of UNESCO, following the First World Computer Congress held in Paris the previous year. An umbrella organization for societies working in information processing, IFIP’s aim is two-fold: to support information processing within its member countries and to encourage technology transfer to developing nations. As its mission statement clearly states, IFIP’s mission is to be the leading, truly international, apolitical organization which encourages and assists in the development, exploitation and application of information technology for the benefit of all people. IFIP is a non-profitmaking organization, run almost solely by 2500 volunteers. It operates through a number of technical committees, which organize events and publications. IFIP’s events range from an international congress to local seminars, but the most important are: • The IFIP World Computer Congress, held every second year; • Open conferences; • Working conferences. The flagship event is the IFIP World Computer Congress, at which both invited and contributed papers are presented. Contributed papers are rigorously refereed and the rejection rate is high. As with the Congress, participation in the open conferences is open to all and papers may be invited or submitted. Again, submitted papers are stringently refereed. The working conferences are structured differently. They are usually run by a working group and attendance is small and by invitation only. Their purpose is to create an atmosphere conducive to innovation and development. Refereeing is less rigorous and papers are subjected to extensive group discussion. Publications arising from IFIP events vary. The papers presented at the IFIP World Computer Congress and at open conferences are published as conference proceedings, while the results of the working conferences are often published as collections of selected and edited papers. Any national society whose primary activity is in information may apply to become a full member of IFIP, although full membership is restricted to one society per country. Full members are entitled to vote at the annual General Assembly, National societies preferring a less committed involvement may apply for associate or corresponding membership. Associate members enjoy the same benefits as full members, but without voting rights. Corresponding members are not represented in IFIP bodies. Affiliated membership is open to non-national societies, and individual and honorary membership schemes are also offered.
Daoliang Li Yande Liu Yingyi Chen (Eds.)
Computer and Computing Technologies in Agriculture IV 4th IFIP TC 12 Conference, CCTA 2010 Nanchang, China, October 22-25, 2010 Selected Papers, Part IV
13
Volume Editors Daoliang Li Yingyi Chen China Agricultural University EU-China Center for Information & Communication Technologies (CICTA) 17 Tsinghua East Road, Beijing, 100083, P.R. China E-mail: {dliangl, chenyingyi}@cau.edu.cn Yande Liu East China Jiaotong University College of Mechanical and Electronic Engineering Shuanggang Road, Nanchang, 330013 Jiangxi, China E-mail:
[email protected] ISSN 1868-4238 e-ISSN 1868-422X ISBN 978-3-642-18368-3 e-ISBN 978-3-642-18369-0 DOI 10.1007/978-3-642-18369-0 Springer Heidelberg Dordrecht London New York Library of Congress Control Number: 2010942867 CR Subject Classification (1998): I.2.11, H.4, C.3, C.2, D.2, K.4.4
© IFIP International Federation for Information Processing 2011 This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, re-use of illustrations, recitation, broadcasting, reproduction on microfilms or in any other way, and storage in data banks. Duplication of this publication or parts thereof is permitted only under the provisions of the German Copyright Law of September 9, 1965, in its current version, and permission for use must always be obtained from Springer. Violations are liable to prosecution under the German Copyright Law. The use of general descriptive names, registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. Typesetting: Camera-ready by author, data conversion by Scientific Publishing Services, Chennai, India Printed on acid-free paper Springer is part of Springer Science+Business Media (www.springer.com)
Preface
I want to express my sincere thanks to all authors who submitted research papers to the 4th IFIP International Conference on Computer and Computing Technologies in Agriculture and the 4th Symposium on Development of Rural Information (CCTA 2010) that were held in Nanchang, China, 22–25 October 2010. This conference was hosted by CICTA (EU-China Centre for Information & Communication Technologies, China Agricultural University); China Agricultural University; China Society of Agricultural Engineering, China; International Federation for Information Processing (TC12); Beijing Society for Information Technology in Agriculture, China. It was organized by East China Jiaotong University. CICTA focuses on research and development of advanced and practical technologies applied in agriculture and aims at promoting international communication and cooperation. Sustainable agriculture is currently the focus of the whole world, and the application of information technology in agriculture has become more and more important. ‘Informatized agriculture’ has been the goal of many countries recently in order to scientifically manage agriculture to achieve low costs and high income. The topics of CCTA 2010 covered a wide range of interesting theories and applications of information technology in agriculture, including simulation models and decision-support systems for agricultural production, agricultural product quality testing, traceability and e-commerce technology, the application of information and communication technology in agriculture, and universal information service technology and service systems development in rural areas. We selected 352 best papers among those submitted to CCTA 2010 for these proceedings. It is always exciting to have experts, professionals and scholars getting together with creative contributions and sharing inspiring ideas which will hopefully lead to great developments in these technologies. Finally, I would like also to express my sincere thanks to all the authors, speakers, session chairs and attendees for their active participation and support of this conference.
October 2010
Daoliang Li
Conference Organization
Organizer East China Jiaotong University
Organizing Committee Chair Yande Liu
Academic Committee Chair Daoliang Li
Conference Secretariat Lingling Gao
Sponsors China Agricultural University China Society of Agricultural Engineering, China International Federation for Information Processing, Austria Beijing Society for Information Technology in Agriculture, China National Natural Science Foundation of China
Table of Contents – Part IV
A Compression Method of Decision Table Based on Matrix Computation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Laipeng Luo and Ergen Liu
1
A Laplacian of Gaussian-Based Approach for Spot Detection in Two-Dimensional Gel Electrophoresis Images . . . . . . . . . . . . . . . . . . . . . . . . Feng He, Bangshu Xiong, Chengli Sun, and Xiaobin Xia
8
A Leaf Layer Spectral Model for Estimating Protein Content of Wheat Grains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chun-Hua Xiao, Shao-Kun Li, Ke-Ru Wang, Yan-Li Lu, Jun-Hua Bai, Rui-Zhi Xie, Shi-Ju Gao, Qiong Wang, and Fang-Yong Wang
16
A New Color Information Entropy Retrieval Method for Pathological Cell Image . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Xiangang Jiang, Qing Liang, and Tao Shen
30
A New Palm-Print Image Feature Extraction Method Based on Wavelet Transform and Principal Component Analysis . . . . . . . . . . . . . . . . . . . . . . . Jia wei Li and Ming Sun
39
A Non-linear Model of Nondestructive Estimation of Anthocyanin Content in Grapevine Leaves with Visible/Red-Infrared Hyperspectral . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . JiangLin Qin, Donald Rundquist, Anatoly Gitelson, Zongkun Tan, and Mark Steele
47
Application of Improved BP Neural Network in Controlling the Constant-Force Grinding Feed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Zhaoxia Chen, Bailin He, and Xianfeng Xu
63
A Semantic Middleware of Grain Storage Internet . . . . . . . . . . . . . . . . . . . . Siquan Hu, Haiou Wang, Chundong She, and Junfeng Wang
71
AE Feature Analysis on Welding Crack Defects of HG70 Steel Used by Truck Crane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Yantao Dou, Xiaoli Xu, Wei Wang, and Siqin Pang
78
An Equilateral Triangle Waveguide Beam Splitter . . . . . . . . . . . . . . . . . . . . Zhimin Liu, Fengqi Zhou, Hongjian Li, Bin Tang, Zhengfang Liu, Qingping Wu, Aixi Chen, and Kelin Huang
89
Analysis and Implementation of Embedded SNMP Agent . . . . . . . . . . . . . Hubin Deng, Guiyuan Liu, and Lei Zhang
96
VIII
Table of Contents – Part IV
Application of Computer Technology in Advanced Material Science and Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Yajuan Liu
103
Application of Interferometry in Ultrasonic System for Vibration . . . . . . . Zhengping Liu, Shenghang Xu, and Juanjuan Liu
108
Automatic Control System for Highway Tunnel Lighting . . . . . . . . . . . . . . Shijuan Fan, Chao Yang, and Zhiwei Wang
116
Comparative Study of Distance Discriminant Analysis and Bp Neural Network for Identification of Rapeseed Cultivars Using Visible/Near Infrared Spectra . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Qiang Zou, Hui Fang, Fei Liu, Wenwen Kong, and Yong He Current Situation and Prospect of Grassland Management Decision Support Systems in China . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Qingwei Duan, Xiaoping Xin, Guixia Yang, Baorui Chen, Hongbin Zhang, Yuchun Yan, Xu Wang, Baohui Zhang, and Gang Li
124
134
Design Method and Implementation of Ternary Logic Optical Calculator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chunzhi Li and Junyong Yan
147
Design of Automatic Cutting and Welding Machine for Brake Beam-Axle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Leping Liu, Mingdong Zhong, and Qizheng Dong
167
Design of Multifunction Vehicle Bus Controller . . . . . . . . . . . . . . . . . . . . . . Zhongqi Li, Fengping Yang, and Qirong Xing
177
Detection of Soil Total Nitrogen by Vis-SWNIR Spectroscopy . . . . . . . . . Yaoze Feng, Xiaoyu Li, Wei Wang, and Changju Liu
184
Development and Application of Tennis Match Video Retrieval Technology in Multimedia Education . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Shehua Cao
192
Fault Diagnosis of Roller Bearing Based on PCA and Multi-class Support Vector Machine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Guifeng Jia, Shengfa Yuan, and Chengwen Tang
198
Health Status Identification of Connecting Rod Bearing Based on Support Vector Machine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Yongbin Liu, Qingbo He, Ping Zhang, Zhongkui Zhu, and Fanrang Kong Investigation of the Methods for Tool Wear On-Line Monitoring during the Cutting Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Hongjiang Chen
206
215
Table of Contents – Part IV
IX
Magnetic-Field-Based 3D ETREE Modelling for Multi-Frequency Eddy Current Inspection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Yu Zhang and Yong Li
221
Measurement of Self-emitting Magnetic Signals from a Precut Notch of Q235 Steel during Tensile Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Lihong Dong, Binshi Xu, and Shiyun Dong
231
Modeling and Performance Analysis of Giant Magnetostrictive Microgripper with Flexure Hinge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Qinghua Cao, Quanguo Lu, Junmei Xi, Jianwu Yan, and Changbao Chu Non-destructive Measurement of Sugar Content in Chestnuts Using Near-Infrared Spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jie Liu, Xiaoyu Li, Peiwu Li, Wei Wang, Jun Zhang, Wei Zhou, and Zhu Zhou
237
246
Nondestructive Testing Technology and Optimization of On-Service Urea Reactor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Xiaoling Luo and Lei Deng
255
Parameters Turning of the Active-Disturbance Rejection Controller Based on RBF Neural Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Baifen Liu and Ying Gao
260
Research of Intelligent Gas Detecting System for Coal Mine . . . . . . . . . . . Hui Chen
268
Research of Subway’s Train Control System Based on TCN . . . . . . . . . . . Qingfeng Ding, Fengping Yang, and Qixin Zhu
279
Shape Detection for Impeller Blades by Non-contact Coordinate Measuring Machine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Shimin Luo
286
Simulation of Road Surface Roughness Based on the Piecewise Fractal Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Zhixiong Lu, Lanying Zhao, Xiaoqin Li, and Jun Yuan
294
Stress Analysis near the Welding Interface Edges of a QFP Structure under Thermal Loading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Zhigang Huang, Xuecheng Ping, and Pingan Liu
306
Study of Intelligent Diagnosis System for Mechanism Wear Fault Based on Fuzzy-Neural Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sanmao Xie
314
X
Table of Contents – Part IV
Study on Autonomous Path Planning by Mobile Robot for Road Nondestructive Testing Based on GPS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Lunhui Xu, Fan Ye, and Yanguo Huang
321
Study on Imitating Grinding of Two-Dimensional Ultrasonic Vibration Turning System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Leping Liu, Wen Zhao, and Yuan Ma
333
Study on Optimal Path Changing Tools in CNC Turret Typing Machine Based on Genetic Algorithm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Min Liu, XiaoLing Ding, YinFa Yan, and Xin Ci
345
Study on the Problem and Countermeasure of Fruit Production Quality and Safety in Yanshan Mountain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Haisheng Gao, Bin Du, and Fengmei Zhu
355
Calculation and Analysis of Double-Axis Elliptical-Parabolic Compond Flexure Hinge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ping’ an Liu, Jianqun Cheng, and Zhigang Lai
361
Surface Distresses Detection of Pavement Based on Digital Image Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Aiguo Ouyang, Chagen Luo, and Chao Zhou
368
The Application Research of Neural Network in Embedded Intelligent Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Xiaodong Liu, Dongzhou Ning, Hubin Deng, and Jinhua Wang
376
The Theoretical Analysis of Test Result’s Errors for the Roller Type Automobile Brake Tester . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jun Li, Xiaojing Zha, and Dongsheng Wu
382
A Type of Arithmetic Labels about Circulating Ring . . . . . . . . . . . . . . . . . Ergen Liu, Dan Wu, and Kewen Cai
390
Application of CPLD in Pulse Power for EDM . . . . . . . . . . . . . . . . . . . . . . . Yang Yang and Yanqing Zhao
398
Application of IDL and ENVI Redevelopment in Hyperspectral Image Preprocessing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Long Xue
403
Design of Integrated Error Compensating System for the Portable Flexible CMMs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Qing-Song Cao, Jie Zhu, Zhi-Fan Gao, and Guo-Liang Xiong
410
Detecting and Analyzing System for the Vibration Comfort of Car Seats Based on LabVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ying Qiu
420
Table of Contents – Part IV
XI
Determination of Pesticide Residues on the Surface of Fruits Using Micro-Raman Spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Yande Liu and Tao Liu
427
Development of the Meter for Measuring Pork Quality Based on the Electrical Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Zhen Xing, Wengang Zheng, Changjun Shen, and Xin Zhang
435
Experimental Investigation of Influence on Non-destructive Testing by Form of Eddy Current Sensor Probe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Fengyun Xie and Jihui Zhou
443
Feasibility of Coordinate Measuring System Based on Wire Driven Robot . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ji-Hui Zhou, Qing-Song Cao, Fa-Xiong Sun, and Lan Bi
450
HSFDONES: A Self-Leaning Ontology-Based Fault Diagnosis Expert System Framework . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XiangBin Xu
460
Nondestructive Measurement of Sugar Content in Navel Orange Based on Vis-NIR Spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chunsheng Luo, Long Xue, Muhua Liu, Jing Li, and Xiao Wang
467
Numerical Simulation of Temperature Field in Selective Laser Sintering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jian Zhang, Deying Li, Jianyun Li, and Longzhi Zhao
474
Numerical Simulations of Compression Properties of SiC/Al Co-continuous Composites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Mingjuan Zhao, Na Li, Longzhi Zhao, and Xiaolan Zhang
480
Simulation and Optimization in Production Logistics Based on eM-Plant Platform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XinJian Zhou, XiangBin Xu, and Wei Zhu
486
Simulation of Transient Temperature Field in the Selective Laser Sintering Process of W/Ni Powder Mixture . . . . . . . . . . . . . . . . . . . . . . . . . . Jiwen Ren, Jianshu Liu, and Jinju Yin
494
Study on Plant Nutrition Indicator Using Leaf Spectral Transmittance for Nitrogen Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Juanxiu Hu, Dongxian He, and Po Yang
504
Study on the Influence of Non-electrical Parameters on Processing Quality of WEDM-HS and Improvement Measures . . . . . . . . . . . . . . . . . . . Guangyao Xiong, Meizhu Zheng, Deying Li, Longzhi Zhao, Yanlin Wang, and Minghui Li
514
XII
Table of Contents – Part IV
Test Analysis and Theoretical Calculation on Braking Distance of Automobile with ABS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Dongsheng Wu, Jun Li, Xiaoping Shu, Xiaojing Zha, and Beili Xu The Detection of Early-Maturing Pear’s Effective Acidity Based on Hyperspectral Imaging Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Pengbo Miao, Long Xue, Muhua Liu, Jing Li, Xiao Wang, and Chunsheng Luo The Effects of Internal and External Factors on the Mechanical Behavior of the Foam Copper . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Longzhi Zhao, Xiaolan Zhang, Na Li, Mingjuan Zhao, and Jian Zhang
521
528
537
Optimum Design of Runner System for Router Cover Based on Mold Flow Analysis Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tangqing Kuang and Wenjuan Gu
543
Design of Tread Flange Injection Mold Based on Pro/E . . . . . . . . . . . . . . . Huilan Zhou
555
Study on the Online Control System to Prevent Drunk Driving Based on Photoelectric Detection Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Lu Liming, Yang Yuchuan, and Lu Jinfu
563
The Design and Simulation of Electro-Hydraulic Velocity Control System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Fengtao Lin
568
Application of Background Information Database in Trend Change of Agricultural Land Area of Guangxi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Xin Yang, Shiquan Zhong, Yuhong Li, Weiping Lu, and Chaohui Wu
575
Reasons of the Incremental Information in the Updating Spatial Database . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Huaji Zhu, Huarui Wu, and Xiang Sun
583
Research on Non-point Source Pollution Based on Spatial Information Technology: A Case Study in Qingdao . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tao Shen, Jinheng Zhang, and Junqiang Wang
592
The Regulation Analysis of Low-Carbon Orientation for China Land Use . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Bikai Gong and Bing Chen
602
A CDMA-Based Soil-Quality Monitoring System for Mineland Reclamation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Dongxian He, Daoliang Li, Jie Bao, and Shaokun Lu
610
Table of Contents – Part IV
Design and Implementation of a Low-Power ZigBee Wireless Temperature Humidity Sensor Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Shuipeng Gong, Changli Zhang, Lili Ma, Junlong Fang, and Shuwen Wang Land Evaluation Supported by MDS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Fengchang Xue Design and Development of Water Quality Monitoring System Based on Wireless Sensor Network in Aquaculture . . . . . . . . . . . . . . . . . . . . . . . . . Mingfei Zhang, Daoliang Li, Lianzhi Wang, Daokun Ma, and Qisheng Ding
XIII
616
623
629
Design of an Intelligent PH Sensor for Aquaculture Industry . . . . . . . . . . . Haijiang Tai, Qisheng Ding, Daoliang Li, and Yaoguang Wei
642
A Simple Temperature Compensation Method for Turbidity Sensor . . . . . Haijiang Tai, Daoliang Li, Yaoguang Wei, Daokun Ma, and Qisheng Ding
650
A Wireless Intelligent Valve Controller for Agriculture Integrated Irrigation System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Nannan Wen, Daoliang Li, Daokun Ma, and Qisheng Ding
659
Evaluation of the Rural Informatization Level in Central China Based on Catastrophe Progression Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Lingxian Zhang, Xue Liu, Zetian Fu, and Daoliang Li
672
GIS-Based Evaluation on the Eco-Demonstration Construction in China . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Lingxian Zhang, Juncheng Ma, Daoliang Li, and Zetian Fu
680
Modeling and Analysis of Pollution-Free Agricultural Regulatory Based on Petri-Net . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Fang Wang, Qingling Duan, Lingzi Zhang, and Guo Li
691
An Online Image Segmentation Method for Foreign Fiber Detection in Lint . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Daohong Kan, Daoliang Li, Wenzhu Yang, and Xin Zhang
701
An Efficient Iterative Thresholding Algorithms for Color Images of Cotton Foreign Fibers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Xin Zhang, Daoliang Li, Wenzhu Yang, Jinxing Wang, and Shuangxi Liu Application of Grey Prediction Model in Rural Informatization Construction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jing Du, Daoliang Li, Hongwen Li, and Lifeng Shen
710
720
XIV
Table of Contents – Part IV
Study on Evaluation Method for Chinese Agricultural Informatization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Xiaoqing Yuan, Liyong Liu, and Daoliang Li
727
Research on Calculation Method for Agricultural Informatization Contribution Rate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Liyong Liu, Qilong Pan, and Daoliang Li
735
An Empirical Research on the Evaluation Index Regarding the Service Quality of Agricultural Information Websites in China . . . . . . . . . . . . . . . . Liyong Liu, Xiaoqing Yuan, and Daoliang Li
742
Hyperspectral Sensing Techniques Applied to Bio-masses Characterization: The Olive Husk Case . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Giuseppe Bonifazi and Silvia Serranti
751
Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
765
A Compression Method of Decision Table Based on Matrix Computation Laipeng Luo and Ergen Liu School of Basic Sciences, East China Jiaotong University, Nanchang, P.R. China
[email protected] Abstract. A new algorithm of attribute reduction based on boolean matrix computation is proposed in this paper. The method compresses the valid information stored in table into a binary tree, at the same time deleting the invalid information and sharing a branch about the same prefix information. Some relative concepts such as local core attributes, local attribute reduction and global core attributes, global attribute reduction are introduced. The conclusions that the global core set is the union of all local core sets and the global attribute reduction sets are the union of respective local attribute reduction sets are proved. The attribute reduction steps of the algorithm are presented. At last, The correctness and effectiveness of the new algorithm are also shown in experiment and in an example. Keywords: Rough Set, equivalence matrix, attribute reduction, information compression.
1 Introduction Rough set theory, introduced by Zdzislaw Pawlak in the early 1980s[1,2], is a new mathematical tool to deal with vagueness and uncertainty. This approach seems to be of fundamental importance to artificial intelligence and cognitive sciences, especially in the areas of machine learning, knowledge acquisition, deci-sion analysis, knowledge discovery from databases, expert systems, decision support systems, inductive reasoning, and pattern recognition[3,4,5]. Attribute reduction is one of the main applications of Rough set. The general problem of finding all reductions are NP-hard, thus it is important for attribute reduction of Rough set to design algorithms with lower price and investigate new computation method. A matrix computation method of Rough set was proposed by the author in[6,7]for information system. A matrix was seen as an internal representation of equivalence relations. By defining the operation of the equivalence matrix, matrices are applied to define dependencies between two subsets of attributes, significance of an attribute ect. The approach presents a series of algorithms and their time complexity of attribute reduction. However, there are still several problems to be solved for the method: (1)The number of objects has great influence on time complexity of these algorithms; (2)These algorithms need too many computations of matrices; (3)The method only discusses matrix computation for information system; (4) How to apply the method to variable precision Rough set model. D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part IV, IFIP AICT 347, pp. 1–7, 2011. © IFIP International Federation for Information Processing 2011
2
L. Luo and E. Liu
With the above-mentioned motivation, in another paper[8],aiming at the problem that how to apply the matrix computation to variable precision Rough set model,we have proposed a measure and computation approach based on matrix about studying Rough set theory. In this paper, we will provide a new insight into attributes reduction in decision table. A compression method of decision table based on matrix computation is proposed.We shall prove the feasibility of the compression method in theory and show its effectiveness in experiment and in an agriculture example.
2 Equivalence Matrix of Decision Table Definition 2.1. Let U be a non-empty finite set of objects and R be equivalence relation on U, We denote the partition gived by the R as follows: U/R= {X1, X 2 ,...X n} . Definition 2.2. Let R be equivalence relation on U, then R is expressed in terms of a binary n×n requivalence matrix MR=[mij]n×n where n= U , mij
⎧1 xi Rx j =⎨ . or ⎩0
Definition 2.3. Let P=(pij)n×n,Q=(qij)n×n be two binary n×n matrices. The intersection P∩Q of matrices P and Q is defined as follows:P∩Q=[tij]n×n,where tij=min{pij,qij}. m
Definition 2.4. Let R={r1,r2,…,rm},then M R = I M ri . i =1
Definition 2.5. LetC,D be equivalence relation and MC =(cij)n×n, MD =(dij)n×n be their matrices respectively. If for arbitrary positive integer i,j, cij≤dij,then M C ≤ M D .
∪
Definition 2.6. Let S=(U,R=C D,V,f) be a decision table, where C is condition attributes and D is decision attributes, Then S is consistent if and only if M C ≤ M D . Definition 2.7. Let S=(U,R=C
∈
∪ D,V,f, M ≤ M ) be a decision table, Given an C
D
attribute a C, then attribute a is nonsignificant in C if M{C−a} ≤ MD .
∪
Definition 2.8. Let S=(U,R=C D,V,f, MC ≤ MD ) be a decision table. The set of all attributes
C ′′ ⊆ C which are significant in S is called the core set of C.
Definition 2.9. Let S=(U,R=C
∪D,V,f, M ≤ M ) be a decision table. A subset T ′ of C C
D
is said to be a attribute reduction of C if and only if
T ′′ ⊂ T ′, then : MT′ > MD .
T ′ satisfies: (1) MT′ ≤ MD ; (2)if
3 Theory Analysis of Matrix Computation Put [R]ij express the element in row i and in column j of MR, where R={r1,r2,…,rm}. Obviously, [R]ij has the following properties:
A Compression Method of Decision Table Based on Matrix Computation
∧ ∧ ∧
3
(1) [R]ij =[r1]ij [r2]ij … [rm]ij(i=1,2…n,j=1,2…n);(2)if [R]ij =1,then for arbitrary rk R, [rk]ij =1;(3) if [R]ij =0,then there at least exist an attribute rk R,such that [rk]ij =0.Referring to properties,we immediately derive the following facts:
∈
Theorem 3.1. Let S=(U,R=C
∈
∪D,V,f, M ≤ M ) be a decision table. A subset T ′ of C C
D
is said to be a attribute reduction if and only if for any [D]ij=0 in MD, [ T ′ ]ij satisfies:(1) [ T ′ ]ij=0;(2)There no exist T ′′ ⊆ T ′, such that [ T ′′ ]ij=0.
∪ D,V,f,
MC ≤ MD ) be a decision table where C={c1,c2,…,cm}.c C is core attribute if and only if there at least exist positive integer i,j,( i=1,2…n,j=1,2…n) such that [D]ij=0, [c]ij=0,but for any b C-c, [b]ij=1. Theorem 3.2. Let S=(U,R=C
∈
∈
∈
∈
Proof. Let c C be core attribute.If there exist some attribute b C(c≠b),such that [b]ij=0,for any[D]ij=0 ,[C]ij=min {[c1]ij,[c2]ij,…[cm]ij}=0, then after deleting attribute b in C,we have M{C−c} ≤ MD .Thus there exist attribute set C ′ ⊆ {C-c},such that C ′ is attribute reduction of S which contradict that c is core attribute in S. Conversely, if there exist positive integer i,j,(i=1,2…n,j=1,2…n), such that [D]ij=0, [c]ij=0,and [b]ij=1 for every b C-c,then after deleting attribute b in C,we have [C]ij=0≠[C-c]ij=1.Thus c is core attribute in S by theorem 3.1.
∈
Definition 3.1. Let S=(U,R=C
∪D,V,f, M ≤ M ) be a decision table. If there exist C
D
positive integer i,j, such that attribute c satisfies [c]ij=0, [C-c]ij=1 when [D]ij=0, [C]ij=0,then attribute c is called local core attribute of decision table S.
∪
Theorem 3.3. If c1,c2…,ck be all local core attribute of S=(U,R=C D,V,f, M C ≤ M D ), then core attribute set C ′ of decision table S is
k
Uc
i =1
i
.That is, C ′ =
k
Uc . i =1
i
Definition 3.2. Core attribute set C ′ of decision table S is called global core attributes.
∪D,V,f, M ≤ M ) be a decision table and C′ be core
Definition 3.3. Let S=(U,R=C
∈
C
D
attribute set.If ak C- C ′ satisfies that there exist positive integer i,j,such that [ C ′ ]ij=1, [C]ij=0 and [ak]ij=0,then attribute set C ′ {ak} is called a local attribute reduction of decision table S. Obviously, local attribute reduction derived by [C]ij=0 can has not only one. All local attribute reduction derived by [C]ij=0 is called a local attribute reduction set.
∪
Definition 3.4. Attribute reduction set T ⊂C of decision table is called global attribute reduction. Theorem 3.4. Let B1,B2…,Bk be all local attribute reduction sets of decision table S=(U,R=C D,V,f, MC ≤MD ),where Bi ={Ai1, Ai2,L, Aiki }(1≤ i ≤ k) and Aij (1 ≤ j ≤ ik ) is
∪
a local reduction.If Tt is attribute reduction,then Tt satisfies:(1) If for arbitrary positive
4
L. Luo and E. Liu k
integer i,j, Bi∩Bj=φ,then Bi∩Bjφ,then
Tt = UAip ,1 ≤ p ≤ ik ; (2) If there exist positive integer i,j, i=1
Tt = UAip ,1 ≤ p ≤ ik where for any Aim,Ajn
∈T , A t
im,Ajn
must satisfy
Aim ∉ B j ٛor Ajn ∉ Bi . k
Proof. (1) Put
Tt = UAip and let C ′ be core attribute set.There at least exist an i =1
b ∈ Tt for any [D]ij=0,[C]ij=0,such that [b]ij=0.That is [ Tt ]ij=0. On the other hand,for arbitrary B⊂Tt , Suppose B ∪ ck = Tt and C′ ∪ ck ∈ Bm By definition3.2 there exist positive integer p,q,such that [C′]pq =1,[ck ]pq =0 While for arbitrary i,j, if attribute
.
。
Bi∩Bj=φ, then [B]pq =[Tt −ck ] ≠0 By theorem 3.1,attribute set B is not attribute reduction of system S. k
(2) If there exist positive integer i,j,Bi∩Bj≠φ, then
Tt = UAip ,1 ≤ p ≤ ik is not the i=1
optimal attribute reduction of system S.It is fact that if Bi∩Bj = A′ and
′ A′ ≠ Ajm ∈ B j ,then A′ ∪Ajm ⊆Tt is not the optimal attribute reduction. Hence, by above (1), if for any
Aim ∈ Bi , A jn ∈ Bj and Aim, Ajn ∈Tt , Aim,Ajn satify
Aim ∉ Bj ٛor A jn ∉ B i then Tt = UAip,1≤ p ≤ ik is the optimal attribute reduction.
4 Compression of Decision Table Because equivalence matrices of attribute set are symmetric, in practical application, we pay attentin to the upper-triangle above the diagonal or the lower- triangle below the diagonal. Let C={c1, c2,…, cr} be condition attribute set and D be decision attribute set. For arbitrary positive integer i,j, we obtain: [C]ij =[c1]ij [c2]ij … [cr]ij.By theorem 3.1,in practical application, we only devote our attention to [D]ij = 0 and [C]ij = 0 in the upper-triangle above the diagonal or the lower- triangle below the diagonal. In this paper,we compress information of [D]ij = 0 and [C]ij = 0 in the equivalence matrix into a binary tree where [c1]ij,[c2]ij,… [cr]ij are orderly arranged and nodes of the tree. Please refer to below for details. First, create the root of the tree, labeled with “null”. Scan the elements of MD a time and find all positive integer i,j which satisfy [D]ij = 0.The corresponding elements of each equivalence relation matrix of condition attributes orderly sorted lead to the construction of the first branch of the tree with r nodes where [c1]ij is linked as a child of the root, [c2]ij is linked to [c1]ij. The rest may be deduced by analogy.
∧ ∧ ∧
∧
A Compression Method of Decision Table Based on Matrix Computation
5
∧
The second [c1]mn,[c2]mn,… [cr]mnwould result in a branch where [c1]mn is linked as a child of the root, [c2]mn is linked to [c1]mn.the rest may be deduced by analogy. Howeve, this branch would share an existing path with other branchs if along the root node, some branch has the common prefix. For example, if [c1]ij=[c1]mn, [c2]ij=[c2]mn, [c3]ij≠[c3]mn, then the first two nodes of the branch which contains [c1]ij, [c2]ij … [cr]ij is the same as the branch which contains[c1]mn, [c2]mn … [cr]mn.The rest branches may be constructed by analogy. By theorem 3.2, definition 3.2, all local core attributes and all local attribute reducion are derived from these branches. By theorem 3.3,3.4, we get global core attributes and global attribute reduction.
∧ ∧
∧ ∧
5 Description of Attribute Reduction Algorithm Let S=(U,R=C
∪D,V,f, M
C
≤ M D ) be a decision table.
Step1: Compute the equivalence matrices of decision attribute set and each of condition attributes ,and arrage the equivalence matrices of each of condition attributes in ordor; Step2: According to [D]ij = 0,compress [c1]ij,[c2]ij,… [cr]ij into a binary tree where [ck]ij,(k=1,2,…r)is node of the tree. Step3: Scan the tree and find the only zero value node in every branch.We get local core of every branch. Core set of system S is union of all local core. Step4: Prune the branch that includes local core and at the same time, retain shareable prefix part. Step5: Travel every branch of binary tree pruned and find local attribute reduction sets of all branchs
∧
Step6: Compute attribute reduction of system S according to {Tt Tt =
UA
ip
,1 ≤ p ≤ ik }
6 Algorithm Analysis The time complexity of algorithm in [7] for finding all core attributes is 2
2
C
2
o( C U ) ,and for finding all attribute reduction is o(2 C U ) . The time complexity of algorithm in this paper for finding all core attributes is at most C +1
C +1
o(2 ) ,and for finding all attribute reduction is at most o(2×2 ) . In general, U >> C ,hence,the method presented in this paper has an advantage over the method in [7]. Next, to compare the two methods, We made the relevant experiments on monks datasets in UCI database. The datasets have one decision attribute, six conditon attributes and 423 records. We did four experiments with the first 100,150,300,423 recordes of standard datasets monks datas. The experiment environment is Petium4 2.1GMHZ,RAM512M, windows XP. The results are as follows:
6
L. Luo and E. Liu Table 1. Run time of two algorithms
5XQ WLPH
after com pression
beforer com pression
'DWD VHW
The chart shows that the method presented in this paper is efficient and scalable for finding core attributes set and attribute reduction sets, and is faser than the method in [7].
7 An Application Example in Agriculture The following is knowledge representation system of cotton diseases. The condition attributes are a-“diseased spot color”, b-“disease site”,c-“disease shape”,d-“feature” and the decision attribute is e-“the type of disease”. {1,2,3} represent the different value of each attribute. Table 2. Knowledge representation system
According to the method proposed in this paper,we can obtain that the core attribute is {a} and the attribute reduction set are {a,b} and {a,c,d}.The conclusion is the same as the other method.That is to say that diseased spot is chief factor to judge the type of disease and diseased spot color, disease site or diseased spot color, disease shape, feature may judge exactly the type of disease.
8 Conclusion In this paper, We further discuss the approach of matrix computation about Rough set and applied it to dicision table. In theory, we have proved the relation between the
A Compression Method of Decision Table Based on Matrix Computation
7
elements of equivalence matrice and core attributes,attribute reduction. At the same time, we suggest an attribute reduction algorithm based on a storage structure of binary tree which can compress the invalid and the same prefix information. The algorithm designed is lower price. We also find that by changing the order of condition attributes sorted, the algorithm is more efficient.
References [1] Pawlak, Z.: Rough Sets. International Jounal of Information and Computer Science 11(5), 341–356 (1982) [2] Pawlak, Z.: Rough Sets: Theoretical Aspects of Reasoning about data. Kluwer Academic Publishers, Dordrecht (1991) [3] Wang, G.Y., Hu, F., Huang, H., Wu, Y.: A granular computing model based on tolerance relation. The Journal of China Universities of Post s and Telecommunications 12(3), 86–90 (2005) [4] Greco, S., Wojna, A.G., Slowinski, R.: Fuzzy rough sets and multiple-premise gradualdecision rules. International Journal of Approximate Reasoning 41(2), 179–211 (2006) [5] Lin, T.Y., Yin, P.: Heuristically fast finding of the shortest reducts. In: Tsumoto, S., Słowiński, R., Komorowski, J., Grzymała-Busse, J.W. (eds.) RSCTC 2004. LNCS (LNAI), vol. 3066, pp. 465–470. Springer, Heidelberg (2004) [6] Guan, J.W., Bell, D.A.: Matrix computational method for information systems. Artificial Intelligence 105, 77–103 (1998) [7] Guan, J.W., Bell, D.A., Guan, Z.: Matrix computation for information systems. Information Sciences 131, 129–156 (2001) [8] Luo, L., Liu, E., Yi, C.: Matrix approach to the study of rough set theory. Systems Engineering and Electrionics 31(4), 859–862 (2009) (in chinese)
A Laplacian of Gaussian-Based Approach for Spot Detection in Two-Dimensional Gel Electrophoresis Images Feng He, Bangshu Xiong, Chengli Sun, and Xiaobin Xia Key Laboratory of Nondestructive Test of Ministry of Education, Nanchang Hangkong University, 330063 Nanchang, P.R. China
[email protected],
[email protected] Abstract. Two-dimension gel electrophoresis (2-DE) is a proteomic technique that allows the analysis of protein profiles expressed in a given cell, tissue or biological system at a given time. The 2-DE images depict protein as spots of various intensities and sizes. Due to the presence of noise, the inhomogeneous background, and the overlap between the spots in 2-DE image, the protein spot detection is not a straightforward process. In this paper, we present an improved protein spot detection approach, which is based on Laplacian of Gaussian algorithm, and we extract the regional maxima by morphological grayscale reconstruction algorithm, which can reduce the impact of noisy and background in spot detection. Experiments on real 2-DE images show that the proposed approach is more reliable, precise and less sensitive to noise than the traditional Laplacian of Gaussian algorithm and it offers a good performance in our gel image analysis software. Keywords: Two-dimensional gel electrophoresis, Spot detection, Laplacian of Gaussian, Morphological grayscale reconstruction.
1 Introduction Proteomic research deals with the systematic analysis of protein profiles expressed in a given cell, tissue or biological system at a given time. In this field, two-dimensional gel electrophoresis (2-DE) is a well-established and widely used technique to separate proteins extracted from sample for identification and analysis of differential expression, according to their isoelectric points and molecular weight [1]. The result of that process is many dark spots on the gel, and each spot represents a protein or a group of proteins. The two-dimensional gel electrophoresis (2-DE) images show the expression levels of several hundreds of proteins where each protein is represented as a spot of grey level values [2]. In order to extract protein spots, image processing techniques can help us to analyze proteins further. Each spot can be characterized by its location and other information, such as area, volume, intensity, etc. Due to the presence of noise, the inhomogeneous background, and the overlap between the spots in 2-DE image, the protein spot detection is not a straightforward process. D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part IV, IFIP AICT 347, pp. 8–15, 2011. © IFIP International Federation for Information Processing 2011
A Laplacian of Gaussian-Based Approach for Spot Detection in 2-DE Images
9
A variety of software packages have been developed for protein spot detection [3]. Many of these packages implement image segmentation methods based on edge detection algorithms such as Laplacian filtering, in conjunction with smoothing operators. However, if a 2-DE image contains artifacts and noise, such as cracks in the gel surface, fingerprints, dust and other pollutions, they will lead to spurious spot detection [4]. The watershed transformation algorithm has also been a popular choice for 2-DE image segmentation [3] [5]. State of the art approaches to spot detection by image segmentation include geometric algorithms [6], parametric spot models [3] and the pixel value collection method [2] [7]. In this article, we introduce an improved protein spot detection approach that is based on Laplacian of Gaussian algorithm, in conjunction with morphological grayscale reconstruction to extract the regional maxima in 2-DE images. The proposed approach extracts the regional maxima of the Gaussian-smoothed gel image by morphological grayscale reconstruction algorithm [8], and uses the second derivative (laplacian) and direction of the gaussian-smoothed gel image as well as neighborhood connectivity properties in determining spot extents. Relative to the traditional Laplacian of Gaussian algorithm, our approach can reduce the impact of noise and avoid the spurious spot detection by using morphological grayscale reconstruction algorithm to restrict regional maxima, and the results are more reliable and precise. The rest of this paper is organized as follows: Section 2 describes the spot detection approach which has been applying in our soft, Section 3 presents the experimental results, and Section 4 presents conclusions and the future work.
(a)
(b)
Fig. 1. (a) Original Image, (b) Inverted Image
2 Algorithm In the following description it is assumed that the original image shows in Fig. 1a is inverted, that is, the image background is dark, and the spots appear as light peaks rising from the background in Fig. 1b. The proposed approach to protein spot detection consists of the following steps: a) Smooth the original gel image; b) Extract the regional maxima of the smoothed image by grayscale reconstruction algorithm;
10
F. He et al.
c) Compute the Laplacian of the smoothed image and define central core regions where the Laplacian pixels are negative in both x and y directions; d) Extract all of spot regions; e) Estimate background density and normalize the valid spot quantifications. 2.1 Image Smoothing Operations During the detection process, the gel images are normally very noisy, so we must implement smoothing filter to reduce high frequency noise in 2-DE image at first. If the image is not sufficiently smoothed, it will find the edges of the spots incorrectly and generate spurious regional maxima. In the proposed approach, we use a Gaussian low-pass filter with a window (e.g. 3x3, 5x5, 7x7, etc) to smooth the inverted image. 2.2 Regional Maxima Extraction Grayscale reconstruction is a very useful operator provided by mathematical morphology. The regional maxima and minima are important features of images; they often represent the corresponding image goal: the regional maxima correspond to the light goal, and the region minima correspond to the dark goal. Grayscale reconstruction turns out to provide a very efficient method to extract regional maxima and minima from grayscale images. Furthermore, the technique extends to the determination of “maximal structures” called h-domes [8]. The h-dome transformation is illustrated on Fig. 2. The h-dome transformation extracts light structures without involving any size or shape criterion. The only parameter (h) is related to the height of these structures. This characteristic is of interest for complex segmentation problems. The h-dome image Dh(I) of the h-domes of a grayscale image I is given by Dh(I)=I-RIδ(I-H) . where
RIδ(I-H)
(1)
is the dilated reconstruction of I from I-h.
Fig. 2. Determination of the h-domes of grayscale image I
In the proposed approach, we can execute h-dome transformation to extract light maximal structures of the smoothed image. It can restrain all maxima, the depth of which are less than or equal to the parameter (h), and reduce the impact of noisy in spot detection.
A Laplacian of Gaussian-Based Approach for Spot Detection in 2-DE Images
11
2.3 Image Laplacian Operations 2-DE images show the expression levels of several hundreds of proteins where each protein is represented as a spot of grey level values. Under optimal density within the area of the spot appears as a monotonically increasing function as illustrated in Fig. 3a. The Laplacian or second derivative of this function is shown in Fig. 3b. We define the central core region of negative values for digital approximations to both partial second derivatives, with respect to x and y directions, as the central core region. The region on the outside of the central core is propagated until it reaches the extent of the positive peaks of the side lobes. This propagated central core region, computed in two dimensions, is then effectively used as a mask for quantitating that spot [5]. This is the reason we want to use the Laplacian for helping to analyze the image.
(a)
(b)
Fig. 3. (a) Cross section of the ideal spot, (b) Laplacian of this ideal spot cross section
After acquired the smoothed image, it is used when computing the digital approximation of the Laplacian of this image. We store the Laplacian direction and magnitude values in two additional images. All laplacian direction image pixels are set to 1, if the Laplacian values are negative in both x and y, otherwise they are set to 0. This directional image defines the initial central core regions of a gel. The central core regions are propagated to adjacent pixels until they reach the maximum value of the Laplacian magnitude image. These final regions are called the propagated central core region and, after some corrections, define the extent of the spot to be quantitated. The next step describes the extraction of each spot region in the proposed approach. 2.4 Spot Region Extraction After the image laplacian operations, the next step is to extract all of the spot regions in sequential raster search of the central core image. The central core regions are propagated to adjacent pixels until they reach the maximum value of the Laplacian magnitude image. These final regions are called the propagated central core region and, after some minor corrections, define the extent of the spot to be quantitated. The steps in finding the final propagated central core for the current spot are enumerated as follows. This algorithm is iterated for each spot as it traverses down the image in a raster pattern. Step1. Find the central core for the current spot in the Laplacian direction image, where the central core regions are set to 1 and the other set to 0. Given a new spot pixel to find all (x, y) pairs that are 4-neighbor connected to this spot with central core pixel, and save this pairs in a list.
12
F. He et al.
Step2. If the central core size is less than the threshold which is the minimum of central core size, then delete the spot, and set that region to 0 in the central core image. Step3. There are situations where some spots are merged into single spots. If the central core size is greater than the threshold which is the minimum to split, then try to split the spot into several spots. Step4. Try to find saturated spots, if the (max Density of the spot)/ (max Density in the image) is greater than percent threshold of the darkest pixel, then try to fill holes in the saturated spot with central core pixels. Step5. Remove concavities which with 0 values between central core pixels with central core value in the current central core spot. Step6. Remove interior central core pixels to speed up the subsequent steps and obtain the edge of the central core. This is useful for segmenting very large spots when using very high pixel resolution. Step7. Propagate the central core to a propagated central core by looking for the maxima away from the center of the spot in all directions. This propagation is terminated by various conditions including running into another spot, noise, etc. If the image is not adequately smoothed, this step will not work very well. Step8. Optimize the propagated central core region. We can fill holes in the propagated central core such that central pixels with 0 values are filled with propagated central core value. Step9. Delete the spurious propagated central core. If the propagated central core region does not contain a regional maximum, so it is a spurious propagated central core, and we could delete it. Step10. Finally, compute the spot features using data from the original image and save the features for this spot in a list of all spots. The spot features include density weighted centroid, standard deviation and covariance spot size, density, area and volume. 2.5 Background Density Estimation and Spot Quantification After all spots are initially segmented, it then performs background correction and normalization on the quantifications. If the background appears relatively uniform, we have hound subtracting global minimum intensity for the gel works sufficiently well. However, the background appears to be spatially varying, we use a smoothing low pass filter to estimate the background [9]. First a rest of image is computed as the original image less the segmented spots with the spots having density value 0. Then the filter is computed over the entire image by moving an averaging window (e.g. 32x32) over the image in a raster, 1 pixel at a time where the mean density is computing in the averaging window at each point not including the 0 values. This background image is used to estimate the background for each spot and correct the spot density. To normalize, we divide each spot intensity on a given gel by the mean spot intensity for that gel, and save the spot features in a list, including location, area, volume, normalized intensity, etc. The researchers can use these spot features to do next analysis, such as spot matching.
A Laplacian of Gaussian-Based Approach for Spot Detection in 2-DE Images
13
3 Experimental Results In this section, the results for detection the protein spots in a gel image by those methods mentioned in Section 2 are presented. In our approach, we firstly inverted the original image, that is, the image background is dark, and the spots appear as light peaks rising from the background. We smoothed the inverted image by Gaussian low-pass filter, if the image is not sufficiently smoothed, it will find the edges of the spots incorrectly and generate spurious regional maxima. Fig. 4a shows the regional maxima labeled by red color. Fig. 4b and Fig. 4c show the Laplacian magnitude image and the directional image which defines the initial central core regions of a gel. The central core regions are propagated to adjacent pixels until they reach the maximum value of the Laplacian magnitude image. These final regions are called the propagated central core region and, after some minor corrections, we can get the optimized propagated central core region in Fig. 4d, is then effectively used as a mask for quantitating that spot, and the background image in Fig. 4e. Finally, we can get the final result as Fig. 4f, described the spot contour by red color.
(a)
(b)
(c)
(d)
(e)
(f)
Fig. 4. The process of the protein spots detection with the improved Laplacian of Gaussian approach: (a) regional maxima on original image, (b) central core regions, (c) laplacian magnitude image, (d) final propagated central core regions that are optimized, (e) background image, (f) final result.
14
F. He et al.
As shown in Fig. 4f, we detected 374 protein spots, almost protein spots can be detected and their contours are very clear based on the proposed approach. Fig. 5 shows the result of the Laplacian of Gaussian algorithm, which detected 506 protein spots without morphological grayscale reconstruction algorithm to restrict regional maxima, however, affected by the stripe noise and artifacts, some of them are the spurious spots. On the other hand, the proposed approach is less sensitive to noisy.
Fig. 5. The final result of traditional Laplacian of Gaussian algorithm
4 Conclusions We present an improved protein spot detection approach which can effective detects and quantifies the protein spots in 2-DE images. The spot detection approach applies morphological grayscale reconstruction algorithm to restrict regional maxima, and the Laplacian of Gaussian algorithm to detect spot regions. The 2-DE image often contains artifacts and noise, and they will lead to spurious spot detection. The morphological grayscale reconstruction algorithm can restrict those spurious regional maxima very well, and reduce the impact of noisy in spot detection. The experimental results show that the proposed approach is more reliable, precise and less sensitive to noise than the traditional Laplacian of Gaussian algorithm and it offers a good performance in our gel image analysis software. Future works include further experimentation, optimization and parallelization of the proposed approach, and its integration in a complete user-friendly software application. Also variation of the proposed approach will be used to detect all spots in heavily polluted 2-DE images. Acknowledgments. This work was supported by Postgraduate Innovation Fund of Jiangxi Province (YC09A112), Postgraduate Innovation Fund of Nanchang Hangkong University (YC2009008), Scientific Research Fund of Jiangxi Provincial Education Department (GJJ09183) and Jiangxi Nature Science Fund (No.2008GZS0032).
References 1. Safavi, H., Correa, N.: Independent Component Analysis of 2-D Electrophoresis Gels. Electrophoresis 29, 4017–4026 (2008) 2. Peer, P., Corzo, L.G.: Local Pixel Value Collection Algorithm for Spot Segmentation in Two-Dimensional Gel Electrophoresis Research. Comparative and Functional Genomics, 1–9 (2007)
A Laplacian of Gaussian-Based Approach for Spot Detection in 2-DE Images
15
3. Berth, M., Moser, F.M.: The State of the Art in the Analysis of Two-Dimensional Gel Electrophoresis Images. Appl. Microbiol. Biotechnol. 76, 1223–1243 (2007) 4. Rye, M.B., Alsberg, B.k.: A Multivariate Spot Filtering Model for Two-Dimensional Gel Electrophoresis. Electrophoresis 29, 1369–1381 (2008) 5. Srinark, T., Kambhamettu, C.: An Image Analysis Suite for Spot Detection and Spot Matching in Two-Dimensional Electrophoresis Gels. Electrophoresis 29, 706–715 (2008) 6. Morris, J.S., Clark, B.N., Gutstein, H.B.: Pinnacle: a Fast, Automatic and Accurate Method for Detecting and Quantifying Protein Spots in 2-Dimensional Gel Electrophoresis Data. Bioinformatics 24, 529–536 (2008) 7. Rye, M.B., Færgestad, E.M., Martens, H.: An Improved Pixel-Based Approach for Analyzing Images in Two-Dimensional Gel Electrophoresis. Electrophoresis 29, 1382–1393 (2008) 8. Soille, P.: Morphological Image Analysis: Principles and Applications, 2nd edn. Springer, Heidelberg (2003) 9. Van Belle, W., Sjøholt, G., Anensen, N.: Adaptive Contrast Enhancement of TwoDimensional Electrophoretic Protein Gel Images Facilitates Visualization, Orientation and Alignment. Electrophoresis 27, 4086–4095 (2006)
A Leaf Layer Spectral Model for Estimating Protein Content of Wheat Grains Chun-Hua Xiao1,2, Shao-Kun Li1,2,3,*, Ke-Ru Wang1,2, Yan-Li Lu2, Jun-Hua Bai2, Rui-Zhi Xie2, Shi-Ju Gao2, Qiong Wang2, and Fang-Yong Wang1 1
Key Laboratory of Oasis Ecology Agriculture of Xinjiang Construction Crop/ Center of Crop High-Yield Research, Shihezi 832003, Xinjiang, China 2 Institute of Crop Science, Chinese Academy of Agricultural Sciences / National Key Facility for Crop Gene Resources and Genetic Improvement, NFCRI, Beijing 100081, China 3 Department of Crop Culture, Institute of Crop Science, Chinese Academy of Agricultural Sciences, Beijing 100081, P.R. China Tel.: +86-010-68918891
[email protected] Abstract. The spectral signatures of crop canopies in the field provide much information relating morphological or quality characteristics of crops to their optical properties. This experiment was conducted using two winter-wheat (Triticum aestivum) cultivars, Jingdong8 (with erect leaves) and Zhongyou9507 (with horizontal leaves). We analysiced the relation between the direction spectral characteristics and the laeves nitrogen content(LNC). The result showed that the spectral information observed at the 0° angle mainly provided information on the upper canopy and the lower layer had little impact on their spectra. However, the spectral information observed at 30° and 60° angles reflected the whole canopy information and the status of the lower layer of the canopy had great effects on their spectra. Variance analysis indicated that the ear layer of canopy and the topmost leaf blade made greater contributions to CDS. The predicted grain protein content (GPC) model by leaf layers spectra using 0° view angle was the best with root mean squares (RMSE) of 0.7500 for Jingdong8 and 0.6461 for Zhongyou9507. The coefficients of determination, R2 between measured and estimated grain protein contents were 0.7467 and 0.7599. Thus, grain protein may be reliably predicted from the leaf layer spectral model. Keywords: wheat canopy, leaf distribution, direction spectra, view angle, model.
1 Introduction The spectral characteristics of a crop canopy are determined not only by biophysical and biochemical features and also plant structural attributes. Leaf optical properties are a main factor for canopy spectral. *
Corresponding author.
D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part IV, IFIP AICT 347, pp. 16–29, 2011. © IFIP International Federation for Information Processing 2011
A Leaf Layer Spectral Model for Estimating Protein Content of Wheat Grains
17
Crop canopy spectral characteristics represent important information needed to guide crop management. Crop composition and structure are difficult to assess by traditional spectral measurement with vertical canopy direction. The direction spectrum includes a great deal of crop canopy information. Multiangle data significantly improved the accuracy of recovering forest parameters when inverting 3-D optical models (Kimes et al., 2002) but forest vertical structure can not be captured accurately using only the 4 spectral bands in the nadir view or all view angles with a single spectral band (Kimes et al.,2006). Directional radiances in or near the principal plane of the sun provides information that leads to more accurate prediction of canopy structure parameters than from other azimuth planes (Gobron et al., 2000). Canopy emissivity increased with increasing view angle due to the greater proportion of vegetation observed at off-nadir view angles; when the proportion of leaves was lower than that of soil, canopy emissivity grew with increasing view angle (Sobrino et al., 2005). It should be noted that these models assume a Lambertian behaviour for soil and vegetation surfaces. Although vegetation surfaces show a near Lambertian behaviour, bare soil surfaces do not and the angular variation on emissivity can not be neglected. Lidars, multiangle radiometers, radars and imaging spectrometers have been identified as systems that can capture information in the vertical dimension. This requires a capability to remotely measure the vertical and spatial distribution of forest structural parameters that are needed for more accurate models of energy, carbon, and water flux over regional, continental, and global scales. Thus, we examined the utility of hyperspectral data for the quantitative characterization of vertical wheat structure. Most remote sensing systems provided an image of the horizontal scope, but could not provide the vertical information on biochemical distribution in a crop canopy, thereby reducing the accuracy of measurement. Multi-angle data can increase the precision of forest parameters (Kimes,et al., 2006). The distribution of tissues in a crop canopy has certain characteristics - biochemical distribution is different because of transfer of matter during the growth stage. Leaves in a wheat canopy are composed of under, middle, upper layer and ear layers (Wang et al.,2004). The spectral characteristic differed among canopy leaves because of the different reflection and scatter, so their effect on canopy spectra was different (Wang et al., 2004). Various biochemical (foliar lignin and nitrogen) and biophysical factors influencing canopy reflectance signatures have been studied in previous works. Information on biochemical parameters is important and the multi-angle spectra which provide information on different directions can facilitate a more exact prediction of biochemical parameters. To date, there are no studies on the relative importance of vertical distribution of wheat leaves that determines canopy reflectance across the shortwave (350–2500nm) spectrum. The contribution of each leaf layer relative to all other factors has also not been adequately determined. Yet, it is the interaction of these factors, including their potential covariance or unique behavior, that must be understood if advances in remote sensing are to be achieved.
18
C.-H. Xiao et al.
In this study, the leaf slice method was used to characterize a wheat canopy, evaluate the spectral response of leaf vertical distribution in the canopy, determine the characteristics of spectral curve for different leaf layers and develop a methodology for predicting the biochemistry of the canopy.
2 Materials and Methods 2.1 Preliminary Experiment A field experiment was conducted in the Experimental Station of the Institute of Crop Science of the Chinese Academy of Agricultural Sciences, Beijing (39º57'55" N, 116º19'46" E) in the 2005-2006 growing season. The soil was a silt clay loam containing 1.16% organic matter, 42.6 mg kg-1alkali-hydrolysable N, 26.5 mg kg-1 available phosphorus and 139.4 mg kg-1 available potassium. Two winter wheat cultivars were used: Jingdong8, an erect leaf plant-type and Zhongyou9507, a lax leaf plant-type. Four N fertilizer (urea, 46%N ) treatments were set up with four randomized replications of each cultivar: N0, no N fertilization, N1, 150 kg hm-2 pure N fertilization, N2, 300 kg hm-2 (rationally fertilized), N3, 450 kg hm-2 (excessively fertilized). The N rates were applied in three splits at pre-sowing (50% of the total amount), reviving stage (25% of the total amount) and jointing stage (25% of the total amount). All the treatments were fertilized with the same amounts of P (P2O5, 144 kg hm-2) and K (K2O, 75 kg hm-2) at pre-sowing. The leaf slice method of wheat canopy: whole (plant) wheat samples 0.5m long and 0.8m wide were chosen; based on the vertical distribution of wheat canopy, the samples were measured off the whole wheat canopy (WWC), ear layer of canopy (ELC), inverse first leaves layer of canopy (ILLC-1), inverse second leaves layer of canopy(ILLC-2), inverse third leaves layers of canopy (ILLC-3) and inverse fourth leaves layer of canopy (ILLC-4). Canopy layers were severed with a scissors from the ear layer to lower layer (Fig. 3).
,
2.2 Measured Traits and Methods All canopy spectral measurements were taken from a height of 1.3 m above ground (the height of the wheat was 90 cm at maturity), under clear sky conditions between 10:00 and 14:00 (Beijing local time), using an ASD FieldSpec Pro spectrometer (Analytical Spectral Devices, Boulder, CO,USA) fitted with a 258 field of view fiber optics, operating in the 350–2500 nm spectral region with a sampling interval of 1.4 nm between 350 and 1050 nm and 2 nm between 1050 and 2500 nm and with spectral resolutions of 3 nm at 700 nm and10 nm at 1400 nm. A 40 cm ×40 cm BaSO4 calibration panel was used for calculation of reflectance. The spectra were measured with view angles of 0, 30,60,90,120, 150, and 180° to the line vertical to the wheat row using the Simple multi-angle spectral measurement equipment (Fig.1) after every layer was removed (N2 treatment). The model spectra were measured with a view angle of 0°
A Leaf Layer Spectral Model for Estimating Protein Content of Wheat Grains
19
using other treatments at ear layer (EL), upper leaves layer (ULL) and lower leaves layer(LLL)(Fig.2). Leaf samples from EL,ULL,LLL were taken almost synchronously with the spectral measurements. Measurements were conducted at jointing, heading, anthesis, milking and waxing stages. These samples were oven-dried at 70 and nitrogen content was determined by the Kjeldahl technique (Bremneretal.,1981) using a B-339 DistillationUnit (BUCHIAnalyticalLtd, Flawil, Switzerland). Wheat grain protein content estimated from the formula : Pro% =6.25 × Nitr (% ).
℃
sun
Ear layer
Upper layer
1/2H H
Lower layer
Fig. 1. Simple multi-angle spectral measurement equipment
1/2H
Fig. 2. Sketch of measured method by layer
2.3 Data Analysis The hyperspectral data were analyzed using the Matlab6.5 software and quantitative data were analyzed using an analysis of variance (ANOVA) procedure.
3 Results 3.1 The Spectral Curves Following Removal of Different Leaf Layers The lower leaves of the canopy changed the spectral reflectivity (Fig.3) with different view angles at 350-700nm, 800-1300nm and 1400-1800nm. This is important for pigment content within the visible wave band (350-700nm). The reflectivity of near infrared (800-1300nm) is influenced by canopy characteristics. The wavebands (1400-1800nm) provide information on the water content. In this paper, we analyzed the characteristics of visible (350-700nm) and near infrared (800-1300nm) wavebands.
20
C.-H. Xiao et al.
Jingdong8
5HIOHFWLRQ
5 HI O HF WL R Q
Zhongyou9507
e e
:DYHOHQWKQP e e e e
:DYHOHQJWKQP e e
e
e e
e e
e
Fig. 3. Spectra of removing the forth leaf layer at different angles for Jingdong8 and Zhongyou9507 wheat
5HIOHFWLRQ
5HIOHFWLRQ
Jingdong8
Zhongyou9507
:DYHOHQJWKQP ,//& ,//&
(/& ,//&
,//& ::&
,//& ::&
:DYHOHQJWKQP ,//& ,//& (/&
,//&
Fig. 4. Spectra of removing different leaf layers at 60°angles for Jingdong8 and Zhongyou9507
Comparing the direction spectral characteristic of the two wheat canopies, the spectral response of Jingdong8 with different view angles was more significant than that of Zhongyou9507 at visible and near infrared wavebands; the max reflectivity of Jingdong8 was 31% higher than that of Zhongyou9507 at 350-700nm and 22.4% higher at 800-1300nm. The spectral curve of doing away with different leaves at the 60° view angle showed that the leaf influenced the curve of visible and near infrared wavebands (Fig.4). The change for Jingdong8 was more significant than for Zhongyou9507; the spectral reflectivity of Jingdong8 was 11.2% higher than that of Zhongyou9507 at 350-700nm and 26.8% higher at 800-1300nm. To further analyze the relation between canopy spectra and leaves, we selected six spectral reflectivities at 450, 550, 670, 980, 1090, 1200nm. 3.2 Analysis of Spectral Reflectivity of Leaf Layers at Different View Angles As shown in Fig.5, for Jingdong8, the spectral reflectivity of the whole canopy was similar with the leaf layer removal treatment, ILLC-4; doing away with ELC and ILLC-1 reduced the reflectivity. The changes in reflectivity at 550nm were more
A Leaf Layer Spectral Model for Estimating Protein Content of Wheat Grains
WWC
21
ELC
ILLC-1
ILLC-2
Fig. 5. Spectral characteristics following removal of different leaf layers in Jingdong8
obvious than at 450 and 670nm. In visible wave bands, changes were evident for 980, 1090, 1200nm in near infrared wave bands. At different view angles, the reflectivity of the 90° view angle was the lowest and with increasing distance from the vertical measurement, the reflectivity increased, especially in visible wave bands than in near infrared. The reflectivity changes due to ILLC-1 and ELC were more obvious than those of other leaf layers. The reflectivity changes due to LLC-3 at 1090nm were 8.8% at 0° view angle and 48.6% for ILLC-1. At 90° view angle, the changes were 5.6 and 40.7%; at 30°, they were 12.9 and 27.7% and at 60° they were 12.2 and 34%, respectively. The change was less pronounced at the 180, 150 and 120° view angles. Compared to the 90° view angle, the reflectivity at 30 and 60° had more information on the lower leaves, which were
22
C.-H. Xiao et al.
important for reflectivity of the canopy. The reflectivity at 0° contained very important information on upper leaves. The canopy spectral reflectivity of Zhongyou9507 was lower than that of Jingdong8 (Fig.6) at1090nm; the reflectivity due to ILLC-3 at 0° view angle changed 4.4%, while that due to ILLC-1 changed14.7%; at 90° view angle, the changes were 5.9 and 46.8%; at 30°, they were 16.9 and 44.4% and at 60° they were 16.8 and 41.2%. Compared with the traditional 90° view angle, the lower leaves were important for canopy spectra at 30 and 60° view angles; the upper leaves were important at 0° view angle although the upper leaves of Zhongyou9507 had less influence on canopy spectra than those of Jingdong8.
WWC
ILLC-1
ILLC-3
ELC
ILLC-2
ILLC-4
Fig. 6. Spectral characteristics following removal of different leaf layers in Zhongyou9507
A Leaf Layer Spectral Model for Estimating Protein Content of Wheat Grains
23
3.3 Analysis of Variance (ANOVA) of Spectral Reflectivity for Different Leaf Layers The ANOVA of mean reflectivities across six wavelengths showed that the reflectivity of Jingdong8 was lower than that of Zhongyou9507 (Table 1). For Jingdong8, the response of canopy spectra due to ILLC-1 and ELC were significant at the 0° view angle; at 30° as for 60°, the effects of ELC, ILLC-1, ILLC-2 and ILLC-3 were significant; at 90°, the ELC and ILLC-1 were important and at 120, 150 and 180° view angles, the ELC and ILLC-1 were significant. The ELC and ILLC-1 were important for canopy spectra and the influence of lower leaves were less. For the view angles of 0°, 90°, 120°, 150°, 180°), the under leaves were less influential; at 30 and 60°, the response of lower leaves were significant. The canopy spectral responses of different leaves for Zhongyou9507 differed from that of Jingdong8; at 0 and 90° view angles, the ELC and ILLC-1 effects were significant at 30 and 60°, the ELC, ILLC-1, ILLC-2 effects were significant. At the same view angle, the ELC and ILLC-1 were more important than the lower leaves. The spectral response of lower leaves was related to the view angle. With vertical and horizontal measurements, the influence of lower leaves were less than that of other view angles. Table 1. Canopy spectral characteristic of different leaf layers of Jingdong8 and Zhongyou9507
Average value Angles°
Treatments
0
WWC ILLC-3 ILLC-4 ILLC-2 ILLC-1 ELC WWC ILLC-4 ILLC-3 ILLC-2 ILLC-1 ELC WWC ILLC-4 ILLC-3 ILLC-2 ILLC-1 ELC
30
60
0.2254 0.2110 0.2046 0.1974 0.1690 0.1508 0.2265 0.2196 0.1951 0.1886 0.1591 0.1487 0.2064 0.2021 0.1816 0.1645 0.1412 0.1225
Jingdong8 a A a A a AB ab AB bc BC c C a A ab A bc AB c AB d BC d C a A a A ab AB bc ABC bc BC c C
Zhongyou9507 0.1657 a 0.1643 a 0.1662 a 0.1573 a 0.1354 b 0.1308 b 0.1713 a 0.1646 a 0.1619 ab 0.1398 bc 0.1276 cd 0.1143 d 0.1784 a 0.1553 ab 0.1503 ab 0.1323 bc 0.1052 c 0.0974 c
A A A A B B A AB AB BC C C A A AB ABC BC C
24
C.-H. Xiao et al.
Table 1. (continued)
Average value Angles°
Treatments
90
WWC ILLC-4 ILLC-3 ILLC-2 ELC ILLC-1 WWC ILLC-4 ILLC-3 ILLC-2 ILLC-1 ELC ILLC-4 ILLC-3 ILLC-2 WWC ILLC-1 ELC WWC ILLC-4 ILLC-3 ILLC-2 ILLC-1 ELC
120
150
180
0.1588 0.1726 0.1451 0.1292 0.1081 0.1040 0.1645 0.1642 0.1511 0.1341 0.1139 0.1051 0.1734 0.1637 0.1547 0.1454 0.1325 0.1166 0.2254 0.1879 0.1883 0.1725 0.1388 0.1360
Jingdong8 a A a AB ab AB ab AB b B b B a A a A ab AB abc AB bc AB c B a A ab AB abc AB abc ABC cd BC d C a A ab AB ab ABC bc BC c BC c C
Zhongyou9507 0.1766 a 0.1657 a 0.1532 abc 0.1276 abcd 0.1137 bcd 0.1084 d 0.1765 a 0.1692 ab 0.1616 abc 0.1432 bcd 0.1255 d 0.1325 cd 0.1679 ab 0.1692 a 0.1568 bc 0.1672 ab 0.1515 c 0.1555 c 0.1678 a 0.1599 a 0.1555 a 0.1464 a 0.1461 b 0.1537 b
A AB AB AB B B A AB AB AB AB B A A A AB AB B A A A A B B
Note: Means followed by different lower case letters differ significantly at P< 0.05; those followed by different upper case letters differ significantly at P <ParsedDataElement>amtype0 <ParsedDataElement>nodeid1 <ParsedDataElement>group11 8 <ParsedDataElement>board_id130 <ParsedDataElement>packet_id1 <ParsedDataElement>voltage 3.25 <ParsedDataElement>temp25. 4 <ParsedDataElement>light74 9
A Semantic Middleware of Grain Storage Internet
75
After mapping, the data is stored into the ontology models as following format, which is the output after semantic annotation. 3.2525.4 25.4 749 From the output we can see the all sensor readings have been endowed formal semantics which provides the basis of semantic interoperability. To verify the integration effectiveness of different grain storages using the semantic middleware prototype, two senor networks are deployed as Fig. 2. Each sensor network has 100 sensor nodes distributed in 2 grain storage bins. Each network has a gateway running the proposed middleware.
Fig. 2. The sensor network deployment structure
Every 30 seconds, the server in Net 1 sends a query to retrieve the readings of a random one of the node 101- 200 which is in the Net 2, vice versa. The observed query results show the data can be access successfully cross the bounds of the subnets, thus the integration is seamless.
76
S. Hu et al.
5 Conclusion and Future Work The semantic integration of the global food storage system is an exciting vision that maximizes the utilization of the potential of data resources on existed food monitoring information systems. The semantic interoperability is a necessary prerequisite for automatic search, retrieval and processing of grain storage sensor data. This paper is a step further towards to unify the domain objects, sensor readings, time/space semantics, lifecycles of grains. The benefits of our work are to automatic the syntax and semantic annotation of sensor data and provide the semantic unification – a solid base for integration of heterogonous grain storage systems to form a global Internet of things on grain storage. As for future work, we are planning to perform more comprehensive performance analysis by integrating more sensor network of grain storage in China and considering more real life scenarios and extending the system to the storage information system of other kinds of food. This effort will be a further step in the direction towards enabling semantic web to access and process the global food storage system. Our paper demonstrates how semantic middleware can be applied to grain storage internet of things. There are at least two advantages of our approach towards building such a future internet. First, RDF as the common information representation enables the semi-structured data sources conveniently handled and additional sources added incrementally without need to modify a global schema. Second, a richer domain model can be provided by formal ontology definition with supporting inheritance hierarchies for classes and properties.
Acknowledgement Project supported by the National High Technology Research and Development Program of China (2008AA01Z208 and 2009AA01Z405), the National Natural Science Foundation of China (60772150), and the Youth Foundation of Sichuan Province (2009-28-419) and the Applied Basic Research Program of Sichuan Province ( 2010JY0013).
References 1. Armstrong, P.: Wireless Data Transmission of Networked Sensors in Grain Storage. In: ASAE Annual International Meeting, Las Vegas, USA, p. 036157 (2003) 2. Rehman, A., Shaikh, Z.A.: Towards Design of Context-Aware Sensor Grid Framework for Agriculture. World Academy of Science, Engineering and Technology 38, 244–247 (2008) 3. Katasonov, A., Kaykova, O., Khriyenko, O., Nikitin, S., Terziyan, V.: Smart Semantic Middleware For The Internet Of Things. In: 5th International Conference on Informatics in Control Automation and Robotics, pp. 169–178. INSTICC, Madeira, Portugal (2008) 4. Yan, L., Zhang, Y., Yang, L.T.: The Internet of Things: From RFID to the NextGeneration Pervasive Networked Systems. Auerbach Publications, FL (2008) 5. Brock, D., Schuster, E.: On the Semantic Web of Things. In: Semantic Days 2006, Stavanger, Norway (2006)
A Semantic Middleware of Grain Storage Internet
77
6. Zhang, W., Kantor, G., Singh, S.: Integrated Wireless Sensor/Actuator Networks in an Agricultural Application. In: SenSys 2004, p. 317. ACM Press, New York (2004) 7. Kim, Y., Evans, R.G., Iversen, W.: Remote Sensing and Control of Irrigation System using a Distributed Wireless Sensor Network. IEEE Trans. Instrumentation and Measurement 57(7), 1379–1387 (2008) 8. Wark, T., Corke, P., Sikka, P., Klingbeil, L., Guo, Y., Crossman, C., Valencia, P., Swain, D.: Transforming Agriculture through Pervasive Wireless Sensor Networks. IEEE Pervasive Computing 6(2), 50–57 (2007) 9. Sheth, A., Henson, C., Sahoo, S.: Semantic Sensor Web. IEEE Internet Computing 12(4), 78–83 (2008) 10. Thirunarayan, K., Pschorr, J.K.: Semantic Information and Sensor Networks. In: 24th Annual ACM Symposium on Applied Computing, pp. 1273–1274. ACM Press, New York (2009) 11. Brunner, J.S., Goudou, J.F., Gatellier, P., Beck, J., Laporte, C.E.: SEMbySEM: a Framework for Sensors Management. In: 1st International Workshop on the Semantic Sensor Web, Herkalion, Greece, pp. 19–34 (2009) 12. Butler, M., Reynolds, D., Dickinson, I., McBride, B., Grosvenor, D., Seaborne, A.: Semantic Middleware for E-Discovery. In: IEEE International Conference on Semantic Computing, pp. 275–280. IEEE Press, New York (2009) 13. Sanchezl, A., Ojo, A., Janowski, T., Estevez, E.: Semantic Interoperability Middleware – Cases and Applications in Electronic Government. In: 3rd International Conference on Digital Information Management, pp. 800–805. IEEE Press, New York (2008) 14. Corradi, A., Montanari, R., Toninelli, A.: Adaptive Semantic Middleware for Mobile Environments. Journal of networks 2(1), 36–47 (2007) 15. Jena, http://jena.sourceforge.net 16. PostgreSQL, http://www.postgresql.org 17. Pellet, http://clarkparsia.com/pellet/ 18. Russomanno, D.J., Kothari, C., Thomas, O.: Building a Sensor Ontology: A Practical Approach Leveraging ISO and OGC Models. In: The 2005 International Conference on Artificial Intelligence, Las Vegas, USA, pp. 637–643 (2005)
AE Feature Analysis on Welding Crack Defects of HG70 Steel Used by Truck Crane Yantao Dou1, Xiaoli Xu2, Wei Wang3,*, and Siqin Pang1 2
1 Beijing Institute of Technology, Beijing, P.R. China Beijing Information Science and Technology University, Beijing, P.R. China 3 College of Engineering, China Agricultural University, Beijing, P.R. China
[email protected] Abstract. In present paper, using Acoustic Emission (AE) as a ultrasonic technique, Welding crack has been investigated on the HG70 steel used by truck crane during three-point bending test was carried out. The study shows that the activity of AE increases greatly during the welding crack initiates and propagates, and the distributing range of AE characteristic parameters can be fixed. AE signals of welding crack initiation and expansion are typical sudden signals which have a different spectrum energy distribution range with that of unstable crack propagation. Keywords: HG70 steel, Welding crack, Acoustic Emission.
1 Introduction As an important loading and transporting machine, truck crane is widely used in various fields of national economy. To achieve high efficiency and large scale production, truck cranes are often overloaded, which result in accidents frequently. The main cause lies in the damage of structural parts, for example, boom fracture brought about by fatigue crack or weld defects. The current popular detecting technologies include macroscopic artificial inspection, ultrasonic testing, magnetic particle testing, radiographic testing, penetrant testing, stress-strain testing and so on, which often fail to have a complete and real-time detection of crane metal structures since they are often used as periodic testing or testing afterwards[4]. Acoustic Emission (AE) technology can just make up the deficiency of conventional detection methods. During the past 20 years, some research works have been carried out to the aerial personnel devices or crane structure[1-3]. The aim of this paper is to investigate the AE signals during three-point bending test of welding specimens made of HG70 steel. The typical characteristic parameters and waveform features of the AE signals are collected and extracted at the three development stages of welding crack, namely as crack initiation, growth and fracture[5], and the crack position was located by way of linear location. *
Corresponding author.
D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part IV, IFIP AICT 347, pp. 78–88, 2011. © IFIP International Federation for Information Processing 2011
AE Feature Analysis on Welding Crack Defects of HG70 Steel Used by Truck Crane
79
2 Preparation 2.1 Specimen Preparation HG70 steel, commonly used in crane booms and legs, is chosen as the experimental material. Table 1 and Table 2 indicate its chemical composition and mechanics properties[6]. Table 1. Chemical composition of HG70
C 0.06
Si 0.29
Chemical Composition (weight fraction %) Mn P S Mo Cr 1.56 0.014 0.008 0.33 0.61
Nb 0.04
V 0.04
Table 2. Mechanics properties of HG70
σ s / MPa
σ b / MPa
σ 5 (%)
AKV / J
570
725
26
69
The structure of welding specimens is shown in Fig 1. The welding interface is located in the middle of the specimens. The specimens are loaded through three-point bending mode.
Welding area
Fig. 1. Structure of welding specimens
2.2 Test Equipment and the Placement of Sensors Computer controlled electro-hydraulic servo universal test machine of WDW4200 Series is adopted for the loading test, with its max. test force being 200kN, and the indicating accuracy: ±0.5%. Besides, AMSY-5 AE-System by Vallen and VSl50-M sensors are chosen for the test. The sensor’s main monitoring frequency ranges from 100 to 450kHz, and the acquired AE signals are amplified by a 34 dB fixed gain preamplifier (AEP4). The background noise under operation is found to be always below 35dB, which indicates that we can choose the voltage threshold as 40dB. With the test objects being small metal specimens, system timing parameters are specified as PDT: 300μs, HDT: 600μs and HLT: 1000μs. As shown in Fig 2, sensor 1 and sensor 2,
80
Y. Dou et al.
placed on the same side of the specimens, provide linear location, which can obtain the AE characteristic parameters, waveform datas and AE locating features of the crack source during the loading time. 2.3 Loading Course and Process The loading test is shown in Fig 3. To eliminate the noise caused by the friction between the specimen and the support roller, a rubber pad is placed between them. At the beginning of the loading, the specimen is preloaded and unloaded three times to achieve a close contact between the specimen, rubber pad and the support roller. During the whole course, the specimens are always at the stage of elastic deformation. The speed of loading and unloading of the test machine is 1mm/min. The final testing aim is to bend the specimens at α ≤ 1300 when a macro crack can be seen by the eye. The AE signals during the whole course are collected and analyzed. F
S1
60 160
S2
Fig. 2. Placement of sensors
Fig. 3. The loading test
3 Analysis of AE Signals 3.1 Analysis of Hits and Counts versus Time As shown in Fig 4, a few AE hits are collected at the first preloading stage with rather low counts rates. When first preloaded (0 200s), the specimen undergoes elastic deformation (disturbance below 1mm) with little release of strain energy and the deformation within the material will vanish after unloading, so only a few AE signals are
~
AE Feature Analysis on Welding Crack Defects of HG70 Steel Used by Truck Crane
81
(a) Hits versus time
(b) Counts versus time Fig. 4. Hits and counts versus time
created which include those created by the deformation of the rubber pad and the friction as well as the agreement between the specimen and the support roller. At the 2nd and 3rd stage of loading, holding load and unloading (400 1100s), no obvious AE signals can be seen. This is partly due to Kaiser Effect, but on the other hand, it signifies there is an agreement between the specimen, rubber pad and support roller where no friction noise is produced. At the last loading (after 1100s), AE hits rates and counts rates increase rapidly, which is caused by the stress concentration in the defect area resulting from the increase of load. In part of the specimens, yield deformation occurs and tiny crack appears and propagates. With the gradual release of internal energy, a large number of AE signals are created.
~
3.2 Analysis of AE Characteristic Parameters versus Time Fig 5 shows the variation of AE characteristic parameters in the loading. At the elastic deformation stage, the value of AE characteristic parameters is rather low, with the amplitude below 60dB, energy below 103eu and the counts below 100. When the loading comes to the stage of plastic deformation and crack initiation and propagation,
82
Y. Dou et al.
(a) Amplitude versus time
(b) Energy versus time
(c) Counts versus time Fig. 5. AE characteristic parameters versus time
AE hits gradually increase with fairly high amplitude, energy and counts. A number of high-amplitude and high-energy hits even occur simultaneously in some short-time periods, which is caused by the propagation of welding crack.
AE Feature Analysis on Welding Crack Defects of HG70 Steel Used by Truck Crane
83
4 Analysis of AE Signal Features at the Welding Crack Initiation and Propagation Stage 4.1 Analysis of Cumulative Hits and Counts Fig 6 exhibits its loading-time curve during the test, whereas Fig 7 & 8 respectively show the cumulative hits and counts of AE signals. A comparative study of the three Figs shows the findings: the curve indicates that the specimen enters the stage of part plastic deformation at around 1130s(I). A tiny crack occurs in the welding area and grows gradually due to the stress concentration, which dwindles the load-carrying cross section. When the loading comes to 1570s (II), the crack expands with a snap and the carry capacity falls instantly. The welding crack expands again at 1865s (III). Fig 7&8 show the two instant increases in the value which coincide with the time of the instant fracture of the metal. The findings prove that AE technology can accurately monitor the expansion of welding crack. 䋳㥋üᯊ䯈᳆㒓
Load(kN)
6
I
II
III
4.2
2.4
0
0
218
654
1090
t˄s˅ 2180
1526
Fig. 6. Loading-time Curve
I
Fig. 7. Cumulative hits versus time
II
III
84
Y. Dou et al.
I
II
III
Fig. 8. Cumulative counts versus time
4.2 Analysis of AE Signal Location Since the initiation and propagation of welding crack is accompanied by rather active AE signals with high amplitude, the amplitude can be amplified to 45dB to filter waves and observe location events. As shown in Fig 9, location events take place mostly in the range of 25~35mm. AE signal location deviation being considered, this range accords with the welding crack location by and large, which sufficiently testify to the location accuracy of AE technology in monitoring welding crack. Fig 10 indicates that location events mainly take place at the time of crack initiation and propagation, which further proves the close relation between middle location events and welding crack.
Fig. 9. AE location
Fig. 10. Loc. Events―Time―X-Loc.
AE Feature Analysis on Welding Crack Defects of HG70 Steel Used by Truck Crane
85
Space filtering ( 25mm ≤ x ≤ 35mm ) and time filtering (t ≥ 1130s) are adopted to extract the typical AE signals of welding crack initiation and expansion. The distribution and range of characteristic parameters are respectively shown in Table 3. Table 3. Range of the AE characteristics of welding crack Characteristics Range Major Range
Amplitude (dB)
~100 45~70, 97~100 45
(eu)
Counts
~ 1~200
Energy
~10
1 2800
35
8
~80000
40
Duration-Time ( μs )
~2040
10
~1200
100
4.3 Analysis of Time Domain Waveform and Frequency Spectrum Features at the Stage of Tip Crack and Crack Propagation The welding crack develops through three stages: initiation of tiny crack, crack stable propagation and rapid fracture[5]. Fig 11 exhibit the waveform of AE signals at the first and second stage. The time domain waveform Fig shows that the AE signals are typical sudden ones. The frequency spectrum energy of the signals ranges from 80kHz to 380kHz with three major frequency bands: the 1st band centers on 100kHz, the 2nd from 150kHz to 200kHz and the 3rd on 300kHz. The 1st and 3rd band see rather high energy and obvious peak value.
(a) Waveform
(b) Frequency spectrum Fig. 11. Waveform and frequency spectrum of initiation of tiny crack
86
Y. Dou et al.
At the moment of rapid fracture, a few signals of full amplitude (almost 100dB) and high energy (107 108eu) are created as shown in Fig 12. The time domain waveform shows that they are continuous signals formed by several sudden signals which are created by several crack sources. These crack sources are caused by the rapid propagation of the welding crack. As shown in the frequency spectrum Fig 12(b), the frequencies are of rich components and well-distributed between 85kHz and 330kHz. The peak frequencies occur at 100kHz and 200kHz. No obvious peak is shown in high frequency band.
~
(a) Waveform
(b) Frequency spectrum Fig. 12. Waveform and frequency spectrum of rapid fracture
5 Conclusion Acoustic emission during three-bending tests of welding specimens of HG70 steel were investigated and it was found that AE hits and counts can coincide with the
AE Feature Analysis on Welding Crack Defects of HG70 Steel Used by Truck Crane
87
bending and cracking of the specimens. Based on AE hits and counts as well as the loading-time curve of the specimens, welding crack can be dynamically monitored and evaluated. By way of linear location, the welding crack in the specimens can be accurately located, and the distribution ranges of various AE characteristic parameters at the stage of crack initiation and propagation are tentatively fixed. The AE signals at the stage of crack initiation and propagation are typical sudden signals whose frequency spectrum energy distributes between 80kHz and 380kHz. Obvious peak value is found at 100kHz and 300kHz. The AE signals at the stage of crack rapid propagation are continuous signals formed by several sudden signals with rich frequency components, but no obvious peak is shown in high frequency band.
Acknowledgments This study was funded by National High Technology Research and Development Program of China (2008AA042801, 2008AA042803).
References [1] Wu, Z., Shen, G., Wang, S.: Characteristics of Acoustic Emission Testing During the Propagating of External Crack on the Crane Box Beam. Nondestructive Testing 30(9), 635–639 (2008) [2] ASTM F9 14-03: Standard Test Method for Acoustic Emission for Insulated and Non Insulated Aerial Personnel Devices Without Supplemental Load Handling Attachments 9(10) (2003) [3] Drummond, G.R., Fraser, K.F., Little, J., et al.: Assessing the structure integrity of crane booms using acoustic emission. In: EWGAE: 25th European Conference on Acoustic Emission Testing Prague, Czech Republic (2002) [4] Wu, Z., Shen, G., Wang, S.: Application status of acoustic emission technology to crane’s nondestructive test. Hoisting and conveying machinary 10, 1–4 (2007) [5] Ennaceur, C., Laksimi, A., Herve, C., Cherfaoui, M.: Monitoring crack growth in pressure vessel steels by the acoustice mission technique and the method of potential difference. International Journal of Pressure Vessels and Piping 83, 197–204 (2006) [6] Sheng, G., Gao, C.: Weldability of HG70 steel Transactions of the china welding institution 25(3), 117–119 (2004) [7] Akbari, M., Ahmadi, M.: The application of acoustic emission technique to plastic deformation of low carbon steel International Congress on Ultrasonics, Universidad de Santiago de Chile (1), 795–801 (2009) [8] Xu, C., Liu, L., Chen, G.: Characteristics analysis of acoustic emission signals from steel specimens under tensile fracture and fatigue crack condition. Journal of China University of Petroleum 33(5), 95–99 (2009) [9] Carpenter, S.H., Gorman, M.R.: A Waveform Investigation of Acoustic Emission Generated during the Deformation and Cracking of 7075 Aluminum. In: Progress in Acoustic Emission VII, The Japanese Society for NDI, Japan, pp. 105–112 (1994)
88
Y. Dou et al.
[10] Qin, G.-d., Liu, Z.-m., Wang, W.-j.: Study on Acoustic Emission Characteristics of 16Mn Steel in Fatigue Test. China Safety Science Journal 15(8), 105–109 (2005) [11] Zhou, J., Mao, H.-l., Huang, Z.-f.: Acoustic emission technique for the detecting of metal fatigue fracture. China Measurement Technology 33(3), 7–9 (2007) [12] Roberts, T.M., Talebzadeh, M.: Fatigue life prediction based on crack propagation and acoustic emission count rates. Journal of Constructional Steel Research 59(9), 681–693 (2003)
An Equilateral Triangle Waveguide Beam Splitter Zhimin Liu1,2,*, Fengqi Zhou1, Hongjian Li2, Bin Tang3, Zhengfang Liu1, Qingping Wu1, Aixi Chen1, and Kelin Huang1 1
School of Basic Sciences, East China Jiaotong University, Nanchang, Jiangxi, 330013, China 2 College of Physics Science and Technology, Central South University, Changsha, Hunan, 410083, China 3 Jiangsu Polytechnic University, Changzhuo, Jiangsu, 213164, China
[email protected] Abstract. In this paper an optical beam splitter based on an equilateral triangle waveguide (ETW) is studied theoretically and numerically. We show that an optical beam splitter bases on ETW formed only when the length of the waveguide and the location of incident light are appropriate. When the length of the ETW is one third of the self-imaging length, the incident light is divided into three identical and symmetrical beams; while the length of the ETW is one nine of the self-imaging length, the incident light is divided into twenty- seven identical and symmetrical beams, if the number is less than twenty-seven, that is because some of images overlap with each other. It is expected that the results obtained here will help to design a new splitter. Keywords: Self-imaging, Equilateral triangle waveguide, beam splitter.
1 Introduction The propagation of light along a waveguide is one of the fundamental and important questions of wave optics. In recent years the splitter and self-imaging of waveguide have been actively discussed. A theoretical and experimental investigation of the selfimaging properties of planar waveguide have been studied [1-3]. Self-images in a rectangular waveguide has been reported [4]. Similar studies [5-6] were performed on square fiber, and it mentioned that round fiber do not have the image transmission characteristics above, that is mainly because of the angular ambiguity and high symmetry. And optical power splitter(OPS) base on multimode interference waveguide has been studied [7-9]. OPS made of photonic crystal waveguide has been studied [10-11]. A few investigators have been discussed the equilateral triangle resonators [12-16] and ray optics model for triangular hollow waveguides [17], but few attentions are paid to investigate the properties of self-images in an ETW. So study on the behavior of light propagation through this waveguide would be of practical interest. In *
This work was funded by the National Natural Science Foundation of China (Grant No. 60708014), the Key Natural Science Foundation of Hunan Province (Grant No. 06JJ2034), the Natural Science Foundation of Jiangxi (2008GQW0017) and the Research Foundation of East China Jiaotong University (08JC04,09JC01).
D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part IV, IFIP AICT 347, pp. 89–95, 2011. © IFIP International Federation for Information Processing 2011
90
Z. Liu et al.
our previous paper [18], we showed that an image transmission through an ETW, in this paper, we study on the behavior of Gaussian beam propagation(GBP) through this waveguide. An optical beam splitter bases on ETW formed by the new study, the results may have convenience in beam splitter application.
2 Theory As shown in our previous paper [18], consider an ETW with side size a and length L (see Fig.1), and the cladding is a perfect reflector, the field vanish at the boundaries. The field inside the waveguide can be expressed by eigenfunction expansion of Helmholtz equation in the triangular section [19].
Fig. 1. Schematic diagram of an equilateral triangle waveguide ⎛ ∞ 4n 2 λ 2 ⎞⎟ E ( x, y, z ) = ∑ a((20n) , n)ψ ((20n) , n ) ( x, y ) exp⎜ − ikz 1 − ( ) ⎜ 3 a ⎟ n =1 ⎠ ⎝ ⎛ ∞ ∞ 4(m 2 + n 2 − mn) λ 2 ⎞⎟ + ∑ ∑ [a((m+ ), n )ψ ((m+ ),n ) ( x, y ) + a((m− ), n )ψ ((m− ),n ) ( x, y )] exp⎜ − ikz 1 − ( ) ⎜ a ⎟ 9 m 2 n n =1 ⎠ ⎝
(1) For integral values of m, n , with the restriction that m > 2n . For the case of m > 2n , there are two degenerate states with different symmetry properties, which can be written as follows [20] (the correct normalizations are included here): ψ ((m− ), n) ( x, y ) =
⎡ ⎛ 2π ( 2m − n) x ⎞ ⎛ 2πny ⎞ ⎛ 2π (2n − m) x ⎞ ⎛ 2πmy ⎞ ⎟⎟ ⎟⎟ − sin ⎜ ⎟ sin ⎜⎜ ⎟ sin ⎜⎜ ⎢sin ⎜ 3a 3a 3 3a ⎣⎢ ⎝ ⎠ ⎝ 3a ⎠ ⎝ ⎠ ⎝ 3a ⎠ 16
2
⎛ 2π (m + n)x ⎞ ⎛ 2π (m − n) y ⎞⎤ ⎟⎟⎥ - sin ⎜ ⎟ sin ⎜⎜ 3a 3a ⎠ ⎝ ⎝ ⎠⎥⎦
(2) ψ ((m+ ), n ) ( x, y ) =
16 ⎡ ⎛ 2π ( 2m − n) x ⎞ ⎛ 2πny ⎞ ⎛ 2π ( 2n − m) x ⎞ ⎛ 2πmy ⎞ ⎛ 2π (m + n)x ⎞ ⎛ 2π ( m − n) y ⎞ ⎟⎟ − cos⎜ ⎟⎟ + cos⎜ ⎟⎟ ⎟ sin ⎜⎜ ⎟ sin ⎜⎜ ⎟ sin ⎜⎜ ⎢cos⎜ 3a 3a 3a 3 3a 2 ⎢⎣ ⎝ 3a ⎠ ⎝ 3a ⎠ ⎝ ⎠ ⎝ 3a ⎠ ⎝ ⎠ ⎝ ⎠
(3)
An Equilateral Triangle Waveguide Beam Splitter
91
=
In the special case of m 2n there is a single nondegenerate state for each n , and the wavefunction is given by ψ ((20n) , n) ( x, y) =
⎡ ⎛ 4πny ⎞⎤ ⎛ 2πnx ⎞ ⎛ 2πny ⎞ ⎟⎟− sin ⎜⎜ ⎟⎟⎥ ⎟ sin ⎜⎜ ⎢ 2 cos⎜ a 3 3a ⎣⎢ ⎝ ⎠ ⎝ 3a ⎠ ⎝ 3a ⎠⎦⎥ 8
2
(4)
The coefficients a ((20n) , n) , a ((m+ ), n ) , a ((m− ), n) are determined by the projection of the initial
distribution onto the waveguide a((20n) , n ) ( x, y ) = ∫∫ dx1dy1ψ ((20n) , n ) ( x1 , y1 ) E1 ( x1 , y1 )
(5a)
a((m+ ), n ) ( x, y ) = ∫∫ dx1dy1ψ ((m+ ), n) ( x1 , y1 ) E1 ( x1 , y1 )
(5b)
a((m−), n ) ( x, y ) = ∫∫ dx1dy1ψ ((m− ), n ) ( x1 , y1 ) E1 ( x1 , y1 )
(5c)
In this paper, the initial light we consider a GBP E1 ( x1, y1 ) = exp[ −
x 2 + ( y − 3a 6)
ω02
(6)
]
(
Where ω 0 is beam waist size of the GBP, The GBP is launched at 0, 3a 6) , instead
(
of the center 0, 3a 3) , in order to avoid any overlap of the images. The coefficients can’t be get analytic solution, which are given by oscillatory numerical integral. L0 = 9a 2 2λ is the self-imaging length of the ETW, the detailed derivation as shown in[18], that is, at this distance (the self-imaging distance) the initial distribution repeats.
3 Numerical Results and Analysis The output field at the back face of the waveguide was numerically calculated according to Eq.(1) with the waveguide size a = 0.2mm ,the wavelength λ = 633nm and GBP with ω 0 = 10μm . Modes m, n should be taken on values for all the possible guided-wave modes, in other words, m max , n max are determined by 1−
4n 2 λ 2 4(m 2 + n 2 − mn) λ 2 ( ) ≥ 0 and 1 − ( ) ≥0 3 a 9 a
But normally, the coefficients decrease rapidly with increasing m, n , the actual number of modes which have to taken into account is much less than m max , n max [1], in this paper m, n = 100 are used in calculation. 3.1 The Field at the Face of z = 0
Firstly, at the incident face of the waveguide z = 0 , that is the incident wave. It is clear that the output field consists very well with the input field both in field intensity
92
Z. Liu et al.
and position from Fig(2). From these we consider that the simulation procedure is correct and feasible.
Fig. 2. The field at the distance z = 0 (right is contour plot)
Fig. 3. The field at the distance z = L0
3.2 The Output Field at the Back Face of Self-imaging Length in ETW
From the theory educing we know L0 = 9a 2 2λ is the self-imaging length of the ETW, which means at this distance the initial distribution repeats. Fig(3) demonstrates the result, the field intensity and position of output field consist with the input field. Similar distributions are also observed at the distance z = nL0 ,where n is an integer. 3.3 The Output Field at the Back Face of Other Length
As shown in our previous paper [18], similar the results at the distance z = L0 3 are also simulated, obviously the initial distribution is splitted into three identical and symmetrical distributions at the back face of the waveguide, and the output fields have 3 fold rotational symmetry. Moreover, the total intensity of the three beams consists with the intensity of the input field, these are displayed in Fig(4). Fig.5(a) demonstrates the similar results when the input field is launched in the other location, for example the location of (0, 3a 4) .However, if the position of the input field is located in the center of the waveguide (0, 3a 3) , the only one output beam is still located the center, Fig.5(b) shows the result, which is because some of images overlap with each other.
Fig. 4. The field at the distance z = L0 3
Fig. 5. The incident wave launched at the
(0, 3a 4) and (0, 3a 3)
An Equilateral Triangle Waveguide Beam Splitter
93
Finally, the field distribution at the distance z = L0 9 are calculated and displayed in Fig(6). The incident light is splitted into nine identical and symmetrical distributions when the incident light is launched at (0, 3a 6) as shown in Fig.6(c); while the incident light is launched at (0, 3a 4) , the initial wave is splitted into twenty-seven identical and symmetrical distributions, Fig.6(d) shows the result. So the number of the output beams is decided by the location of the incident light. We consider the initial wave can split into twenty- seven at this distance, if the number is less than twenty-seven, that is because some of images overlap with each other. (Note that: Fig.6(d) takes GBP with a waist size of ω 0 = 6 μm to avoid superposition of the field).
Fig. 6. The field at the distance z = L0 9 : (c) the incident wave launched at the (0, 3a 6) ; (d) launched at the (0, 3a 4)
A splitting of the field distributions in waveguide is well known phenomenon, this phenomenon can be used for beam splitters. The main difference between the waveguide beam splitter and other beam splitter is that it can provide many output beams with the same intensity. Another nice property of the ETW beam splitter or the difference between the ETW beam splitter and the other waveguide beam splitter is that the ETW beam splitter can provide 3 n (n = 0,1,2 ") beams with the 3 fold rotational symmetry. So the interesting application of the results as demonstrated in this work is useful in designing a new beam splitters.
4 Conclusion Summarizing, from above discussed, an optical beam splitter bases on ETW formed when the length of the waveguide and the location of identical light are appropriate. When the length of the ETW is one third of the self-imaging length, the incident light is divided into three identical and symmetrical beams; while the length of the ETW is one nine of the self-imaging length, the incident light is divided into twenty- seven identical and symmetrical beams, if the number is less than twenty-seven, that is because some of images overlap with each other. It is expected that the results obtained here will help to design a new splitter.
94
Z. Liu et al.
References [1] Ovchinnikov, Y.B.: Revivals of light in a planar metal waveguide. Optics. Communications 182, 35–43 (2000) [2] Ovchinnikov, Y.B., Pfau, T.: Revivals and oscillations of the momentum of light in a planar multimode waveguide. Physical. Review. Letters 87, 123901 (2001) [3] Ovchinnikov, Y.B.: A planar waveguide beam splitter. Optics.Communications 220, 229–235 (2003) [4] He, S.L., Ao, X.Y., Romanov, V.: General properties of NM self-images in a strongly confined rectangular waveguide. Applied. Optics 42, 4855–4857 (2003) [5] Wu, C.Y., Somervell, A.R.D., Barnes, T.H.: Direct image transmission through a multimode square optical fiber. Optics. Communications 157, 17–22 (1998) [6] Wu, C.Y., Somervell, A.R.D., Haskell, T.G.: Optical sine transformation and image transmission by using square optical waveguide. Optics. Communications 175, 27–32 (2000) [7] Sun, Y.L., Jiang, X.Q., Yang, J.Y., Tang, Y., Wang, M.H.: Experimental demonstration of two dimensional multimode-interference optical power splitter. Chinese Physics Letters 20, 2182–2184 (2003) [8] Han, Z.H., He, S.L.: Multimode interference effect in plasmonic subwavelength waveguides and an ultra-compact power splitter. Optics Communications 278, 199–203 (2007) [9] Zhang, Y.W., Liu, L.Y., Wu, X., Xu, L.: Splitting-on-demand optical power splitters using multimode interference waveguide with programmed modulations. Optics Communications 281, 426–432 (2008) [10] Ma, Z.T., Ogusu, K.: Power splitter based on cascaded multimode photonic crystal waveguides with triangular lattice of air holes. Optics Communications 282, 3473–3476 (2009) [11] Li, W., Xu, X.M.: An ultra-short double-wavelength optical power splitter for two waveguides operation based on photonic crystal multimode interference. Optics Communications 5, 69–73 (2010) [12] Huang, Y.Z.: Eigenmode confinement in semiconductor microcavity lasers with an equilateral triangle resonator. In: Proceedings-SPIE The international society for optical., vol. 239, p. 3899 (1999) [13] Guo, W.H., Huang, Y.Z., Wang, Q.M.: Resonant frequencies and quality factors for optical equilateral triangle resonators calculated by FDTD technique and the Padeapproximation. Photonics Technology Letters, IEEE 12, 813 (2000) [14] Huang, Y.Z., Guo, W.H., Wang, Q.M.: Analysis and numerical simulation of eigenmode characteristics forsemiconductor lasers with an equilateral triangle micro-resonator. Journal of Quantum Electronics, IEEE 37, 100 (2001) [15] Wysin, G.M.: Resonant mode lifetimes due to boundary wave emission in equilateral triangular dielectric cavities. Journal of optics A: Pure and Applied. Optics 7, 502–509 (2005) [16] Huang, Y.Z., Guo, W.H., Yu, L.J., Lei, H.B.: Analysis of semiconductor microlasers with an equilateral triangle resonator by rate equations. Journal of Quantum Electronics 37, 1259–1264 (2001) [17] Isaac, G., Khalil, D.: Ray optics model for triangular hollow silicon waveguides. Applied Optics 45, 7567 (2006)
An Equilateral Triangle Waveguide Beam Splitter
95
[18] Liu, Z.M., Zhu, K.C., Tan, B., Hao, Z.Q., Wen, W.: Image transmission through a mental equilateral triangle waveguide. Chinese Journal of Quantum Electronics 24, 253–256 (2007) (in chinese) [19] Doncheski, M.A., Robinett, R.W.: Quantum mechanical analysis of the equilateral triangle billiard: periodic orbit theory and wave packet revivals, 8 (2003), arXiv: quantph/0307063 [20] Lin, S.L., Gao, F., Hong, Z.P., Du, M.L.: Quantum spectra and classical orbits in twodimensional equilateral triangle billiards. Chinese. Physics. Letter 22, 9–11 (2005)
Analysis and Implementation of Embedded SNMP Agent Hubin Deng, Guiyuan Liu, and Lei Zhang College of Information Engineering, East China Jiaotong University, Nanchang, P.R. China
[email protected] Abstract. With the extensive application of network devices and the rapid development of embedded technology, network management of embedded devices becomes increasingly complicated. SNMP management is the most widely used network management system(NMS). Most of the network components used in enterprise network have built-in network agents that can respond to an SNMP network management system. By analyzing the SNMP (Simple Network Management Protocol) and base on NET-SNMP development kit, discuss the construction of MIB modules and code conversion and complete embedded SNMP Agent extension. Through network management tools to verify the SNMP agent on the network management functions of the embedded device. Keywords: Embedded technology, Network management, SNMP Agent, MIB modules, NET-SNMP development kit.
1 Introduction Along with the popularization of the network application and the network equipment, it is also gradually increasing to the network management demand. Because Simple Network Management Protocol(SNMP) has obtained the widespread application in the industrial world by its simplicity. The TCP/IP major part router and the switchboard all support SNMP in the protocol standard certain main management information database (MIB). In addition, in other private network equipment management domain, the SNMP network management has also obtained the widespread application. In the SNMP management model, the management station is carries on the management and the monitoring center to AGENT, Agent is managed to the equipment to carry on the monitoring and front end the operation network management. Therefore, in the network equipment, to increase SNMP the network management to act AGENT adapts the network supervising and managing development essential work. This article takes SNMP and the existing system resources as a foundation, analyzes the embedded SNMP proxy software with emphasis of the function module, with the aid of opens source tool development package NET-SNMP, has constructed the MIB storehouse module, and has developed the embedded equipment SNMP agent software. D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part IV, IFIP AICT 347, pp. 96–102, 2011. © IFIP International Federation for Information Processing 2011
Analysis and Implementation of Embedded SNMP Agent
97
2 Simple Network Management Model Simple Network Management Protocol (SNMP) is a UDP-based network protocol. It is used mostly in network management systems to monitor network-attached devices for conditions that warrant administrative attention. SNMP management is also referred to as Internet management. It’s called SNMP management because it has matured to the level that it manages more than the Internet, for example, intranet and telecommunication networks. Any network that uses the TCP/IP protocol suite is an ideal candidate for SNMP management. SNMP network management system can manage even non-TCP/IP network elements through proxy agents. The SNMP management consist four elements: Network Management System (NMS), Agent, Management Information Base (MIB), and Simple Network Management Protocol (SNMP). It uses the concept of Client/Server application. A network management system (NMS) executes applications that monitor and control managed devices. On each manageable equipment, a SNMP Agent is running. This agent manages information relating to the equipment which is stored in a local database called the MIB.And the SNMP Protocol is used to connect the NMS and the Agents. SNMP is a protocol built on the top of UDP/IP: The architecture specifies the management messages between the management system and the management agents. 5 types of SNMP messages or SNMP requests can be exchanged (SNMPv1) between a SNMP agent and a SNMP manager: 1. Obtaining the current value of a MIB object managed by an agent: request getrequest (GET). 2. Obtaining the current value of the next MIB object managed by an agent: request get-next-request (GETNEXT). 3. Updating of the current value of a MIB object managed by an agent: request set-request (SET). 4. Sending back the value of a MIB object managed by an agent: request getresponse . It's the answer to a GET, GETNEXT or SET request. One can see that SNMP is a command/response protocol without state. 5. Signal/alarm emitted by an agent to a manager: message trap (TRAP). Agent
Management get-request
get-response 8'33257 get-next-request get-response set-request
8'33257 8'33257
get-response trap 8'33257
Fig. 1. Five types of SNMP Operation
98
H. Deng, G. Liu, and L. Zhang
3 MIB Module Design and Code Conversion SNMP itself does not define which information a managed system should offer. Instead, SNMP uses an extensible design, where the available information is defined by management information baes(MIB). MIB describe the structure of the management data of a device subsystem; they use a hierarchical namespace containing object identifiers (OID). Each OID identifies a variable that can be read or set via SNMP. MIB use the notation defined by ASN.1.
Fig. 2. Definiton(skeletal) of SMI for SNMP v2
In telecommunications and computer networking, Abstract Syntax Notation One (ASN.1) is a standard and flexible notation that describes data structures for representing, encoding, transmitting, and decoding data. It provides a set of formal rules for describing the structure of objects that are independent of machine-specific encoding techniques and is a precise, formal notation that removes ambiguities. ASN.1 is defined as a set of rules used to specify data types and structures for storage of information. Basic Encoding Rules (BER) is defined for the transfer syntax by the ASN.1 syntax. The syntax to create a SMIB module is referred to the description section in SMI(Structure of Management Information). In this paper, MG-SOFT is used to define a MIB document in ASN.1. MG-SOFT's MIB tools are quite mature and widely used for SNMP development and testing. Its MIB Browser can not only read and write MIB, but also receive the Trap sent by SNMP agents; MIB Compiler can be used to check the legality of MIB, which is useful for the compose of MIB. This MIB document is then transformed into the C language source file by MIB2C.MIB2C is a useful software tool in NET-SNMP. Follow these steps to convert: First of all copy the MIB module definition files to the mibs directory: cp./modulename /usr/local/share/snmp/mibs; Then, run MIB2C command: /Usr/local/bin/mib2c modulename to translate; Finally, MIB2C generated in the current directory two C source files: modulename.h and modulename.c. These two documents are made under the MIB library
Analysis and Implementation of Embedded SNMP Agent
99
module conversion, and also need to be added to the NET-SNMP SNMP Agent Software to extensions of the source code.
4 SNMP Entity and AGENT Extension The SNMP Agent consists of a dispatcher, a message processing subsystem, and a securtiy subsystem. The SNMP Agent consists of a dispatcher, a message processing subsystem, and a securtiy subsystem. The SNMP message processing subsystem of an SNMP engine interacts with the dispatcher to handle version-specific SNMP message. It contains one or more message processing models. The version is identified by the version field in the header. The security and access control subsystem provides authentication and privacy protection at he message level. The access control subsystem provides access authorization security. SNMP Entity
Fig. 3. SNMP Agent Entity
100
H. Deng, G. Liu, and L. Zhang
There is only one dispatcher in an SNMP engine, but it can handle multiple versions of SNMP message. It performs three sets of functions. First, it sends message to and receives message from the network. Second, it determines the version of the message and the interacts with the corresponding message processing model. Third, if provides an abstract interface to SNMP applications to deliver an incoming PDU to the local application and to send a PDU from the local application to a remote entity. To complete the SNMP Agent to a embedded device takes the steps as follows: In the net-snmp-5.4.2.1 directory configure: ./ Configure - host = arm-linux target = arm --With-cc = /armv4l-unknowlinux-gcc - -with-ar = arm-linux-ar --disable-shared - with-endianness = little-With-mibmodules = "moudlename". In the net-snmp-5.4.2.1 directory, compile: make. In net-snmp-5.4.2.1/agent directory, you can see the generated snmpd process, as shown below: Copy the snmpd program to the development board "/ usr / bin" directory and start with the following command: /usr/bin/snmpd –V –c /etc/snmpd.conf.
Fig. 4. Run the snmpd programe
5 Conclusion The SNMP protocol was developed to facilitate the network management. In this paper the free software package, NET-SNMP is used to extend the NET-SNMP agent to an embedded system under the ARM-LINUX OS in order that the embedded can be controlled remotely by SNMP NMS.NET-SNMP makes it possible to integrate an embedded system into a network and to manage it with SNMP managers like MGSOFT. The Management station's IP is 192.168.2.4,and the embedded equipment's IP is 192.168.2.120.The testing result of verifying the embedded SNMP agent shows as in Figure 5.
Analysis and Implementation of Embedded SNMP Agent
101
Fig. 5. Verify the SNMP agent
Acknowledement This work was sponsored by the technology proliferation plan item of Nanchang city (Caiqi[2008], NO.68), science and research foundation of East China Jiaotong University(09XX05).
References [1] Rose, M.T., McCloghrie, K.: RFC1155 Structure and identification of management information for TCP/IP-based internets (May 01, 1990) (for the complete set of data types) [2] Case, J.D., Fedor, M., Schoffstall, M.L., Davin, C.: RFC1157 Simple Network Management Protocol (SNMP) (May 01, 1990) (Obsoletes RFC1098) [3] McCloghrie, K., Rose, M.T.: RFC1213 Management Information Base for Network Management of TCP/IP-based internets: MIB-II (March 01, 1991) (Obsoletes RFC1158) (Updated by RFC2011 RFC2012, RFC2013) (Also STD0017) [4] Case, J., McCloghrie, K., Rose, M., Waldbusser, S.: RFC1905 Protocol Operations for Version 2 of the Simple Network Management Protocol (SNMPv2).SNMPv2 Working Group (January 1996) [5] Subramanian, M.: Network Management Principles and Practice. Higher Education Process Pearson Education, Beijing (2001) [6] Comer, D.E.: Automated Network Management Systems. Machinery Industry Press, Beijing (2008) [7] Wang, S., Li, T.: Application of SNMP on VxWorks Embedded Operation System. Micro Computer Information 21(5), 86–87 (2004)
102
H. Deng, G. Liu, and L. Zhang
[8] Richardson, P.C., Xiang, W., Mohamad, S.: Performance Analysis of a Real·-time Control Network Test Bed in a Linux–based System with Sporadic Message Arrivals. IEEE Transactions Oil Industrial Informatics 2(4), 231–241 (2006) [9] Fu, G., Yang, W.: Research on scheme of intelligent network management. Journal of Xi’an University of Engineering Science and Technology 19(1), 89–92 (2005) [10] Stallings, W.: SNMP and SNMPv2.The infrastructure for network management. Communications Magazine, IERE 36(3), 110–115 (2008)
Application of Computer Technology in Advanced Material Science and Processing Yajuan Liu* SoftSchool, East China JiaoTong University, Nanchang, P.R. China Tel.: +86-791-7046184; Fax: +86-791-7046185
[email protected] Abstract. Computer technology is an actual system model, which is largely unaffected by experimental conditions, time and space constraints, and is of great flexibility. Nowadays, computer technology has thoroughly penetrated in the various areas of material processing and research, which becomes one of the important frontiers in the field of material manufacturing industry. At the same time, material science and technology are also developing rapidly and constantly giving birth to the new industrial field, such as nanotechnology, optoelectronic, magnetic electronic technologies, which are inseparable of computer technology. Hence, in this article, the application of computer technology in advanced material science and processing, which includes material science database, computational material science, computer-aided design or processing etc are reviewed. Keywords: Computer Technology, Material Processing, Material Science.
1 Introduction With the continuously deepening research of material science, material science occupies an important position in the national economy; however, material science is still an immature interdisciplinary, which mainly depends on the facts and the experience of the current study. The systematic studies need a very long process[1]. Computer as a modern tool plays an increasingly significant role in various areas of the world, which has penetrated into many fields. With respect to the material science and engineering, the computer is also becoming a very important tool and becomes one of the reasons for the accumulation of the rapid development of material science. For example, computer technology has been widely used in the variety field of material forming technology, including application in liquid forming, plastic forming, polymer material forming, powder forming et al, which can basically provide a qualitative description toward to quantitative prediction for material processing[2-3]. Furthermore, computer application in material science is the trend of multi-scale simulation and integration[4-5]. In this *
Corresponding author.
D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part IV, IFIP AICT 347, pp. 103–107, 2011. © IFIP International Federation for Information Processing 2011
104
Y. Liu
article, the application of computer technology in material science and processing including material science literature search practice, computational material science, computer-aided design and processing are mainly introduced.
2 Computer Technology in Material Science Databases A reasonable choice of material, accurate design and scientific processing are directly impacting on product cost and quality, which even affect the social development and progress. In recent years, with computer technology development, particularly in the development of database technology, material science databases in scientific research has become increasingly emphasized and get more and more widely used. A sample material database was illustrated in Figure 1.
Fig. 1. A sample of material database
Material science database can be divided into the data type, numeric type as well as the map database. Otherwise, it is also divided into the online and offline database type. The literature database is of mainly online services, while numerical databases were more used in off-line. Moreover, according to the point view of material, it can be classified into metallic and nonmetallic material databases. Public material databases have been constructed every year and the developed countries continue to make the information on this strategic. Numeric material database has been established in China since 1992 and has accumulated in recent years[6]. Though after several decades of development and accumulated, however, the following deficiencies still exist: (1) The lack of the data of mechanical properties. In the development of metallic material database, it focuses on the iron and steel materials,
Application of Computer Technology in Advanced Material Science and Processing
105
however, with respect to the light alloys such as aluminum, magnesium alloys, it was still lack of the mechanical properties' database.(2) The lack of the data of material processes performance. Cold and hot processing of material data is rarely found. (3) Although some scientific databases of materials have built at home and abroad, its research and development still lags far behind the actual rate of application requirements. Generally, network, standardization, intelligence and commercial trends will make the application material database, extended to broader areas of material research and development.
3 Computational Material Science Computational material science (CMS) is a typical interdisciplinary of material science and computer science, which is the material scientific research about the “computer design” and “computer experiment” in material composition, structure, performance, service performance[7]. According to the literatures[8], CMS could be mainly included into two aspects: one is calculation and simulation, which is starting from the experimental data, through the establishment of mathematical models and numerical calculations to simulate the actual process; the other is the computer added material design, which is directly through the theoretical model to calculate, predict or design the new structure and properties of material. Therefore, CMS is a bridge to connect theory and experimental material. It is well known that the material composition, structure, performance and service performance are the four elements of material research. The traditional research is based material experimental results in the laboratory, which is an experimental science. However, with the requirements of high material performance increasing, especially because the material sciences research object is constantly changing spatial scale into small, the micro-level studies do not reveal the nature of material properties, nano-structures and atoms scale and even electronic level become the studied content when the functional material are studied[9]. Therefore, material research is increasingly dependent on high-level testing technology, the research difficulty and costs are getting much higher. In addition, the service performance is increasing attention in material research, which is to study the interaction of material and service environment and its impact on material performance. As the material was serviced in an increasingly complex environment, laboratory studies of service performance have become more and more difficult. In short, it was difficult that the new and modern material research and development relying solely on laboratory experiments to conduct material research. Computer simulation technology, however, according to the basic theory, from the inclusive concept of a virtual environment, micro-, meso-, macro-scale, multi-level research on the material in the computer, but also can simulate the ultra-high temperature, high pressure material under extreme environments such as service performance to simulate material properties under service conditions, failure
106
Y. Liu
mechanism, so as to realize the performance of material in service improvements and material design. Thus, in the field of modern material science, the computer "experiment" has the same important position in research methods. It is necessary to point that CMS is closely related to computer development. In the past, even if the use of large computers is also very difficult to calculate a number of materials, such as material, quantum mechanic calculations, now it can complete this on PC. In addition, with the continuous progress of CMS and mature, material, computer simulation and design is no longer a hot research topic of just theoretical material physics and material scientists calculated, but also will become an important research tool of general-purpose material researchers.
4 Computer Aided Design and Manufacture The initial processing of computer used for complex engineering analysis and calculation is followed by the development process in the modern industrial product design, such as the aircraft surface flow field calculation, the stress analysis of the complex structure[10]. The system can do a lot of complex calculations in a very short period of time, and it is possible for many programs for rapid analysis and evaluation to choose the best design. Material processing CAD can be divided into casting CM, plastic forming CAD, welding forming CAD, injection molding CAD, as well as mold CAD[11]. The computer simulation of casting process was carried out earlier and the technically is more mature, which has been into the micro-macro simulation stage. From the early 90's in 20 centuries, it has launched the computer simulation of micro-morphology, in which it can simulate the nucleation, growth the process of casting solidification process of forecasting[12-14]. After years of research and development, a large number of casting process simulation software has been the commercialization, which is shown in Table 1[15]. Table 1. Overview of main foreign casting special software Software name Mavls software
Developer Alphacast software Ltd
Flow-3d
Flow science, Inc
ProCast
UES software,Inc
Cast CAE4
Finland
Function Predicted melt flow temperature, pressure, velocity distribution, macro-and micro-shrinkage, dendrite arm spacing, steady-state temperature distribution Automatically predicted solidification shrinkage, binary segregation and tracking of surface defects formation of micro-structure such as porosity, pore aggregation Calculated solidification shrinkage, formation of 3-D view
Application of Computer Technology in Advanced Material Science and Processing
107
5 Conclusion In summary, material science is a cross-emerging for the development of immature discipline. At present, its research is largely depended on the facts and the experience. The systematic studies need a very long process. The computer is becoming an extremely important tool, which is one of the important reasons for rapid development of material science. The use of computers for design of new material has gradually been recognized and used. However, this understanding and effort are still insufficient. Facing the future, computer simulation technology, material calculation and design will become an inevitable trend.
References 1. Flaszka, W.G., Paro, J.A., Kivivuori, S.O.J.: Computer-aided forging design using model material simulation. J. Mater. Process. Technol. 24, 403–409 (1990) 2. Joshi, K., Lauer, T.W.: Impact of information technology on users’ work environment: A case of computer aided design (CAD) system implementation. Inf. Manage. 34, 349–360 (1998) 3. Sapuan, S.M.: A knowledge-based system for material selection in mechanical engineering design. Mater. Des. 22, 687–695 (2001) 4. Gates, T.S., Odegard, G.M., Frankland, S.J.V., Clancy, T.C.: Computational materials: Multi-scale modeling and simulation of nanostructured materials. Compos. Sci. Technol. 65, 2416–2434 (2005) 5. Hao, S., Liu, W.K., Moran, B., Vernerey, F., Gregory, B.O.: Multi-scale constitutive model and computational framework for the design of ultra-high strength, high toughness steels. Compos. Methods in Appl. Mech. Eng. 193, 1865–1908 (2004) 6. http://www.csdb.cn/viewdb.jsp?uri=cn.csdb.material 7. Sokrates, T.P.: Frontiers in computational materials science. Compos. Mater. Sci. 2, 149–155 (1994) 8. Richard, P.M.: Computational materials science -a personal perspective of an industrial scientist. Compos. Mater. Sci. 2, 161–167 (1994) 9. Hafner, J.: Atomic-scale computational materials science. Acta Mater. 48, 71–92 (2000) 10. Wan, H.: Application of computers in materials science. Sci. Technol. of Aotou Steel Corp. 31, 6–9 (2005) 11. Bata, G.L., Salloum, G.: Computer integrated material processing (CIMP) -A generic application of CAD/CAM technology. Mater. Des. 8, 220–228 (1987) 12. Jerald, R.B., Sanjay, J., John, W.D., Srinivas, P.: Application of computer-aided engineering techniques to tooling for castings. J. Manuf. Syst. 11, 215–223 (1992) 13. Im, Y.T.: A computer-aided-design system for forming processes. J. Mater. Process. Technol., 89–90, 1–7 (1999) 14. Karima, M., Richardson, J.: A knowledge-based systems framework for computer-aided technologies in metal forming. J. Mechan. Work. Techn. 15, 253–273 (1987) 15. Cao, H.J., Song, Y.P., Wang, W.Y.: The application and development of Computer Simulation of Casting Process. J. Henan Univ. Sci. and Techn.: Nat. Sci. 27, 5–8 (2006)
Application of Interferometry in Ultrasonic System for Vibration Zhengping Liu, Shenghang Xu, and Juanjuan Liu School of Mechatronical Engineering, East China Jiaotong University, Nanchang, 330013, P.R. China
[email protected] Abstract. Vibration signals are important state parameters for mechanical equipments’ status monitoring and fault diagnosis. In this paper, in order to overcome the limitations of the traditional vibration measurement med1ods and instrument, a new non-contacting vibration method based on ultrasonic for vibration detection in special environment was presented. The mainly researched in this paper were the circuit for ultrasonic transmitting, receiving, algorithm and the module based on LabVIEW for signal analysis and processing. New algorithm was adopted in the system design. The measurement for vibration signals, which may have higher accuracy, was based on ultrasonic wave of different frequency. Experiments were carried on for proving the theory and the result was expected, verifying the reliability and feasibility of the system. Keywords: Vibration Signal, Ultrasonic, Signal Processing, Fault Diagnosis.
1 Introduction Vibration signals are important state parameters for mechanical equipments status monitoring and fault diagnosis. The method of vibration signals measurement is very limited. Such as displacement sensor, speed sensor and acceleration sensor, they are limited of itself. Most of the sensors are mounted to measure the objects. Testing vibration on Eddy current sensor is a noncontact system, but the distance is very limited. The traditional vibration measurement med1ods and instrument can not achieve effective measurement in HTHP industrial environment. A new non-contacting vibration method based on ultrasonic for vibration detection is completed high precision, long-distance measuring in special environment.[1] To overcome the limitations of the traditional vibration instruments, a noncontacting vibration method based on ultrasonic for vibration detection in HTHP environment was presented in this paper. The noncontact system of vibration measurement is based on optical interferometry. Ultrasonic has advantages such as high frequency, shorter wave-length, steady direction of propagation, and easily obtainment of directional and focused ultrasound beam. The principle of the vibration measurement system is based on Doppler. A continuous wave ultrasonic beam is transmitted toward the vibrating surface, and the ultrasound signal reflected by this moving surface is sensed by a second transducer. The received ultrasound signal is D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part IV, IFIP AICT 347, pp. 108–115, 2011. © IFIP International Federation for Information Processing 2011
Application of Interferometry in Ultrasonic System for Vibration
109
phase modulated. Therefore, features can be extracted from the vibration signal by analyzing the factors that affect the variations of traffic load.
2 Method The method of the vibration measurement system is based on the phase. The schematic diagram of the ultrasonic system for vibration measurements is given in the Fig.1., y(t) and y(r) are respectively the transmit signal and the receive signal. In operation, the ultrasonic transmitter transmits y(t) continuously and the ultrasonic receiver will simultaneously receives y(r) reflected by the object. Besides, the w is the angular frequency of ultrasonic wave, the C is speed of ultrasonic wave, and the L(t) is distance between the ultrasonic transducers and the measured object.[1] The transmitted and the received wave forms are given by
y (r ) = sin( wt + φ )
(1)
y (t ) = sin wt
(2)
The distance that the signal spreads from transmitting transducer to receiving transducer is 2 L(t). The L(t) depends on effective vibration length d(t). The change of phase Φ(t) is determined by the formula,
φ (t ) =
2 L(t )
λ
∗ 2π
(3)
The effective vibration length is determined by the formula,
d (t ) = ΔL(t ) =
Δφ (t ) λ 2π 2
(4)
The λ is the wavelength of the ultrasonic wave used in this system; the Φ(t) is the change of phase and it is changed among 0 to 2π ; the Φ(t) depends on the effective vibration length d(t). The peak amplitude dmax is peak amplitude of the ultrasonic wave. The θ has some repeated and the d(t) is not judged when the d(t) exceeds the length of λ/2. So the d(t) is bounded by λ/2 and the D(t) that it .the d(t) is bounded by λ/2 that is defect. To overcome this limitation, an effective algorithm was selected and low frequency wave and high frequency wave was adopted in this paper. The low frequency wave has large wavelength, so that the measurement range was increased. The high frequency wave has high frequency, so that the measurement precision is improved. The total displacement D(t) is determined by the formula, D(t)=2 d (t ) = ΔL(t ) =
θ Δφ (t ) ∗λ =n∗λ + ∗λ 2π 2π
(5)
The θ1 is the phase shift and its value usually cycles between zero to 2π. As shown in the Fig.2., the D is the total displacement, the θ1 is the phase, the λ1 is the wavelength
110
Z. Liu, S. Xu, and J. Liu
and the f1 is frequency of the low frequency wave. The total displacement D(t) is determined by the formula,
θ1 ∗ λ1 2π
D (t ) = D1(t) =
d
L
(6)
d
R
T D
Fig. 1. Schematic diagram of ultrasonic system for vibration measurements
The θ2 is the phase, the λ2 is the wavelength and the f2 is frequency of the low frequency wave. The θ2 is a greater phase shift than phase frequency of the f1 because the f2 is greater than the f1. The integer number N2 of wavelength of frequency f2 signal can be calculated from N2=Int[D1/λ2]. The total displacement D(t) is determined by the formula, D(t)=D2(t)= N 2 ∗ λ 2 +
θ2 ∗ λ2 2π
D
2π θ1
λ1 2π
n
θ2
λ2 Fig. 2. Schematic diagram of the method
(7)
Application of Interferometry in Ultrasonic System for Vibration
111
The final equation concluded from the above equations is shown as follows, D(t)= Int[D1/ λ 2 ]* λ 2 +
= Int[
θ2 θ θ 1 ∗ λ 2 = Int[ 1 ∗ λ1 ∗ ]* λ 2 + 2 ∗ λ 2 2π λ2 2π 2π
θ1 f 2 c c θ2 ∗ ]* + ∗ 2π f 1 f 2 2π f 2
(8)
where c is the speed of sound. The frequencies of 10 kHz and 40kHz are chosen in the experiment. So, the maximum distance is 1.70 cm.[1],[2],[3],[4]
3 The Hardware System Architecture The hardware of the vibration measurement system consists of two ultrasonic transducers, for transmitting and receiving the signals, and the circui for ultrasonic transmitting and receiving signal processing. The part completes signal acquisition and processing function automatically. 3.1 Design of Transmission Circuit The design of the circuit for ultrasonic transmitting adopted NE555 time-based circuit and also the peripheral circuits and the more harmonic oscillator circuit.The circuit of NE555 time-based worked for the setting and reset alternately repeatedly on without the steady-state operation mode. The 3 pin of output terminal is outputting utalternate with low level and high level, the output waveform approximated rectangular wave. There are many kinds of ultra harmonics of rectangular, therefore, the circuit without the steady-state operation mode can be called multi vibrator as self-excitation.[4],[5] As shown in the Fig.3., the circuit for ultrasonic transmitting consists of the circuit of NE555 time-based and the correlative circuit. The output of capacitor C1 is unchanged. The 2 pin of NE555 outputs low level, the 3 pin of NE555 outputs high level and inward transistors are in off condition. The 7 pin is impending and the capacitor C1 is charged by VDD. Then, the voltage of C1 is building up. The VC1 reaches the threshold level 2VDD/3 after a time of T1. So, now the circuit of NE555 time-based has overturned and reseted. Then the 3 pin outputs low level, and the inward transistors are in conducting state. The 7 pin outputs low level, the capacitor C1 is discharging. When the VC1 is droping to VDD/3 after T2, the circuit of NE555 time-based has overed and reseted, and the 3 pin outputs high level and the 7 pin is impending again, circulate down so. So the 3 pin can output a rectangular wave. The T1 of high-level outputs charging time is determined by the formula,
t1 = −(R1 + R2 )C1 ln[(VDD − 2VDD / 3) /(VDD −VDD / 3)] = ( R1 + R2 )C1 ln 2 = 0.693( R1 + R2 )C1
(9)
112
Z. Liu, S. Xu, and J. Liu
VCC C3 100pF
8
R1 4
5K
7 R2
6
1.5K
NE555
3
2
C1 100pF
1
5
R 20K
C2 100pF
UCM40T
Fig. 3. The circuit for ultrasonic transmitting
The T2 of low-level outputs charging time is determined by the formula,
t 2 = 0.693R 2 C1
(10)
The cycle time of rectangular wave is determined by the formula,
T = t1 + t2 = 0.693( R1 + 2 R2 )C1
(11)
The oscillation frequency f ,
f = 1 / T ≈ 1.44 /( R1 + 2 R2 )C1
(12)
The dutyfactor D of output pulse is determined by the formula,
D = ( R1 + R2 ) /( R1 + 2 R2 )
(13)
The output is a square wave signals when the R2 is much greater than the R1 and the duty factor D equals to 50%. The oscillation frequency f is up to R1 and R2 to determine. Center-frequencys of the ultrasonic transmitter are respectively 40 kHz, 10 kHz. When the center-frequency is 40kHz, the C1 is 0.1μF, and the resistances of R1 and R2 are respectively 0.6k Ω and 1.5k Ω . When the center-frequency is 10kHz, the C1 is 0.1μF, Resistances of R1 and R2 are respectively 1k Ω and 72k Ω . 3.2 Design of Receiving Circuit As show in the Fig.4., the circuit for ultrasonic receiving, that in which signal amplification, filtering, DC signal eliminating are included, has function of signal processing, avoiding the shortcomings of feeble signal and noise. The 2 stage signal amplification
Application of Interferometry in Ultrasonic System for Vibration
113
system was adopted in this experiment. The signal that is processed is deduced to PC computer. The signal can be shown by the software system. The circuit for ultrasonic receiving can be used in combination with the circuit for ultrasonic transmitting. The circuit is composed of four parts: the amplifying circuits A1 and A2, the rectifying circuit R3, the filter circuit RC, and the comparator circuit A4. The filter circuit has frequency selectivity that it amplifies 40 kHz signal and the yield value equals to 5. Multistage filter circuit was adopted. The half-wave rectifying circuit consists of operational amplifiers. The cut-off frequency of the filter circuit is 40 kHz. The comparator circuit also consists of operational amplifiers.
4 Testing The system for experiment was composed of three parts: the hardware of the system, the interface circuit and the software of the system. +15
C D 5 C
R
6
R
R
R out2
+
R
5
7 R
C
R PR
6
out2 +
5
7 R
6
out2
5
7
+
D C+
6
out2
7
OUT
+
R PR GND
Fig. 4. The circuit for ultrasonic receiving
The experimental facility included a flexible manipulator, a vibration exciter, two ultrasonic transducers, various circuits and a data acquisition card. The vibration source consists of the flexible manipulator that was driven by vibration exciter. The vibration exciter was adjusted among 20 to 20 kHz.The date was gathered by the multifunction data acquisition card which is made in the NI Company. The system of software can realize the real-time signal ongoing acquisition and storage. The change of phase is transformed into the visualization function graphic of D(t). The software extracted effectively characteristic value of the vibration signal.4,5 The frequency of the excitation signal in the experiment was selected at 71.2Hz. The irregular original signal was provided in Fig.5. In addition, Fig.6. gives the changing law of phase-shifts. After the collection and analysis of the phase-shifts change, completed vibration signal will be obtained, as well as the characteristic value. However, when the amplitude of the flexible manipulator is too large, the changing signal of phase shift signals confusion and cannot be measured then, as show in Fig.7.
114
Z. Liu, S. Xu, and J. Liu
Fig. 5. Original signal oscillogram
Fig. 6. Phase-shifts D(t) oscillogram
Fig. 7. Phase-shifts D(t) oscillogram when the amplitude is too large
5 Conclusion With the new system, which is based on vibration detection with ultrasonic wave, designed in this paper, remote objects can be detected on the vibration, overcoming the limitations that contact measurement has, such as piezoelectric sensors etc. The new system can also be applied to low frequency vibration measurements, with a much wider measurement range than the eddy current sensor. Here, no accessory is needed to be mounted on the vibrating object, avoiding the influence of vibration objects. The measurement technique based on the ultrasonic wave is a supplement for non-contact vibration measurements. It has many advantages such as low cost, convenient operation, high test speed, and so on. So that this measurement technique has much potential applicative value for condition monitoring of some industrial machinery equipments, especially for application in vibration measurement of fine structure or in some special working environment, such as high temperature, high pressure, dust, strong corruption,
Application of Interferometry in Ultrasonic System for Vibration
115
non-Contact and so on. Certainly, there are some shortcomings of this measurement technique. Great restrictions are still existed for the amplitude of vibration, not being solved completely. Continue study and further improvement are still needed in the future.
References 1. Meng-Hsiang, Y., Huang, K.N., Huang, C.F.: A Highly Accurate Ultrasonic Measurement System For Tremor Using Binary Ampltude-Shift-Keying And Phase-Shift Method. Biomedical Engineering-Applications 15(2), 15–21 (2003) 2. Ngoi, B.K.A., Venkatakrishnan, K.: An AcoustoOptic Vibrometer for Measurement of Vibration inUltraPrecision Machine Tools. Int. J. Adv. Manuf. Technol. 16, 830–834 (2000) 3. Soon, W.H., Ho, C.L., Yoon, Y.K.: Non-contact Damage Detection of a Rotating Shaft Using the Magnetostrictive Effect. Journal of Nondestructive Evaluation 22(4), 141–150 (2003) 4. Fernando, F., Enrique, B.: Member, An Ultrasonic Ranging System for Structural Vibration Measurements. IEEE Transactions on Instrumentation and Measurement 40(4), 1991–1997 (1991) 5. Matar, O.B., Remenieras, J.P., Bruneel, C., Roncin, A., Patat, F.: Noncontact Measurement of Vibration Using Airborne Ultrasound. IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control 45(3), 626–634 (1998)
Automatic Control System for Highway Tunnel Lighting Shijuan Fan, Chao Yang, and Zhiwei Wang School of Mechatronic Engineering, East China Jiaotong University, Nanchang, P.R. China
Abstract. To solve the problems such as driving safety and energy consumption existing in tunnel lighting control system, an automatic control system of tunnel lighting is designed based on stepless control method. The tunnel lighting control model is established based on “Specifications for Design of Ventilation and Lighting of Highway Tunnel (China)”. Simulation experiment of tunnel lighting control based on the established stepless control model is completed with Matlab. Compared with theoretical luminance data, simulation results show that the automatic control system can meet the luminance requirements of actual tunnel lighting, the error can be controlled less than 2%. Compared with HPS (high pressure sodium) lamps and LED (light-emitting diode) lamps with the consideration of maximum lighting value, the stepless controlled LED lamps can save more than 80% and 35% energy than HPS lamps and LED lamps respectively, and can save more than 20 % energy than 4-steps controlled LED lamps. Keywords: tunnel lighting, automatic control system, stepless control method, continuous light tuning, energy-saving.
1 Introduction China social economy and traffic undertaking are developing rapidly, tunnel traffic is becoming more and more necessary and important in mountainous areas of China, but the operating cost of tunnel traffic is huge, how to improve traffic safety performance and reduce operating cost of the tunnel traffic has become a focus issue that the China's transport department concerned. Tunnel lighting is an indispensable part to ensure driving safety and normal operation in tunnel traffic, and also is a key factor to reduce tunnel interior energy consumption [1]. Therefore, the corresponding design specifications about tunnel lighting are issued in various countries, such as CIE (Commission International d'Eclairage), BS (Britain lighting standards) and IES (Illuminating Engineering Society of North America), etc[2]-[4]. To establish a safe, economical and energy-saving tunnel lighting system has important significance for sustainable development of China’s highway engineering. Design quality of tunnel lighting control system determines whether tunnel lighting design is excellent or not. The existing control methods include manual control, sequential control and automatic control methods. Manual and sequential control methods are easy to implement and more stable and reliable in practice, but the tunnel interior luminance can be not adjusted along with the changes of weather, traffic volume and vehicle speed, the both methods hardly have any energy-saving effect, as a result, much electric energy is wasted [5]-[6]. D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part IV, IFIP AICT 347, pp. 116–123, 2011. © IFIP International Federation for Information Processing 2011
Automatic Control System for Highway Tunnel Lighting
117
Automatic control methods are widely used in modern tunnel lighting system, which can be divided into hierarchically control method and stepless control method according to adjustment continuity of tunnel light. At present, hierarchical control methods are often used in tunnel lighting control system in China before the application of LED lamps in tunnel lighting, but some problems exist in the methods, the main problems are: 1) the automatic control levels can be done are only from 2 to 3 because of the limited routing circuit, the parameters of environmental luminance, traffic volume and vehicle speed are only considered with maximum values at design stage, as a result, the lighting luminance of each section in tunnel is always in maximum state, lighting efficiency is evidently low, electric energy consumption is great; 2) the contradictions with driving safety and tunnel monitor arise during the operation course [7]-[8]. In the paper, an automatic control system for tunnel lighting is designed based on the characteristics of LED lamps, especially the characteristics of control easiness compared with other lamps, which made the luminance in tunnel be adjusted dynamically along with the changes of environmental luminance, traffic volume and vehicle speed, and thus continuous tuning of tunnel lighting is achieved. The control system not only ensures operation safety of the tunnel, but also realizes energy-saving.
2 Control System Structure In order to meet the demands of tunnel lighting and energy-saving better, stepless control method is adopted in the tunnel lighting control system. The control system is composed of vehicle detectors, luminance detectors, data converters, lighting control computer, dimming controllers and LED lamps. The structure block diagram is shown in Fig. 1. Environmental luminance, traffic volume and vehicle speed information are collected by vehicle detectors and luminance detectors. The data converted by data converters from collected information are sent to lighting control computer installed in tunnel control room. According to the predetermined dimming logic in lighting control computer, the luminance of each section in tunnel is calculated, the required
Vehicle detector
Data Converter Lighting Control Computer
Luminance detector
Data Converter
LED
LED
Dimming Controller
Dimming Controller
Data Converter
Fig. 1. Block diagram of tunnel lighting auto-control
118
S. Fan, C. Yang, and Z. Wang
dimming values for dimming controllers of each node are calculated in the control computer in accordance with the preceding luminance calculated and power curve of LED lamps. The control computer sent control commands to dimming controllers, LED lamps power are adjusted by the commands received, and fed relevant information back to lighting control computer, and continuous light tuning of tunnel lighting is achieved.
3 Tunnel Lighting Control Strategy The control strategy of tunnel lighting system is designed based on “Specifications for Design of Ventilation and Lighting of Highway Tunnel” [9] (hereinafter referred to as the “Design Specifications”), the principle is that the mathematical models of each section in tunnel are established based on the adaptation curve of the luminance in tunnel (as shown in Fig. 2), tunnel exterior environmental luminance, traffic flow and vehicle speed information. In accordance with the established mathematical models, the dynamic dimming control of LED lamps is conducted, and the luminance in tunnel is very close to the adaptation curve, energy-saving is achieved with optimal control effect. Stepless control is not absolute continuous dimming, but a more refined hierarchical control method. Environmental parameters with maximum values are generally considered in hierarchical control at tunnel lighting design stage, the effect of the changes of environmental parameters is neglected. Sometimes automatic control is achieved whose control levels are only from 2 to 3 because of the limited routing circuit. Stepless control is closer to continuous dimming by adopting more refined levels [10]. When planning tunnel lighting, five sections have to be considered: access zone, entrance zone, transition zone, interior zone and exit zone. There are different lighting requirements for different zones. In order to meet requirements of human eyes adaptation to luminance, logarithm dimming method for LED lamps is adopted: single-lamp
/6 /$
adaptation curve
Luminance cd/m 2
/WK
/WU G
/WU /WU
6
$ Access zone
3
/LQ
(
P
Entrance zone Transition zone Interior zone
Fig. 2. Theory demand curve of tunnel lighting
Automatic Control System for Highway Tunnel Lighting
119
256-grade dimming, so that the whole tunnel lighting looks like a linear regulator dimming. Meanwhile, the threat of driving safety due to abrupt luminance change can be avoided by gradual control method. Triggering light dimming too often is not only conducive to the human eyes adaptability, and to some extent, service time of lamps is also reduced. Therefore time-triggered model is resorted during the control process: information of environmental luminance, traffic flow and vehicle speed is collected and required luminance is calculated at interval of 3~5 minutes. In this paper, the mathematical models of the luminance of each section in tunnel are established on the data from “Design Specification” by a linear regression. 1) The mathematical model of the entrance zone: one-rank linear regression is adopted in MATLAB to make statistics analysis of the data from “Design Specifications” and fitted regression equations of luminance discount coefficient of the entrance zone are obtained. The luminance discount coefficient of the entrance zone under different traffic volumes and vehicle speeds is calculated by formula (1).
⎧0.0004v − 0.0085 Q ≤ 700 ⎪ 0.54v + 0.0002Q(v − 1.1) − 12.91 ⎪ k=⎨ 1700 ⎪ ⎪⎩0.0006v − 0.0107 Q ≥ 2400
700 < Q < 2400 .
(1)
where v is vehicle speed, Q is traffic volume. The luminance of entrance zone is calculated by the formula (2).
Lth = k ⋅ L20 ( S ) .
(2)
where Lth is the entrance zone luminance (cd/m2); L20(S) is the tunnel exterior environmental luminance (cd/m2). The entrance zone luminance under different traffic volumes and vehicle speeds is calculated by formula (3). ⎧(0.0004v − 0.0085) × L20 ( S ) Q ≤ 700 ⎪ ⎪ 0.54v + 0.0002Q (v − 1.1) − 12.91 × L20 ( S ) Lth = ⎨ 1700 ⎪ ⎪⎩(0.0006v − 0.0107) × L20 ( S ) Q ≥ 2400
700 < Q < 2400 .
(3)
2) The mathematical model of the interior zone: the interior zone luminance is relevant to traffic volume and vehicle speed, the tunnel exterior environmental luminance has no effect on it. In order to reduce the luminance calculation error, the second-order linear regression is adopted to fit the interior zone luminance values. The interior zone luminance under different traffic volumes and vehicle speeds is calculated by formula (4). ⎧0.0013v 2 − 0.135v + 4.95 Q ≤ 700 ⎪ 2 2 ⎪158v + 0.09v Q − 19534v − 4.88vQ + 4.75Q + 808250 Lin = ⎨ 700 < Q < 2400 . 170000 ⎪ ⎪0.0022v 2 − 0.1838v + 5.425 Q ≥ 2400 ⎩
(4)
120
S. Fan, C. Yang, and Z. Wang
3) The transition zone luminance depends on the entrance luminance. The transition zone is composed of three sections: Ltr1, Ltr2, Ltr3, and the corresponding luminance of each section is calculated by Ltr1=0.3Lth, Ltr2=0.1Lth, Ltr3=0.035Lth, respectively, according to the “Design Specifications”. 4) The exit luminance is 5 times of the interior zone luminance, Lout=5Lin.
4 Main Control Program Flow Chart The flow chart of the main tunnel lighting control program is shown in Fig. 3, the steps are: (1) Initializing sub-modules of the system, reading luminance information
Begin
Initialize Modules
Read Data
Manual Control Enable˛
d 0 ⎪ 2 ⎪⎪ a = ⎪ ⎨ ⎨ y ⎪ ⎪⎪ x2 + , y ≤ d 0 h ⎪ ⎩ ⎪ ⎧ r ⋅ sign (a ), a > d ⎪ fhan = − ⎪ ⎨ a ⎪ ⎪⎩ r d , a ≤ d ⎪⎩
(3)
2.2 Extended-State Observer (ESO)
⎧e = z1 − y ⎪ z = z + h ⋅ ( z − β e) ⎪1 1 2 01 ⎨ ⎪ z2 = z2 + h ⋅ ( z3 − β 02 ⋅ fal ( e, 0.5, δ ) + b0 u ) ⎪⎩ z3 = z3 + h ⋅ ( − β 03 ⋅ fal (e, 0.25, δ ))
(4)
Where, h is the control cycle. 2.3 Output of Nonlinear Feedback (NF)
⎧e1 = v1 − z1 , e2 = v2 − z 2 , ⎨ ⎩u0 = β1 ⋅ fal (e1 , α1 , δ ) + β 2 ⋅ fal (e2 , α 2 , δ ) ⎧⎪ ε α ⋅ sgn(ε ), ε > δ
fal (ε , α , δ ) = ⎨
⎪⎩ε / δ 1−α , ε ≤ δ
(5)
(6)
Where, the sgn(ε ) is the symbolic function. And the other parameters can be found in [4-5].
Parameters Turning of the Active-Disturbance Rejection Controller
263
3 ADRC Optimized by RBF Neural Network The structure of ADRC based on RBF neural network control system is shown in Figure 2.
Fig. 2. Schematic of ADRC based on RBF Neural Network
3.1 RBF Neural Network
The RBF neural network is presented by J.Moody and C.Darken in the 1980s which is a feed-forward network with three layers [6-7]. It possesses the capability of local adjustment and can approximate any continuous function in any accuracy. The structure of a typical RBF neural network is shown as Figure 3.
Fig. 3. Schematic of a RBF neural network
264
B. Liu and Y. Gao
Radial vector of RBF network adopts multivariate Gau-ssian function as radial basis function in this paper.
h j = exp(−
Xi − Cj 2b 2j
) j = 1, 2, … m
(7)
Where h j is the output of hidden unit; b j is the width of hidden unit; C j is the center of hidden units; X i is the output vector; ⋅ is Euclid norm. For an input pattern X i , the output of single output networks are given by m
yout (k ) = ∑ ω j h j
(8)
j =1
Where
ωj
is the weight between the
jth node in the hidden layer and the output unit;
m is the number of hidden unit. 3.2 Control of RBF Neural Network 1) The input and output of controller. RBF neural network can get more information from speed regulating system to adapt the chances of condition with effective measures. The structure of neural network is determined by genetic algorithm. The vector of input is:
X (k ) = [r ( k ), e(k ), e(k ) − e(k −1) / T , ∑ e(k )]T Where r (k ) is the given input;
(9)
∑ e(k ) is the sum of error.
2) The parameters of neural network Array of parameter denotes all the parameters of the hidden unit connected with output layer in RBF network. Applications of RBF network have much difficulty in determining the array and the number of the hidden unit. The selection of hidden unit has a very huge influence on mapping ability and effectiveness of network. If there are too few hidden unit, the network can’t complete the mission of classification and mapping function; if there are too many hidden units, influence generalization ability and learning efficiency. In order to improve the performance of whole system, it is necessary to seek more suitable learning algorithm of RBF network to determine array of parameter; and the number of hidden unit. 3) Target function of neural network J=
1 ( y (k ) − yout (k )) 2 2
Where y (k ) is the output of the target; yout (k ) is the output of the network;
(10)
Parameters Turning of the Active-Disturbance Rejection Controller
265
4) Self-turning of the parameters in NF Use the gradient descent method to turn the parameters in NLESF: ∂y xc (1) ∂u ∂y Δβ 2 = η e( k ) xc (2) ∂u ∂y Δβ3 = η e( k ) xc(3) ∂u Δβ1 = η e(k )
η
(11)
is the learning rate.
4 Simulation and Analysis In this paper, the simulation model adopts a nonlinear discrete system, whose transfer function is described by rin( k ) = 0.5* sign(sin(0.2π * t )) + (12) 0.25* sign(sin(0.2π * t − π / 2)) Suppose input signal is rin(k ) = 0.5* sign(sin(0.2π * t )) + 0.25* sign(sin(0.2π * t − π / 2))
(13)
We adopt proposed algorithm and controller in the simulation. The outputs of ADRC controller based on RBF identification are shown in Figure 4 . The simulation results are shown the outputs of the identification network can match the output of the closed-loop controlled plant well. The adaptive turning of NF in ADRC controller parameters are shown in Figure 5. At the same time Jacobian information of identification is shown in Figure 6 in simulation. From Figure 7, we can see that the system output trace the reference information well while using ADRC based on BP neural network. And in Figure 8, we can clearly see that the method of PID control cannot trace the reference signal as well as the previous method.
Fig. 4. Output of the target and the network
266
B. Liu and Y. Gao
Fig. 5. Jacobian information of identification
Fig. 6. Adaptive turning curve of ADRC parameters
Fig. 7. Response of ADRC based on RBF Neural Network
Parameters Turning of the Active-Disturbance Rejection Controller
267
Fig. 8. Resonse of PID
5 Conclusion Conventional PID controller can hardly work well at different operating condition. A novel controlling method-ADRC based on Radial Basis Function (RBF) Neural Network (NN) is presented in this paper. The controller has advantages of both selflearning capability of neural network and simplicity of ADRC. During the practice, ADRC based on RBF neural network has the superiority such as strong robustness, simple theory. Simulation result shows that the proposed controller has adaptability, strong robustness and satisfactory control performance in the nonlinear and time variable system.
References [1] Suresh Kumar, A., Subba Rao, M., Babu, Y.S.K.: Model reference linear adaptive control of DC motor using fuzzy controller. In: IEEE Region 10th Conference on TENCON 20082008, TENCON 2008, November 19-21, pp. 1–5 (2008) [2] Buchnik, Y., Rabinovici, R.: Speed and position estimation of brushless DC motor in very low speeds. In: Proceedings of 23rd IEEE Convention on Electrical and Electronics Engineers in Israel 2004, pp.317–230 (2004) [3] Pan, Y., Furuta, K.: Variable Structure Control with Sliding Sector for Hybrid Systems. In: International Workshop on Variable Structure Systems, VSS 2006, June 5-7, pp. 286–291 (2006) [4] Ji, Z.-c.,Shen, Y.-x., Xue, H.: Study on the adaptive fuzzy control for brushless DC motor. In: Proceedings of the CSEE 2005, vol. 25(5), pp.104–109 (2005) [5] Ren, y., Zhou, L.-m.: PMSM Control Research Based on Particle Swarm Optimization BP Neural Network. In: International Conference on Cyberworlds 2008, September 22-24, pp. 832–836 (2008) [6] Han, J.: From PID to Active Disturbance Rejection Control. IEEE Transactions on Industrial Electronics 56(3), 900–906 (2009)
Research of Intelligent Gas Detecting System for Coal Mine Chen Hui School of Mechanical and Electrical Engineering, East China Jiaotong University, 330013, Nanchang, China Tel.: 0791-7046122
[email protected] Abstract. According to statistic data of China in recent years, gas explosion accounted for above 70% in all coal mine accidents. Frequent gas explosion accidents have caused great losses of lives and property. Therefore gas detection and monitoring system is needed to serve as a safety device in coal production. In this paper, an intelligent gas detecting system is designed. This detection instrument adopts SCM AT89S52 as its control hardcore and uses catalytic combustion type gas sensor element MC112 as the sensor for gas (CH4) detecting. The main functions of this system are as follows: monitoring the real-time concentration of CH4 and displaying the concentration value; emitting sound and light alarm signals if the CH4 concentration value is beyond the alarm value inputted by keyboard panel; using serial communication port to transmit data to the host computer above ground. The software debugging and hardware simulating of the system above are also implemented at the same time. Keywords: Data collection, Sensor, Coal mine, Serial communication, SCM.
1 Introduction As the most important source of energy in China, Coal consumption is about 70% of all the energy consumption [1,2]. However, coal mine accident happened frequently in China, lots of people suffered from these disasters. Among all the accident, gas leakage lead to gas explosion is the main reason of these accidents [3,4]. So, it’s significantly important to develop gas monitoring safety system in coal production enterprises. In this paper, a gas detection and monitoring system is presented, its main functions including Real-time monitoring gas concentration, transmitting dynamic safety operation parameters underground coal mine, automatic warning about danger before accident happens and providing useful information on rescuing and evacuating people or equipment to decision makers.
:
2 Overall Design of Gas Detecting System Gas detecting system should meet some specific needs: it can monitoring combustible gas concentration underground coal mine, warning against over standard gas concentration and transmitting real time data to host computer above ground [5, 6]. The gas D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part IV, IFIP AICT 347, pp. 268–278, 2011. © IFIP International Federation for Information Processing 2011
Research of Intelligent Gas Detecting System for Coal Mine
269
detecting system in this paper adopts single-chip microcomputer as control computer; the overall schematic diagram of system is shown in Figure 1. The reason for selecting single-chip microcomputer as a control core is that it possesses advantages of small size, high reliability, low price which made it very suitable to be used in industries of intelligent instrument and real time control field [7].
Fig. 1. Schematic diagram of gas detection system
The operating interface of system is shown in Figure 2. Number at upper right corner shows the default or user-defined gas concentration value, number at upper left corner shows detected gas concentration value. One alarm lamp is equipped. All functions are controlled by keys arranged on the control panel, including POWER key, RESET key, DATA COLLECTION key. Other keys including ten number keys, ADJUST VALUE key and ENTER key are used to change threshold values.
Fig. 2. Operation interface diagram of system
270
C. Hui
Basic operating procedures are as follows: Firstly press POWER key, the system initialized. Press DATA COLLECTION key, LED at upper right corner displays the threshold value 1.00; User can customized threshold value by press ADJUST VALUE key and ten number keys, then press ENTER to confirm the change. System starts to detect gas concentration and display these parameters on upper left LED area, at meantime transmit real-time data by RS-485 to host computer above ground.
3 Hardware System Design of Gas Detection System The Hardware architecture of system mainly including main control unit, sensors and signal amplifier circuit, A/D converter module, sound-light alarming circuit, keyboard and display module, serial-communication module. 3.1 Main Control Unit Featured by high integration level, small size and low prices, Single chip microcomputer has been widely used in a broad range of industrial applications including process controlling, data collection, electromechanical integration, intelligent instrument, household appliances and network technology, and significantly improved the degree of technology and automation. Two factors are taken into account here in chip selecting, first one is anti-interference ability, the poor working conditions and complex operating situations in mine tunnel increase the interferences in SCM application systems, so the SCM must have high
Fig. 3. Main control unit
Research of Intelligent Gas Detecting System for Coal Mine
271
resistance to outside interference; second one is the performance-price ratio of the SCM. Considering the aforementioned factors, we adopted the AT89S52 developed by ATMEL as main control unit, and the final scheme of main control circuit consists of AT89S52, timer and reset circuit [8,9]. As shown in figure 3. 3.2 Sensors and Signal Amplifying Circuit 3.2.1 Sensor Selection A crucial issue in gas detecting system design is how to select gas sensors. Common gas sensors are metal oxide semiconductor such as tin oxide, zin coxide, titanium oxide and aluminum oxide. Problems encountered with these sensors are lack of flexibility, poor response times and operated at elevated temperature [10]. A new method of ch4 detecting based on infrared techniques was presented in recent years, but it is still in progress and much work should be done before it can be applied to solve the practice problems [11].
Fig. 4. Outside view and internal circuit of MC112
This system adopted catalytic combustion type gas sensor MC112 developed by SUNSTAR group to measure the gas (ch4) concentration. Figure 4 shows the outside view and internal circuit of MC112, table 1 lists the main technology parameters of MC112. MC112 gas detector exploits catalytic combustion principle; the two-arm bridge is comprised of measure and compensate components pairs. When it is exposed to combustible gases, measure components resistance RS increased and transmit output voltage parameter through measuring bridge, the voltage parameter is directly proportional to the gas concentration value. The compensate component works as temperature compensation and reference. Main features of MC112 include good repeatability, work stably, reliability, linear output voltage, and quick response. The mine safety rules stated that if methane gas concentration exceeds 1%, safety system should make an alert, if gas concentration exceeds 2%, all people must evacuate immediately. Since the detecting range of MC112 for low concentration methane is 0%-2%, it is suitable for measuring low concentration methane in the coal mine.
272
C. Hui
Table 1. Main technical parameters of MC112
Rated voltage (V)
3.0
Rated current (mA)
@100
1%CH4 1%Butane 1%Hydrogen Linearity (%)
>14 >30 >24 0~5
Response time (90%)
Less than 10 sec.
Renewal time (90%) Application environment Storage environment Size
Less than 30 sec. -20℃~+60℃ π 2⎠ 2 ⎝ ⎩⎪
Amplification factor
k =
ω
(8)
M:
M = N (cos kl4 −
α k
sin kl4 )
1 cos kl5
(9)
N is area factor and k is wave number. According to formula (7) and formula (9), the simulative curve is shown in Fig.14.
Fig. 14. The curve about kl4 and
M
Study on Imitating Grinding of Two-Dimensional Ultrasonic Vibration Turning System
341
kl4 = 2.8416 , M max = 10.72 , l4 = 0.115m , kl5 = 1.275 , l5 = 0.052m , kl5 < π 2 , kx0 = π 2 − α 2 , x0 = 0.052m . Nodal plane places l5 = 0.052m from the big end of the cone.
6 Two-Dimensional Ultrasonic Vibration System Dynamics Simulation Piezoelectric ceramic uses Solid5 in ANSYS element library to establish solid model [9]. Solid5 is used in thermal, magnetic, acoustic, piezoelectric and analysis of three-dimensional coupled field. This unit has eight nodes, and each node has six degrees of freedom. As for the piezoelectric coupling analysis, Solid5 has large deformability. For others used Solid95. Its unit has 20 nodes and each unit also has X,Y,Z freedom. Solid95 is higher than Solid45, and it doesn’t reduce the accuracy of the case. Piezoelectric ceramic disks are made of PZT-8, and they are thickness polarization. The polarized piezoelectric ceramic is anisotropic, so it is different from isotropic materials. The dielectric constant matrix, the piezoelectric constant matrix and elasticity matrix are given as follows. Dielectric constant matrix
[ε ] F m :
0 ⎤ ⎡904 0 ⎢ [ε ] = ⎢0 904 0 ⎥⎥ ⎢⎣0 0 576 ⎥⎦
Piezoelectric constant matrix
[ e]
c
m3
(10)
:
0 −4 ⎤ ⎡ 0 ⎢ 0 0 −4 ⎥⎥ ⎢ ⎢ 0 0 13.4 ⎥ [ e] = ⎢ ⎥ 0 0 ⎥ ⎢ 0 ⎢ 0 10.3 0 ⎥ ⎢ ⎥ 0 ⎦ ⎣10.3 0
(11)
342
L. Liu, W. Zhao, and Y. Ma
Elasticity matrix
[C ] N m 2 : ⎡13.7 6.8 7.3 0 0 0 ⎤ ⎢ 0 13.7 7.3 0 0 0 ⎥ ⎢ ⎥ ⎢ 0 0 12.1 0 0 0 ⎥ 10 [C] = ⎢ ⎥×10 ⎢ 0 0 0 3.1 0 0 ⎥ ⎢ 0 0 0 0 3.1 0 ⎥ ⎢ ⎥ ⎣ 0 0 0 0 0 3.4⎦
(12)
The horn is a axial symmetry revolving body, and uses bottom-up model. The points generate lines, and lines generate a axial plane. The model of the two-dimensional ultrasonic vibration system is obtained by the axial plane rotated 360°around the axis. Free mesh is used to generate finite element mesh, which is shown in Fig.15.
Fig. 15. Finite element mesh
Mode extraction method is subspace iteration method, and the modal number is 10. In modal analysis, side surface of all nodes between piezoelectric ceramic and front shroud constrain Y displacement, as shown in Fig.16. The electrode voltage (V = 0) is coupled on upper and lower surface, which is shown in Fig.17.
Fig. 16. Constrained Y displacement
Study on Imitating Grinding of Two-Dimensional Ultrasonic Vibration Turning System
343
Fig. 17. Coupled the electrode voltage
Through the modal operation, the free resonance frequency is 22255Hz in the short-circuit state. This is close to the theoretical calculation of 20kHz. By modal expansion analysis, The relative displacement nephogram is shown in Fig.18.
Fig. 18. Relative displacement nephogram
7 Conclusion Traditional ultrasonic vibration turning mechanism and the pros and cons of the processing are analysed in this paper. On this basis, the imitating grinding of twodimensional ultrasonic vibration turning is proposed. The structure of two-dimensional ultrasonic vibration transducer is different from the structure of traditional longitudinal ultrasonic transducer. Necessary grinding track and honing track can be obtained by changing the voltages that apply to piezoelectric ceramic. Meanwhile, the design methods of transducer and horn are given, and which realize the transducer and horn modal analysis by using finite element software. The research improves the machining precision, and makes the application of this technology possibly in ultraprecision machining.
344
L. Liu, W. Zhao, and Y. Ma
Acknowledgement. The authors gratefully acknowledge the financial support provided by the University-Industry Cooperation Project of Jiangxi Education Department (Grant No. GJJ09005).
References 1. Cao, F.: Ultrasonic Machining Technology. Chemical Industry Press, Beijing (2005) 2. Yuan, J., Li, X., Chang, D.: Press Ultrasonic Elliptical Vibration Cutting. Aeronautical Manufacturing Technology 4, 92–93 (2005) 3. Shaomoto, E., Moriwaki, T.: Study on Elliptical Vibration Cutting. Manufacturing Technology 43(1), 35–36 (1994) 4. Lu, Z., Yang, L.: Analysis and Simulation of Mechanism in Ultrasonic-Vibration Cutting Based on Dynamic Fracture Mechanics. Journal of Harbin Institute of Technology 40(9), 1400–1401 (2008) 5. Wu, H., Cao, L., Liu, J.: Analysis of Kinematic Geometry on Face Grinding Process on Lapping Machines. Mechanical Engineering 38(6), 144–146 (2002) 6. Chang, Q., Liu, Y., Ye, S.: Computer Simulation Technology for Precision Lapping Process of Workpiece. Computer Simulation 24(5), 249–250 (2007) 7. Tian, Y.B., Zhou, L., Shimizu, J.: Elimination of Surface Scratch/Texture on The Surface of Single Crystal Si Substrate in Chemo-mechanical Grinding (cmg) Process. Applied Surface Science 255(7), 4205–4211 (2009) 8. Lin, S.: Ultrasonic Transducer Principle and Design. Science Press, Beijing (2004) 9. Qi, J., Li, Z., Zhao, C., Huang, W.: Linear Ultrasonic Motor with Two Degree of Freedom Using Longitudinal and Bending Vibration Modes. Maxon Motor. 12, 18–19 (2005)
Study on Optimal Path Changing Tools in CNC Turret Typing Machine Based on Genetic Algorithm Min Liu1, XiaoLing Ding1,*, YinFa Yan1, and Xin Ci2 1
College of Mechnical and Electronic Engineering, Shandong Agricultural University, DaiZong Street NO.61, Tai’an City, China
[email protected],
[email protected] 2 Tai'an Falcon CNC Machine Co., Ltd., Tai’an City, China
Abstract. This paper is aimed to find the optimum path of CNC turret typing system to reduce the changing tools times and optimize tool movement routes to make up for the deficiency of CNC Turret Typing machine production efficiency. An uncertainty polynomial model is raised based on the asymmetric traveling salesman problem. And genetic algorithm (GA) is used to solve the path optimization problem. The optimization of path can minimize the moving tools times. Furthermore, the optimization problem is simplified to shortest distance between points. Fitness function, selection operator, crossover operator, mutation operator and other genetic operations are studied in this paper. In addition, the greedy crossover operator, the elite preservation strategy and the self-adaption strategy are imported in GA, which enhance the ability of finding the optimum and speed the efficiency. Finally, MATLAB simulation testifies that the algorithm is valid. The experiment result shows that the GA can shorten processing time and can reduce the air travel effectively without changing the machine's hardware through reasonable arrangement of the changing and moving tools path. As a result, the efficiency and precision of CNC turret typing system was improved availably. Keywords: CNC turret, Genetic algorithm, Path optimization.
1 Introduction To meet the demands of tracking and controlling the product quality for steel structure producers and understanding the products information for users, steel structure information identification has been paid more and more attention. In our country, engraving, stylus printing and typing are three main ways of steel structure information identification. Among them, engraving mainly adopts NC milling cutter engraving, laser mark and EDM shaping technology. For NC milling cutter engraving, the words, fonts and stroke can be adjusted by NC. But there are still some problems, for example, huge investment, short tool life and high expense. Laser mark, this technology is widely used *
Corresponding author. Tel.: 0538-8249833.
D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part IV, IFIP AICT 347, pp. 345–354, 2011. © IFIP International Federation for Information Processing 2011
346
M. Liu et al.
in marking the small workpieces. However, when the word depth call for larger, the technology is very difficult to achieve and high cost [1]. EDM shaping is a mature and high precision technology, which demands greater coordinate travel, high efficiency and larger cost for the large steel structure [2]. Another stylus printing using dotmatrix marking method has higher printing speed and lower cost, but the lower print recognition rates and poorer visual effects, which is prone to cause mark misunderstand [3]. Steel structure typing, which make stamping characters more clear, is a kind of technique using ramjet power to impact the characters. Besides, the typing depth is easy to control, the tool is long life and low cost. At present, CNC turret typing is an advanced typing technology. The meachine with CNC turret typing can achieve a two-way rotary and nearby change tools anywhere [4]. The generation and optimization of the tools path in stamping process plays an important role. Efficient tool path optimization algorithm can reduce the tool traveling distance and can improve production efficiency. In the process of generating numerical code, the point is optimized by shortening the changing and moving path to reduce idle travel time, then can improve processing efficiency. Therefore, it is worth studied to select the reasonable processing path. The CNC turret typing system is characterized by high speed, high precision and wide moving range. The specific location and methods of the relative motion between tool and workpiece were described via tools path, including the effective stamping locus and the auxiliary locus [5]. The auxiliary locus which cost some processing time is mainly used for turret rotating positions and tool orientation, although they are not directly involved in shaping the workpiece. So optimize the auxiliary locus is needed. Due to the large auxiliary locus, the arrangement sequence is closely related with the processing path. It is very difficult to use the traditional method to search optimal path. In this paper, the path optimization based on the evolutionary algorithm is used to plan the auxiliary locus. On the basis of the reduction of the changing tools amount and optimization tools locus, The problem of optimal path changing tools is discussed and solved from another perspective. This assumption, not considering the machine fault and tool wear, permit us to believe only one same specification tool in CNC turret machine.
2 Basic Principle of Optimal Path Changing Tool and Modeling The changing tools time is longer than fast moving tools in CNC Processing, so it is very important to reduce the tools changing times to improve the processing speed. For the above reasons, the combinatorial optimization system is called for the least tools changing times firstly. The steps are as follows. Firstly, each tool changing one time must process the same characters completely without repeat and miss. Secondly, the tools change through turret nearby rotating to process others characters. How to arrange the processing paths to assure the least spare travel bring on a problem, which can be come down to the asymmetric traveling salesman problem (TSP) with additional constraint (that is, the least tool changing times). For example, a businessman wants to sell goods in n cities. How to choose a shortest path make the businessman go through the whole cities only once and return starting point [6]. TSP is a typical NP complete
Study on Optimal Path Changing Tools in CNC Turret Typing Machine
347
problem, which is easy to describe but difficult to handle. The classical and improved GA is an effective method for solving TSP so the method has been widely used. GA is a global search capability algorithm and is derived from evolutionism based on genetic selection and nature elimination of Darwin ism and genetics. The algorithm is leaded in selection, crossover and mutation, etc. Genetic processing is repeatly application in the population which contains all possible solutions. In the process, offspring groups adapt to the environment more than previous. The optimal individual can be viewed as the near optimal solution by decoding [7]. In CNC turret typing system, processing code is analyzed firstly. Then the paths after the tool changing once are as a path group. After changing the correspondence tool, the tool moves along the shortest air travel paths begaining with a given character. The next character is processed until all the same characters are processed, so circulates. Here, tools play a TSP role in fact, and character points play the role of cities. Furthermore, the goal is the shortest of tool’s air travel in processing.
3 Optimal Path Changing Tools Based on GA 3.1 About GA GA is a predominant all-round optimizing method in species evolution processin nature. GA makes question’s solution into a certain amount of initial chromosomes (initial solution) firstly. Secondly, the chromosomes are put into the problem environment. The well-adjusted chromosomes will be selected by the “survival of the fittest ” principle. Then the above chromosomes take the selection, crossover and mutation operation. Thus a new generation and more adapt to the environment of chromosomes will appear. So following evolution, the optimal solution will converge to an individual which most adapts to the environment finally. GA searches from one group to another and the method does not easily fall into the local optimal solution. In addition, the search process only depends on individual fitness function, so the method is very good robustness and extensive adaptability. 3.2 Overall Design of Optimal Path Changing Tool In this paper, GA is used to optimize the CNC turret changing tools path. Path optimization was simplified as the optimization from point to point by an encoding method based on path. The greedy crossover operator is designed, which has advantage of inheriting good chromosome from parent individuals and improving search capability. Furthermore, this algorithm uses norm geometry selection method and the elite preservation strategy to protect the outstanding individual structures of group, and the algorithm adopts the adaptive strategy to adjust the mutation probability dynamically, so the efficiency of the algorithm is improved greatly. The path optimization program of CNC turret typing system includes coding module, selection, crossover and mutation module, GA controls parameter module, initialization module, decoding module, fitness calculation module and so on. The overall design flowchart is shown in Fig.1.
348
M. Liu et al.
processing code analysis motion command tool path muster population initialization
mutation fitness evaluation
termination or not ?
crossover N
selection
Y
similarly most optimal solution new processing code Fig. 1. Flow diagram of the program
3.3 Realization of GA The working process of the GA mainly includes three types of operation, such as the chromosome coding and structure primitive group, the fitness calculation method and the selection, crossover, mutation operation method. The maximum number of generations termination criterion is adopted in GA and selected the viable path with the best fitness. 3.3.1 Chromosome Coding and Structure Primitive Groups The path coding form includes binary representation, adjacency representation, ordinal representation, path representation, matrix representation, edge representation and so on. GA mostly use particular gene representation composed of binary code, but for TSP scheduling problem, path representation is used usually. This is the nature and simplest method in path coding and it is a real number coding. Changing tools serve as the processing technology routing according to the combinatorial optimization goal. Tool is changed one time as a process group which is expressed as a chromosome group. Grefenstette path representation is adopted in coding chromosome which can ensure that any individual has practical significance [8]. The path representation describes as follows. The Nth changing tool consists one process group W. The processing sequence in W is assumed in formula (1). o = (t1 , t 2 , t 3 ,..., t n ) .
(1)
The rule is that once the process completed must move away from the group. The corresponding operation sequence of the Ith process t i is mi (1 ≤ mi ≤ n − i + 1) in all
Study on Optimal Path Changing Tools in CNC Turret Typing Machine
349
unprocessed groups W − {t1 , t2 ..., ti −1} . The operation sequence is as for the process group coding in the chromosome. So a list M of chromosome is got after dealing with all steps in formula (2). M = (m1 , m2 , m3 ,..., mn ) .
(2)
3.3.2 Fitness Function The fitness is the only criterion to judge whether the individual is good or not in GA. Optimal path changing tools in CNC turret typing system uses minimum tool moving locus as target function on the premise of the least changing tools times. Therefore, the corresponding fitness is shown in formula (3) for any chromosome T. F (t ) = 1 / ∑ i
∑
d ij u ij .
(3)
j
In the above equation, d ij shows the distance between the character i and j. u i j shows whether the tool travels from the point i to j. If "yes", its value is 1, "no" is zero. 3.3.3 Design the GA Operator GA operator includes selection, crossover and mutation operator. Reasonably choosing operator can prevent premature convergence and can accelerate the search speed. In the operation of GA, these operators make the group ceaseless evolving and move towards the optimal solution gradually. 1 . Selection operator. The operator is used to perform the operation of selecting the ƻ superior and eliminating the inferior for the individuals in group. Then the operator can determine how to select some individuals to inherite from the father generations to child. Due to the best chromosome not always appear in the last generation, the best chromosome is preserved at the beginning. Once the new group finds a better chromosome which is used to replace the original. After completion of the evolution, the last chromosome can be regarded as the optimization result. Therefore, this algorithm uses norm geometry selection and elite preservation strategy which can protect the excellent individual structure in evolutionary process. The selection mathematical model is described in formula (4). Suppose the number of individuals is n, the fitness value of one individual i is F(i) , so the probability of i selected is Psi . n
Psi = F i / ∑ F j . j =1
(4)
Ps i reflects the fitness value of individual i occupying the proportion in the sum of fitness at the whole group. 2 . Crossover operator. Due to use Grefenstette path coding, genotype individual ƻ from GA is able to correspond a practical processing sequence. So we can use usually single point or multi-points crossover operator which determines the global search ability of GA. Many kinds of crossover operators are combined use in the algorithm. In
350
M. Liu et al.
particular, GA is characterized by poor convergence, so the greedy crossover operator is introduced to accelerate convergence speed. 3 . Mutation operator. The role of mutation operator is to maintain the population ƻ diversity and improve algorithm local searching ability. The mutation operation adopted in this algorithm random produces a character number as the first processing character. In this way, the typing path is changed according to the above method. Then the phenomena of premature convergence can be prevented by adaptively adjusting the mutation probability. 3.4 Operation Parameter The operation parameters selected in GA include group size, individual coding length, crossover probability, mutation probability, biggest evolutionary algebra, etc., which influence the GA performance greatly. For the specific problem, whether the parameter setting is fitting or not, is judged by multi-Processing convergence condition and the quality of solutions.
4 Application and Sample If one tool of the CNC turret center has 31 points, now we should optimize the process coding to make the path shortest after processing the same characters under the condition of not changing tool. The parameters of optimization path algorithm are as following. The group size is 40, maximum evolutionary algebra is 500, crossover probability is 0.6, mutation probability is 0.05, self-adapted mutation parameter k=1, the elite number is 2. At last, using GA optimization toolbox (GAOT) to simulate the whole trajectory. GAOT in MATLAB is used to verify the GA efficiency. GAOT’s genetic operations is very flexible and provide reliable, extensible exploration platform for application and research GA. Program is as following, then the optimized path is shown in Fig.3. clear all global distMatrix t=[1300 2302;3659 1415;4107 2144;2712 1799;3688 1635;3826 1156;2238 1229;5196 1044;2312 790;3386 570;3007 2970;2562 3756;2688 1991;2361 1776;1332 2695;3715 2678;3918 2179;4061 2370;3780 2212;3676 2578;4029 2838;4263 2931;3429 1908;3507 2376;3394 2643;3439 3201;2935 3240;3140 3550;2545 2357;2778 2826;2370 2975]; sz=size(t,1);distMatrix=dists(t,t); xFns ='cyclicXover uniformXover partmapXover'; xOpts=[2;2;2]; mFns ='inversionMutation adjswapMutation shiftMutation swapMutation threeswapMutation'; mOpts=[2;2;2;2;2];termFns='maxGenTerm'; termOps=[500];selectFn='normGeomSelect'; selectOps=[0.08];evalFn='tspEval'; evalOps=[ ];
Study on Optimal Path Changing Tools in CNC Turret Typing Machine
351
bounds=[sz];gaOpts=[1e-6 1 1]; startPop=initializeoga(80,bounds,'tspEval',[1e-6 1]); [x endPop bestPop trace]=ga(bounds,evalFn,evalOps,startPop,gaOpts,termFns ,termOps,selectFn,selectOps,xFns,xOpts,mFns,mOpts); bestPop trace plot(trace(:,1),trace(:,2)); hold on plot(trace(:,1), trace(:,3)); figure(2) clf A=ones(sz,sz); A=xor(triu(A),tril(A));[xg yg]=gplot(A,t); clf h=gca;hold on ap=x; plot(t(x(1:sz),1),t(x(1:sz),2),'r-'); plot(t([x(1),x(sz)],1),t([x(1),x(sz)],2),'g-'); plot(xg, yg,'b.','MarkerSize',24);
Fig. 2. Tool path chart after using standard GA to optimize
Fig. 3. Tool path chart after using improved GA to optimize
352
M. Liu et al.
Fig. 4. Tool path optimum and average value after using GA to optimize
The simulation result is shown in Fig.3 and Fig.4. Fig.2 shows tool path after using standard GA to optimize, the total distance is 18572.231687. The tool path after using improved GA is shown in Fig.3, the total distance is 16116.820139. The air travel distance shorten 13.22% compared with the standard GA. Obviously, the import of the greedy crossover operator, the elite preservation strategy and the self-adaption strategy improved the standard GA very successfully in this paper. Furthermore, the ability of finding optimum is improved and the improved GA obtains the optimum at the thirtieth generation first time. From Fig.3 it can be seen that tool path is improved obviously after optimization. CNC turret path is optimized on the premise of minimum changing times, because changing tools will waste more time than fast moving. As a result, the redundant tool locus is reduced and the processing time is saved greatly. Optimal path changing tool in CNC turret typing machine based on GA obtained a good optimize effect.
5 Conclusions In this paper, we describe the practical problems in CNC turret typing system and use the improved GA to find optimal tool path. According to give priority to change tools at the least number of path optimization, the problem is converted to TSP problem with constraint condition firstly. Then the selection, crossover and mutation operator and fitness value are studied separately. A view of the simulation results can be obtained from here. This method used GA succeeded in optimizing the CNC turret typing path. These operations can effectively solve the tools air travel problem in the multi-point processing. It can not only shorten the processing time but also make up for the deficiency of production efficiency of CNC Turret Typing machine. It can be seen that this method is effective, feasible on the path optimization problems and has laid a foundation for further research. But, by virtue of introducing special constraint, genetic algorithm generality is restricted which is the main emphasis of the future research. Acknowledgement. The authors would like to acknowledge ShanDong Provincial Education Department Scicence & Technology Project for their financial support (J09LD22). The authors also would like to acknowledge Tai'an Falcon CNC Machine Co. for providing the excellent environment and the experiment facility.
Study on Optimal Path Changing Tools in CNC Turret Typing Machine
353
References 1. Fang, Q., Cai, D., Qian, B.: Marking Path Optimization Algorithm for Laser Marking and Application. Applied Laser 8, 290–293 (2009) 2. Xie, Z., Xiao, Y., Han, Z.: EDM Lettering Technology of Large-scale Workpiece’s Surface. Journal of Chongqing University 5, 14–17 (2003) 3. Chen, x., Xu, w.: Design of Print Head of Steel Plate Pin-printer and Analysis on the Priciple of Its Print Sign Forming. Metallurgical Equipment 4, 53–56 (2003) 4. Luo, H., Dong, H.: GA-based Approach to Optimize Cutters Arrangement on the Turret Post of Turning Center. Agricultural Machinery 2, 172–175 (2007) 5. Erkan, Ü., Mehmet, E.T., Selçuk Halkaci, H.: An artificial immune system approach to CNC tool path generation. J. Intell. Manuf. 20, 67–77 (2009) 6. Zhou, Z., Ding, T.: Research on Holes Machining Path Planning Optimization with TSP and GA. Design and Research 7, 30–32 (2007) 7. Chen, J.C., Zhong, T.X.: A hybrid-coded genetic algrithm based optimisation of non-productive paths in CN machining. The International Journal of Advanced Manufacturing Technology 20, 163–168 (2002) 8. Yu, W., Fu, J., Chen, Z.: Approach optimizing tool paths arrangement based on genetic algorithm. Journal of Zhejiang University (Engineering science) 40, 2117–2121 (2006) 9. Zhang, L., Wu, T.: Path optimization on laser drilling based on genetic algorithm. Mechanical & Electrical Engineering Magazine 24, 77–79 (2007) 10. Qu, J., Xiao, S.: PCB drilling path optimization based on ant-coIony aIgorithm. Mechanical & Electrical Engineering Magazine 24, 56–58 (2007) 11. Ma, Z.M., Huang, L.: The Optimization Algorithm of Holes Machining Path With Datum Hole. Machine Tool & Hydraulics 36, 28–32 (2008) 12. Xuan, D., Li, Z.: Component Placement Process Optimization for Chip Shooter Machine Based on Genetic Algorithm. Journal of XI’AN JiaoTong University 42, 295–299 (2008) 13. Ramli, R., Abu Qudeiri, J., Yamamoto, H.: Tool Path of Lathe Machine in Flexible Transfer Lines by using Genetic Algorithms. In: IESM 2007, Beijing-China, May 30-June 2 (2007) 14. Ugur, A.: Path planning on a cuboid using genetic algorithms. Information Sciences 178, 3275–3287 (2008) 15. Xing, L.-N., Chen, Y.-W.: A hybrid approach combining an improved genetic algorithm and optimization strategies for the asymmetric traveling salesman problem. Engineering Applications of Artificial Intelligence 21, 1370–1380 (2008) 16. Singh, A., Baghel, A.S.: A new grouping genetic algorithm approach to the multiple traveling salesperson problem. Soft Compute 13, 95–101 (2009) 17. Makhanova, S.S., Batanovb, D., Bohezb, E., Sonthipaumpoonc, K., Anotaipaiboona, W., Tabucanond, M.: On the tool-path optimization of a milling robot. Computers & Industrial Engineering 43, 455–472 (2002) 18. Lee, S.-G., Yang, S.-H.: CNC Tool-Path Planning for High-Speed High-Resolution Machining Using a New Tool-Path Calculation Algorithm. Int. J. Adv. Manuf. Technol. 20, 326–333 (2002) 19. Castagnetti, C., Duc, E., Ray, P.: The Domain of Admissible Orientation concept: A new method for five-axis tool path optimisation. Computer-Aided Design 40, 938–950 (2008) 20. Oysu, C., Bingul, Z.: Application of heuristic and hybrid-GASA algorithms to tool-path optimization problem for minimizing airtime during machining. Engineering Applications of Artificial Intelligence 22, 389–396 (2009)
354
M. Liu et al.
21. Zhuaa, L., Gang, Z., Han, D., Xiongb, Y.: Global optimization of tool path for five-axis flank milling with a conical cutter. Computer-Aided Design 42, 903–910 (2010) 22. He, W., Lei, M., Bin, H.: Iso-parametric CNC tool path optimization based on adaptive grid generation. Int. J. Adv. Manuf. Technol. 41, 538–548 (2008) 23. Moriwaki, T., Tangjitsitcharoen, S., Shibasaka, T.: Development of intelligent monitoring and optimization of cutting process for CNC turning. International Journal of Computer Integrated Manufacturing 19, 473–480 (2006)
Study on the Problem and Countermeasure of Fruit Production Quality and Safety in Yanshan Mountain Haisheng Gao, Bin Du, and Fengmei Zhu School of Food Science and Technology, Hebei Normal University of Science and Technology, 066000 Qinhuangdao, P.R. China
Abstract. In this paper, the main problem and countermeasure of fruit production quality and safety were studied in Yanshan Mountain. The subjective factor was fruit farmer quality and safety consciousness faint, and the objective factors were: severe orchard environmental pollution, faint fruit post-harvest handling, lag in process technology, fruit quality determination and market inspection system is not yet perfect. The measures of improving fruit quality and safety production in Yanshan Mountain were proposed. On the premise that adjusting and optimizing regional distribution, building and improving quality and safety production management system, controlling post-harvest handling and circulation contamination, implementing organic strategy, establishing Green Silicon Valley. Keywords: Yanshan Mountain, Fruit quality, Fruit safety production, Problem, Countermeasure.
1 Introduction
~
,
~
The center of Yanshan mountain is 39°45′ 41°N 115°40′ 119°50′E. Yanshan Mountains are a major mountain range north of the North China Plain, in northern Beijing, Tianjin and Hebei Province. The length of Yanshan Mountain is 330 Km from east to west, 120 Km from south to north. Yanshan Mountain has a marked continental monsoon climate. The average annual sunlight is 2600 2800 hours. The mean annual precipitation of this region is 607.5mm. It is suitable to plant fruit trees in the hilly region and the plain. The main fruits are: apple, pear, grape, hawthorn, chestnut, walnut, jujube, peach, apricot, plum, cherry and so on [1]. With the rapid development of science and technology and modern industry and the progress of human civilization, it is becoming increasingly clear of the awareness for the industrial pollutants and drug residue by food chain are harmful to human health and it is increasingly intensive for demand of fruit safety. What is more, the agricultural product quality and safety is the hot issue of government attention, social attention and global focus. The fruit product quality and safety is also the key problems of implementing the agriculture and rural economy structure adjustment and improving international competitiveness of fruits. So, it is significant for improving fruit production quality and promoting fruit industry development by carrying out fruit production quality and safety in Yanshan Mountain and strengthening environment control. This
~
D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part IV, IFIP AICT 347, pp. 355–360, 2011. © IFIP International Federation for Information Processing 2011
356
H. Gao, B. Du, and F. Zhu
paper for the first time reports the problem and countermeasure of fruit production quality and safety in Yanshan Mountain.
2 The Problems of Fruit Production Quality and Safety in Yanshan Mountain 2.1 The Weak Quality and Safety Awareness of Most Fruit Farmers The fruits production makes the family as the basic unit in China nowadays. The decision subject of production management is farmer. They were lack of safety awareness and the legal system because they were restricted by the cultural level and technical condition, and stimulated by the economic interests. In orchard management, they fertilize more chemical fertilizer and less organic fertilizer, more nitrogen fertilizer and less phosphate and potash fertilizer. They rely mainly on chemical pesticides for pest control. And in the eradication of pests but also kill the natural enemies of pests by improper application. They have to use high toxic pesticide residues due to the increasing pest resistance. All of this leads to rapid deterioration of the orchard ecological environment and over standards of fruits pesticide residue. 2.2 The Growing Problems of Environmental Pollution in Orchard In recent years, with the development of modern industry, the dramatic increase of industrial wastewater, exhaust and residue emissions leads to deterioration of the orchard ecological environment and poor fruit growth and development and lower fruits production. Accumulation of toxic substances in fruits results in decreased safety and quality, even some fruits can not eat. In addition, the extensive use of agricultural chemicals, fertilizer and herbicide in some production areas causes the dramatic decrease of atmosphere, water and soil quality and the deterioration of ecological environment, and fruits serious pollution. 2.3 The Faint Treatment of Fruits Commercial Post-harvest The fresh fruits achieve fast development recently. The fresh fruits post-harvest would dehydrate and shrink, the flesh would brown. The fruits would be rot in preservation and transportation. Relatively, the cultivation before harvest was taken seriously in China fruits production, and the post-harvest treatment was neglected. The treatment of fruits commercial post-harvest and cold chain circulation seriously lag. The facilities condition of preservation and transportation is poor. There are some pollution problems in the treatment of fruits commercial post-harvest and preservation, transportation and sale. For instance, not strict fruits grade, simple and rough package, improper preservation and transportation management measures and unscientific disease control in preservation. All of this not only aggravate the fruits decay losses in post-harvest circulation, but also affect fruits quality and safety [2]. 2.4 The Relative Backward Processing Technology In addition to juice and wine production, other fruits production is based mainly on middle and small enterprises in China at present. These enterprises have poor processing
Study on the Problem and Countermeasure of Fruit Production Quality and Safety
357
facilities conditions, low processing capacity, defective fruit as raw material, low scientific and technological content in production, low added value and unstable quality. There are irregular operations and not complete cleaning and sterilization in some fruits production enterprises. These lead to microbiological contamination by yeast, lactic acid bacteria and mould, further lead to over standard of microorganism. At the same time, there are some problems such as irregular additive application, over standard in fruits production. The cleaner in fruit washing and the disinfectant in packaging sterilization also cause chemical hazards. 2.5 Imperfect Quality Determination and Market Inspection System The building of test organizations lags in China. Most provinces can not detect the pesticide residues. The building of fruit market test organizations is very weak. There are few instruments and inspectors in farmers market and fruits market. In addition, the detection methods were still lagging behind, the detection efforts are not enough, and the detection standard can not conform to international standard.
3 The Countermeasures of Fruit Production Quality and Safety in Yanshan Mountain 3.1 Adjusting and Optimizing Regional Distribution The regional culture is a basic condition achieving fruits safe production. Planting fruit trees in unsuitable areas is not only affecting species characteristics, but applying complicated fertilization and spraying because of pursuit of high quality. These increase pollution links and make fruits safety worse. The important mark of fruits trees production modernization is adjusting structure, optimizing fruits trees distribution, improving product composition and meeting market demand. The optimizing and adjusting fruits trees structure is adjusting species structure. On the one hand, the specific measures are: with the combination of mountain and plain green, selecting less pollution hilly and sandy beach land, developing adaptable ecology forest and forest net, fruit crop intercropping, controlling the development of large amount of fruits, developing famous dried fruits, adjusting early maturing, maturing, late maturing variety, supplying reasonable market. On the other hand, regional adjustment is the crucial measure. In accordance with the principles of tree species selected for suitable planting site, gradually realizing regional culture of fruits trees. 3.2 Building and Improving Quality and Safety Production Management System The actual situation of current fruits production is thousands of households decentralized management. Building the perfect safety production management system, regulating production activities and technical measures, reducing the infection of harmful substances and achieving safety production are important. The four measures are as follows:
358
H. Gao, B. Du, and F. Zhu
1. The key technologies are environmental technology system, production technology system and quality standards technology system. 2. The environmental quality monitoring system includes the non-pollution environmental monitoring system of water, soil, atmosphere and Living creature. 3. The regulations systems include building produce quality and safety detection and monitor system and produce quality regulations system. 4. The service system includes organizing or supporting the management system combined with farmers, firms, technology demonstration base and market, turning service for agricultural production into service for agricultural products, integrating the agricultural efficiency with economic benefits. 3.3 Controlling Post-harvest Treatment and Circulation Contamination The fruits post-harvest mechanization commercial treatment accounts for 5% at present, while the fruits post-harvest losses account for 20%. Post-harvest treatment is the bottleneck of restricting fruits into high ranking market [3]. Therefore it is very important for us to intensify fruits post-harvest treatment, formation production cooperatives, implement the measures—decentralized production, centralized grading and packaging and centralized sale, enhance the competitiveness of Chinese fruits in international market. The major measures are as follows: the first is pre-cooling. The pre-cooling is essential for fruits post-harvest in place of production. The role of precooling is dispersing heat of fruits, lowering metabolic intensity and maintaining the freshness of fruits. The second is grading. Excluding disease fruits and mechanical injury fruits, then reasonable grading was applied according to fruits shape, color, freshness for transportation, storage and sale. The third is washing and waxing. The steps are washing the dirt, pesticide, oil and germ by fruits detergent, then showering using water and drying in the air. The wax of fruits was destroyed by washing. The fruits would loss water in storage and transportation. So treating the fruit surface by wax, shellac, starch film and sucrose membrane. This treatment could maintain the freshness and nutrition of fruits and beautify fruits. The fourth is storage and preservation. The fruits would be rot by the germ infection, so using the preservative is essential before storage. The natural biological agents and herbal extracts, are selected as antiseptic, preservative and packaging. The fifth is package. The package is favorable to protect fruit, facilitate transportation and storage and facilitate the purchase. It is important for package to guarantee fruits safety transportation and sale. Especially export fruit, the package quality and tactics are crucial. The carton should be treated using disinfectant. The sixth is proposing food quality traceability system and food market access system to effectively protect the legitimate rights and interests of producers and consumers. In the circulation, the industrialization management should be paid attention to. Building industrialization mechanism is the important content of developing modern fruit production and also is the essential measure of reducing fruits pollution links, building fruits safety production system. Therefore, the first is to develop value added processing and market building, nurture leading enterprises. The second is to strengthen the connection of leading enterprises, base and farmers. The third is to build the matching management system and operation mechanism, realize scale operation, scientific management and social service, build the new pattern of forest
Study on the Problem and Countermeasure of Fruit Production Quality and Safety
359
industry development of production, supply and sale, maintain the sustainable development of fruits industry. 3.4 Implementing Organic Strategy There are two patterns of global agricultural development: On the one hand, the conventional agriculture is extensive use of fertilizers, pesticide. The pursuit of this pattern is economic objectives. Conventional agriculture is the main form of agriculture nowadays. On the other hand, Organic agriculture is the form that excluding or strictly limiting the use of chemical compounds. The pursuit of organic agriculture is combination of economic and ecological benefits. Organic agriculture is the development direction of modern agriculture [4,5]. The product of organic agriculture is organic food. Organic food is a kind of sustainable food. The goal is environmental protection, safety, health, quality. Organic strategies must be implemented and organic fruit production must be progressively developed. So the safety of fruits is ensured fundamentally. Implementing the organic strategies is a gradual process. On the basis of improve pollution-free fruit and green fruit, developing organic fruits step by step. The current main tasks are as follows: the first is improving environment and environmental protection; the second is reducing fertilizer application, promoting application of organic fertilizers and microbial Fertilizers; the third is promoting biological control technology, using biological pesticides or non-pollution pesticides, improving the credibility of the fruits in Yanshan Mountain. 3.5 Establishing Green Silicon Valley The development of fruits safety production should learn experience from “Silicon Valley”, and strengthen scientific and technological research and practical technical reserves. Building green fruits production base, and making it as fruits production test with high technology, the base of senior technical personnel, the place of practical technology, the model base of organic fruits and safety production. Depending on the example of green Silicon Valley, and promoting the development of national fruits safety production, ensuring the long-term development of fruits production in Yanshan Mountain.
4 Conclusions In a word, Yanshan Mountain has unique advantage of producing high quality fruits. However, there are low level of farmer organization and deep traditional business. Industrialization is the primary measure of activating fruits industry development. In accordance with the requirements—uniform production technology, uniform product standards, uniform monitoring method, uniform management measures and uniform pesticide regimen. Building a group of green fruits production export demonstration garden and promoting the development of national fruits safety production. Strengthening the construction of fruits pesticide residue monitoring network system and building regular fruits pesticide residue monitoring system are crucial. Putting Emphasis on fruits production base, fruits wholesale market and fruit operation
360
H. Gao, B. Du, and F. Zhu
market circulation, equipping with instruments and equipment in processing, enriching the test personnel, improving test level, expanding test project, implementing the market access system. The aim is implementing the measures such as emphasis on investment, building base, controlling heading enterprises, connecting farmers and developing market, and putting Yanshan fruits industry business into the development track of industrialization.
References 1. Science and Technology Department of Hebei Province: Characteristic industry science and technology development planning in rural areas in Yanshan mountain, Hebei Province. Development planning writing group. pp. 1–3 (2008) 2. Haisheng, G., Xiyan, Z., Runfeng, L.: Review of post-harvest treatment and preservation technologies of fruit and vegetable. Trans. CSAE 2, 273–278 (2007) 3. Haisheng, G., Runfeng, L., Xiufeng, L.: Advance in Past-havest Commoditization Processing Technologies of Horticultural Products. Food Sci. (in Chinese) 9, 627–631 (2008) 4. Xiaogui, S., Qingdang, G.: Discussion of Fruit Security in Shandong Province. Nonwood forest res. 1, 68–70 (2004) 5. Fengtian, Z., Yang, Z.: China Food Security: Problems and Policy Measures. China Soft Sci. 3, 16–20 (2003)
Calculation and Analysis of Double-Axis Elliptical-Parabolic Compond Flexure Hinge* Ping’ an Liu1, Jianqun Cheng1, and Zhigang Lai2 1
School of Mechanical and Electronic Engineering at East China Jiaotong University, Nanchang 330013, China
[email protected] 2 Jiangxi Technical College of Manufacturing, Nanchang, Jiangxi, 330095, China
Abstract. A double-axis flexure hinges combined with elliptical and parabolic curve is presented in this paper. The design equations for its compliance computation have been deduced by application of Castigliano’s displacement theorem after the anlysis of the structure, which were analyzed by the use of the software MATLAB, and comfirm the validity of the model with finite element analysis software ANSYS. Keywords: flexure hinge, mixed flexure hinge, compliance, Castigliano’ theorem.
1 Introduction Flexible hinge eliminates its idling running and mechanical friction by using small angle elastic deformation and self-recovery characteristics in the transmission process, and can get highter resolution of displacement. It is been widely used in various occations, such as small angular displacement, high-precision rotation of occasions, especially in precision measurement, positioning and other fields because of its small-volume, gapless, no mechanical friction,no lubrication, in-gage and high sensitivity of the transmission structure [1-3]. People have been made a series of studies on the single-axis oval, chamfered, parabolic and hyperbolic flexure hinge [4-5] since 1965 Paros and Weisbord derived circular flexure hinge flexibility formul. However, they have done little reaserch about the compound flexure hinge in addition to Chengui Min’s [6] on the right circular and oval compound flexure hinge. Elliptical flexible hinge has a lager range of rotation but lower accuracy, and Parabolic with small range of rotation but high precise compared to the Circular incision hinge; for both accuracy and range of motion exercise, combining the elliptic and parabolic type flexure hinge will take the advantages of both flexible hinges, so this paper will present elliptic - parabolic compound flexible hinge. *
Supported by Natural Science Foundation of Jiangxi, China, NO. 2008GZC005.
D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part IV, IFIP AICT 347, pp. 361–367, 2011. © IFIP International Federation for Information Processing 2011
362
P. Liu, J. Cheng, and Z. Lai
2 Elliptie-Parabolic Compound Flexible Hinge Elliptic and parabolic confound flexible hinge is composed of two half elliptical and parabolic form, as shown in figure1, oval-side as a fixed end and the Parabolic side as the rotation-side, and make the following assumptions: (1) (2) (3) (4)
Elliptic and parabolic incision have the same thickness,and both are t. Elliptic and parabolic are tangented at the intersection, or smooth over. Two incision length in the X direction are m. Ellipse major axis m is greater than the minor axis n, when m=n,that is circurlar and parabolic compound flexible hinge.
Fig. 1. Elliptic-parabolic hinge pro/e model
Fig. 2. The structure parameters of compound hinge
3 Flexibility Formulas Assumptions based on cantilever small deformation,bending deformation is generated by the force and moment. The impact of axial load has considered, while the impact of shear and torsion were not. The structure parameters of axial elliptic - parabolic compound flexure hinge were shown in Figure 2,and the force analysis as shown by
Calculation and Analysis of Double-Axis Elliptical-Parabolic Compond Flexure Hinge
363
Figure 3, the flexibility formulas of compound hinge were deduced based on mechanical card's second theorem:
Fig. 3. The force analysis of compound hinge
⎧θz1 ⎫ ⎧Cθ −M C12 0 ⎫⎧Mz1 ⎫ ⎪ ⎪ ⎪ ⎪⎪ ⎪ ⎨y1 ⎬ = ⎨ C12 CY 0 ⎬⎨Fy1 ⎬ ⎪x ⎪ ⎪ 0 0 CX ⎭⎪⎪⎩Fx1 ⎭⎪ ⎩1⎭ ⎩
(1)
According to reciprocal theorem known θz1 =C11Mz1 +C12Fy1,it can be obtained from the left of formula (1) with the use of cartesian displacement:
⎧ ∂u ⎪θz1 = ∂Mz1 ⎪ ⎪⎪ ∂u ⎨ y1 = ∂Fy1 ⎪ ⎪ ∂u ⎪ x1 = ∂Fx1 ⎪⎩
(2)
By the material mechanics, without considering the shear and torsion, the deformation energe of the compound flexible hinge is:
U=
⎤ 1 ⎡ FX 12 MZ2 dx + dx ⎥ ⎢∫ ∫ 2 ⎣ l EA( x) EI z ( x) ⎦ l
and F x = F x 1 , M z = M z1 + Fy1 x , A( x) = bt( x) , I ( x )
=
(3)
(b t ( x ) ) / 1 2 3
364
P. Liu, J. Cheng, and Z. Lai
The coordinate system as shown in figure 2, variable thickness equations of dual-axis flexure hinge as follow:
⎧ ⎡ n 2⎤ ⎪⎪ t +2⎢n−m 2mx−x ⎥ ⎣ ⎦ 0≤x≤m t(x) =⎨ 2 ⎪t +2⎡2p( x−m) ⎤,m<x≤l ⎪⎩ ⎣ ⎦
(4)
To the type of substitution (1) (2) (3) can obtain the flexibility formulas: ⎧ ⎛ 2n + t ⎞ ⎫ ⎪ −π −t(4n + t) − (8n + 4t) Arc tan ⎜⎜ ⎟ Arc tan( 2m p ) ⎪ −t(4n + t) ⎟⎠ 1 ⎪ ⎪ t ⎝ CX = ⎨m* + ⎬ EB ⎪ 4n −t(4n + t) 2 pt ⎪ ⎪ ⎪ ⎭ ⎩
CY =
3 4 3 2 2 3 4 ⎛ −4nt −t2 ⎞ 12 ⎧⎪ 3 16c ( −80c + 24c t +8c (3+ 2π)t + 4c(1+ 2π)t +πt ) 3 (2n +t)3 (6n2 − 4nt −t2 ) +m Arc tan⎜ ⎟ ⎨m 2 5 3 2 ⎜ ⎟ EB ⎪ t 16c t ( 4c +t ) ⎝ ⎠ ⎩ 4n3 ( −t ( 4n +t ) ) 2
m (1 1 2 p 2 m 4 + 5 6 p m 2 t − t 2 ) + + 32 p 2t 2 (4 pm 2 + t)2
Cθ − M
(5)
(1 2 p m 2 + t ) A r c ta n ( 3 2
64 p t
(
2m
5 2
)
2 ⎛ 2n + t ⎞ m 12 pm + 5t 6n ( 2n + t ) 12 ⎧⎪ 6n 2 + 4nt + t 2 = − m Arc tan + + ⎨m 2 5 2 ⎜ ⎟ EB ⎩⎪ t ( 2n + t )( 4n + t ) 2 ⎝ −4nt − t 2 ⎠ 8t 2 4 pm 2 + t −4nt − t 2 2
(
)
(
)
t
p ⎫ )⎪ ⎪ ⎬ ⎪ ⎪⎭
(6)
2m p ⎫ )⎪ ⎪ t ⎬ 5 ⎪ 2 16 pt ⎪⎭
3 Arc tan(
(7) From the formulas of elliptical-parabolic compound hinges which have deduced, it can easily see that the flexibility is inversely proportional to the thickness B and modulus of elasticity E. 3.1 The Verification of Closed-Loop Compliance Formula Design two dual-axis elliptical – parabolic compound hinges, its parameters as follows: thickness B is 5mm, length of oval long axle 5mm, the minor axises 3mm and 5mm, the thinnest thickness in the middle is 1mm, the total width 10mm. FEM experimental parameters are: modulus of elasticity: E = 110 ×109Nm- 2
,Poisson ratio: μ= 0. 3.
parabol parameters: p=1/16 Fx1 = Fy1 = 1 N Mz1 = 1 N.m.
Calculation and Analysis of Double-Axis Elliptical-Parabolic Compond Flexure Hinge
365
Fig. 4. The ansys geometric model of compound
Models with different parameters in the finite element software to verify compliance formulas, choose the shell63 finite element, making finite element analysis to compound flexure hinge( As shown in figure 4), analytical method(AM) and finite element results were shown in Table 1. Known from the data in the table, finite element method(FEM) and the exact formula is less than 10% deviation.
Table 1. Analytical and finite element calculation about flexibility
3.2 Performance Analysis From the formulases (5) to (7),it is easy to know that the formula of flexibility is inversely proportional to modulus E and thickness B; the relationship between the CX, CY, Cθ-M and the short axis length n incision thickness t were shown in figure 5-7.
、
366
P. Liu, J. Cheng, and Z. Lai
Fig. 5. The relationship between Cx and n
、t
Fig. 6. The relationship between Cy and n
、t
Fig. 7. The relationship between Cθ-M and n
、t
(1) We can see from figure 5, the relationships between the flexibility of x-axis direction Cx and t, n are in the overall smooth, keep the t unchanged, with the increase of n, Cx will decrease linearly; if the n unchanged, Cx decrease with t increases and changes in relatively fast as t in the range of 0.2-0.4, Cx changes in relatively fast ,in conclusion the stiffness of X axis is large relatively.
Calculation and Analysis of Double-Axis Elliptical-Parabolic Compond Flexure Hinge
、
367
,
(2) The relationship between flexibility Cy and t n as shown in Figure 6 Cy will reach maximum when t equals to 0.2,and Cy will change sharply as t in the range of 0.2-0.4 and decrease smoothly when t is greater than 0.4, Cy has no significant change with n. (3) From the figure 7,we can see that the Cθ-M obtains maximum when t equals to 0.2 and decreases as n increase,especiallly when t in the range of 0.2-04, it decreases sharply; when t is greater than 0.6,the Cθ-M decreases slowly and close to zero. From the anlysis above all,this conpound hinge has large rigidity in the direction of x –axis and good flexibility in y axis direction and rotate around Z axis,so it is very suitable as rotation flexibility hinge.
4 Conclusions This paper presents a new type of compound hinge oval - parabolic compound flexible hinge. And analized its structure, its flexibility formulas were deduced based on card's second law and vertified by the finite element software,find the errors are within 8%, which demonstrates the correctness of the flexibility formulas. Then analyzed the relationship between the flexibility and the thickness B and the short axis lengh n with matlab software,and got some conclusions, it could be a theoretical references in future industry and agriculture fields.
References 1. Li, J., Wang, J.-W., Chen, K., and so on: Based on generalized geometric error model of the micro-robot accuracy analysis. Tsinghua University (Natural Science) 4(5), 28–32 (2000) 2. Fu, P., Zhou, R.-K., Zhou, S.-Z., et al.: Synchrotron radiation beam line in the flexible hinge of [J]. Optics and Precision Engineering 9(1), 67–70 (2001) 3. Sun, L.-N., Li, M.: A kind of two-dimensional micro-nano positioning stage design and analysis. Optics and Precision Engineering 14 (2006) 4. Feng, C., Xia, J.: Biaxial elliptical flexure hinge design and calculation. Engineering Mechanics 24(4) (April 2007) 5. Chen, J., Liu, P.: Cosh flexible hinge stiffness is derived. Mechanical Design and Manufacturing (7) (July 2008) 6. Chen, G., Jia, J.-Y.: Straight round elliptical flexure hinge of compound. Machine Design and Research 21(4), August 2
Surface Distresses Detection of Pavement Based on Digital Image Processing Aiguo Ouyang1, Chagen Luo1, and Chao Zhou2 1
Key Laboratory of Conveyance and Equipment, Ministry of Education, East China Jiaotong University, Nanchang, China 2 Jiangxi Agricultural University, Nanchang, China
Abstract. Pavement crack is the main form of early diseases of pavement. The use of digital photography to record pavement images and subsequent crack detection and classification has undergone continuous improvements over the past decade. Digital image processing has been applied to detect the pavement crack for its advantages of large amount of information and automatic detection. The applications of digital image processing in pavement crack detection, distresses classification and evaluation were reviewed in the paper. The key problems were analyzed, such as image enhancement, image segmentation and edge detection. The experiment results of the commonly used algorithms forcefully supported following conclusion: the noise in pavement crack images is effectively removed by median filtering, the histogram modification technique is a useable segmentation approach, the canny edge detection is an ideal identification approach of pavement distresses. Keywords: Digital Image processing, Crack detection, Edge detection, Image segmentation.
1 Introduction Maintenance of pavements is an important aspect for the Departments of Transportation all over the world. The first step towards maintenance is the identification of pavement distresses and their documentation for further action. Pavement distresses are visible imperfections on the surface of the pavements. Accurate evaluations will result in a better chance that resources will be distributed normally. Thus, yield a better service condition [1]. Pavement could be evaluated through the different types of distress experienced, such as cracking, disintegration and surface deformation. At present, there were various methods for conducting distress surveys, recording and analysing distress survey data [2]. Pavement engineers had long recognized the importance of distress information in quantifying the quality of pavements. Traditionally, pavement condition data are gathered by human inspectors who walk or drive along the road to assess the distresses and subsequently produce report sheets, but it is high cost and time consuming. Worse still, work has to be done along fast moving traffic. Such condition would endanger the safety of the personnel involved. Finally, large differences will D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part IV, IFIP AICT 347, pp. 368–375, 2011. © IFIP International Federation for Information Processing 2011
Surface Distresses Detection of Pavement Based on Digital Image Processing
369
exist between the actual condition and evaluation results because of the subjectivity of the evaluation process. In the wake of tedious manual measurements and safety issues, a variety of types of methods have been devised to identify the cracks on pavements apart from the crude process of manual inspection. Image processing, ultrasonic detection and infrared detection, the most widely reported of the automated methods is that known as WiseCrax. Wisecrax [3] is an example of a commercially available device that uses infrared imaging to detect cracks on pavements. A camera is mounted on a vehicle with that takes pictures of the pavements continuously. Images are processed off-line overnight at the office workstation by a unique open architecture process using advanced image recognition software. Typical survey vehicle configuration consists of one or more downward-facing video cameras, at least one forward facing camera for perspective, and any number of additional cameras for the capture of right-of-way, shoulder, signage, and other information depending on agency requirements (as shown in Figure 1). New methods are being devised to identify cracks more efficiently and get closer towards perfection as good as the human eye and are still in the process. Ye et al. [4] presented a crack width detection method to extract the crack image of pavement distresses and calculate the crack width. Li et al. [5] proposed a pavement crack image analysis approach based on the image dodging to improve the reliability of the pavement crack recognition. Existing automatic real-time detection systems focus on the low identification rate and classification difficulty [6]. First, a lot of noise brings in pavement images caused by the road environment itself. Second, it is lack of easy and effective identification and classification algorithm [7]. Although some auto-inspection Survey Vehicle In-Office
Data Acquisition By Image Sensor
Stotrage Devices
Image Interpretation
Output devices
Supplemental Devices
Digitizing Devices
In-office or On Board Computer
Reporting Result
Lengend:
Real-time Processing or On Board Computer Processing In-office Processing
Fig. 1. Elements of a pavement imaging system
Use
370
A. Ouyang, C. Luo, and C. Zhou
system is in application at present, the system with surface-scan camera has the problem of low distinguish, and the dynamic collecting image clarity is not very ideal [8]. No method has achieved completely satisfactory results [9].
2 Significance of Pavement Distress Information Pavement distresses are visible imperfections on the surface of pavements. They are symptoms of the deterioration of pavement structures. Agencies that have implemented a Pavement Management System (PMS) collect periodic surface distress in formation on their pavements through distress surveys [10]. Distress information takes a vital role in quantifying the quality of pavements. This information has been used to document present pavement condition, chart past performance history and predict future pavement performance [11]. Pavement distress information is also broadly used as the only quality measure of pavements in many PMS. This is particularly true for systems used by local governments and in urban areas where roughness measurements are not performed because of a lack of equipment availability, high cost or a lack of relative applicability.
3 Pavement Image Analysis The purpose of image processing is to extract the distress features from the pavement image. Preprocessing is done by removing extraneous features that have higher pixel intensities than the mean pixel intensity in the image. In this process, all the pixels representing paint striping and surface textures brighter than the average background gray level are surpassed to the background. The effect of pavement image processing is illustrated by the following example. This study utilised full programming language software MATLAB 7.10.0 (Mathwork, Natick, MA, USA) to enable a series of MATLAB statements to be written into a file and then execute them with a single command. 3.1 Image Enhancement Image enhancement is applied in an attempt to remove noise in pavement images. Noise reduction is one aspect of preprocessing phase of crack detection process. Filtering is the most common form of noise reduction. Median filtering is one of the most commonly used preprocessing techniques for crack detection today. Median filtering is therefore applied as pavement image enhancement technique as suggested by Jitprasithsiri [12]. The median is much less sensitive than the mean to extreme values. Median filtering is better able to remove these outliers without reducing the sharpness of the image. Such a property of median filtering was explained by a classic example of salt and pepper noise which was a random addition of black and white pixels into a gray scale image. The result of median filtering in figure 6 indicated that noise was removed aptly by the median filtering compared with other operators.
Surface Distresses Detection of Pavement Based on Digital Image Processing
371
Fig. 2. Original pavement grayscale image
Fig. 3. Grayscale image with salt and pepper
Fig. 4. Laplacian operator
Fig. 5. Gaussian operator
Fig. 6. Result of median filtering
Fig. 7. Log operator
Fig. 8. Sobel operator
Fig. 9. Prewitt operator
372
A. Ouyang, C. Luo, and C. Zhou
3.2 Image Segmentation Image segmentation is the crucial step in automatic image distress detection and classification (e.g., types and severities) and has important applications for automatic crack sealing [13]. The segmentation approach chosen was based on a histogram modification technique. This histogram modification was achieved through iterative clipping. At each stage in iterative clipping, more pixels were assigned to the background. This process continues until only distress features were left. The end result as shown in Figure 12 was an image in which the distresses were distinct and easily separable from the background. At this point, a threshold value could be determined automatically in order to isolate the distress features from the background. Transformation function in Figure 11 presented that the threshold value was near 0.6.
1400 1200 1000 800 600 400 200 0 0
50
100
150
200
255
Fig. 10. Histogram of original image
Output intensity value
1 0.8 0.6 0.4 0.2 0 0
0.2
0.4 0.6 0.8 Input intensity value
Fig. 11. Transformation function
1
Fig. 12. Result of histogram modification
3.3 Canny Edge Detection The Canny edge detector is a good edge detector among traditional edge detection algorithms [14]. The gray scale pavement image was first smoothed using a Gaussian
Surface Distresses Detection of Pavement Based on Digital Image Processing
373
filter with a specified standard deviation to reduce noise (see figure 13), then gradients were calculated to determine edge points. Using two threshold values 0.5 and 0.4, different correlated edge points were linked together. The strength of the method is its ability to detect edges in the presence of noise and to detect weak edges. Preliminary results from canny edge detection in figure 14~16 showed that the optimum identification of distress was dependent on the parameters used in the algorithm and that the optimum parameters varied with each image. There was a problem of coming up with false distress boundaries when a very high standard deviation in Gaussian filtering was used. The distress may seem wider than it actually was resulting in false severity level detection.
Fig. 13. Grayscale image by gaussian filtering
Fig. 14. Threshold value=0.5
Fig. 15. Threshold value=0.4
Fig. 16. Result of canny edge detection
3.4 Pavement Distress Classification and Evaluation Many protocols and definitions exist for pavement distresses classification and evaluation. One of the most widely used protocols is Strategic Highway Research Program Long-Term Pavement Performance (SHRP-LTPP) protocol [15]. SHRP-LTPP first classifies the type of cracks according to their orientations, locations, and shapes and quantifies the severity and extent of the cracks according to their properties, such as width, length, and areas. Another important protocol is the World Bank’s universal cracking indicator (UCI) [16]. UCI defined a single number to indicate the severity of all the cracks in a pavement segment. For a single crack such as a longitudinal, transverse, or diagonal crack, its indicator was defined as the product of
374
A. Ouyang, C. Luo, and C. Zhou
its width and length. For the block and alligator cracks, their indicators were defined by the area that contains the block or alligator cracks. The unified crack index (ASTM STP 1121) is a protocol similar to UCI. The standard crack density can be automatically determined by dividing the number of pixels for the cracks by the number of the total pixels of the pavement segment.
4 Conclusions and Perspectives As mentioned above, digital images lend themselves to automated analysis because of the ability to analyze variations in grayscale as those variations relate to pavement features. A major force behind the move toward digital imaging of pavements is the opportunity to reduce distress data from those images through automated methods. With the fast development of analog videotaping, using digital imaging to capture pavement surface is becoming the preferred method. With advances in image sensors and computer technologies, the automation of data collection and analysis is a major goal of contemporary pavement management. Several automated analysis algorithms for pavement distress are in use and others are under development.
Acknowledgements The authors gratefully acknowledge the financial support provided by National Science and Technology Support Program (2008BAD96B04), Key Laboratory of Ministry of Education for Conveyance and Equipment, East China Jiaotong University (Grant No. 09JD10).
References 1. Kim, J.: Development of a Low-Cost Video ImagingSystem for Pavement Evaluation. Oregon State University. Ph.D. Thesis (1998) 2. Cheng, H.D., Miyojim, M.: Automatic pavementdistress detection system. Journal of Information Sciences 108, 219–240 (1998) 3. T.R.B.E. NCHRP synthesis 334: Automated Pavement Distress Collection Techniques. Transportation Research Board of the National Academies, Washington, D.C. (2004) 4. Ye, G.R., Zhou, Q.S., Lin, X.W.: Measurement of Surface Crack Width Based on Digital Image Processing. Journal of Highway and Transportation Research and Development 27, 75–78 (2010) (in Chinese) 5. Li, Q.Q., Hu, Q.W.: A Pavement Crack Image Analysis Approach Based on Automatic Image Dodging 27, 1–5 (2010) (in Chinese) 6. Wang, R.B., Wang, C., Chu, X.M.: Developments of Research on Road Pavement Surface Distress Image Recognition. Journal of Jilin University (Engineering and Technology Edition) 32, 91–97 (2002) (in Chinese) 7. Zhang, J.: Study on Pavement Crack Identification and Evaluation Technology Based on digital Image Processing. Chang’an University. Ph.D. Thesis (2004) (in Chinese)
Surface Distresses Detection of Pavement Based on Digital Image Processing
375
8. Chu, X.M., Yan, X.P.: The Automatic Search of Pavement Surface Distress Image Based on on-line Learning. In: International Conference on Transportation Engineering 2007 (ICTE 2007), vol. 246, pp. 3282–3287 (2007) 9. Tsail, Y.C., Kaul, V., Russell, M.M.: Critical Assessment of Pavement Distress Segmentation Methods. Journal of Transportation Engineering 136, 11–19 (2010) 10. Haas, R., Hudson, W.R., Zaniewski, J.: Modem Pavement Management. Krieger Publishing Company, Malabar (1994) 11. Shahin, M.Y.: Pavement Management for Airport, Roads and Parking Lots. Chapman & Hall, New York (1994) 12. Jitprasithsiri: Development of a New Digital Pavement Image Processing Algorithm for Unified Crack Index Computation. University of Utah. Ph.D. Thesis (1997) 13. Mustaffara, M., Lingb, T.C., Puanb, O.C.: Automated Pavement Imaging Grogram (APIP) for Pavement Cracks Classification and Quantification-A Photogrammetric Approach. The International Archives of the Photogrammetry. Remote Sensing and Spatial Information Sciences 37, 362–372 (2008) 14. Canny, J.: A computational approach to edge detection. IEEE Transactions on Pattern Analysis and Machine Intelligence 8, 679–698 (1986) 15. Hawks, N.F., Teng, T.P., Bellinger, W.Y., Rogers, R.B., Baker, C., Brosseau, K.L., Humphrey, L.C.: Distress Identification Manual for the Long-Term Pavement Performance Project, SHRP-P-338. National Research Council, Washington, D.C. (1993) 16. Paterson, W.D.: Proposal of Universal Cracking Indicator for Pavements. Transportation Research Record 1455, TRB, National Research Council, Washington, D.C., pp.69–76 (1994)
The Application Research of Neural Network in Embedded Intelligent Detection Xiaodong Liu1, Dongzhou Ning1, Hubin Deng2, and Jinhua Wang1 1
Compute Center of Nanchang University, 330039, Nanchang, Jiangxi, China
[email protected] 2 East China Jiaotong University, 330039, Nanchang, Jiangxi, China
[email protected] Abstract. Neural network which can adapt the sample data by training has good fault-tolerance and can be used in the field of intelligence widely. In the embedded system, restricted to the resources and the capacity of processor, the neural network application has a series of problems, such as losing timelines and the system could be collapsed easily. This article discusses how to use limited memory, processor and external equipment resources to achieve the neural network algorithm for improving the universality of detection system and adaptive ability in the embedded intelligent measuring system. Keywords: Neural Network, Embedded System, Intelligent Detection.
1 Introduction Most embedded detective devices are developed by using the fixed parameter method, that is, the mapping relationship between test data and test conclusion is fixed. However, in practical application, the input conditions of detection and the testing standards may need to be changed, which may make the existing detection system inadaptable. The neural network owns the ability of self-adaptation and self-learning, therefore, by studying on the sample data, it can determine the mapping relationship between the input and the output. And if the neural network is applied to the detection system, it may adjust the mapping relationship automatically by training samples, which can make the detection system more flexible and adaptable. Application of the algorithm is shown in the figure 1 [1].
Detection data acquisition
Training sample data Testing sample data
Processor
Detection conclusion
Neural network
Fig. 1. Application of neural network D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part IV, IFIP AICT 347, pp. 376–381, 2011. © IFIP International Federation for Information Processing 2011
The Application Research of Neural Network in Embedded Intelligent Detection
377
Because of the constraints of the resources of embedded portable device and the capacity of processor, computational complexity should be minimized in the application of neural network algorithm, thus to improve the efficiency by centralizing resources to deal with necessary tasks.
2 Neural Network Platform Design Great success has already been achieved in the research and application of neural network, however, there is no perfect theory as a guide in the aspect of network structure development and design, so network parameters are adjusted just only on the basis of former researchers’ designing experience and experiment analyses. The functions of neural network platform development are network parameter settings, network training, network training analysis, network testing, network prediction etc. In the network structure designing and the training algorithm, because of the resources constraints in embedded system, we should use as simple and effective methods as possible. The following is the description of issues needed to be considered in the design: 2.1 The Choice of Network Model 2-layer linear perceptron model and 3-layer BP network model are provided and the user can choose one of them in the network parameter settings in accordance with the complexity of the detection system. According to the Almighty Approaching Theorem, a 3-layer BP neural network with a hidden layer can approximate to any continuous function of bounded domain with arbitrary precision as long as there are enough hidden layer nodes[2]. Therefore in terms of the function, 3-layer BP neural network can meet most sophisticated detection systems’ requirements, while for some detection systems which are simple mapping, linear perceptron model with a small amount of calculation and fast speed can be chosen so as to maximize the efficiency of the systems. 2.2 The Choice of the Number of Input and Output Nodes The number of the input nodes is determined by testing item. The number of the output nodes is generally determined by the number of the testing conclusions. But if there are a lot of testing conclusions, the number of the output nodes could become great, then, the amount of calculation also increases. So when there are a lot of testing conclusions, the testing conclusions can be encoded to binary code, then output nodes take the number greater than or equal to log2N, where N is the number of conclusions. 2.3 The Choice of the Number of Hidden Layer Nodes If the network model is BP network, the number of the hidden layer nodes needs to be set. Generally speaking, if there are too few hidden layer nodes, the network can not study well, and much more training will be needed and, meantime the training will not be highly precise; while too many nodes will bring issues such as a large amount of calculation, long training time and reduced network fault-tolerance capacity. Because there is no corresponding theoretical guidance, most designs are determined by
378
X. Liu et al.
combining the experience and trial calculation. System uses the default settings as the empirical formula (1).
n1 = n + m + 2 .
(1)
In the formula (1), “n1” is the number of the hidden layer nodes; “n” is the number of the input nodes; “m” is the number of the output nodes. If the training effect of experience value is not good, you can re-set the number of the hidden layer nodes in the parameter settings and do many experiments in order to achieve faster convergence rate. In order to know the training efficiency, you should output the training number in training process, changes of the network errors, changes of network weights and the training time and so on, so as to a make a report for the network analyses. 2.4 The Choice of Transfer Function In the application of detection, most of the outputs are the field data and test conclusions. The field data are collected by the acquisition system; detection conclusions are obtained from the neural network algorithm. The detection conclusions are indicated with two-value, so the transfer function of output layer generally uses the sigmoid function or the hard limit function. The sigmoid function formula is shown as following.
f ( x ) =
1 1 +
e
−
x
.
(2)
2.5 The Choice of Initial Weight Initial weight is generated randomly by the random function, and its value range is (-0.5, 0.5).
2.6 The Choice of the Learning Rules Because of the particularity of embedded systems, we should try to select the learning algorithm with simple calculation and low memory consumption. The perceptron network model structure is simple, and the overall amount of calculation is not large, so the weight correction algorithm of BP model is mainly considered. Conventional BP learning algorithm has large amount of calculation and long training time, and the slope of sigmoid function is close to 0 when the input is large, therefore the increase amplitude of gradient of learning algorithm is very small ,which may make the modification process of network weights almost at a standstill and the training very inefficient. To reduce the amount of calculation and to improve the efficiency, we may adopt a flexible BP learning algorithm. We only need to consider the symbol of gradient to the error function, rather than the increase amplitude of gradient. The symbol of gradient determines the direction of weight updating, and the weight size change is determined
The Application Research of Neural Network in Embedded Intelligent Detection
379
by an independent “update value” (based on the former “update value”)[2].The iterative process of weight correction example is shown below:
ω ( t + 1) = ω ( t ) + αΔω ( t ) × sgn(
∂E t ). ∂ω ( t )
(3)
In the formula (3)[2], Δω (t) is the previous update value, and the initial value Δω (0) is set by the actual application. α is the learning parameter, using variable step-size learning, that is, if it is in two successive iterations, the symbol of the partial derivative of error function to a weight remains unchangeable, then α is 1.2; otherwise α is 0.8[2]. It is faster to converge by using this algorithm, which can also be achieved effectively in embedded system.
2.7 The Preprocessing of Input and Output Data in Training Sample In the network learning process, because the transfer function of neuron is a bounded function, while in the detection, as the dimension and unit of input test item data may be different, some values are very large, while some values are very small, which has a great influence on the network.. To prevent some neurons from reaching saturation state, and meantime make the larger input fall in the region where gradient of neuron activation function is large, input and output vectors need to be normalized before the training. Namely:
x' i =
xi - xi min . xi max - xi min
(4)
In the formula (3): ximax, ximin are the maximum value and minimum value of each input component of number i neuron; xi, x'i are respectively the former and later input component after pretreatment of number i neuron. The input data values after normalization processing range between 0 and 1. Normalization processing requires a large number of mathematical operations. But if there is no huge difference among the input data values, the normalization is not required to reduce the computational complexity. Therefore, we may use whether to normalize the input data or not as the network training option parameter and make choice before the training. If the neural network training is unsuccessful, or if the training time is too long, we need to analyze the process of training and to train after readjusting the network parameters. If the training results are satisfactory, we may use the testing sample data to test the function of the network.. Successful testing demonstrates that the construction of the network is completed, and then the network parameters and the weights of each Layer of the network are saved as files, ready for testing procedures.
3 Input and Output Platform Design In the application of neural network, we need to input a large number of training data and output the training process and the training results, which requires a higher demand of input and output. To reduce the cost of portable detect device and the burden of
380
X. Liu et al.
volume, we may use two kinds of solution methods. One method is training on a host computer independently, and we download the network weights and network parameters of success training to the target computer in form of files for the detection program to recover the network; The other method is sharing the rich resources of input and output device of remote computer by using the remote training mode. In the above two methods, the training process of the former method is conducted on PC machine, which doesn’t take up resources of embedded device, and its advantage is that the high-speed PC-CPU and the large capacity memory can greatly enhance the training speed, while the disadvantage is that the neural network parameter settings and the training are independent of detection system, which is inconvenient for users to modify and update the system; the second method integrates network training and testing, completed by the target machine, with remote computer only serving as sharing input and output devices. Because of limitations of embedded CPU and memory, the training becomes slow, and even collapses when the network structure is complex, or the amount of calculation is huge. But the system is flexible, and more convenient to be maintained and updated. Because in neural network design, requirements of application have already been considered, and we have adopted more effective and simpler methods in the design of network structure and training algorithm; besides, the network training is the preliminary work of establishing detection system, and it will not be run after the network is built successfully, so its speed does not affect the work efficiency of the detection system. For these reasons, we use the second method to design. To reduce the development, maintenance and use costs, we use B / S structure in the system, embedding a small WEB server in device, all of the user interface using static or dynamic web pages and being viewed through remote browser, and the data interaction and generation of dynamic web pages being realized by using high efficient CGI programming. The overall structure of the system is shown in figure 2: ,QWHUQHW,QWUDQHW EURZVHU 1HWZRUN HTXLSPHQW 7&3,3 6LJQDO DFTXLVLWLRQ
+DUGZDUH GULYHU
(PEHGGHG V\VWHP 6&$50/,18; 1HXUDO QHWZRUN SODWIRUP
:(% VHUYHU
Fig. 2. Overall system structure diagrams
4 Conclusion In the detection system, using neural network algorithms can increase the system flexibility and self-adaptability. When testing standards or project data change, we can use new training data to train the network again and get the new network parameters for updating system rapidly. Besides, the system is based on remote detection technology, making the system updating more convenient. And that also reduces the requirements
The Application Research of Neural Network in Embedded Intelligent Detection
381
of input and output of portable detect device and the cost of hardware and software development; if the test site environment is poor, the remote detection could show more on its advantages. But the small embedded WEB server in the system may accordingly occupy more resources .Moreover, due to the inherent delay of network communication, the display of testing data and testing results have a certain delay relative to field data, so it is only suitable to the situation in which on-site environmental data are stable. If the on-site environment changes fast, you can add display components in the embedded device, which can have the devices test offline, and neural networks building and the system maintaining and updating could be conducted remotely.
References
应信号处理 经网络技术
1. Bernard, W., Samuel, D.S.: Adaptive Signal Processing( 自适 ). Machinery Industry Press, Beijing (2008) 2. Tian, Y.: Hybrid Neural Network Technology(混合神 ). Science Press, Beijing (2009)
The Theoretical Analysis of Test Result’s Errors for the Roller Type Automobile Brake Tester Jun Li, Xiaojing Zha, and Dongsheng Wu School of Mechanical and Electronic Engineering, East China Jiaotong University, Nanchang 330013, P.R. China
Abstract. The main testing parameter of the roller brake tester is the braking force. Actually, there are some differences in results even if the same vehicle is tested on the same tester. So it will bring trouble to evaluate the braking performance accurately. Based on force analysis, the mathematical model of the roller opposite force type automobile brake tester is built in this article. And then the factors of influencing braking force value will be analyzed by theoretical calculations Taking the instance of the Jianghuai light truck and using the mathematical model analyze influencing factors for testing results and calculate error value. The results showed that adhesion cofficients of between testing wheels and rollers and between non-testing wheels and floor, structure of the roller brake testor have an influece on testing results of braking force and structure of the roller brake testor is the least. The theoretical analysis provide references for comparison test among testing equipments of the braking performance. Keywords: Roller brake tester, Mathematical model, Adhesion coefficient, Braking force, Error.
1 Introduction Automobile braking performance is one of the most important factors in vehicle safety performances[1][2]. It is also a key indicator and an essential inspection item of vehicle safety performance test. Currently, the automobile braking performance is tested mostly through the roller opposite force type brake tester according to GB7258-2004 named “Safety specifications for power-driven vehicles operating on roads” However, the same car in different tester for the braking performance test, the results often not consistent, even contrary. So that, it can't evaluate the vehicle's braking performance whether reached the national standards. Based on the above situation, it is necessary to approach the influencing factors of testing results for the parameters of the testers, and establish its mathematical model of automobile testing on the roller brake tester. Thus, it can analyze the relationship of the vehicle testing results from different testers, and it can obtained more consistent evaluations of the vehicle's braking performance. D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part IV, IFIP AICT 347, pp. 382–389, 2011. © IFIP International Federation for Information Processing 2011
The Theoretical Analysis of Test Result’s Errors
383
2 Mathematical Model of Automobile Testing on the Roller Brake Tester According to the braking test of automobile on the roller tester, the force analysis is processed in two ways: force analysis of front wheels test and rears wheel test. So the mathematics theoretical models are built individually. When analyzing, it is assumed that the center of front and rear wheels is located in the same level. And the influences caused by rolling resistance and resilient wheels on force-measuring system are ignored. 2.1 Mathematical Model of Front Wheels Braking Test [3] When the automobile’s front wheels brake, the force analysis of the roller brake tester is as shown in Figure 1.
D----wheel diameter(mm); d----roller diameter(mm); α----formed angle; G1, G2----load of the front wheels and rear wheels(N);
、N ----normal force of the front and rear rollers to front wheels(N); 、F ----tangential force of the front and rear rollers to front wheels(N);
N1 Fb1
2
b2
φ1----adhesion coefficient between testing wheels and rollers;
φ2----adhesion coefficient between non-testing wheels and floor; Xb2----horizontal reaction force of floor to rear wheels(N) Fig. 1. The force analysis of front wheels braking test on the roller tester
According to the principle of mechanical equilibrium, relations can be built as below:
∑ FX = 0 Fb1 cos α + Fb 2 cos α + N1 sin α − N 2 sin α − F0 = 0
(1)
∑ FY = 0 − Fb1 sin α + Fb 2 sin α + N1 cos α + N 2 cos α − G1 = 0
(2)
With the increasing of Fb1 when the front wheels brake force is being tested, if the front wheels don’t slip backwards, the maximum value of Fb1 and Fb2 will be:
384
J. Li, X. Zha, and D. Wu
Fb1max = N1 ⋅ ϕ1 and Fb 2 max = N 2 ⋅ ϕ1 . Meanwhile, F0 = X b 2max = G2 ⋅ ϕ 2 . So Eq.1 and Eq.2 can be written as:
N1 =
G2ϕ2 (ϕ1 sin α + cos α ) − G1 (ϕ1 cos α − sin α ) (1 + ϕ12 ) sin 2α
(3)
N2 =
G2ϕ2 (ϕ1 sin α − cos α ) + G1 (ϕ1 cos α + sin α ) (1 + ϕ12 ) sin 2α
(4)
Under the premise of no moving backwards, the vehicle’s front wheels maximum brake force should be:
Fbf max = Fb1max + Fb 2 max = ( N1 + N 2 )ϕ1 = According to Eq.3, if ϕ2 =
G2ϕ12ϕ 2 + G1ϕ1 (1 + ϕ12 ) cos α
(5)
G1 (ϕ1 cos α − sin α ) then N1 = 0 , the tested wheels G2 (ϕ1 sin α + cos α )
will leave the roller and move backwards. Meanwhile we can get:
N2 =
G1 ϕ1 sin α + cos α
(6)
If the tested wheels are locked after leaving the roller tester, the maximum brake force of the front wheels is:
Fbf′ max = N 2 ⋅ ϕ1 =
G1ϕ1 ϕ1 sin α + cos α
(7)
2.2 Mathematical Model of Rear Wheel Braking Test When the vehicle’s rear wheels brake force is tested, the force analysis of the roller brake tester is as shown in Fig.2.
Fig. 2. The force analysis of rear wheel braking test on the roller brake tester
The Theoretical Analysis of Test Result’s Errors
385
Similarly, according to the principle of mechanical equilibrium, relations can be built as below:
∑ FX = 0 N1 sin α + Fb1 cos α + Fb 2 cos α − F0 − N 2 sin α = 0
(8)
∑ FY = 0 N1 cos α + N 2 cos α + Fb 2 sin α − Fb1 sin α − G2 = 0
(9)
If the rear wheels don’t slip backwards, have Fb1 max = N1 ⋅ ϕ1 , Fb 2 max = N 2 ⋅ ϕ1 and F0 = X b max = G1 ⋅ ϕ 2 . So it can be gotten that:
N1 =
G1ϕ2 (ϕ1 sin α + cos α ) − G2 (ϕ1 cos α − sin α ) (1 + ϕ12 ) sin 2α
(10)
N2 =
G1ϕ 2 (ϕ1 sin α − cos α ) + G2 (ϕ1 cos α + sin α ) (1 + ϕ12 ) sin 2α
(11)
The maximum brake force of the rear wheels is
Fbr max =
G1ϕ12ϕ2 + G2ϕ1 (1 + ϕ12 ) cos α
According to Eq.10, it can be known that N1 = 0 when ϕ2 =
(12)
G2 (ϕ1 cos α − sin α ) , G1 (ϕ1 sin α + cos α )
then
N2 =
G2
ϕ1 sin α + cos α
(13)
If the tested wheels locked after leaving the roller tester, the maximum brake force of the rear wheels is:
Fbr′ max = N 2 ⋅ ϕ1 =
G2ϕ1 ϕ1 sin α + cos α
(14)
3 Analysis on Influencing Factors of Testing Results [4] [5] 3.1 The Adhesion Coefficient between Testing Wheels and Rollers According to GB/T13564-2005 “the roller opposite force type automobile brake tester”, if vehicle is lighter than 3 tons, the roller tester’s diameter should be 245 mm and the center distance should be 430 mm. Take a light truck HFC1061KS for example, its tires model type is 7.50-16, and the individual weight of front and rear axles are 1716 kg and 1412 kg. Considering the actual testing situation, let ϕ1 = 0.6 ~ 0.95 and ϕ 2 = 0.7 . Substituting them in Eq.5, the calculation results are shown in Fig.3 as below.
386
J. Li, X. Zha, and D. Wu
4
1.6
x 10
The relationship between the value of φ1 and front and rear brake force
1.5
Fb(N)
1.4 1.3 1.2 front Fbf
1.1 1 0.6
rear F br
0.65
0.7
0.75 0.8 φ1
0.85
0.9
Fig. 3. The relation curves of the front and rear wheel braking forces with the increase of φ1
From Fig.3, it can obviously tell that the difference of braking force between ϕ1 = 0.6 and ϕ1 = 0.95 is up to about 4000 N, which reflects that the braking performance is affected by φ1. Because the rollers may be worn more or less after a long use, the value of φ1 will decrease. Regarding the front and rear brake force when ϕ1 = 0.75 as reference values, some comparisons are made as Tab.1 between
ϕ1 = 0.65 and ϕ1 = 0.75 . Table 1. The comparison of the front and rear brake force between ϕ1 = 0.65 and ϕ1 = 0.75
Front Fbf Rear Fbr
ϕ1 = 0.65
ϕ1 = 0.75
12324 N 11458 N
13488 N 12695 N
Relative error δ 8.63% 9.75%
It can be concluded that with the increase of adhesion coefficient between wheels and rollers, the braking force test capability of front and rear wheels will significantly increase as well. Hence, a good value of φ1 which must be ensured through a number of ways is a basic premise for braking force test. 3.2 The Adhesion Coefficient between Non-testing Wheels and Floor Similarly, taking the light truck and the same tester as an example, the front and rear braking force will be calculated. According to GB7258-2004, the value of φ1 should not be less than 0.75. So it is taken as 0.75 on the calculation. Considering the actual testing situation, φ2 is taken between 0.4 and 0.8. The results are calculated and shown in Fig.4.
The Theoretical Analysis of Test Result’s Errors
1.45
x 10
4
387
The relationship between the value of φ2 and front and rear brake force
1.4 1.35
b
F (N)
1.3 1.25 1.2 1.15 front F bf
1.1
rear F br
1.05 0.4
0.5
0.6
φ2
0.7
0.8
Fig. 4. The relation of the front and rear brake force with the increase of φ2
From Fig.4, we can tell that the test result differs when we take different values of the adhesion coefficient between non-testing wheels and floor. Furthermore there is a proportional relationship between the value of φ2 and braking force. The difference of braking forces between ϕ 2 = 0.4 and ϕ 2 = 0.8 is up to about 2000 N, which has a great impact on the evaluation of braking performance. In test stations, if the ground cannot be cleaned up or maintained in time, it will lead to the decrease of adhesion coefficient, and then affect the evaluation on braking performance. For example, the adhesion coefficient of terrazzo floor is less than 0.6. Regarding the front and rear brake force when ϕ 2 = 0.7 as reference values, some comparisons are given as Tab.2 between ϕ 2 = 0.7 and ϕ 2 = 0.6 . Table 2. The comparison of the front and rear brake force between ϕ 2 = 0.6 and ϕ 2 = 0.7
Front Fbf Rear Fbr
ϕ2 = 0.6
ϕ 2 = 0.7
12906 N 11988 N
13487 N 12694 N
Relative error δ 4.31% 5.56%
By the Eq.6, Eq.7, Eq.13 and Eq.14, it is known that when the φ2 takes a particular data ( N1 = 0 ) the testing braking force is minimum. It means the vehicle is slipping on the tester. That may be one reason why the vehicle with good brake performance cannot pass the test. So in the actual test, it is specified that the adhesion coefficient between non-test wheels and ground cannot be less than 0.7.
388
J. Li, X. Zha, and D. Wu
3.3 Structural Factors of the Roller Brake Tester After investigation we found that the sizes of testers which are being used in different test institutions have differences, although GB/T13564-2005 has made provisions for the standard of tester. According to the force analysis in Fig.1 and Fig.2, it can be known that the formed angle α is related to the wheel diameter D, the roller diameter d and the roller center distance L. Supposing that D is a constant value (taking the light truck as an example), we can discuss the relationship between the formed angle α and braking force through combining with the actual parameters of different testers. The formed angle α is proportional to the front and rear braking force, as shown in Fig.5.
4
1.45
x 10
The relationship between the value of α and front and rear brake force
1.4
Fb(N)
1.35 1.3
1.25 1.2 29
front Fbf rear Fbr
30
31
32 Į
33
34
35
Fig. 5. The braking force of front and rear wheels with the change of the formed angle α
Also taking the light truck as an example, applying relevant parameters of three different testers and combining Eq.5 with Eq.12, the braking force of front and rear wheels are shown as below: Table 3. The front and rear brake force of JAC light truck on the different testers
Type
d(mm)
L(mm)
α(°)
Fbf(N)
Fbr(N)
1
200
390
29.6
13293.9
12512.7
2
240
470
34.5
14027.5
13203.2
3
245
430
31.0
13487.0
12694.4
In Tab.3 the data of Type 3 is the ruled data in GB/T13564-2005. If taking them as reference values, the relative errors are -1.43% in Type 1 and 4.01% in Type 2.
The Theoretical Analysis of Test Result’s Errors
389
From Fig.5 and Tab.3, the conclusion is made: when the same vehicle is tested in different testers, the formed angles are different due to the different structures. It causes the errors in test results. Therefore, the structure of the brake tester can be an impact for the braking force test as well.
4 Conclusions By taking a light truck as an example, the mathematical model is established through force analysis. Combining the specific parameters of tester, the test results of braking force can be theoretically analyzed and calculated by the model. The conclusions can be made as below: 1. The testing results are affected by the adhesion coefficient between the tested wheels and rollers, by the adhesion coefficient between non-tested wheels and ground, and by the structure of the brake roller tester. Among them, the structure of brake tester contributes relatively less affection to the braking force. 2. Due to the difference of the structure of testers in different places, it makes some differences in braking force test results. They belong to structural errors and are unavoidable in the testing process. From Tab.3 and Fig.5 we can see that the formed angle has less influence in test results. It indicates that doing comparative tests of different test devices is feasible and the test results are comparable. 3. To reduce the test errors resulting from the structure of the testing equipment, the formed angles α can be modified through the mathematical model if they are too large or too small.
Acknowledgement The work is supported by the Department of Transportation of Jiangxi Province (2009T0053).
References 1. Zhou, D., Tao, S., Li, W.: The Establishing and Analysis of Stress Model about the Roller Brake Tester and the Flat Brake Tester. Machinery Design & Manufacture (1), 50–52 (2005) 2. Lan, Y., Hao, D.: The analysis of advantage and disadvantage between the roller brake tester and the platform brake tester. Highways & Transportation in Inner Mongolia (3), 28–30 (1998) 3. Jiao, G., Kong, X., Liu, G.: Contrastive Analysis of Static Inspection and Dynamic Inspection of Automobile Brake Performance. Journal of Northeast Forestry University 34(5), 93–94 (2006) 4. Yu, Z.: Automobile Theory. China Machine Press, Beijing (2006) 5. Huang, X.: Analyze the Key Factors that Influence the Braking Force Test. Northeast Forestry University, Ha Erbin (2006)
A Type of Arithmetic Labels about Circulating Ring Ergen Liu, Dan Wu, and Kewen Cai School of Basic Sciences, East China Jiaotong University, Nanchang, P.R. China, 330013
[email protected] Abstract. The graph composed with several rings is a kind of important and interesting graph, many scholars studied on the gracefulness of this kind of the graph, The reference [1] is given the gracefulness of
m
kinds
C 4 with one
common point. In this paper, we researched the arithmetic labels of four kinds graph:
C8,1,n , C8, 2, n , C8,3,n , C8, 4,n , and we proved they are all
(d ,2d ) --arithmetic graph. Keywords: Arithmetic graph, labeling of graph, C 8,i ,n .
1 Introduction The graph in this paper discussed are undirected, no multiple edges and simple graph, the unorganized state of definitions and terminology and the symbols in this graph referred to reference[2][3]. There are two kinds of labels of the graph: one is the reduced label, is to say that in order to get the label of one edge you should reduce the endpoints of the edge; the other is additive label, for the same you should acttive the endpoints of one edge to get the label of the edge. For example, the well-known of “Gracefulness” is reduced. the “Compatible labels” is additive. In 1990, B.D.Achaya and S.M.Hegde import the concept of “Arithmetic lables”(referred to reference[2]), which is a more extensive additive label, it have applied value on solution to question of the joint ventures in rights and obligations. Definition 1.1. For G = (V , E ) , if there is a mapping f ( called the vertices label ) from V (G ) to the set of nonnegative integer N 0 , meet:
(1) f (u ) ≠ f (v)
,which u ≠ v ,and u,v ∈ V (G) ;
, , ,k + (q − 1)d }.
( 2) {f (u ) + f (v) uv ∈ E (G )} = {k k + d
,
Then we call graph G is a ( k d ) --arithmetic graph. D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part IV, IFIP AICT 347, pp. 390–397, 2011. © IFIP International Federation for Information Processing 2011
v of the
A Type of Arithmetic Labels about Circulating Ring
391
2 Main Results and Certification Theorem 2.1. C8,1,n is a ( d ,2d ) --arithmetic graph. Proof. As the graph shown on fig.1. v2(0)
v2(1)
v6(0)
v4(0)
v4(1)
v(2n−1)
v6(1)
v1(0) v1(1) v3(0)
v7(0)
v5(0)
v1(2) v3(1)
v1(n) v(3n−1)
Fig. 1. The Graf
v(6n−1)
v1( n−1)
v7(1)
v5(1)
v(4n−1)
v(7n−1)
v(5n−1)
C8,1,n
Label all vertices as follows:
f (v1(i ) ) = 8id (i = 0,1,2,
, n) ,
f (v 2( i ) ) = 8id + d (i = 0,1,2,
, n − 1) ,
f (v 3(i ) ) = 8id + 3d (i = 0,1,2,
, n − 1) ,
f (v 4( i ) ) = 8id + 6d (i = 0,1,2,
, n − 1) ,
f (v 5(i ) ) = 8id + 2d (i = 0,1,2,
, n − 1) ,
f (v 6( i ) ) = 8id + 5d (i = 0,1,2,
, n − 1) ,
f (v 7(i ) ) = 8id + 7d (i = 0,1,2,
, n − 1) .
Now we proof that the mapping We can see the mapping
{
f is arithmetic labeling of C8,1,n .
f meet f (u ) ≠ f (v) which u ≠ v and u, v ∈ V (C8,1,n ) .
Next we prove f (u ) + f (v ) uv ∈ E (C8,1, n )} is an arithmetic progression in the way of mathematical induction. When n = 1 Then f (v1( 0 ) ) = 0 , f (v 2( 0) ) = d , f (v 3( 0) ) = 3d , f (v 4( 0) ) = 6d , f (v 5( 0) ) = 2d ,
f (v 6( 0) ) = 5d , f (v 7( 0) ) = 7 d , f (v1(1) ) = 8d .
{
Therefore f (u ) + f (v) uv ∈ E (C8,1,1 )}
{
= f (v1( 0 ) ) + f (v 2( 0) ), f (v1( 0) ) + f (v3( 0 ) ), f (v3(0) ) + f (v5( 0) ),
f (v 2(0 ) ) + f (v 4( 0) ), f (v5( 0) ) + f (v7( 0) ), f (v 4(0 ) ) + f (v6( 0) ), f (v 6( 0) ) + f (v1(1) ), f (v 7( 0) ) + f (v1(1) )}
,,,,, , ,
= {d 3d 5d 7d 9d 11d 13d 15d } is an arithmetic progression, and the common difference is 2 d .
392
E. Liu, D. Wu, and K. Cai
Suppose when n = k − 1 , we know
{f (u) + f (v) uv ∈ E (C
} = {d , d + 1 × 2d , d + 2 × 2d ,
, d + (8k − 9) × 2d }
8,1, k −1 )
is an arithmetic progression, and the common difference is 2 d . Then when n = k
{f (u) + f (v) uv ∈ E (C = {f (u ) + f (v) uv ∈ E (C
8,1, k
)}
} ∪ {f (v1( k −1) ) +
8,1, k −1 )
f (v 2( k −1) ), f (v1( k −1) ) + f (v3( k −1) ),
( k −1) ( k −1) f (v3( k −1) ) + f (v5( k −1) ), f (v 2 ) + f (v 4 ), f (v5( k −1) ) + f (v7( k −1) ), ( k −1) (k ) f (v 4( k −1) ) + f (v6( k −1) ), f (v6( k −1) ) + f (v1( k ) ), f (v 7 ) + f (v1 )}
= {d , d + 1 × 2d , d + 2 × 2d ,
, d + (8k − 9) × 2d } ∪ {d + (8k − 8) × 2d ,
d + (8k − 7) × 2d , d + (8k − 6) × 2d , d + (8k − 5) × 2d , d + (8k − 4) × 2d , d + (8k − 3) × 2d , d + (8k − 2) × 2d , d + (8k − 1) × 2d }
,
,
= {d d + 1 × 2d d + 2 × 2d
, ,d + (8k − 1) × 2d}
is an arithmetic progression, and the common difference is 2d . In sum for the arbitrary n ∈ N 0 , the mapping f : V (C 8,1,n ) → N 0 meet: (1) f (u ) ≠ f (v) when
{
u ≠ v and u, v ∈ V (C8,1,n ) ;
,
,
, ,d + (8n − 1) × 2d}.
(2) f (u ) + f (v) uv ∈ E (C8,1,n )} = {d d + 1 × 2d d + 2 × 2d Therefore C8,1, n is a ( d ,2d ) --arithmetic graph. Theorem 2.2. C8, 2,n is a ( d ,2d ) --arithmetic graph.
Proof. As the graph shown on fig.2. v 3( 0 )
v 5( 0)
v 3(1)
v1( 0)
v1(1 )
v2( 0)
v2(1 ) v4( 0 )
v 6( 0)
v (3n−1)
v 5(1)
v (2n−1)
v2( 2)
v (4n−1)
v 6(1)
Fig. 2. The Graf
v1( n )
v1( n−1)
v1( 2)
v4(1)
v (5n−1)
C 8, 2 , n
v2( n ) v (6n−1)
A Type of Arithmetic Labels about Circulating Ring
393
Label all vertices as follows:
f (v1(i ) ) = 7id (i = 0,1,2, f
(v 2(i ) )
, n) ,
= 7id + d (i = 0,1,2,
, n) ,
f (v3(i ) ) = 7id + 3d (i = 0,1,2,
f (v 4(i ) ) = 7id + 6d (i = 0,1,2,
, n − 1) ,
= 7id + 2d (i = 0,1,2,
, n − 1) ,
f
, n − 1) ,
(v5(i ) )
f (v 6(i ) ) = 7id + 5d (i = 0,1,2,
, n − 1) .
Now we proof that the mapping f is arithmetic labeling of C8, 2 ,n We
can
see
u , v ∈ V (C 8 , 2 , n ) .
the
mapping
f
meet
f (u ) ≠ f (v) which u ≠ v and
{
Next we prove f (u ) + f (v) uv ∈ E (C8, 2, n )} is an arithmetic progression in the
way of mathematical induction. When n = 1 Then f (v1( 0) ) = 0 , f (v 2( 0 ) ) = d , f (v3( 0) ) = 3d , f (v 4( 0 ) ) = 6d , f (v 5( 0 ) ) = 2d ,
f (v 6( 0) ) = 5d , f (v1(1) ) = 7 d , f (v 2(1) ) = 8d .
{
Therefore f (u ) + f (v) uv ∈ E (C8, 2,1 )}
{
= f (v1( 0) ) + f (v 2( 0) ), f (v1(0) ) + f (v3( 0) ), f (v3(0 ) ) + f (v5( 0 ) ),
f
(v 2(0 ) )
+ f (v 4( 0) ), f (v5( 0) ) + f (v1(1) ), f (v 4(0 ) ) + f (v6( 0) ), f (v 6( 0) ) + f (v 2(1) ),
f (v1(1) ) + f (v 2(1) )}
,,,,, , ,
= {d 3d 5d 7 d 9d 11d 13d 15d }
is an arithmetic progression, and the common difference is 2d . Suppose when n = k − 1 , we know
{f (u) + f (v) uv ∈ E (C
} ={d , d + 1 × 2d , d + 2 × 2d ,
8, 2 ,k −1 )
, d + (7k − 7) × 2d }
is an arithmetic progression, and the common difference is 2d . Then when n = k
{f (u) + f (v) uv ∈ E (C = {f (u ) + f (v) uv ∈ E (C
8, 2 , k
)}
} ∪ { f (v1( k −1) ) +
8, 2 , k −1 )
f (v3( k −1) ),
f (v3( k −1) ) + f (v5( k −1) ), f (v 2( k −1) ) + f (v 4( k −1) ), f (v5( k −1) ) + f (v1( k ) ), f (v 4( k −1) ) + f (v 6( k −1) ), f (v 6( k −1) ) + f (v 2( k ) ), f (v1( k ) ) + f (v 2( k ) )}
394
E. Liu, D. Wu, and K. Cai
d + (7k − 7) × 2d } ∪ {d + (7k − 6) × 2d ,
= {d , d + 1 × 2d , d + 2 × 2d ,
d + (7 k − 5) × 2d , d + (7k − 4) × 2d , d + (7k − 3) × 2d , d + (7k − 2) × 2d , d + (7 k − 1) × 2d , d + 7 k × 2d }
,
,
= {d d + 1 × 2d d + 2 × 2d
, ,d + 7k × 2d }
is an arithmetic progression, and the common difference is 2d . In sum for the arbitrary n ∈ N 0 , the mapping f : V (C8, 2,n ) → N 0 meet: (1) f (u ) ≠ f (v) when u ≠ v and u, v ∈ V (C8, 2 ,n ) ;
,
{
,
(2) f (u ) + f (v) uv ∈ E (C 8,2,n )} = {d d + 1 × 2d d + 2 × 2d
, ,d + 7n × 2d } .
Therefore C8, 2,n is a (d ,2d ) --arithmetic graph. Theorem 2.3. C8,3,n is a (d ,2d ) --arithmetic graph .
Proof. As the graph shown on fig.3. v1(0)
v2(0)
v 0( 0)
v1(1)
v2(1)
v 0(1)
v 3( 0)
v4(0)
v 3(1)
v1( n−1)
v1(2)
v1(n )
v (0n−1)
v 0( 2)
v4(1)
v (2n−1)
v (3n−1)
v 3( 2)
Fig. 3. The Graf
v 0( n )
v (4n−1)
v 3( n )
C 8, 3, n
Label all vertices as follows:
f (v0( i ) ) = 8id (i = 0,1,2,
, n) ,
f (v3( i ) ) = 4id + 3d (i = 0,1,2,
, n) ,
f (v 4(i ) ) = 8id + 2d (i = 0,1,2,
, n − 1) .
f (v1( i ) ) = 4id + d (i = 0,1,2,
, n) ,
f (v 2( i ) ) = 8id + 6d (i = 0,1,2,
, n − 1) ,
Proof in imitation of Theorem 2.1. For the arbitrary n ∈ N 0 , the mapping f : V (C8,3, n ) → N 0 meet: (1) f (u ) ≠ f (v) when u ≠ v and u, v ∈ V (C8,3,n ) ;
{
,
,
, ,d + (6n + 1) × 2d} .
(2) f (u ) + f (v) uv ∈ E (C8,3,n )} = {d d + 1 × 2d d + 2 × 2d
A Type of Arithmetic Labels about Circulating Ring
395
By (1), (2) we know the mapping f is arithmetic labeling of C 8,3, n . Therefore
C 8,3, n is a (d ,2d ) --arithmetic graph. Theorem 2.4. C8, 4, n is a (d ,2d ) --arithmetic graph.
Proof. As the graph shown on fig.4. v 2( 0 )
v 2(1 )
v (2 n − 1 )
v 2( 2 )
v 2( n )
v 1( 0 )
v 1(1 )
v 1( 2 )
v 1( n − 1 )
v 1( n )
v 3( 0 )
v 3( 1 )
v 3( 2 )
v
( n−1) 3
v 3( n )
v 4( 0 )
v 4(1 )
v (4 n − 1 )
v 4( 2 )
Fig. 4. The Graf
v 4( n )
C 8, 4 , n
Label all vertices as follows:
f (v1(i ) ) = 5id (i = 0,1,2,
f (v 2( i ) ) = 5id + d (i = 0,1,2,
, n) ,
f (v3(i ) ) = 5id + 3d (i = 0,1,2,
, n) ,
f (v 4(i ) ) = 5id + 2d (i = 0,1,2,
, n) , , n) .
Proof in imitation of Theorem 2.2. For the arbitrary n ∈ N 0 , the mapping f : V (C 8, 4,n ) → N 0 meet: (1) f (u ) ≠ f (v) when u ≠ v and u, v ∈ V (C8, 4,n ) ;
{
,
,
, ,d + (5n + 2) × 2d}.
(2) f (u ) + f (v ) uv ∈ E (C8, 4,n )} = = {d d + 1 × 2d d + 2 × 2d
By (1), (2) we know the mapping f is arithmetic labeling of C8, 4, n . Therefore
C 8, 4, n is a (d ,2d ) --arithmetic graph.
3 Labels of Some Special Graph In order to explan the correctness of the aforementioned labels, we give the arithmetic labeling of C8,1,3 C8, 2, 3 C8, 3,3 and C8, 4,3 .
396
E. Liu, D. Wu, and K. Cai
(1) The arithmetic labeling of C8,1,3 on fig.5.
d
6d
5d
9d
14d
17d
13d
22d
21d
0
24d 8d 3d
2d
16d 11d
7d
10d
15d
Fig. 5. The Graf
19d
23d
18d
C8,1,3
(2) The arithmetic labeling of C8, 2,3 on fig.6.
3d
2d
10d
9d
17d
16d
0
7d
14d
21d
d
8d
15d
22d
6d
5d
13d
12d
Fig. 6. The Graf
20d
19d
C8, 2 , 3
(3) The arithmetic labeling of C8,3,3 on fig.7.
d
6d
0
5d
14d
8d
3d
2d
7d
9d
22d
16d
10d
Fig. 7. The Graf
13d
24d
11d
C8,3,3
18d
15d
A Type of Arithmetic Labels about Circulating Ring
397
(4) The arithmetic labeling of C8, 4,3 on fig.8.
d
6d
11d
16d
0
5d
10d
15d
3d
8d
13d
18d
2d
7d
12d
Fig. 8. The Graf
17d
C8, 4 , 3
References [1] Ma, K.: Gracefulnss. The press of technology, Beijing (1991) [2] Chartrand, G., Lesniak, L.: Graphs and Digraphs. Wadsworth and Brooks/Cole, Monterey (1996) [3] Bondy, J.A., Murty, U.S.R.: Graph Theory with Applications. Elsevier North-Holland (1976) [4] Liu, E.-g., Wu, D., Cai, K.-w.: Two Graphs Arithmetic Labeling. Journal of East China Jiaotong University 26(5), 89–92 (2009) [5] Achaya, B.D., Hegde, S.M.: Arithmetic graphs. Journal of Graph Theory 18(3), 275–299 (1990)
Application of CPLD in Pulse Power for EDM Yang Yang and Yanqing Zhao School of Mechanical and Electronical Engineering, East China Jiaotong University, Nanchang 330013, China
[email protected] Abstract. In order to improve the precision and surface quality of Electrical Discharge Machining (EDM), the paper studies the application of complex programmable logic device (CPLD) used in pulse power for EDM, according to the characteristics of the device, using VHDL language input and schematic input method to design control circuit for EDM pulse power. Keywords: EDM, CPLD, VHDL language, pulse power.
1 Introduction EDM is a special processing method, which makes use of the continuous pulse spark discharge generated by two poles in working fluid, relying on each discharge generates partial and instantaneous high-temperature to ablate down metallic material gradually, cutting into necessary shapes, it is also called as discharge machining or electric erosion processing [1]. Pulse power supply as an important part of machine tools of EDM, provides the required breakdown voltage for processing media, and provides energy to ablate metal after breakdown, its performance is well or not will determine the stability of processing equipment and the height of production efficiency [2, 3]. With the development of EDM application technology, the research on technology of pulse power is more and more in-depth. In the 1980s, it advanced the high and low pressure complex pulse and combshaped pulse (pulse group). Subsequent, it advanced the controlled current rising edge pulse and equal energy pulse, and nowadays has proposed to a new technology which can achieve single discharge pulse detection. Now people study of the pulse power variable parameters more refined, parameter adjustment and digital level higher. In this case, if it continues to use discrete components or small and medium-scale integrated circuits as the basic original, obviously doesn’t meet the design requirements of power nowadays with high integrated, fast changing, good controllability. From 80 to 90 during the 20th century made the high and low pressure complex, comb-shaped pulse (pulse packet), the subsequent rising edge of control current, such as the energy pulse, and has proposed to achieve a single discharge pulse detection technology, people Pulse variable parameters of a more detailed, parameter adjustment and further quantify the number of higher degree. In this case, the EDM pulse power supply circuit such as to continue to use discrete components and small and mediumscale integrated circuits as the basic original, obviously does not meet today's power highly integrated, fast changing, good controllability of the design requirements. D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part IV, IFIP AICT 347, pp. 398–402, 2011. © IFIP International Federation for Information Processing 2011
Application of CPLD in Pulse Power for EDM
399
2 Characteristics of CPLD Devices The first programmable logic device - PLD was produced in the 1970s, and it’s output structure is programmable logic macrocell, because the design of hardware architecture was completed by the software, which like that workers designed the interior structure after the house completed, so its design is more flexible than pure software on digital circuits. But its structure is too simple so that they only can achieve smaller circuits. To make up for the flaw that PLD only can design smallscale circuit, in the middle of 1980s, the complex programmable logic device was invented– CPLD [4, 5]. With programming flexibility, high integration, design and development period is short, wide scope of application, development tools is advanced, low costs of design and manufacturing, the experience of the designers requirements low, standard products without testing, confidentiality, price moderate, etc, and it can realize largescale circuit design, so the complex programmable logic device – CPLD is widely used in prototype of product design and production (In general under 10,000 pieces). Almost all the situation where small scale general digital circuits were applied can applied CPLD device. In recent years, because advanced integration process was Adopted, CPLD device was produced in quantity and its costs continue to drop. Integrated density, velocity, and has been greatly improved. And its integrated density, velocity, capability has been greatly improved. Because it has a lot of advantages, such as high reliability, excellent electromagnetic compatibility (EMC) and micro-power consumption, CPLD has been widely used in the field of numerical control. What is particularly important is that CPLD has the function of online programming (ISP) (according to different requirements, just a software reprogramming, you can complete the online programming), which makes function update and system debugging more convenient and greatly shorten the product development cycle, also it is great convenience to design and modify.
3 EDM Pulse Power Circuit Structure This pulse power takes CPLD and MOSFET as the core component. Through the programming on the CPLD, it can realize the steering impulse adjustment. This pulse takes the driving of circuit applied to the MOSFET as the power switch, and provides pulse power for the processing circuit. The power schematic is shown in Figure 1. CPLD can construct digital logic integrated circuits for users’ own needs, and it is suitable to be used to realize each kind of operation and combinatory logic. Furthermore, it can achieve large-scale circuit design for its many kinds of characteristics, such as programming flexibility, high integration, advanced develop kit, low design cost and so on. ALTERA’s MAXII series chips are high speed and support for the global maximum input clock frequency up to 304 MHz. By using the chip to carry on the pulse control signal disposition, it can output the disposition information for the external instrumentation, simultaneously may receive the superior information.
400
Y. Yang and Y. Zhao
Fig. 1. Schematic diagram of pulse power
4 CPLD Devices of Control Circuit for Pulse Power Design Pulses control program principle framework is shown in Figure 2. It set the required time parameters for pulse high and low, mainly through the multiplexer time-sharing operation to the next level. The multiplexer selection is carried out by the output adjusting. When time parameters through the multiplexer, it immediately starts the counter to count, and loop checks whether the given parameters are equal. If they are equal, the program sends a corresponding pulse to the retainer, and triggers the inverter for inverting operation. Pulses are from the inverter output.
Fig. 2. Pulses control program principle framework
Application of CPLD in Pulse Power for EDM
401
Pulses control program flow chart is shown in Figure 3. Pulse parameters on-time and off-time represent the pulse width and pulse interval and set by the DIP switch, according to the value of signal sel, put the values of on-time and off-time to variables time-count. If time-count is 0, then signal sel is negated, and sel takes counter-action in the next clock, at the same time take anti-variable temp, and the value assigns to the output signal out-value. If time-count is not 0, then enter the next statement. Every clock cycle, the variable Counter plus one, and compared with the variable timecount. The role of variable Counter is allocating time as a counter with the given pulse parameters. When Counter equals time-count, input signal sel is negated, and initialize the Counter, then take anti-signal temp to output values.
Fig. 3. Pulses control program flow chart
In the VHDL design, two common objects are signal and variable [6], signal is a means of date exchange in the design entity, using signal objects can connect design entity to form module, representative of a hardware circuit in the hardware connection, sometimes the signal will be integrated into register, while variable mainly stores temporary data locally, it is a local variable. Within a process, signal processing is delayed, namely only after the arrival of the next clock for signal processing, but the variable processing is real-time. Therefore, after entering the process, the signal of On-time and Off-time need to be converted to variable form. Variable of Time-Count judging and variable of Count computing are real-time. When Time-Count equals 0, that is, the high pulse or low pulse is a clock cycle, and the program directly jumps out from process. The minimum output pulse increased to a clock cycle by the pulse entering output value .after using QUARTUSII simulation, waveform is shown in Figure 4.
402
Y. Yang and Y. Zhao
Fig. 4. Pulses width 10ns and inter-pulse 50ns simulation graph
CPLD uses the source crystal oscillator to provide clock pulses, taking into account the conduction ability of MOSFET .we choose 100M crystal which can provide minimum pulse period of 10ns.
5 Conclusions The designed control circuit has many advantages, such as stable performance, strong anti-interference ability, precise pulses, good scalability, fully embodies the characteristics of CPLD devices. The CPLD devices applied to the design of EDM pulse power, improved the flexibility of circuit design, reduced PCB area, increased power system reliability, shortened product development cycle, and reduced design costs and application costs, it can meet the increasingly complex electronic processing equipment control circuit applications.
References [1] Liu, J., Zhao, J., Zhao, W.: Special machining, 4th edn., pp. 7–61. Mechanical Industry Press (2004) [2] Gao, C., Song, X., Liu, Z.: Research on a Micro Energy Pulse Power Supply for MicroEDM. Aviation Precision Manufacturing Technology, 9–11 (1997) [3] Han, F., Yamada, Y., Kawakami, T., Kunieda, M.: Experimental Attempts of SubMicrometer Order Size Machining Using Micro-EDM. Precision Engineering 30(2), 123– 131 (2005) [4] Liao, Y., Lu, R.: CPLD digital circuit design. Tsinghua University Press, Beijing (2001) [5] Xue, Z.: Application of CPLD in pulse power of EDM. Electromachining & Mould (1), 20–21 (2003) [6] Zhao, S., Yang, F., Liu, J.: VHDL and computer interface design. Tsinghua University Press, Beijing (2004)
Application of IDL and ENVI Redevelopment in Hyperspectral Image Preprocessing Long Xue School of Mechanical and Electronical Engineering, East China JiaoTong University, Nanchang 330013, China
[email protected] Abstract. In the last few years, hyperspectral imaging technique has had a bright future for application on nondestructive detection of agricultural products. However, in experiment, there are usually dozens or even hundreds of hyperspectral images. It needs to spend lots of time in preprocessing all the hyperspectral data. Software ENVI is a kind of image processing software, although the users can define their own algorithms and apply to the opened bands in ENVI or the bands and spectra operation of the images, it doesn’t provide program to batch process. In this paper, the custom function “normalized” of Interactive Data Language (IDL) is applied to achieve batch processing on multiple hyperspectral images. Keywords: ENVI, redevelopment, hyperspectral image.
1 Introduction Software ENVI (The Environment for Visualizing Images) is image processing software. It is developed by the sciences of remote sensing using Interactive Data Language (IDL). It combines the latest spectral image processing and image analysis technology with an intuitive, user-friendly interface to help you get meaningful information from imagery. It can fast, conveniently, correctly extract information from geography and special image data. Therefore, it has been widely used in science research, environment protection, meteorology, agriculture, mineral and petroleum exploration. Since hyperspectral image combines the character of image and spectrum, so it is applied in agricultural product quality, food safety and nondestructive detection. At home, Hong Tiansheng et al. built prediction model of Chinese pear quality based on hyperspectral imaging technique and artificial neural network [1]. Zhao Jiewen et al. detected the subtle bruises on apple using hyperspectral image [2]. Liu Muhua et al. nondestructively detected soluble solids content and contamination on navel orange surface by hyperspectral laser-induced fluorescence imaging [3-4]. At abroad, M. S. Kim et al. detected fecal contamination on apple, cantaloupes using hyperspectral fluorescence imagery [5-8]. Juan Xing et al. detected surface defects on tomatoes by a hyperspectral imaging system [9]. In experiment, there are usually tens to hundreds of acquired hyperspectral images, and the images need to be preprocessed. Although ENVI software offers interactive computing functions of spectral bands and spectrum, the same preprocessing of hundreds of images takes considerable time and is a very boring job. D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part IV, IFIP AICT 347, pp. 403–409, 2011. © IFIP International Federation for Information Processing 2011
404
L. Xue
This article uses IDL to write a batch program module based on ENVI, which can achieve the normalization preprocessing of the spectral data.
2 IDL and ENVI Redevelopment 2.1 IDL Introduction IDL is the ideal, timesaving solution for data analysis, data visualization, and software application development. The users can quickly and easily use this software to convert the data to images. The converted images can be simple color, full color or threedimensional images. As ENVI software provides the function interface of IDL, so the users can use the built-in IDL functional components, IDL users function, or a custom program. However, these functions must be kept in a directory of ENVI path list to be automatically compiled. 2.2 Spectral Data Preprocessing Because the uneven distribution of light source intensity under the different wave bands and the presence of dark current noise of camera, so the images may contain more noises in some wave bands, therefore, the hyperspectral images must be normalized. Under the same experimental condition, the image of a stand white plate was acquired. The dark image was acquired by completely covering the lens with its cap. The normalization preprocessing is achieved according to equation (1):
I norm = Where:
I sample − I dark I reference − I dark
(1)
I norm ——the pixel value of the image after normalized preprocessing; I sample ——the pixel value of a sample image; I reference ——the pixel value of standard white plate;
I dark ——the pixel value of the dark image. 2.3 ENVI Redevelopment The most common method of ENVI software redevelopment is using IDL to develop small application program modules, and call these modules. Meantime, these modules must be stored in the directory “X:\RSI\IDLXX\products\enviXX\save_add” to be compiled automatically. In this article, the hyperspectral files to be processed were saved in the directory “X:\original\”, and a total of n hyperspectral files were named according to numerical order (such as DH1.raw, DH2.raw, ..., DHn.raw). In the folder,The standard white plate hyperspectral image (white.raw) and the dark hyperspectral image (black.raw) were saved in the same folder. After preprocessing, the hyperspectral files were stored in the directory “x: \ normalized \ folder”, the program code is as follows:
Application of IDL and ENVI Redevelopment in Hyperspectral Image Preprocessing
405
; Building the patch processing program module of hyperspectral images “normalized”. The function is to normalize the original images. pro normalized ; Setting the path of hyperspectral files to be processed. inpath=' X:\original\' ; Setting the path of hyperspectral files having been processed. outpath=' X:\normalized\' ; Defining the array “filename” which was used to store the file name of dark image and standard white plate image. filename = ['black.raw', 'white.raw'] ; read the hyperspectral files, L represents the data is numerical type. for j=1L,nL do begin ; Converting data from numerical type to character type num=string(format='(i0)',j) ; Saving the path and file name of a specific hyperspectral file with “in_name” in_name0=inpath+num+'.raw' ; Printing the file name in IDL screen to express the processing course print,in_name0 ; Opening the file, option “no_realize” represents that the file will not be opened in available band list, and “fid” is the number of the file. envi_open_file, in_name0,/no_realize, r_fid=fid ; Opening the dark and standard white plate images. in_name1=inpath+filename[0] in_name2=inpath+filename[1] envi_open_file, in_name1,/no_realize,r_fid=fid2 envi_open_file, in_name2,/no_realize,r_fid=fid3 ; Accessing file’s information, “ns” represents the sample number of the file, “nl” represents the number of lines, “nb” represent the number of bands, “xstart” and “ystart” represents the pixel
406
L. Xue
coordinates of the top left corner in the image and the default value is (1,1). envi_file_query, fid, interleave=interleave,ns=ns, nl=nl, nb=nb,dims=dims, data_type=_datatype, xstart=_xstart, ystart=_ystart,SPEC_NAMES=SPEC_NAMES ; “dims” is the spectral range to be processed. “Ssn” represents the first file, and “esn” represents the last file. “sln” represents the first line of the file, and “eln” is the last line. pos=lindgen(nb) dims=[-1l,ssn,esn,sln,eln] envi_file_query, fid2, interleave=interleave2 envi_file_query, fid3, interleave=interleave3 ; Setting the command blocks. tile_id=envi_init_tile(fid,pos,num_tiles=num_tiles, interleave=interleave>1,xs=dims[1],xe=dims[2],ys=di ms[3],ye=dims[4]) tile_id2=envi_init_tile(fid2,pos,num_tiles=num_tiles, interleave=interleave>1,xs=dims[1],xe=dims[2],ys=di ms[3],ye=dims[4]) tile_id3=envi_init_tile(fid3,pos,num_tiles=num_tiles, interleave=interleave>1,xs=dims[1],xe=dims[2],ys=di ms[3],ye=dims[4]) ; Processing each block data, and saving the file in the directory “x:\normalized\” openw, _unit, outpath+num+'.raw', /get_lun for i=0L, num_tiles-1 do begin data = envi_get_tile(tile_id, i) data2 = envi_get_tile(tile_id2, i) data3 = envi_get_tile(tile_id3, i) data4=(float(data)-float(data2))/( float(data3)float(data2)) writeu, _unit,data4 endfor free_lun, _unit ; Building the head file.
Application of IDL and ENVI Redevelopment in Hyperspectral Image Preprocessing
407
envi_setup_head, fname=outpath+aa, ns=dims[2]dims[1]+1,nl=dims[4]-dims[1]+1, nb=nb,data_type=4, offset=0, interleave=interleave,xstart=_xstart+dims[1], ystart=_ystart+dims[3],descrip=' Normalized file', /write ; Releasing the command block. envi_tile_done, tile_id envi_tile_done, tile_id2 envi_tile_done, tile_id3 envi_file_mng,id=fid,/remove envi_file_mng,id=fid2,/remove envi_file_mng,id=fid3,/remove ; Printing the number of a file which has been processed. print,num close,/all,/force endfor envi_batch_exit end The image of pear at 680nm before preprocessing is shown in Figure 1. From figure 1, the obvious strip in the image is found easily. Figure 2 show the image of pear at 680nm after preprocessing, the quality of image is improved greatly.
Fig. 1. Image of pear at 680nm before preprocessing
408
L. Xue
Fig. 2. Image of pear at 680nm after preprocessing
Figure 3 show the average visible and near-infrared spectra in the 500 to 950nm region of the pear, (a) before preprocessing, (b) after preprocessing.
(a) Before preprocessing
(b) After preprocessing Fig. 3. The average visible and near-infrared spectra in the 500 to 950nm region of the pear
Application of IDL and ENVI Redevelopment in Hyperspectral Image Preprocessing
409
3 Conclusion By using IDL language and ENVI redevelopment, an extensible remote sensing system with perfect functions can be developed. Nowadays, this system is further expansion because more and more research achievements are integrated in it by scientific researchers. The strong abilities of IDL language and ENV redevelopment make the scientific researchers can release from boring programming work to invest more time and effort to actual research.
References [1] Hong, T., Qiao, J., Wang, N., et al.: Non-destructive inspection of Chinese pear quality based on hyperspectral imaging technique. Transactions of the CSAE 23(2), 151–155 (2007) [2] Zhao, J., Liu, J., Chen, Q., Saritporn, V.: Detecting Subtle Bruises on Fruits with Hyperspectral Imaging. Transactions of the Chinese Society for Agricultural Machinery 1(31), 106–109 (2008) [3] Guo, E., Liu, M., Zhao, J., Chen, Q.: Nondestructive Detection of Sugar Content on Navel Orange with Hyperspectral Imaging. Transactions of the Chinese Society for Agricultural Machinery 39(05), 91–93, 103 (2008) [4] Xue, L., Li, J., Liu, M.: Detecting Pesticide Residue on Navel Orange Surface by Using Hyperspectral Imaging. Acta Optica Sinica 28(12), 2277–2280 (2008) [5] Kim Moon, S., Lefcourt, A.M., Chen, Y.-R.: Ns-scale time-resolved laser induced fluorescence imaging fordetection of fecal contamination on apples. In: SPIE 2004, vol. 5587, pp. 190–197 (2004) [6] Vargas, A.M., Kim, M.S., Tao, Y., Lefcourt, A., Chen, Y.-R.: Safety Inspection of Cantaloupes and Strawberries Using Multispectral Fluorescence Imaging Techniques. ASAE Paper, No. 043056. St. Joseph. ASAE, Mich. (2006) [7] Kim, M.S., Lefcourt, A.M., Chen, Y.-R.: Automated detection of fecal contamination of apples based on multispectral fluorescence image fusion. J. Food Engineering 71, 85–91 (2005) [8] Lee, K-J., Kang, S., Kim, M.S.: Hyperspectral imaging for detecting defect on apples. ASABE Paper No. 053075. St. Joseph. ASABE, Mich. (2005) [9] Xing, J., Ngadi, M., Wang, N.: Wavelength Selection for Surface Defects Detection on Tomatoes by Means of a Hyperspectral Imaging System. ASAE Paper, No. 063018 St. Joseph. ASAE, Mich. (2006)
Design of Integrated Error Compensating System for the Portable Flexible CMMs Qing-Song Cao, Jie Zhu, Zhi-Fan Gao, and Guo-Liang Xiong College of Mechanical and Electrical Engineering, East China Jiaotong University, Nanchang 330013, China
Abstract. During the working of portable flexible CMM, the mechanical arm’s gravity and operating load generate dynamic strain, the thermal distortion also results in its length change. The caused errors would severely deteriorate the measuring accuracy. An error compensating system of the mechanical arm is designed in this paper, based on the monitor of dynamic strain and environmental temperature. The mathematical model of portable flexible coordinate measuring machine (CMM) was established.For the dynamic strain, four resistance strain gauges are sticked symmetrically on the root of each mechanical arm to monitor the dynamic strain. Then length error caused by the measuring arm’s gravity or external operation load can be calculated through the analysis of stress. Strain compensation can be realized, when they are fed back to the measurement procedures. For thermal distortion, the temperature values acquired by the single-linear digital thermometer are substituted in the correction formula to calculate the relevant compensation dosage. Finally, these compensation dosages are feedbacked to the measuring program to have these errors compensated continually. The integrated error compensation system can improve the measurement accuracy of portable flexible CMM effectively. Keywords: flexible arm, coordinates measuring machine, strain, temperature, error compensation.
1 Introduction The traditional orthogonal coordinate measuring machines (CMMs) have the advantage of simple kinematical model and high measuring accuracy. However, with the development of the modernization of industry and manufacture, the limitations of traditional perpendicular CMMs also increasingly exposed to us, which can not meet the application requirements in many cases. The flexible CMMs connect arm through revolting joint by turn, which is a new type of multiple degree and non-Cartesian system. It has the advantage of smaller volume, lighter weight, larger measuring range, portable and barrier-free measure compared with the traditional orthogonal CMMs. It is particularly suitable for the on-site mobile measure and it has gradually become the developing trend of the measurement field. However, due to its totally new structure, the mathematical modeling procedure is no longer simple on the one hand. On the other hand, the error of each joint will be seriously magnified by the measuring arm, and it will seriously affect the measurement accuracy. Therefore detecting and compensating for this error is very necessary. D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part IV, IFIP AICT 347, pp. 410–419, 2011. © IFIP International Federation for Information Processing 2011
Design of Integrated Error Compensating System for the Portable Flexible CMMs
411
For flexible CMMs, the rotation angle of each joint, along with the mechanical parameters including length of each measuring arms and offset of every two joints etc, are substituted the kinematical model built by D-H coordinate transform method. Then the measurement of 3-D coordinates can be realized. Therefore, the precision of 3-D coordinates is directly influenced by the angle error of each joint and the length error of every mechanical arm, to ensure the measuring accuracy error compensation system must be estblished. Ye et al. [1, 2] considered error of the structural parameters during the machining and assembling process as the largest error source, and correct ideal model of the CMMs to get its error model. In this way, the measurement accuracy is greatly improved. Wang et al. [3] analyzed various factors that affect measurement results in detail, and consider that the processing and assemble error, the environmental temperature and strain, etc. affect the measurement precision of the CMMs. Wang [4] analyzed the impacts on the measuring results caused by external environment such as temperature, humidity and external power, and presented several measures that should be taken to reduce them.
2 The Principle of Compensation for Integrated Error 2.1 Mathematical Model of the CMM Portable CMM adopts the whole serial revolted-joint structure, the relative position and posture between every two rods is the basis of coordinate measuring. And coordinate system of each rod should be built for the analysis of this relationship between them [5]. In 1955 Denavit and Hartenberg put forward the famous D-H homogeneous transformation method. By this method, 6 DOF coordinate system for structure schematic diagram of the CMM is built, as shown in Fig.1. Z3
X2 X3
L5
Y2 L4
L3
L6 Y4
L2
X4 L7
Z1 X1
X5 Z5
L8
L1 Z6
L10
Y6 Z7 L9 Y7
Fig. 1. Coordinates of the portable CMM
Z0 X0
412
Q.-S. Cao et al.
The coordinate transformation matrix of coordinates from number i+1 to number i is signed asTi-1i. A CMM with n joints, the coordinate transformation matrixes of coordinate system are as below.
Tnn −1 , Tnn−−12 ,..., T21 , T10 If the position of one point in the final coordinate system is known, then coordinates of the measuring head in the base coordinate system can be obtained through continuous transformation of coordinate systems. On the basis of the coordinate systems built above and D-H coordinate transformation principle, the coordinate transformation matrix of the measuring head with respect to the base coordinate system can be expressed as
⎡1 ⎢0 A = T10T21T32T43T54T65T76 = ⎢ ⎢0 ⎢ ⎣0
⎡ cos θ1 0 sin θ1 ⎢ sin θ 1 − cos θ 1 1 ×⎢ ⎢ 0 0 0 ⎢ 0 0 ⎣ 0 ⎡ cos θ3 ⎢ sin θ 3 ×⎢ ⎢ 0 ⎢ ⎣ 0
0 sin θ 3 0 − cos θ 3 1 0 0 0
⎡ − sin θ5 ⎢ cos θ 5 ×⎢ ⎢ 0 ⎢ ⎣ 0
0 cos θ5 0 sin θ 5 1 0 0 0
L3 cos θ1 ⎤ ⎡ cos θ 2 L3 sin θ1 ⎥⎥ ⎢⎢ sin θ 2 × L2 ⎥ ⎢ 0 ⎥ ⎢ 1 ⎦ ⎣ 0
0⎤ 1 0 0 ⎥⎥ 0 1 L1 ⎥ ⎥ 0 0 1⎦ 0 0
L4 cos θ 2 ⎤ 0 − cos θ 2 L4 sin θ 2 ⎥⎥ ⎥ 1 0 0 ⎥ 0 0 1 ⎦ L6 cos θ 3 ⎤ ⎡ − sin θ 4 0 − cos θ 4 L7 cos θ 4 ⎤ L6 sin θ 3 ⎥⎥ ⎢⎢ cos θ 4 0 − sin θ 4 L7 sin θ 4 ⎥⎥ × ⎥ L5 ⎥ ⎢ 0 −1 0 0 ⎥ ⎢ ⎥ 1 0 0 1 ⎦ ⎣ 0 ⎦ L9 cos θ5 ⎤ ⎡ cos θ 6 0 si n θ 6 L10 sin θ 6 ⎤ L9 sin θ 5 ⎥⎥ ⎢⎢ sin θ 6 0 cos θ 6 L10 cos θ 6 ⎥⎥ × ⎥ L2 ⎥ ⎢ 0 1 0 0 ⎥ ⎢ ⎥ 1 0 0 1 ⎦ ⎣ 0 ⎦ 0
sin θ 2
(1)
As can be deduced from this mathematical model, the measurement result depend on the arms length Li and the joint’s rotating angle θi , the error of the arms length and the joint’s rotating angle affected measuring accuracy directly. 2.2 Principle of Compensation for Temperature Error
During the measurement procedure, the environmental temperature should be kept as constant as possible. If not, the deviation of the temperature from the reference of 20 will result in difference between real length and ideal length of each rod, which
℃
Design of Integrated Error Compensating System for the Portable Flexible CMMs
413
should not be neglected. For this kind of difference, temperature element is sticked to the rod to monitor its temperature, and then get it processed to get the relevant compensation dosage. The formula of the compensation dosage is ΔLi = (T − 20 ) × α i
(2)
In this formula, T is the environmental temperature, αi is thermal expansion coefficient of the material. So when the affection of thermal variation on the measuring results is considered, and the compensation dosage is added to the ideal mathematical model, we can get the mathematical model with errors of temperature variation as follow ⎡1 ⎢0 A' = T10T21T32T43T54T65T76 = ⎢ ⎢0 ⎢ ⎣0 ⎡cos θ1 ⎢ sin θ 1 ×⎢ ⎢ 0 ⎢ ⎣ 0
0 sin θ1 1 − cos θ1 0 0 0 0
⎡ cos θ3 ⎢ sin θ 3 ×⎢ ⎢ 0 ⎢ ⎣ 0
0 sin θ3 0 − cos θ3 1 0 0 0
⎡ − sin θ5 ⎢ cos θ 5 ×⎢ ⎢ 0 ⎢ ⎣ 0
0 cos θ5 0 sin θ 5 1 0 0 0
0 1 0 0
( L3 + ΔL3 ) cos θ1 ⎤ ⎡cos θ 2 ( L3 + ΔL3 ) sin θ1 ⎥⎥ ⎢⎢ sin θ 2 × ⎥ ⎢ 0 L2 + ΔL2 ⎥ ⎢ 1 ⎦ ⎣ 0
0 0 ⎤ 0 0 ⎥⎥ 1 L1 + ΔL1 ⎥ ⎥ 0 1 ⎦ 0 sin θ 2 0 − cos θ 2 1 0 0 0
( L4 + ΔL4 ) cos θ 2 ⎤ ( L4 + ΔL4 )sin θ 2 ⎥⎥ ⎥ 0 ⎥ 1 ⎦ ( L6 + ΔL6 ) cos θ3 ⎤ ⎡ − sin θ 4 0 − cos θ 4 ( L7 + ΔL7 ) cos θ 4 ⎤ 0 − sin θ 4 ( L7 + ΔL7 ) sin θ 4 ⎥⎥ ( L6 + ΔL6 ) sin θ 3 ⎥⎥ ⎢⎢ cos θ 4 × ⎥ ⎢ 0 ⎥ −1 0 0 L5 + ΔL5 ⎥ ⎢ ⎥ 1 0 0 1 ⎦ ⎣ 0 ⎦ ( L9 + ΔL9 ) cos θ5 ⎤ ⎡ cos θ 6 0 si n θ6 ( L10 + ΔL10 ) sin θ 6 ⎤ ( L9 + ΔL9 ) sin θ 5 ⎥⎥ ⎢⎢ sin θ 6 0 cos θ 6 ( L10 + ΔL10 ) cos θ 6 ⎥⎥ × ⎥ ⎥ ⎢ 0 L2 + ΔL2 1 0 0 ⎥ ⎥ ⎢ 1 0 0 1 ⎦ ⎦ ⎣ 0 (3)
2.3 Principle of Compensation for Strain Error
In the measurement procedure, the bending deformation on measurement arm will result in the extra rotating angle of each joint. As the angle deviation would be seriously magnified through the arm, the impact caused must not be neglected. The bending deformation caused by the operation force can be detected by strain gages sticked symmetrically around the measuring arm. The bridge type is most frequently used, which can amplify signal effectively. The circuit diagram for strain measurement is shown in fig.2.
414
Q.-S. Cao et al.
strain B
strain A
R-ǻR R+ǻR
U0 Amplification
A/D conversion
MCU
R
R
E
Fig. 2. Block diagram of strain-measuring circuit
As is shown in the figure, the bridge is in state of equilibrium, and there are no signals output when the measuring arm is not forced. Two strains are sticked on the adjacent bridge arms. The resistance of strain A gets bigger and that of strain B decreases when the arm is exposed to stress, the bridge is no longer balanced, and there is voltage signal output in this situation. As resistance strain is used in this paper, this signal is very weak, and it must be amplified first so that it can be treated afterwards. According to the above shows that can block, amplified output voltage of bridge road can be expressed as U0 =
K 0 ⋅ K ⋅ E ⋅ ΔL 2⋅ L
(4)
Where, E is the input voltage, K is the sensitivity coefficient of the strain gauge selected, L is the length of the measuring arm. So the length variation of measuring arm can be obtained and expressed as ΔL =
2 ⋅U 0 ⋅ L E ⋅ K ⋅ K0
(5)
And the corresponding extra rotation angle of the joint is ⎞ L ⋅ E ⋅ K ⋅ K0 ⎛ L ⎞ −1 ⎛ Δθ = cos −1 ⎜ ⎟ ⎟ = cos ⎜ ⎝ L + ΔL ⎠ ⎝ 2 ⋅ L ⋅U 0 + L ⋅ E ⋅ K ⋅ K0 ⎠
△
(6)
Namely, this θ is the extra rotation angle of each joint caused by the measuring arm’s own gravity or external operation force. And it is used as the compensation
Design of Integrated Error Compensating System for the Portable Flexible CMMs
415
value and is substitute to mathematical model of the CMMs. So mathematical model with strain error can be expressed as ⎡1 ⎢0 A'' = T10T21T32T43T54T65T76 = ⎢ ⎢0 ⎢ ⎣0
0 1 0 0
0 0⎤ 0 0 ⎥⎥ 1 L1 ⎥ ⎥ 0 1⎦
⎡cos(θ1 + Δθ1 ) ⎢ sin(θ + Δθ ) 1 1 ×⎢ ⎢ 0 ⎢ 0 ⎣
0 sin(θ1 + Δθ1 ) L3 cos(θ1 + Δθ1 ) ⎤ ⎡cos(θ2 + Δθ2 ) 0 sin(θ2 + Δθ2 ) L4 cos(θ2 + Δθ2 ) ⎤ 1 − cos(θ1 + Δθ1 ) L3 sin(θ1 + Δθ1 ) ⎥⎥ ⎢⎢ sin(θ2 + Δθ2 ) 0 − cos(θ2 + Δθ2 ) L4 sin(θ2 + Δθ2 ) ⎥⎥ × ⎥ ⎢ ⎥ 0 0 L2 0 1 0 0 ⎥ ⎢ ⎥ 0 0 1 0 0 0 1 ⎦ ⎣ ⎦ ⎡cos(θ3 + Δθ3 ) 0 sin(θ3 + Δθ3 ) L6 cos(θ3 + Δθ3 )⎤ ⎡− sin(θ4 + Δθ4 ) 0 − cos(θ4 + Δθ4 ) L7 cos(θ4 + Δθ4 )⎤ ⎢ sin(θ + Δθ ) 0 − cos(θ + Δθ ) L sin(θ + Δθ ) ⎥ ⎢ cos(θ + Δθ ) 0 − sin(θ + Δθ ) L sin(θ + Δθ ) ⎥ 3 3 3 3 6 3 3 ⎥ ⎢ 4 4 4 4 7 4 4 ⎥ ×⎢ × ⎢ ⎥ ⎢ ⎥ 0 1 0 L5 0 0 0 −1 ⎢ ⎥ ⎢ ⎥ 0 0 0 1 0 0 0 1 ⎣ ⎦ ⎣ ⎦ ⎡ − sin(θ5 + Δθ5 ) 0 cos(θ5 + Δθ5 ) L9 cos(θ5 + Δθ5 ) ⎤ ⎡cos(θ6 + Δθ6 ) 0 si n(θ6 + Δθ6 ) L10 sin(θ6 + Δθ6 ) ⎤ ⎢ cos(θ + Δθ ) 0 sin(θ + Δθ ) L sin(θ + Δθ ) ⎥ ⎢ sin(θ + Δθ ) 0 cos(θ + Δθ ) L cos(θ + Δθ ) ⎥ 5 5 5 5 9 5 5 ⎥ ⎢ 6 6 6 6 10 6 6 ⎥ × ×⎢ ⎢ ⎥ ⎢ ⎥ L2 0 1 0 0 1 0 0 ⎢ ⎥ ⎢ ⎥ 0 0 0 1 0 0 0 1 ⎣ ⎦ ⎣ ⎦
(7)
3 Design of the Compensation System for Integrated Errors 3.1 Overall Scheme
To achieve compensation for errors caused by thermal change and strain, a compensation system is designed in this paper, which consists of the acquisition of temperature and strain, the calculation of compensation dosage and the modification of corresponding mathematical model. The overall diagram is shown in Figure 3.
Temperature detector
Strain Gauge
Thermal value
Strain value
Processing program 1 on PC
Processing program 2 on PC
Compensation dosage 1
Compensation dosage 2
Mathematical model To achieve
Fig. 3. Overall diagram of the error compensation system
Modified mathematical model
416
Q.-S. Cao et al.
3.2 Design of the Error Compensation System of Temperature
As we know, most materials expand as the temperature increases and contract when they are cooled. The flexible arm CMMs are operated by people and are often used in workshop. The environmental temperature changes and the heat created by the electronic devices in the CMMs would result in length change of measuring arm. Therefore, measures have to be taken to modify these errors. A new single-wire enhanced digital thermometer DS18B20 [6], which is produced by the DALLAS Corporation, is used in this paper. The measuring rang of temperature of this sensor is -55 ~ 125 and the measurement accuracy is ± 0.5 among -10~ 85 . As the sensor adopts a single-wire structure, the information can be sent or output by a single-wire interface, and the PC can read, write and convert the temperature through an I/O line. In addition, the sensor can achieve multiple points of temperature measurement, thus it can reduce the amount of line for temperature measurement [7]. The temperature measurement circuit is shown in figure 4.
℃
℃
℃,
Fig. 4. Measurement circuit of temperature
The measuring circuit of thermal variation is very simple, and its function is mainly accomplished by software. The program diagram is shown in figure 5.
Design of Integrated Error Compensating System for the Portable Flexible CMMs
417
Reset
Precision setting
Temperature value error CRC correct Temperature processing
Temperature display
Sending data
End
Fig. 5. Program diagram of thermal measurement
The temperature obtained from the single-wire digital thermometer DS18B20 is sent to computer through serial communication, and then it is compared with the ideal temperature of 20 to get the corresponding length change, using the thermal expansion coefficient of the selected material. On the PC, LabVIEW is used to obtain and calculate the compensation dosage of the corresponding temperature through serial communication. The program diagram is shown in figure 6.
℃
Fig. 6. Program diagram of thermal error compensation
418
Q.-S. Cao et al.
3.3 Design of the Error Compensation System of Strain
The arm of flexible CMM usually adopts long and thin circular pipe, so it is easy to bend and deform under its own gravity and the exterior operation power. Previous studies showed that it creates bending deformation of more than 10″at the two ends of the arm is very common. This is one of the most important factors limiting the measuring accuracy of this kind of CMMs. The bending and deformation of CMM’s each arm is not only in relation to the structural parameters of the arm, but also to the extension of the arm. It is difficult to calculate the amount of the bending and deformation of the arm precisely by theoretical calculation. So it is necessary to paste numbers of strain gauges to monitor deformation on each arm, and then make corresponding program to compensate for the errors immediately. The bending and deformation of measuring arm is equal to increase the rotation angle of each joint additionally. Strain gauges are pasted symmetrically on the arm to monitor the bending and deformation along the axis in this paper. The strain gauges are linked as form of half bridge, and circuit of strain measurement is shown in figure 7.
Fig. 7. Circuit of strain measurement
The amplified output voltage of the bridge is an analog quantity, and it must be converted to by A/D converting circuit as the MCU can only obtain and deal with the digital quantity. The A/D converting circuit is shown in figure 8. The programs compiled with LabVIEW on the PC for strain measurement is just like that of the temperature, we can replace the formula with the formulae 6 above. Take into account fully, the error of environmental thermal variation and dynamic strain was compiled with LabVIEW thus the Complete error compensation system was established. It is also the key technology to improve the measurement accuracy.
Design of Integrated Error Compensating System for the Portable Flexible CMMs
419
Fig. 8. A/D convertion circuit
4 Conclusion Firstly, the ideal mathematical model of the portable CMM is established in this paper, and then the mathematical model with errors is established on the basis of analysis of parts of the error sources. At last an error compensation system based on environmental thermal variation and dynamic strain is designed from two aspects of hardware and software, the integrated error compensation system can improve the measurement accuracy of portable flexible CMM effectively.
References 1. Ye, D., Huang, Q. C., Che, R. S.: Error modeling for multi-joint coordinate measuring machine. J. Optics and Precision Engineering 7(2), 92–96 (1999) 2. Ye, D., Che, R.S.: Modeling and error analysis for Human-simulating multi-joint coordinate measuring machine. Journal of Harbin Institude of Technology 31(2), 28–32 (1999) 3. Wang, X.Y., Liu, S.G., Zhang, G.X.: Mathematical Model and error Analysis of the Articulated Arm Flexible CMM. J. Nanotechnology and Precission Engineering 3(4), 262–267 (2005) 4. Xu, J.X.: The develop of portable Articulated Arm CMM. D. Research institute of mechanical science, Beijing, (2007) 5. Li, Z.W., Wang, C.J., Zhou, X.H.: Mathematical Model and Parameter Calibration of the multi-joint coordinate measuring machine. J. Mechanical Processing Technology and Equipment (6), 35–36 (2006) 6. Denavit, J., Hartenberg, R.: A kinematic notation for lower pair mechanism based on matrices. J. ASME Journal of Applied Mechnics 22(6), 215–221 (1955) 7. Chen, T.: Application of Single Chip Processor and C language programming. China Machine Press, Beijing (2008)
Detecting and Analyzing System for the Vibration Comfort of Car Seats Based on LabVIEW Ying Qiu Key Laboratory of Conveyance and Equipment, Ministry of Education School of Mechanical and Electronical Engineering, East China Jiaotong University, Nanchang, P.R. China
[email protected] Abstract. In this research, LabVIEW 8.5 is taken as the software platform. Sensors, data acquisition cards are combined as its hardware. A detecting and analyzing system for the vibration comfort of car seats is developed, in which the functions as to the collecting and processing of vibrating signals, the access of data and the display of images are achieved. The system can also analyze the dynamic characteristics and the comfort of the occupant-seat system. By choosing different seat characteristics according to different vehicle environments, the best comfort can be obtained. In addition, it also provides some favorable references to the designing of safe and comfortable automobile seats. Keywords: Vehicle seat, LABVIEW, Dynamic characteristics, Vibration detection.
1 Introduction Car seat is an important part of the car. Its main function is to support the driver and the passenger's body, to reduce the impact on the body aroused by rough road surface and to attenuate the resulting vibrations. With the improvement of living standard and the rapid development of automobile technology, the demand for automotive comfort is higher and higher. Automotive comfort includes static and dynamic comfort. The former mainly involves size parameters, surface quality and the regulation characteristics; and the latter is largely related to vibration characteristics. In mechanical vibration test, because numerous complex test equipment are needed, a simple test may even require great manpower and large quantity of material resources, not to mention some complex work like the vibration test on car seats. With the development of computer and software technology, it is taking over the traditional method and is becoming the trend in the area of testing. In this article, a development platform based on LabVIEW software (a virtual instrument platform initiated by American National Instruments Company) is constructed. Necessary sensor, signal conditioners and data acquisition cards are taken as the main hardware to build a detecting and analyzing system for the seat vibration test. Results from experiments show that the suggested D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part IV, IFIP AICT 347, pp. 420–426, 2011. © IFIP International Federation for Information Processing 2011
Detecting and Analyzing System for the Vibration Comfort of Car Seats
421
system can effectively fulfill the work of acquiring and processing seat vibration signals, analyzing its dynamic characteristics and assessing its comfort rate.
2 Overall System Structure and Working Principle
,
Using virtual test instruments the Virtual vibration testing and analyzing system completes the work of measuring, data analysis and processing of mechanical vibration signals. Totally resorting to computer software, it successfully achieves the functions of vibration signal acquisition, display, access and processing. The system structure is illustrated in Figure1.
Fig. 1. Schematic diagram of system structure
Hardware for the system includes: acceleration sensor, signal amplifier suitable for transfer, data acquisition card and PC machines. Its software is realized through graphic-oriented software LabVIEW. The seat vibration test and analysis system first transforms the vibration signals detected by the sensor into analog signals, and then passed it to the signal conditioning. Next, the analog electronic signal is converted into digital signal by data acquisition card and finally is passed to the vibration test and analysis computer software for analysis. The system realizes the seat vibration signal testing and data analysis via computer, which can completely replace the conventional signal analyzer and many other hardware devices to complete a variety of signal analysis and processing.
3 Designing of System Functions Based on the needs of seat vibration comfort testing, the main functions of this system include vibration signal acquisition and display, signal analysis and processing, dynamic characteristics analysis and comfort-rating assessment on seat vibration. Block diagram of the system function is shown in Figure 2.
422
Y. Qiu
seat vibration detecting and analyzing system
vibration signal acquisition
dynamic signal analyzing and processing characteristics analysis
comfort rating assessment
Fig. 2. Block diagram of the system function
4 Implementation and Testing of System Functions Front panel of the seat vibration test system is shown in Figure 3, the functional modules can be switched freely through the option cards to achieve the functions of signal acquisition, signal processing, dynamic analysis and evaluation of comfort. 4.1 Vibration Signal Acquisition Module The system completes the external circuit signal acquisition through the DAQ assistant acquisition module and NI ELVIS external board. The acquisition results are shown in Figure 3.
Fig. 3. Front panel of the seat vibration test system
4.2 Signal Analysis and Processing Module In order to eliminate the interference aroused by external signals on the vibration signals, the acquired signals must be pretreated. The system mainly makes some
Detecting and Analyzing System for the Vibration Comfort of Car Seats
423
filtering and spectrum analysis on analog signals mixed with noise, in order to complete the module test. The mathematics algorithm for filtering is rather complicated, but LabVIEW has modularized the filtering function, so it is quite easy to complete the filtering program with these functions. Signals mixed with noise have been filtered and a comparison between the source signal and the filtered signals will clearly demonstrate the effect as shown in Figure 4.
Fig. 4. Effect of filtering
Besides filtering the signals, the module also has many other functions including frequency domain analysis, window processing, correlation analysis, FFT transform and etc. 4.3 Dynamics Analysis Module The main function of car seats is to support the driver and the passenger's body, to reduce the impact of uneven road surface and to attenuate the resulting vibrations passed to the people, so as to provide a comfortable and safe riding condition. The kinetic parameters --- natural frequency and the damping ratio of the car seat, have a major impact on the damping properties of the seat. Thus, take a representative 1-DOF and 3-DOF occupant-seat model for dynamic analysis. Suggestions on how to choose dynamic parameters are also presented. The structure of 1-DOF and 3-DOF occupant-seat model is shown in figure 5. In 1-DOF model, the human body is considered as a rigid body on the seat. Its frequency response function is: | H (ω ) |=
1 + (2ξγ ) 2 (1 − γ 2 ) 2 + (2ξγ ) 2
(1)
Where: ξ = C /( 2 KM ) is the damping ratio, γ = ω / K / M is the frequency ratio. The seat damping ratio and frequency ratio can be adjusted by changing the parameters of the car seat, so as to improve the frequency response of the seat.
424
Y. Qiu
Fig. 5. Structure of 1-DOF and 3-DOF occupant-seat model
In 3-DOF model, the human body is considered as a 2-DOF dynamic model on the seat. Suppose: ω n = K /( M 1 + M 2 ) , ω n1 = K1 / M 1 , ω n 2 = K 2 / M 2 , γ = ω / ω n ,
γ 1 = ω / ω n1
,
γ 2 = ω / ω n 2 , ξ = C (2 K ( M1 + M 2 )) , ξ1 = C1 (2 K1M 1 ) ,
ξ 2 = C2 (2 K 2 M 2 ) , μ1 = M 1 ( M 1 + M 2 ) , μ2 = M 2 ( M 1 + M 2 ) , The frequency response function is: H (ω ) =
1 + j 2ξγ (1 + j 2ξγ ) − ( μ1 B1 + μ 2 B2 )
,
(2)
1 + j 2ξ 2 γ 2 1 + j 2ξ1γ 1 . B2 = 2 2 1 − γ 1 + j 2ξ1γ 1 1 − γ 2 + j 2ξ 2γ 2 When analyzing the simulation in LabVIEW, the parameters in the model are set on the following basis: 1) “Test Method of Car Seat Dynamic Comfort", in which the standard is M = 51kg ; 2) statistic results from domestic human body vibration test, in which M1 = 29.8kg, M2 = 5.5kg, K1 = 224.8N/cm, K2 = 133.1N/cm, C1 = 3.9N / (cm.s), C1 = 1.9N / (cm.s). The frequency response function of the two models is related not only to the parameters of the human body model but also to the natural frequency and damping ratio of the seat. Figure 6 shows the system frequency response curve under different natural frequency and damping ratio of 1-DOF model, by which reasonable parameters can be chosen. Where: B1 =
Fig. 6. Frequency response curve under different damping ratio and natural frequency
Detecting and Analyzing System for the Vibration Comfort of Car Seats
425
The module can also analyze the displacement response curve of human body vibration by simulating different car environment and road condition. When different seat parameters are selected, the changes of human body vibration displacement can be directly observed, which will provide favorable reference for the design of safe and comfortable seats. 4.4 Seat Comfort Assessment Module The function of data analysis module is to analyze the collected and pretreated signals, so that the specific values of comfort can be obtained and car seat comfort can be measured. Data is processed using total RMS value of weighted acceleration comparison method by the data analysis module. Taking into account individual’s different sensitivity to the frequency of vibration, this method converts RMS acceleration which frequency ranges within 1 ~ 80Hz into RMS acceleration within sensitive frequency (4 ~ 8Hz, or 1 ~ 2Hz) by multiplying a different frequency weighting, then calculates the total acceleration RMS, and then sets the standard for evaluation: whether the corresponding limit exceeds ISO2631 sensitivity limit of the frequency or length of time. As for the composite vibration condition which human body withstand from X, Y, Z three directions, we can first obtain respective frequency weighted RMS by calculating the acceleration spectra in each vibration direction, and then calculate the combined weighted acceleration value Aω.
Aω = (1.4axω ) 2 + (1.4a yω ) 2 + (azω ) 2
(3)
Where: the factor of 1.4 is the most sensitive in the body within the same frequency range corresponding vertical and horizontal and vertical curve ratio. "Total Travel Value Method" has been introduced in the draft of ISO2631/CD1991, and the relationship between the acceleration value and the subjective feeling of the occupant has been given, as is shown in Table 1. Table 1. Relationship between Acceleration Value and Subjective Feeling of the Occupant
(m/s )
Weighted Acceleration Value 2.0
~ ~ ~ ~
2
Subjective Feeling of the Occupant Comfortable slightly uncomfortable somewhat uncomfortable Uncomfortable very uncomfortable Unbearably Uncomfortable
The vibration signals from X, Y and Z direction are measured by the sensor. Then they are converted into vehicle vibration comfort indicators Cω and combined weighted RMS Aω through the program operation. Finally, the comfort condition is displayed on the front panel.
426
Y. Qiu
5 Conclusion This system is actualized through a powerful graphical programming language LabVIEW and the existing hardware devices. It can measure seat vibration, process data, analyze seat dynamics characteristic and evaluate the driving comfort. Compared with traditional method, this system has the following advantages: (1) Computer software program has replaced the traditional spectrum analyzers and other hardware devices. This system validates real-time analysis and processing, reduces test costs and shortens the test cycle. (2) Resorting to the existing signal processing modules of LabVIEW to deal with vibration signals, the process is quicker and more accurate, and thus complex mathematic calculation is avoided. (3) By making a dynamic simulation analysis on the human body - seat model, it provides a favorable reference to the designing of safe and comfortable seats. Based on ISO2631 standard, seat driving comfort is also evaluated. Acknowledgement. This work is supported by the Research Foundation of ECJTU (NO. 01308115), and Key Laboratory of Conveyance and Equipment.
References 1. Boileau, P. E.: A body Mass Dependent Mechanical Impendence Model for Application in vibration Seat Testing. J. Journal of Sound and Vibration 253(1), 243–264 (2002) 2. National Instruments Corp LabVIEW User Manual, Austin, Texas, USA, pp. 27–444 (1998) 3. National Instruments Corp BridgeVIEW and LabVIEW G Programming Reference Manual, Austin, Texas USA, pp. 27–566 (1998) 4. Dai, X., Sun, H., Ou, J.J.: Virtual instrument design for mechanical vibration measurement. In Chinese. China Measurement & Testing Technology 34(4), 92–95 (2008) 5. Meng, Y.M., Gao, F.W., Duan, J.L., Li, S.P., Huang, B.P.: The research of vibration measurement and analysis system based on LABVIEW. in Chinese. Journal of Guangxi University (Nat. Sci. Ed.) 32(2), 114–117 (2007) 6. Qian, Y., Zhou, Y.P., Chen, J.W.: Analytical System of Vibration Measurement Based on LabVIEW. In Chinese. Journal of Wenzhou University, 58–61 (2005) 7. Chen, X.W., Liu, Y.: Realization of test analysis for vibration signal based on LabVIEW. In Chinese. Electronic Measurement Technology, 108–120 (2008)
Determination of Pesticide Residues on the Surface of Fruits Using Micro-Raman Spectroscopy Yande Liu1,2 and Tao Liu1 1
School of Mechanical and Electronical Engineering, East China Jiaotong University, Nanchang, China 2 Institute of Optics-Mechanics-Electronics Technology and Application (OMETA), East China Jiaotong University, Nanchang, China
Abstract. A simple, rapid and environmentally friendly method was developed for micro-Raman spectroscopy determination of pesticide residues on the surface of fruits. Raman spectra of fruits, pesticides and pericarps sprayed by pesticide solutions were acquired using a laser power of 14 mW at excitation wavelength of 780 nm. From the Raman spectra, the residual pesticides could be distinguished and determined through the characteristic Raman peaks. The overall results indicted that micro-Raman spectroscopy is a potential tool to determine the pesticide residues on the surface of fruits for fruit quality and safety control. Keywords: micro-Raman spectroscopy, fruit, pesticide residue.
1 Introduction Pesticides are defined by the United Nations Food and Agricultural Organization (FAO) as substances or mixtures intended to prevent, destroy, repel or mitigate any pest, including insects, rodents and weeds [1]. Currently, pesticides play a critical role in protecting fruit crops. However, large amount of pesticides are used in fruits production making it unsafe and endangering humans and animals. One of the most common used pesticides are the organophosphates, which kill insects and mites by attacking the central nervous system, and consequently pose a threat to human beings and animal lives [2]. Nowadays, fruit safety and quality has been attracted more and more attention. In many countries, strict tolerance levels are set up to ensure public safety. The current analysis methods for determining the residual pesticides in fresh fruits are based on chemical analysis, such as liquid chromatography-mass spectrometry (LC-MS) [3], liquid chromatography-tandem mass spectrometry (LC-MS/MS) [4, 5], gas chromatography-mass spectrometry (GC-MS) [6], capillary electrophoresis-mass spectrometry (CE-MS) [7] and ultra-high-performance liquid chromatography (UHPLC) [8], etc. However, all of the methods mentioned above are time-consuming, not environmentally friendly and inconvenient enough for the process of sample preparation. Hence, it was very necessary to develop a simple, rapid and low cost method for the determination of pesticide residuals in fruits. D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part IV, IFIP AICT 347, pp. 427–434, 2011. © IFIP International Federation for Information Processing 2011
428
Y. Liu and T. Liu
Raman spectroscopy is a fast technique compared to classical chromatography. It is a powerful analytical tool that provides wonderful advantages including non-destructive, low cost and ultrasensitive characterization down to single molecular level [9]. Shende et al. [10] used surface-enhanced Raman spectroscopy (SERS) to analyze pesticides on fruit surfaces. Zhou et al. [11] recorded several spectra of fruits with pesticides using near infrared Fourier transform Raman spectroscopy (FT-Raman). Several Raman spectra of some fruits and pesticides on the surface of fruits were acquired and compared with at two excitation wavelengths of 514.5 and 1064 nm by Zhang et al. [12]. Micro-Raman has its unique characters compared to other Raman techniques, since it allows analysis of materials with a spatial resolution of several microns focused by an optical microscope. It has been developed for a rapid, environmentally friendly and low cost generation procedure, and has been applied in many fields, such as gems [13, 14], fibers [15, 16] and geology [17, 18], etc. In this paper, a new attempt to determine the residual pesticides on the surface of fruits using micro-Raman spectroscopy was reported. The objective was to record the Raman spectra of fruits and pesticides in order to study the feasibility of using micro-Raman spectroscopy to determine the residual pesticides on the surface of fruits.
2 Materials and Methods 2.1 Apparatus A Nicolet DXR Raman Microscope (Thermo Fisher Scientific Corp., Madison, WI, USA), equipped with a liquid nitrogen cooled CCD detector and a 14 mW maximum power diode laser, that emits at 780 nm, was employed for confocal micro-Raman spectra acquisition using 2 ml standard glass chromatographic vials (12 mm×32 mm) as sample cells. The laser was focused within the sample using an inverted microscope setup equipped with a 10× ultra long working distance objective. The scattered signal was then recorded at a 180° backscattering geometry and dispersed by a single monochromator using a 400 grooves/mm diffraction grating and a laser line filter used as a beam splitter to lead the laser beam into the microscope assembly and also to reject the Rayleigh scattered light on the returning path. Spectrometer was controlled with OMINIC software (Thermo Nicolet Corp., Madison, WI, USA). 2.2 Sample Preparation and Chemicals The samples, including apples, pears and oranges fruits in the experiment, were purchased directly from the Nanchang fruit market. The chemicals, dimethoate standard (O,O-dimethyl- S-(N-methylcarbomoylmethyl) phosphorodithioate) solid (99.1% w/v), chlorpyrifos standard (O,O-diethyl-O-(3,5,6-trichloro-2-pyridyl) phosphorothioate) solid (99.5% w/v) and malathion standard (S-(1,2-dicarbethoxyethyl)-O,O-dimethyl dithiophosphate) solution (99.1% w/v) were purchased from bzwz corporation (Beijing, China). Two emulsifiable concentrate commercial pesticide formulations, chlorpyrifos (40% w/w) and malathion (40% w/w) were obtained directly from the Nanchang pesticide market. Before recording the micro-Raman spectra, all the fruits were cleaned carefully. Firstly, fruits were flushed several times with fresh water to get rid of the dirt left on the
Determination of Pesticide Residues on the Surface of Fruits
429
surface, and then wiped with cotton and alcohol. After that, small pieces of pericarps from the cleaned fruits were taken off, and then sprayed by a little amount of pesticide formulations. At last, filter papers were used to wipe off any left over pesticides from the pericarps, and let the samples dry naturally. 2.3 Micro-Raman Procedure When it was time to record the micro-Raman spectra, the resolution and accumulating time were fixed at 1.93 cm-1 and 100 scans per spectrum. The micro-Raman spectra of clean pericarps of apples, pears, oranges and four standard pesticides were collected, firstly. These spectra were used as standards in the comparison database. Then the spectra of pericarps sprayed by the pesticide formulations were collected and compared with those in the database to identify the trace amount of pesticides on the surface of fruits. All the spectra mentioned above were carried out in chromatographic glass vials from 3390 to 110 cm-1 with the diode laser power employed fixed at 14 mW.
3 Results and Discussion 3.1 Raman Spectra of Fruits
1005
1155
1530
Based on the measuring conditions mentioned above, the micro-Raman spectra of clean pericarps of apple, pear and orange fruits are shown in Fig. 1. Each of fruits has its characteristic peaks. Oranges mainly have the Raman peaks at 1530, 1155 and 1005 cm-1 attributed to C=C stretching, C-C stretching and C-CH3 bending of carotene [19]. However, both apple and pear fruits have combined peaks around 1600, 1439, 1136 and 850 cm-1. It is estimated that these peaks are due to the combination of Raman scattering peaks by nourishment compositions of apple and pear. Because each of them contains very complicated compositions as there are usually different amount of water, sugar, vitamin, fat, protein, amino acid, as well as other components and elements such as calcium, iron and zinc, etc. At present, these fruits still can not be identified from the Raman peaks, but it has differences in the peaks attributed to the different compositions.
1800
1600
1400
1200 R a m a n s h if t / c m
850
765
850
1078 1060
1158
1295
1439
a p p le
1136
1332
1600 1617
1524
pear
1657
Raman intensity (cps)
o ran g e
1000
800
600
-1
Fig. 1. Raman spectra of orange, pear and apple fruits
430
Y. Liu and T. Liu
3.2 Raman Spectra of Pesticides
800
Cl
N
Cl
676
C2H5O P O C2H5O S Cl
1000
158 530
1276 1238 1100
1453 1447
2938
400 200
1569
600
3052
Raman intensity (cps)
1200
340
630
Raman spectra of three standard pesticides, including dimethoate, chlorpyrifos and malathion, are shown from Fig. 2 to Fig. 4 in the 3390-110 cm-1 region. All pesticides have the same structure with P=S bonds, which display a strong band in the 600-700 cm-1 region. This band is clearly distinguishable for compounds without benzene rings, but several of the vibrations due to the benzene ring overshadow it as in chlorpyrifos spectrum, where two overlapping bands appear at 630 and 676 cm-1 in Fig. 2 [20]. The most intense bands in the chlorpyrifos Raman spectrum are those present at 340 and 630 cm-1 due to N-cyclopropyl bending and ring deformation, respectively. Other less intense bands located at 158, 676, 1276 and 1569 cm-1 are due to P-O vibration, ring breathing, C-H bending and C=C stretching. C-H stretching modes are observed from 2800 to 3100 cm-1.
0 3000
2500
2000 1500 1000 -1 Raman shift / cm
500
Fig. 2. Raman spectrum and molecular structure of chlorpyrifos
14 00 C H 3O
S
O S
C C H2
NHCH3 171
P
2924
10 00
C H 3O
255
8 00
400
903 768 696
2 00
1156
1640
4 00
1403 1330
6 00 2837
Raman intensity
12 00
494
651
In Fig. 3 of dimethoate Raman spectrum, P=S bond displays a strong band at 696 cm-1. The most intense band in the spectrum is located at 575 cm-1 due to C-S vibration. Other less intense bands located at 171, 494, 768, 903, 1156, 1330 and 1640 cm-1 are due to P-O vibration, P-S vibration, S-P-O backbone mode, C-C backbone stretching, C-C stretching, N-H in-plane bending and symmetric C=O stretching. The intense bands appearing around 2936 cm-1 are due to CH2 and CH3 deformation.
0 3000
2500
2 000 1500 1000 -1 R a m a n s h ift / cm
500
Fig. 3. Raman spectrum and molecular structure of dimethoate
Determination of Pesticide Residues on the Surface of Fruits
431
652
200
159 392
400
496
1451
600
1733
Raman intensity (cps)
800
859
O CH3O S H P S C C O CH2CH3 CH3O C C O CH2CH3 H2 O
1000
1110 1022
1200
2936
In Fig. 4 of malathion spectrum, P=S bond displays a strong band at 652 cm-1. The bands located at 159, 496, 859, 1022, 1110 and 1733 cm-1 are due to P-O vibration, P-S vibration, C-O-C stretching, C-H in-plane deformation, C-C stretching and C=O stretching [21]. The intense bands appearing in 1451 and 2800 to 3100 cm-1 are due to CH2 and CH3 deformation and C-H stretching.
0 3000
2500
2000 1500 1000 -1 Raman shift / cm
500
Fig. 4. Raman spectrum and molecular structure of malathion
From the three figures mentioned above, each of pesticides could be distinguished using their characteristic peaks shown in Table 1. These bands could be employed for determining pesticide residues on the surface of fruits. Table 1. Characteristic peaks of chlorpyrifos, dimethoate and malathion Pesticide Chlorpyrifos Dimethoate Malathion
Characteristic peaks (cm-1) 1569, 1276, 676, 630, 340 1640, 651, 494 1733, 1451, 652
3.3 Raman Spectra of Pesticide Residues on the Surface of Fruits Acquirements of Raman spectra of fruits with pesticides were performed with the same measuring conditions as before. Fig. 5 and Fig. 6 show the Raman spectra of chlorpyrifos and malathion formulations residues on the surface of oranges. From Fig. 5, it can be obviously seen that the spectrum of chlorpyrifos residue on the surface of orange not only contains the strong peaks of orange itself, but also contains the characteristic modes of chlorpyrifos which located at 676, 630 and 340 cm-1. Similarly, in Fig. 6, the characteristic modes of malathion at 1451, 859, 652 and 496 cm-1 can be seen in the spectrum of its residue on the surface of orange. Based on the results, it was concluded that the method of using micro-Raman spectroscopy could be utilized for automatic and intelligent determination of the pesticide residues on the surface of fruits.
Y. Liu and T. Liu
1155
1528
432
1004
(a )
676
340
630
Raman intensity (cps)
(b)
(c)
16 0 0
1 40 0
12 00
1 00 0 8 00 -1 R am a n s h ift / cm
6 00
40 0
2 00
1155
1528
Fig. 5. Raman spectra of (a) orange’s pericarp, (b) chlorpyrifos formulation (40% w/w) left on the surface of orange and (c) standard chlorpyrifos solid
Raman intensity (cps)
1004
(a )
1800
1600
1400
496
859
1733
1451
652
(b )
(c ) 1200 1000 -1 R a m a n s h ift / c m
800
600
400
Fig. 6. Raman spectra of (a) orange’s pericarp, (b) malathion formulation (40% w/w) left on the surface of orange and (c) standard malathion solution
4 Conclusions The micro-Raman spectra of apples, pears, oranges, several pesticides and pesticide residues on the surface of oranges were successfully recorded using a micro-Raman spectrometer. The characteristic peaks of pesticides can be seen from the spectra of pesticides left on the surface of fruits. Micro-Raman spectroscopy is a potential tool to determine the pesticide residues for fruit quality and safety control with rapid, non-destructive, low cost and environmentally friendly advantages.
Determination of Pesticide Residues on the Surface of Fruits
433
Acknowledgements The authors gratefully acknowledge the financial support provided by National Science and Technology Support Program (2008BAD96B04), Natural Science Foundation of Jiangxi Province (2008GQN0029, 2007GZN0266), Special Science and Technology-Support Program for Foreign Science and Technology Cooperation Plan (2009BHB15200), Technological expertise and academic leaders training program of Jiangxi Province (2009DD00700).
References 1. UN Food and Agricultural Organization: International Code of Conduct on the Distribution and Use of Pesticides, Rome, Italy (2002) 2. Tuormaa, T.E.: Adverse Effects of Agrochemicals on Reproduction and Health: a Brief Review from the Literature. J. Nutritional Environ. Med. 5, 353–366 (1995) 3. Picó, Y., Font, G., Moltó, J.C., Mañes, J.: Pesticide Residue Determination in Fruit and Vegetables by Liquid Chromatography–mass Spectrometry. J. Chromatogr. A 882, 153–173 (2000) 4. Jansson, C., Pihlström, T., Österdahl, B., Markides, K.E.: A New Multi-residue Method for Analysis of Pesticide Residues in Fruit and Vegetables Using Liquid Chromatography with Tandem Mass Spectrometric Detection. J. Chromatogr. A 1023, 93–104 (2004) 5. Frenich, A.G., Vidal, J.L.M., López, T.L., Aguado, S.C., Salvador, I.M.: Monitoring Multi-class Pesticide Residues in Fresh Fruits and Vegetables by Liquid Chromatography with Tandem Mass Spectrometry. J. Chromatogr. A 1048, 199–206 (2004) 6. Guan, H.X., Brewer, W.E., Garris, S.T., Morgan, S.L.: Disposable Pipette Extraction for the Analysis of Pesticides in Fruit and Vegetables Using Gas Chromatography/Mass Spectrometry. J. Chromatogr. A 1217, 1867–1874 (2010) 7. Juan-García, A., Font, G., Juan, C., Picó, Y.: Pressurised Liquid Extraction and Capillary Electrophoresis–mass Spectrometry for the Analysis of Pesticide Residues in Fruits from Valencian Markets, Spain. Food Chem. 120, 1242–1249 (2010) 8. Lacina, O., Urbanova, J., Poustka, J., Hajslova, J.: Identification/Quantification of Multiple Pesticide Residues in Food Plants by Ultra-high-performance Liquid Chromatography-time-of-flight Mass Spectrometry. J. Chromatogr. A 1217, 648–659 (2010) 9. Armenta, S., Quintás, G., Garrigues, S., de la Guardia, M.: Mid-infrared and Raman Spectrometry for Quality Control of Pesticide Formulations. Trends Anal. Chem. 24, 772–781 (2005) 10. Shende, C.S., Inscore, F., Gift, A., Maksymiuk, P., Farquharson, S.: Analysis of Pesticides on or in Fruit by Surface-enhanced Raman Spectroscopy. In: Proc. SPIE, vol. 5587, pp. 170–176 (2004) 11. Zhou, X.F., Fang, Y., Zhang, P.X.: Raman Spectra of Pesticides on the Surface of Fruits. J. Light Scatt. 16, 11–14 (2004) 12. Zhang, P.X., Zhou, X.F., Cheng, A.Y.S., Fang, Y.: Raman Spectra from Pesticides on the Surface of Fruits. J. Phys. 28, 7–11 (2006) 13. Sudheer, S.K., Pillai, V.P.M., Nayar, V.U.: Characterization of Laser Processing of Single-crystal Natural Diamonds Using Micro-Raman Spectroscopic Investigations. J. of Raman Spectrosc. 38, 427–435 (2007)
434
Y. Liu and T. Liu
14. Palanza, V., Martino, D.D., Paleari, A., Spinolo, G., Prosperi, L.: Micro-Raman Spectroscopy Applied to the Study of Inclusions within Sapphire. J. Raman Spectrosc. 39, 1007–1011 (2008) 15. Doherty, B., Miliani, C., Berghe, I.V., Sgamellotti, A., Brunetti, B.G.: Micro-Raman Spectroscopic Study of Artificially Aged Natural and Dyed Wool. J. Raman Spectrosc. 39, 638–645 (2008) 16. Colomban, P., Dinh, H.M., Riand, J., Prinsloo, L.C., Mauchamp, B.: Nanomechanics of Single Silkworm and Spider Fibres: a Raman and Micro-mechanical in Situ Study of the Conformation Change with Stress. J. Raman Spectrosc. 39, 1749–1764 (2008) 17. Rull, F., Martinez-Frias, J., Sansano, A., Medina, J., Edwards, H.G.M.: Comparative Micro-Raman Study of the Nakhla and Vaca Muerta Meteorites. J. Raman Spectrosc. 35, 497–503 (2004) 18. Łodziński, M., Wrzalik, R., Sitarz, M.: Micro-Raman Spectroscopy Studies of Some Accessory Minerals from Pegmatites of the Sowie Mts and Strzegom-Sobótka Massif, Lower Silesia, Poland. J. Mol. Struct., 744–747, 1017–1026 (2005) 19. El-Abassy, R.M., Donfack, P., Materny, A.: Rapid Determination of Free Fatty Acid in Extra Virgin Olive Oil by Raman Spectroscopy and Multivariate Analysis. J. Am. Oil Chem. Soc. 86, 507–511 (2009) 20. Skoulika, S.G., Georgiou, C.A., Polissiou, M.G.: FT-Raman Spectroscopy—Analytical Tool for Routine Analysis of Diazinon Pesticide Formulations. Talanta 51, 599–604 (2000) 21. Quintás, G., Garrigues, S., de la Guardia, M.: FT–Raman Spectrometry Determination of Malathion in Pesticide Formulations. Talanta 63, 345–350 (2004)
Development of the Meter for Measuring Pork Quality Based on the Electrical Characteristics* Zhen Xing1, Wengang Zheng2, Changjun Shen1,2, and Xin Zhang1,2 1
Beijing Research Center for Intelligent Agricultural Equipment, 100097, Beijing, China 2 National Engineering Research Center for Information Technology in Agriculture, 100097, Beijing, China {xingz,zhengwg,shencj,zhangx}@nercita.org.cn
Abstract. The excitation frequency on the electrical characteristics of pork had been identified by the experiments of the electrical characteristics and the theory of electrical characteristics, the relationship between the evaluation of quality of pork and the impedance characteristics was established, and a nondestructive portable meter for measuring pork quality was developed. The functions of each hardware and software design of the meter were described in detail. The result shows that the conductivity of the pork increases with the level of corruption by the experiment of the measurement selected sample of pork. The pork quality meter based on the electrical characteristics provides the measurement method and apparatus of a low-cost, rapid, qualitative assessment of pork quality. Keywords: the electrical characteristics, pork quality, a portable meter, a lowcost.
1 Introduction Food safety become the focus of attention when people’s living standard have been rising, The meat is the main food in china; China is one of the biggest countries for the production of livestock and poultry meat and its products. According to The People's Republic of China Statistical Yearbook 2008, China's pork production in 2007 was 42.878 million tons, the amount of production and consumption all ranks the world. People are cautious about meat consumption when meat safety incidents, such as, the injection meat or the clenbuterol, are frequently exposed [1]. In order to eliminate consumer concern for meat product, some approaches should be taken, such as, improving the moral quality of the farmers and producers, and preventing non-healthy meat to target into the market. So the measuring meter for fast detection of meat quality is an important role in the quarantine before meat onto the market. *
The paper was supported by the National High Technology Research and Development Program("863"Program) of China (2007AA10 Z212).
D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part IV, IFIP AICT 347, pp. 435–442, 2011. © IFIP International Federation for Information Processing 2011
436
Z. Xing et al.
Currently, there are a lot of pork quality detection methods. For example, the fat of the pig ketene body and the lean meat were graded by using spectroscopy or ultrasonic detection in the abroad; the nutritional composition in meat and meat quality changes are studied by using the equipment of near-infrared spectroscopy or the chemo metrics in the domestic; the study group of Ding[2] had been studied the nondestructive optical detection methods for the fatty tissue, and measured the content of the Hemoglobin and the myoglobin in different depths of muscle; the study group of Huang [3,4] had been studied the optical parameter properties of agricultural products, the detection of fresh pork and deep water of the fresh pork. In addition, the meat quality of detection methods, based on dielectric properties or electrochemical properties, had been developed rapidly. For example, Zhang [5] had been studied temperature effect on the dielectric properties of the lunch pork roll; Kent [6] soaked the meat in water at different times, attained different the water content, and studied the water content effect on the dielectric properties of chicken, scallops and pork. Zhang [7] had been studied the salt effect on the permittivity and dielectric loss factor of the pork. Currently, the dielectric properties of the raw meat and meat products had been studied in abroad, especially the microwave dielectric properties, however, detecting the quality of pork by using the dielectric properties has only begun in the domestic. A new sensor based on the transmission line impedance theory was designed by using the coaxial transmission line technology, and the meter for the meter for measuring pork quality based on this new sensor was developed in this paper; it can be more convenient, fast and efficient measurement of pork quality, and ensure the safety of the pork quality.
2 Sensor Design The sensor consists of a power supply unit, a high-frequency signal source, a signal processing unit, a coaxial transmission line and measuring probes. The structure is as shown in Figure 1.
Power unit
Signal processing Circuit
High-frequency Oscillator
Transmission Line
Output Fig. 1. The principle structure of the sensor
Measuring Probe
Development of the Meter for Measuring Pork Quality
437
2.1 Measuring Principle Coaxial transmission line measurement technique is that the electrical characteristics of biological tissue (such as conductivity, dielectric constant, etc.) was transferred into high-frequency electromagnetic wave reflection coefficient, the relationship between the electrical characteristics of biological tissue and high frequency electromagnetic wave reflection coefficient, the electrical characteristics of biological tissue can be obtained by measuring the radiation coefficient. When the power supply was at work, a fixed-and high-frequency signal, generated by the crystal oscillator, transmitted through the coaxial transmission line to the probe which was inserted into the testing material. Due to the impedance’s mismatching, parts of the signals would return and superpose with the latter input ones. Consequently, it set up a doable way of associating a stabilized amplitude voltage with the coaxial transmission line and the probe, which is available to the quality detecting. 2.2 Signal Processing Circuit According to signal detection theory, the signal processing circuit of accuracy, stability, sensitivity directly determines the overall system performance, therefore, the design of the signal processing circuit is crucial, it must eliminate noise and amplify useful signal. Aiming at catering to this need, the measurement circuit can be simplified as two-stage amplification. The signal processing circuit is shown in detail in Figure 2.
Fig. 2. Block diagram of signal processing circuit
438
Z. Xing et al.
Observing that the first-stage used difference amplifier to perform differential operation, and the input sinusoidal signal was changed into DC signal, the function of the current-conveyor was to ensure the working probe is held at virtual ground when it was connected at the impedance. In addition, difference amplifier has the advantage of eliminating the Zero Drift. The second-stage amplifier was simply to give rise to the former-step’s output signal, which aimed at improving stability of the testing system. The parameters of every component are as follows.
3 System Architecture The system architecture is shown in the figure 3. The low-power MCU MSP430F149 is used as the core of the system acquisition and control, its power consumption and input leakage current (up to 50nA) is relatively low in the industry, the MSP430F149 is an industrial grade product, stable performance, high reliability. There are many work modes of the MSP430F149; the system power consumption is reduced by selecting a low power operating mode, so the MSP430F149 is especially suitable for handheld applications. It can reduce the number of external components and the cost of the instrument because of its 12-bit A / D converter. The keyboard and LCD modules for the operator provide a friendly human-machine interface; the real-time clock module can record the system running time; the memory module can save operating parameters and measurement data; the sensor interface is using a convenient PS / 2 interface.
RTC module
Sensor interface
LCD module
MSP430F149 Memory module
Key module
Power module Fig. 3. The system architecture
3.1 The Power Module The meter is using two 1.5V batteries or a rechargeable battery, but the number of external components and the MSP430F149 must work at 3.3V, so it must increase the value of the power module. The circuit of the increasing voltage is shown in the figure 4. The chip of MAX1724, which character is a low quiescent current and high efficiency, is used in the power module.
Development of the Meter for Measuring Pork Quality
439
Fig. 4. The schematic of power
3.2 The RTC Module The RTC module is designed for ensuring the system time synchronization, the schematic of the RTC is shown in the figure 5. The PCF8563 is a CMOS Real-Time Clock and calendar optimized for low power consumption. A programmable clock output, interrupt output, and voltage-low detector are also provided.
Fig. 5. The schematic of the RTC
3.3 The Memory Module The 24LC02B is organized as one block of 256 X 8-bit memory with a 2-wire serial interface. Low-voltage design permits operation down to 2.5V, with standby and active currents of only 1 μA and 1 mA, respectively. its feature is low power consumption, simple external circuit, and write protection. The principle circuit of the
440
Z. Xing et al.
memory module is shown in the figure 6. The pins of A0, A1, and A2 connect to the ground, its address is 000, and the pin of the WP connects to the ground, so the chip can be all written. The pins of the SCL and the SDA connect respectively 10K resistor, and make the level of the SCL and the SDA high, so can conveniently read and write.
Fig. 6. The circuit of the memory module
4 System Software Design The system software is modular in design, and they include a data measurement and processing module, a data storage module, a serial interface module, a keyboard and a LCD module, and act. The events are handled at the before platform and backstage. When the system is power on, initializes kinds of parameters, and then enters into the waiting stage; when receiving the measurement command, the system begins to measure, process, and storage the data.
5 Experiment and Discussion 5.1 Experiment Analysis The test material was obtained from swine fillet and qualified by the quarantine system. The raw meat was fed to the meat grinder by inches to chop into fillet emulsion specimen, which were uniform in composition, within five times’ wring. Using the electronic balance divided the emulsion into 10g per portion by weight. Being intended to optimize the quality of test sample, delicately adding the demonized water into it until well-mixed was suggested. Afterwards, semi finished products were placed in the experimental beaker to prepare emulsion-aqueous solution of 10% and 20% respectively.
Development of the Meter for Measuring Pork Quality
441
Fig. 7. Sensor output voltages within distinguished water-addition samples
The curvilinear was shown in Figure 7, the points of each group fluctuated around a certain value. With different water content added, the output results varied from one sample to another. Apparently, the more water it contained, the higher output voltage it may afterward lead to, and vice versa. Taking the 10%-water-addition sample for an instance, we got the average of 3.10 and standard deviation of 0.01457 by calculating. The value was 0.47%, which indicated a low dispersion degree. Similarly, that of the 20%-water-addition sample was 0.49%. So we dare to say, with the same testing sample, different sensors independently applied have obtained tolerably similar results, whose service performance and effectiveness can meet the need of measuring. 5.2 Discussions Based on the research of the coaxial transmission line theory, the sensor and the meter for measuring pork quality were designed. The measurement results show that the stability and consistency of the sensor meet measurement requirements. The plan of a low-power, battery-powered portable meter is designed. The characters of the MSP430F149 are low power consumption and real-time wake-up function, so the power consumption of the measuring meter of pork quality is rather lower, and this feature is ideal for portable meter. A friendly human-machine interface and standard PS/2 interface are very convenient for users. In addition, the pork quality includes food quality, nutritional quality, technical quality, health quality and human quality, etc. At present, the quality of pork was inspected mainly through PH value, odor, color and the content of the water, the detection methods of using the PH value, odor, color were not in time in measuring the pork quality. The content of the water in the pork impacted on the pork quality was only considered in this paper; other parameters were ignored in this paper. If the pork quality was comprehensive responded, we must considered other factors, such as the PH value, odor, color, act. So the system of the pork quality evaluation is not put forward, and requires further study.
442
Z. Xing et al.
References 1. Goldenberg, A.A., Lu, Z.: Automation of meat pork grading process. J. Computers and Electronics in Agriculture. 16(2), 125–135 (1997) 2. Ding, H., Duan, Y., He, T.: Spectral Research On The Effect of Optical Propa Gation in The Fat. J. Spectroscopy and Spectral Analysis 15(6), 19–24 (1995) 3. Ji, R., Wang, Z., Huang, L., et al.: Study of Measurement Methods for Detecting the Water Content in Deep Layer of Fresh Meat. J. Modern Scientific Instruments 1, 119–121 (2006) 4. Ding, Q., Wang, Z., Huang, L., et al.: Development of portable bio-impedance spectroscopy system for measuring porcine meat quality. J. Transactions of the CSAE 12(25), 138–144 (2009) 5. Zhang, L., Lyng, J.G., Brunton, N., et al.: Dielectric and thermo physical properties of meat batters over a temperature range of 5-85 . J. Meat Science 68(2), 173–184 (2004) 6. Kent, M., Peymann, A., Gabriel, C., et al.: Determination of added water in pork products using microwave dielectric spectroscopy. J. Food Control 13(3), 143–149 (2002) 7. Zhang, L., Lyng, J.G., Brunton, N.P.: The effect of fat, water and salt on the thermal and dielectric properties of meat batter and its temperature following microwave or radio frequency heating. J. Journal of Food Engineering 80, 142–151 (2007)
℃
Experimental Investigation of Influence on Non-destructive Testing by Form of Eddy Current Sensor Probe Fengyun Xie* and Jihui Zhou School of Mechanical and Electronical Engineering, East China Jiaotong University, Nanchang 330013, China
[email protected] Abstract. Eddy current testing is a kind of non-destructive testing (NDT) method based on the principle of electromagnetic induction. The probe is the key of eddy current non-destructive inspection and its design or combination of form will influence detectability. A complete set of eddy current testing system is designed for standard testing copper and defect copper. Under the same liftoff distance and the same high frequency excitation signal, designed three types of probe, which are single probe single coil, single probe double coil and double probe double coil, are used to carry out NDT experiment. Experimental results show that sensitivity and resolution of detection system are obvious difference among the different form of eddy current sensor probe. Probe shape and coil winding are improved according to the experimental results. Corresponding improved probe is adopted to carry out NDT experiment at the same condition. Experimental results show that detectability is enhanced significantly. Keywords: Eddy current testing, Probe, Non-destructive testing (NDT), Sensitivity.
1 Introduction Eddy current testing based on the principle of electromagnetic induction is a kind of NDT method for testing metal semi-finished and metal components. It has been more and more widely used for its no need touch, high-speed of testing, easy automation, suitable for online testing and high detection sensitivity of surface defect. For example, YANG1et al, considered identification of corrosion fringe in pulsed eddy current non-destructive testing, He2 et al, considered pulsed eddy current technique for defect detection in aircraft riveted structures. However, the form of sensor probe has a great influence on the detection performance. In recent years, it is a research focus of many scholars how to improve the sensitivity of eddy current probes. But the discussion only is in the influence of one aspect, rather than the system. For example, Ren3 et al, proposed to increase the probe diameter or adapt big pie-type probe to improve the detection sensitivity, but it will reduce detection sensitivity of small defects. Kim4 et al, considered a dual-electromagnetic sensor *
Corresponding author. Tel.: 0086-791-7046135; Fax: 0086-791-7046122.
D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part IV, IFIP AICT 347, pp. 443–449, 2011. © IFIP International Federation for Information Processing 2011
444
F. Xie and J. Zhou
system for weld seam tracking of I-butt joints. Li5 et al, analyzed detection principle and influence factors of eddy current. Yu6 et al, designed a novel pulsed eddy current testing probe based on 3D magnetic field measurement. Gao7 et al, studied on structure optimization of eddy current probe, and obtained three influence factors on eddy current sensitivity: exciting coil of eddy current is stronger than slot width, and slot width is stronger than detection coil of eddy current. In this paper, we summarize the previous discussion, full unscramble measurement principle of eddy current. The procedure is: to design a placed device of eddy current testing; to use different forms of probe to carry out comparative test to the standard copper and defect copper. The specific objectives are: to get the relations between probe form and sensitivity of NDT; to put forward recommendations for improvement of the probe.
2 Testing Principle and Methods of Eddy Current Eddy current testing is a kind of NDT method based on the principle of electromagnetic induction, which is applied to conductive materials. Testing principle is the following: the signal generator provides high frequency alternating current to the probe in the detection coil, detection coil produces an alternating magnetic field, and the specimen produces eddy current. Eddy current is affected by the properties of the specimen, and in turn it changes impedance (voltage) of coil, then oscilloscope output impedance (voltage) of coil. Testing process includes picked up signal, signal amplification, signal processing, eliminated interference signal, and display and record test results. There are two kinds of detection methods. One is a double-coil high frequency reflection, whose schematic diagram is shown in Fig. 1.it has a high detection sensitivity of surface defect, but it is lower defect sensitivity with deeper below the surface; The other is double-coil low frequency transmission, whose schematic diagram is shown in Fig. 2.
Fig. 1. Schematic diagram of low frequency reflection
Fig. 2. Schematic diagram of high frequency transmission
Two methods have the same testing process. Signal generator produced alternating current flows through the coil, when the probe moves to the specimen, the eddy current will decrease. Impedance of the coil is displayed and recorded by the meter.
Experimental Investigation of Influence on Non-destructive Testing
445
Using auxiliary equipment to do comparative analysis of results from Standard specimens and defect specimens can obtain NDT results.
3 Theoretical Basis for Experimental Study The workpiece signal of detected is from impedance of the detection coil or the induced voltage changes of secondary coil in eddy current testing. We use long straight solenoid of cylindrical conductor as the object of research, suppose the radius of cylindrical conductor as a less than the solenoid inner radius b, and suppose turns of per unit length as n. In the cylindrical conductor (0 2 standard deviation) 8.11%
600
T. Shen, J. Zhang, and J. Wang
Fig. 6. The non-point source pollution risk distribution map of the study area
3 Conclusion and Outlook This study describes the development process and the basic method of the study of the non-point source pollution and introduces the application of the spatial information technology in the non-point source pollution research. On this basis, we collect a variety of spatial data of the study area and use the MUSLE equation to divide the study area into safe region, low risk region, middle risk region and high risk region. The results show that the more serious regions of the non-point source pollution in the study area distribute in mountainous and hilly areas, where is steeper, has more developed river system and is prone to runoff, so these area has higher risk to bring non-point source pollution. In this study, we use the spatial model to study the non-point source pollution. It is showed that the spatial model requires few amount of data, saves time and labor, is simple and with full scientific. Although it can not calculate a specific non-point source pollution load, it gives the risk area of the non-point source pollution. So this study provide a strong basis to facilitate the non-point source pollution management and control for environmental managers and policy makers. We can see that the spatial information technology is a powerful tool to study the non-point source pollution and will be indispensable and play an important role in the future.
Acknowledgments Project supported by the Science and Technology Projects of Qingdao (09-1-1-53-nsh and 08-2-1-36- nsh).
Research on Non-point Source Pollution Based on Spatial Information Technology
601
References [1] Wang, X.Y.: Theory and Methods of Quantitative Study on Non-Point Pollution. Journal of Capital Normal University (Natural Science Edition) 17, 91–95 (1996) [2] Vladimir, N., Gordon, C.: Handbook of Nonpoint Pollution Sources and Management, 1st edn. Van Nistrand Reubhold Company, New York (1981) [3] National Water Quality Inventory: Report to Congress Executive Summary, USEPA, Washington (1995) [4] Jin, X.C., Liu, S.K., Zhang, Z.S.: China Lake Environment, 1st edn. China Ocean Press, Beijing (1995) [5] Wang, X.Y.: Non-point source pollution and its management, 1st edn. China Ocean Press, Beijing (2003) [6] Wischmeier, W.H., Smith, D.D.: Predicting rainfall erosion losses. a guide to conservation planning. Agriculture Handbook No. 537, U.S. Department of Agriculture, USDA, Washington (1978) [7] Arnold, J.G., Allen, P.M., Bemhardt, G.A.: A comprehensive surface-groundwater flow model. Journal of Hydrology 14, 47–69 (1993) [8] Xia, Q.: Calculation of non-point source pollution load in the watershed mode. China Environmental Science 5, 23–30 (1985) [9] Shan, B.Q., Yin, C.Q., Bai, Y.: Study on phosphorus load from a watershed with rainfall simulation method. Acta Scientiae Circumstantiae 20, 33–37 (2000) [10] Cai, C.F., Ding, S.W., Shi, Z.H.: Prediction on soil nutrients losses at typical small wate rshed of three gorges area with gis. Journal Of Soil And Water Conservation 5, 9–12 (2001) [11] Bao, Q.S., Wang, H.D.: China’s non-point source pollution of water environment research and prospects. Scientia Geographica Sinica 16, 66–71 (1996) [12] Liu, F., Wang, H.D., Liu, P.T.: Watershed non-point source pollution identification and quantification in the Yuqiao Reservoir Basin. Acta Geographica Sinica, 329–339 (1988) [13] Liu, Y., Zhang, T.Z., Chen, J.I.: Discussion on charge policy for governance of agricultural non-point source pollution in dianchi watershed”. Journal of Xiamen University (Natural Science) 11, 787–790 (2003) [14] Pan, G.X., Jiao, S.J., Li, L.Q.: Effect of longterm fertilization practices on mobility of phosphorus in a huangnitu paddy soil receiving low P input in the Taihu lake region, Jiangsu province. Chinese Journal of Environmental Science 3, 91–95 (2003) [15] Yan, W.J., Bao, X.: Study on agricultural movement of Chaohu lake basin and non-point source pollution. Journal of Soil and Water Conservation 12, 128–132 (2001) [16] Wang, S.P., Yu, L.Z., Xu, S.Y.: Research of non-point sources pollution loading in suzhou creek. Research of Environmental Sciences 27, 20–23 (2002) [17] Zhen, Y., Wang, X.J.: Advances and prospects for nonpoint source pollution studies. Advances in Water Science 13, 105–110 (2002) [18] Sivertun, A., Reinelt, L.E., Castensson, R.: A GIS method to aid in non-point source critical area analysis. International Journal of Geographical Information Systems 2, 365–378 (1988) [19] McElroy, A.D., Chiu, S.Y., Negben, J.W.: Loading functions for assessment of water pollution from non-point sources, 1st edn. US Environmental Protection Agency, Washington, DC (1976)
The Regulation Analysis of Low-Carbon Orientation for China Land Use Bikai Gong and Bing Chen Institute of Land Reclamation and Ecological Restoration, China University of Mining and Technology (Beijing); Engineering Research Center of Mining Environment & Ecological Safety, Ministry of Education, 100083 Beijing, China
[email protected] Abstract. Greenhouse gas emission reduction has become the responsibility and consensus of all mankind’s development, the way of non-sustainable land use results in a lot of greenhouse gas emissions in the context of low-carbon economy. On the basis of revealing impact of land use on carbon emissions and status of carbon emissions, this paper proposes the regulation proposals of land use based on low-carbon emission: make carbon emission’s list of land use, optimize land use structure and layout, and strengthen land use control, intensive land use, soil and water conservation and ecological protection, management of agricultural land, woodland, grassland and wetland, control the scale of construction land and pace of expansion, establish compensation mechanism of land ecology. The research has great significance for China forming and further improving the policy system of land use of low-carbon emission. Keywords: land use; low-carbon economy; regulation; China.
1 Introduction “Our future energy – creating low-carbon economy” of the British Government in 2003 first proposed the concept of low-carbon economy, the core idea is to obtain more economic output with less energy consumption[1]. Then low-carbon economy attracted international attention, it advocates resource conservation, friendly environment and sustainable development, and reduces high-carbon energy consumption of coal, petroleum and others possibly, improve energy efficiency substantially, large-scale utilize renewable and low-carbon energy, large-scale develop emission reduction technology of greenhouse gas, build low-carbon society and maintain ecological balance. The development of low-carbon economy is global revolution involved in mode of production, lifestyle, values, national interests and human destiny, as well as an inevitable choice of the global economy being transferred from high-carbon energy into low-carbon energy[2, 3]. The development of low-carbon economy is not only a global effective way to climate warming, but also an inevitable choice of China saving energy, reducing emission and participating in international cooperation. China is in period of rapidly developing industrialization and urbanization, the coal –dominated energy structure D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part IV, IFIP AICT 347, pp. 602–609, 2011. © IFIP International Federation for Information Processing 2011
The Regulation Analysis of Low-Carbon Orientation for China Land Use
603
results in China’s emission of carbon dioxide in the forefront of the world. In 19992003, the emission of carbon dioxide increased by 170 million tons, and China has become the world’s second largest emission country[4]. On the research of lowcarbon economy, some China scholars proposed many useful ideas. Zhuang Guiyang showed that the essence of low-carbon economy is the problem of high-energy efficiency and clean energy structure, the core is the innovation of energy technology and system[5]. Zhang Kunmin analyzed the facing challenge of China developing low-carbon economy from the view of energy, and thought energy strategy of supporting sustainable economic and social development needs to be build[6, 7]. Xia Kunbao showed that low-carbon economy is the only way to achieve sustainable urban development; it needs to carry out mode of low-carbon production and consumption, and greatly develop circular economy and clean production[8]. Low-carbon economy seems to be the problem of environmental technology’s application and industrial structure’s optimization, but the problem of land use essentially. China has been concerned about impact of land use on global change for a long time, and proposed policy advice on the regulation of land use based on impact of land use on carbon emissions and status of carbon emissions. Conventional regulated policies of low-carbon, with the use of policies of land use structural optimization, and the introduction of regulated measures of land use, have a great significance of China performing and successfully achieving the promise of voluntary emission reduction.
2 Impact of Land Use on Carbon Emissions Glaeser and Kahn analyzed the relationship between carbon emissions and land use, and thought constrains on land use are more strict, carbon emissions’ level of the residents living are lower[9]. The change of land use is an important uncertainty factor in the estimates of carbon emission. On the one hand, the change of land use changes the types of ecosystem directly, and then changes the net primary productivity of ecosystem and their inputs of soil organic carbon; on the other hand, the change of land use changes soil physical and chemical properties potentially, and then changes soil respiration’s sensitivity coefficient to temperature changes. Land use plays an important role in increasing global atmospheric carbon dioxide, and China’s change of land use has great impacts on carbon emission of terrestrial ecosystem. The way of long-term non-sustainable land use (deforestation, reclamation of grassland, transformation of marshes, etc.) results in the release of carbon from terrestrial ecosystem, and becomes one of the main reason of atmospheric carbon dioxide rising, only second to fossil fuel burning. According to the calculations of the World Resource Organization and well-known experts on the carbon cycle, in the global carbon emissions from 1850 to 1998, land use change and its carbon emissions accounted for 1/3 of the total emissions of human activities affect; while the cumulative carbon emissions of the whole country’s land use change were 10.6PgC from 1950 to 2005, accounting for 30% of the total artificial carbon source emissions, as well as 12% of carbon emissions of global land use change at the same period[10]. There are direct and indirect carbon emissions from the classification of carbon emissions of land use. Direct carbon emissions can also be subdivided into carbon
604
B. Gong and B. Chen
emissions of changed and maintained land use types. The former refers that the change of land use or cover type results in carbon emissions caused by the change of ecosystem type, such as deforestation, reclaiming land from lakes, construction land expansion and so on; the latter refers to carbon emissions driven by the change of land management and carbon sink of ecosystem, including farming, grassland degradation, nutrient input and changes of cropping systems. Indirect carbon emissions mainly refer to all man-made carbon source emissions carried on each type of land use, including the heating of neighborhood, exhaust from traffic land, and process emissions of mining land; they are spatial intensities and distributive effects of artificial carbon source emissions on the different land use types.
3 The Status of Carbon Emission Reduction of Land Use In 2006, China’s disposable energy consumption accounted for more than 16% of the world, and the emissions of carbon dioxide were more than 20% of the world, equal with per capita emissions in the world. This indicated that, in the process of industrialization and urbanization, the intensity of carbon emissions is a little high, while energy consumption will continue to grow, with not much room for carbon emissions. Describe the statue of carbon emissions of China’s land use based on direct and indirect carbon emissions of land use, carbon emissions effect of land use types, and regional carbon emissions. 3.1 Direct Carbon Emissions of Land Use From 1980 to 2005, terrestrial ecosystem has presented obvious carbon sinks, and the average annual level of carbon sinks were about 154 to 167 million tons of carbon. From the view of classification structure, vegetation and soil carbon pools both presented the functions of carbon sinks, the average annual carbon sinks of vegetation was 108 to 121 million tons, and the capacity of soil carbon sinks was weak with 1/3 of vegetation carbon sinks. From the view of ecosystem types, the function of forest carbon sinks had the obvious effect on the whole terrestrial ecosystem, about 2/3 of the whole terrestrial ecosystem. From the view of spatial pattern, the carbon sinks of land use had more significant effect in east, south and north China, and there was more significant effect of carbon emissions of land use in northeast and southwest China, while there was less effect in northwest China. 3.2 Indirect Carbon Emissions of Land Use Take the comprehensive level of carbon emissions of China’s four departments in 1995 for an example, the total emissions of carbon dioxide were 2.642 billion tons, methane was 32 million tons, and the carbon dioxide equivalent was about 3.3 billion tons. In 2005, the total emissions of carbon dioxide were 5.55 billion tons, methane was 38 million tons, and the carbon dioxide equivalent was about 6.34 billion tons. In the 80th of last century, artificial source emissions were as 3 times as the storage of terrestrial ecosystem, but 10 times in 2005. Therefore, the increase of artificial resource carbon emissions was much faster than the promotion of carbon’s absorptive capability of terrestrial ecosystem[10].
The Regulation Analysis of Low-Carbon Orientation for China Land Use
605
3.3 Carbon Emission Effect of Land Use Types Without considering the situation of carbon’s absorption, the general carbon emission intensity of agricultural land is 0.37 tons per hectare, below the international average level. The main reason is that China traditional agriculture paid attention to the use of organic fertilizer and straw, so the effect of organic carbon storage of agricultural soil was remarkable. Forest is an important carbon sink, the carbon emission intensity is 0.06 tons per hectare and the carbon absorption intensity is 0.49 tons per hectare. Since the 70th or 80th of last century, the strong advocating of afforestation has resulted in forest volume rising, large young forest had obvious carbon absorption effect of growth, stronger than at the same period of timber harvesting, firewood collection, disaster interference and other effects. For construction land, there were the highest volume and intensity of carbon emissions. The emission intensity of construction land was 55.8 tons per hectare in all. That means that there will be 149.8 times of carbon emission when transforming one hectare of agricultural land into construction land; while transforming one hectare of forest land into construction land, it will increase 929 times of carbon emissions. From the view of different regions, there were the highest carbon emission intensity of construction land in north and east China, about 81.2 tons per hectare and 65.3 tons per hectare; the general level in northeast, middle south and southwest China, about 48.8 tons per hectare, 46.5 tons per hectare and 49.1 tons per hectare; lower level in northwest China about 33.9 tons per hectare. In a word, there are greater carbon emissions of per unit area of construction land in areas of high levels of heavy industrialization and big population density. From the view of the internal structure of construction land, there was highest carbon emission intensity in industrial and mining land, about 196 tons per hectare; followed by traffic land, about 43.7 tons per hectare; the smallest was urban and rural residential land, only about 8.3 tons per hectare[10]. 3.4 Carbon Emissions of Regional Land Use As the world’s major energy consumer, China’s carbon emissions are not only reflected on volume’s growth, but also spatial pattern change of carbon emissions. From the view of the change of large regional system, the carbon emissions in the east have dominated the country all the time; the proportion of central region’s carbon emissions in the country showed the trend of flatting to down; and the proportion in the western region was small but basically maintained an upward trend[11]. Specifically as follows: there were high carbon emissions in most regions of east and north China, as well as Pearl River Delta, Yangtze River Delta, middle-south cities of northeast China and Chengdu-Chongqing Urban Agglomeration. There were lowcarbon emissions or carbon balances in northwest China and most of Qinghai-Tibet Plateau. While the main areas of carbon sinks were in most of southwest, middlesouth, southeast China, part of northeast China, and Tianshan-Qingling zones. Besides, the afforestation areas in north and part of northwest China had a positive effect of carbon sinks.
606
B. Gong and B. Chen
4 The Regulated Suggestions of Land Use on the Basis of Low-Carbon As the world’s largest developing country, China’s economic development rely on resources consumption of fossil energy greatly, that resulted in increased carbon emissions and heave environmental pollution, these issues have seriously affected the quality of economic growth and the sustainability of development. The economic and social development drove China’s carbon emissions, but the policy of increasing sinks such as commonly ecological protection and afforestation had no significant influence on containing carbon emissions increase. While the potential of carbon emission reduction after the optimization of land use structure in low-carbon economic development and macro-control is 1/3 of conventional low-carbon policy. China has been highly concerned about energy conservation and land use on the impact of global change, on the basis of revealing the effect mechanism of carbon emissions of land use, form land use structure and layout of low-carbon emissions though innovating techniques of land use planning. That had great significance for further improving the policies of carbon emission reduction, particularly forming China’s policy system of land use of low-carbon emissions. 4.1 Make the List of Carbon Emissions of Land Use At present, home and abroad related standard of calculating carbon emissions are mainly on the basis of “2006IPCC Guidelines for National Greenhouse gas inventories”, and “Initial National Communication on Climate Change” based on the domestic status in 1994, recently the new “Second Communications on Climate Change” has not yet to be introduced. The above files or standards have been difficult to meet the actual demand of the land department making decision, so carbon emission analysis of land use fitting the situations need to be made according to China’s unique classification system of land use, vegetation and soil. Overall China natural and social-economic carbon emissions of land use for nearly 20years, form carbon emission lists of fitting land use and its vegetation characteristics, on line with the current classification system of land use, and develop appropriate standards, in order to provide theoretical basis for macroscopic policymaking department to carry out low-carbon land use plan. 4.2 Establish Matchable System of Low-Carbon Land Use Policy Currently, developing low-carbon economy and taking the road of low-carbon ecology is an inevitable direction of transforming in many cities, especially resourcebased cities. In order to change the way of low-carbon land use and set up awareness of low-carbon land, China also needs to propose fitting matchable system of land use policy based on low carbon from two views of carbon emission reduction and increased carbon sinks. The policy of increased carbon sinks includes optimization of land use structure, water and soil conservation, ecological protection, and the management of forestland, farmland, grassland and wetland; the policy of carbon emission reduction includes optimization of land use structure, carbon emission reduction of agricultural land and construction land, mechanism’s construction of land ecological compensation and others.
The Regulation Analysis of Low-Carbon Orientation for China Land Use
607
(1) Optimize the structure and layout of land use From the view of emission reduction of increased carbon sinks of various land management policy, there are greatest potential for carbon emission reduction of land use structure’s optimization. Innovating planning techniques of land use can further optimize the structure and layout of land use. Restricting non-sustainable way of land use, increasing the scale of forest area, controlling the reduction of cultivated land, grassland, marsh, beach and the expansion of construction land, and promoting the transformation from unused land into woodland, grassland and cultivated land, will produce more positive effects for carbon storage of China ecosystem. (2) Strengthen land use control China has been in period of rapid development of industrialization and urbanization, large scale infrastructure construction should be carried out by the way of low power, high efficiency and low-carbon emissions. Many cities are in the time of rapidly expanding, necessary controls and regulations need to be carried out for development of urban traffic and urban land at the same time[12]. Use urban dynamic emulation analysis, and fully stimulate traffic distance and carbon emissions of different granting schemes in the supply of urban land, in order to achieve balanced distributions of all kinds of land in the urban and scientifically plan to reduce urban carbon emissions. Besides, promote industrial low-carbon by the policy of supplying land. Through land participating in macro-control, strengthen land review, strictly control blind expansion of “three high” industries (high intensity of energy consumption, high carbon emissions, high pollution), and encourage the development of low-carbon industry. When increasing approvals of construction land, local government will give priority to protect industrial land of low-carbon economy, contain project sites of excess capacity and duplicate construction. Build cyclic economic parks and lowcarbon developed test areas of land use. Accelerate establishing demonstration industrial parks, industrial parks and economic development parks of energy-saving low-carbon. Overall urban and rural construction with the idea of low-carbon economy, and develop new characteristic industrialization and urbanization fitting local realities. (3) Strengthen the saving of intensive land use Strengthen the saving of intensive land use, reduce emission of architectures. The saving of intensive land use not only reduces the consumption of land resources directly, but also produces more advanced architecture models, more efficient operated ways and more scientific streaming configurations though being replaced by more abundant elements. These will reduce the output of energy consumption and carbon emission per unit. (4) Strengthen water and soil conservation and ecological protection In important areas, sensitive areas and ecologically fragile areas of affecting the national ecological security, strictly control land desertification and water and soil erosion, and strengthen water and soil conservation and ecological protection; in the desert and Gobi areas, speed up the greening process of land and afforestation, newly increase areas of planting forest and forest carbon sinks.
608
B. Gong and B. Chen
(5) Strengthen management of agricultural land Develop new agriculture of organic, ecology, and efficiency; reduce the use of chemical fertilizers and pesticides. The potential for carbon emission reduction of agricultural land’s management can not be ignored. According to the study, the carbon sequestration capacity of agricultural soil has increased 0.014%per year in the last few years with the improvement of the quality of cultivated land. Deeply study the carbon sequestration of farmland ecosystem, mitigate climate changes through biological and ecological carbon sequestration. Optimize farming system, promote the use of organic fertilizers, and improve the content of organic matter in soil. Straw returning to field increase soil nutrients and reduce soil erosion by wind and water. With materials of livestock manure, crop straw and other agricultural organic wastes, relying on the technology of recycling economy, achieve the exchange of each agricultural by-products and recycling use of wastes, and turn organic wastes into resources. (6) Strengthen management of forestland, grassland and wetland The studies show that, forest vegetation is the most effective carbon sink in the earth, the annual net carbon uptake of the global average per hectare of forest vegetation is about 0.26~0.39t. Developing forestry is an important measure to develop low-carbon economy, plant trees and grasses, strengthen management of wetland, and expand carbon sinks. Since the 80s of the 20th century, due to the large scale of afforestation and returning farmland to forest, animal husbandry, and lakes, the level of carbon storage in terrestrial ecosystems has improved significantly, 1/4~1/3 of artificial carbon emissions at the same period were absorbed[10]. Deeply study the carbon sequestration of grassland and forest ecosystem, mitigate climate changes through biological and ecological sequestration. Since 1949, there have been great achievement of afforestation, but the existing level of forest’s maintenance and management should be improved, in order to reach the maximize of making oxygen from forest. For the main content of afforestation, speed up the construction of ecological barriers, continue to promote the construction of shelterbelts of upper and middle yangtze river and three-north area, increase forest coverage, and improve territory virescence. (7) Control the scale of construction land and the speed of expansion The requirements of coordinating urbanization and low-carbon are defining the scale of construction land, spatial scope and expanded pace of the overall urban planning in the level of urban planning. And the space of construction land are divided through the analysis of eco-environment capacity and development space in the level of urban planning, including the balanced mold of the whole shape of low-carbon, dynamic balance of low-carbon land use, green system’s design of low-carbon road and others[13]. (8) Establish the mechanism of land ecological compensation The total values of land resources include not only economic value, but also ecological value and social value[14]. Land ecological compensation includes that the state could expropriate tax of environmental resources, in order to ensure the government provide public service functions of ecological products of land resources. Besides, people who suffers the loss or pays the economic costs because of public
The Regulation Analysis of Low-Carbon Orientation for China Land Use
609
interests (ecological protection) should be given fair compensation. Through accurately measuring and calculating areas of land for different use, combining with other empirical data, ecological value formed by the status of land use can be calculated.
References 1. Zhang, Q., Ye, X.-p., Chen, G.-w.: Low-carbon urban planning:a new vision. City Planning Review 34(2), 13–18 (2010) 2. Zhang, G.-f., Liu, Q.-w.: A study on low carbon economy policy based on equity theory. China Coal 36(1), 6–8, 12 (2010) 3. Huang, W.-s.: On the low-carbon tourism and the creation of low carbon tourist attractions. Ecological Economy (11), 100–102 (2009) 4. The carbon emission report of world-bank[EB/OL], http://i.cn.yahoo.com/billasx/blog/p22/ 5. Zhuang, G.-y.: The way and potential of China’s low-carbon economy development. Studies in International Technology and Economy 8(3), 8–12 (2005) 6. Zhang, K.-m.: China’s role,challenges and strategy for the low carbon world. China Population Resources and Environment 18(3), 1–7 (2008) 7. Zhang, K.-m.: Development of low-carbon economy is China’s internal demand. Theoretical Horizon (2), 26–28 (2010) 8. Xia, K.-b.: Development of low-carbon economy,urban sustainable development. Environmental Protection (3), 33–35 (2008) 9. Glaeser Edward, L., Kahn Matthew, E.: The greenness of cities:carbon dioxide emissions and urban development. Journal of Urban Economics 67(3), 404–418 (2010) 10. Carbon emissions: issues of land use regulation and control[EB/OL] (December 25, 2009), http://www.mlr.gov.cn/tdsc/lltt/200912/t20091228_131048.htm 11. Lei, Z., Huang, Y.-x., Li, Y.-m., et al.: An investigation on spatial changing pattern of co2 emissions in China. Resources Science 32(2), 211–217 (2010) 12. Pan, H.-x.: Urban spatial structure towards low carbon:new urban transport and land use model. Urban Studies 17(1), 40–45 (2010) 13. Han, Q., Liu, H.-l.: Low-carbon eco-towns planning and research. Development of Small Cities & Towns (12), 73–78 (2009) 14. Gong, B., Deng, L., Hu, Y., et al.: Study on ecosystem service and its value assessment of cultivated land of reservoir inundated—a case study of Huangjinping hydropower station. Resource Development & Market 23(12), 1085–1088 (2007)
A CDMA-Based Soil-Quality Monitoring System for Mineland Reclamation Dongxian He1, Daoliang Li2, Jie Bao3, and Shaokun Lu1 1 Key Lab. of Agriculture Engineering in Structure and Environment, College of Water Conservation and Civil Engineering, China Agricultural University, Beijing, P.R. China 2 College of Information & Electrical Engineering, China Agricultural University, Beijing, P.R. China 3 College of Engineering, China Agricultural University, Beijing, P.R. China
[email protected] Abstract. After the mining of mineral resource, mineland resources utilization became increasingly complicated. Soil and planting information can be used as the indicators to evaluate mineland reclamation level. In order to obtain the soil and planting information in real time, a CDMA-based soil-quality monitoring system was developed to wirelessly monitor air/soil environment and plant image in remote or local area. As a system test, a monitoring system equipping one network camera and four sensors including air temperature, relative humidity, soil temperature, and soil water content was deployed at a national mineland reclamation demonstration in Fuxin, Liaoning and communicated with a remote server in Beijing using CDMA service with IPsec-based VPN connection. Via a one-year testing, the CDMA-based soil-quality monitoring system as a dynamic infrastructure and available tool showed a good performance for mineland reclamation quality evaluate. Keywords: CDMA service, Mineland reclamation quality evaluate, Soil water content, VPN connection.
1 Introduction After the mining of mineral resource, mineland resources utilization became increasingly complicated. In order to obtain the mineland detail information in real time for the reclamation in land administrative department of mining enterprise concerned, soil and planting information changes of mineland area is very important and can be used as the indicators to evaluate mineland reclamation level. How to get the indicator information and how to sign the reclamation quality is concerned to sustainable development of mineland reclamation utilization. Therefore, efficiency and low cost way to get mineland reclamation quality change is essential and required for mining enterprise and land administrative department. Mining land reclamation by the quality indicator technology developed in recent decade is based on wireless network technology, GIS, and remote sensing technology D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part IV, IFIP AICT 347, pp. 610–615, 2011. © IFIP International Federation for Information Processing 2011
A CDMA-Based Soil-Quality Monitoring System for Mineland Reclamation
611
can be used as the integration charge of indicators of reclamation quality index. In the past 10 years, mineland reclamation in the quality indicator technology has carried out a significant development and has achieved remarkable achievements [1-8]. Therefore, development of a dynamic monitoring system of mineland reclamation for obtaining the soil and planting information in an efficiency and low cost way is becoming required and becoming more important.
2 Materials and Method 2.1 System Configuration The CDMA-based soil-quality monitoring system consists of a CDMA module with IPsec-based VPN function (InRouter210C, Beijing Inhand Co., China), a network bullet camera (IP7330, Vivotek Co., Taiwan), a web-server-embedded chip (PICNIC2.0, TriState Co., Japan), switching hub (TL-SF1005D, TP-Link Tech. Co., China), a sensor module, and a solar power supply module (SPV-DCRC-20W, Beijing Sangpu Co., China) (Fig. 1). The sensor module is including temperature (PT1000, Hayashi Elect. Co., Japan), relative humidity (CHS-UPS, TDK Co., Japan), soil temperature (PT1000, Hayashi Elect. Co., Japan) and soil water content (EC-5, Decagon Co., US). The power supply module consists of a power control device, two 20W monocrystalline silicon photovoltaic panels, a 75AH battery (LC-X1275ST 75AH, Panasonic Co. Japan), and a timer (ZN48T-12V, Symore Co. China). The realtime data and images collected by the sensors and network camera are dynamically and wirelessly transported to a remote server via TCP/IP. The CDMA-based soilquality monitoring system was operated between appointed duration by a timer.
Fig. 1. The CDMA-based soil-quality monitoring system
612
D. He et al.
2.2 Topological Architecture Each CDMA-based soil-quality monitoring system as an isolated local area network is connected to Internet via CDMA service by Telecom China (Fig. 1). That means the monitoring system as a sensing network node can be constructed to a large-scale wireless sensing network under CDMA signal covered areas. The monitoring system is also easy to increase sensor channels and network cameras as well as the CDMA bandwidth depending. In this monitoring system, four sensor channels were integrated into the webserver-embedded chip within one LAN IP and each camera has a LAN IP. All data and images are communicated by dynamic HTTP file and identified dynamically and storied by one to five remote servers. The CDMA module will be dynamically connected to the remote servers via IPsec-based VPN security technology. In order to identify the specified the remote servers, the remote VPN routers have to use a dynamic domain connection or a fixed global IP to support the remote VPN calling. The monitoring systems deployed anywhere will become to a local network connection because of the IPsec-based VPN tunnels connected. Therefore, all accredited users or clients can visit or manage the wireless sensing devices anywhere and anytime under Internet environment. 2.3 Testing Environment A CDMA-based soil-quality monitoring system was deployed in a national mineland reclamation demonstration zones in Fuxin, Liaoning province for obtaining air and soil environment, and plant images. In this testing, one remote server with IPsec-based VPN router (BV-601, Nesco Co., China) is deployed in China Agricultural University located in Beijing, China. The testing experiments were conducted for one year.
3 Results and Discussion 3.1 Dynamic Image Transportation A Java applet program was installed in the remote server to download image from the network camera via HTTP every minute and stored in the remote server. In order to monitor more areas, the network camera has setup in two benchmark level for measuring height of plants and leaf area index (Fig. 2). The image processing results
Fig. 2. The dynamic images obtained by the CDMA-based soil-quality monitoring system
A CDMA-Based Soil-Quality Monitoring System for Mineland Reclamation
613
will be described by other papers and omitted in this paper. All accredited users can visit or manage the network camera modules and any clients can visit the real-time or historical images from website anywhere and anytime under Internet environment. Therefore, this network camera module with an enough-quality dynamic image requirement at low cost is suitable to the monitoring system. 3.2 Dynamic Environment Data Transportation The environmental data such as temperature, relative humidity, soil temperature, and soil water content were dynamically storied or issued in webpage or Extensible Markup Language (XML) file by a special JAVA applet program in the remote servers (Fig. 3). The 10-bit analog signals of the sensors were obtained by the webserver-embedded chip without any memory and firmware. Therefore, the legacy problem of the USB-based or RS232-based devices caused from the firmware update can be solved in this monitoring system. The environmental data were also used to combine with image processing results such as the shape and color features to construct an artificial neural network model based on back-propagation algorism for identifying the growth measurement and nutrition detection. The sensor module is easy to increase the sensor channels with a little influence in data communication, because the narrow CDMA bandwidth will strongly influence the image communication in this monitoring system.
Fig. 3. Environmental data measured by the CDMA-based soil-quality monitoring system
3.3 Network Communication and Electronic Consumption The monitoring system were communicated with the remote server in Beijing by a 20-30 Kbps access speed with over 25-27 signal quality level of CDMA services. The real-time image with 640*480 pixels resolution captured by the network camera and
614
D. He et al.
compressed to JPEG file were dynamically transported to remote server during the system operation. The all data and images obtained at same time were identified by IP which is transferred to take turns and with omitted disconnecting IP. In this case, less than 2% of packet loss rate was resulted. Local user can visit the remote server via LAN, remote users can visit the remote server through Internet. These two methods have nothing to do with the CDMA. Therefore, remote user access speed is limited by the bandwidth of Internet hotspot such as ADSL network in this case. The remote server can support simultaneous accesses of 20 remote users. However, the numbers of concurrent local users do not have measurements and no restrictions on. The monitoring system using 12V DC power supply and timed operation by solar supply system consisting of two 20W solar panels and a 75Ah battery were consumed about 22W capacity with 1.8A electric current. This power supply module can drive the monitoring system for about 10 hours in sunshine day and for about 5 hours in cloudy day. Therefore, four hours operation divided to two hours in morning or afternoon were connected in this system testing.
4 Conclusion The real-time environment data and images were dynamically collected in remote area by the CDMA-based soil-quality monitoring system. This CDMA-based soilquality monitoring system based on web-server-embedded technology and CDMA service with IPsec-based VPN function as a node infrastructure is useful to construct a ubiquitous wireless sensing network in high-security for mineland reclamation.
Acknowledgments This work has been supported by International Technology Cooperation Program (2007DFA91050).
References 1. Bradshaw, A.: The use of natural processes in reclamation — advantages and difficulties. Lands. Urb. Plan. 51, 89–100 (2000) 2. He, D.X., Hirafuji, M., Fukastu, H., Yang, Q.: An environmental measurement system using wireless networks and web-server-embedded technology. In: Progress of Information Technology in Agriculture, pp. 541–545. China Agriculture Press, Beijing (2003) 3. Chena, J.C., Changb, N.B., Shieh, W.K.: Assessing wastewater reclamation potential by neural network model. Eng. Appl. Artif. Intell. 16, 149–157 (2003) 4. Fukatsu, T., Hirafuji, M.: Field monitoring using sensor-nodes with a web server. J. Robot. Mechatron. 17, 164–172 (2005) 5. Sevkiye, S.T.: An analysis on the efficient applicability of the land readjustment (LR) method in Turkey. Habitat. Int. 31, 53–64 (2007)
A CDMA-Based Soil-Quality Monitoring System for Mineland Reclamation
615
6. Anderson, J.D., Ingram, L.J., Stahl, P.D.: Influence of reclamation management practices on microbial biomass carbon and soil organic carbon accumulation in semiarid mined lands of Wyoming. Appl. Soil Ecol. 40, 387–397 (2008) 7. Philip, A.T., Helmers, D.P., Kingdon, C.C., McNeil, B.E., de Beurs, K.M., Eshleman, K.N.: Changes in the extent of surface mining and reclamation in the central appalachians detected using a 1976-2006 Landsat time series. Remote Sens. Environ. 113, 62–72 (2009) 8. Li, F.H., Keren, R.: Calcareous sodic soil reclamation as affected by corn stalk application and incubation: a laboratory study. Soil Sci. Soc. Chin. 19, 465–475 (2009)
Design and Implementation of a Low-Power ZigBee Wireless Temperature Humidity Sensor Network Shuipeng Gong1, Changli Zhang1,2, Lili Ma1, Junlong Fang1, and Shuwen Wang1 1
Northeast Agricultural University, Harbin, Heilongjiang Province, P.R. China, 150030,
[email protected] 2 Northeast Agricultural University, Harbin, Heilongjiang Province, P.R. China, 150030, Tel.: +86-451-55190456,
[email protected] Abstract. The key technology of greenhouse facilities is the monitoring of environmental parameters. Now, monitoring system of greenhouse is based on wire transmission. It is complicated to route wire and difficult to maintain. Also its reliability and anti-interference performance will degrade because of heat, light and acid. This paper leads a low-power and short range ZigBee technique into greenhouse monitoring system. In order to compose intelligent network sensor system, the paper analyses the composition of network nodes power consumption, proposes low-power design method both in hardware and software. The paper selects CC2430 module which composed of transceiver and microprocessor, and use SHT15 temperature humidity sensor. After hardware and software debugging, this wireless network can acquire and transmit data of greenhouse temperature and humidity accurately and rapidly, and this system resistant to stable work, tight structure, large loads and low power consumption. Keywords: Greenhouse, CC2430, ZigBee, Low-power, Temperature humidity sensor, Wireless network.
1 Introduction Now, monitoring system of greenhouse is based on wire transmission. It is complicated to route wire and difficult to maintain. Also its reliability and antiinterference performance will degrade because of heat, light and acid. As wire transmission brings many disadvantages, more and more monitoring systems use wireless sensor networks. ZigBee plays an important role in wireless sensor networks. ZigBee as a priority in wireless sensor networks because its advantages, such as low power consumption, high tolerance, Ad Hoc Networks [1]. In normal ZigBee network nodes use portable power (battery). Its harsh environment and large quantities make it very difficult to replace. So reducing the power consumption and extending the reliable working hours of sensor nodes become one of the key considerations. To compose intelligent sensor network system, the paper analyzes the consumption of the network and proposed low-power consumption methods both form hardware and software. D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part IV, IFIP AICT 347, pp. 616–622, 2011. © IFIP International Federation for Information Processing 2011
Low-Power ZigBee Wireless Temperature Humidity Sensor Network
617
2 System Design The system uses a tree topology and consists of coordinator, router and terminal node. Each router and terminal node carries temperature humidity sensor. The coordinator connects PC by RS-232 serial ports. It collects all nodes’ information and sends them to PC so that the manager can monitor and manage the information. Figure 3 shows the framework of the monitor system.
Fig. 1. The system uses a tree topology. This shows the framework diagram of the monitor system.
3 Low-Power Hardware Designs The hardware of the entire network includes coordinator and sensor nodes. Coordinator is powered by USB as coordinator connects with PC. So the low power strategy of the system mainly use in ZigBee nodes. ZigBee nodes are composed of processor module, wireless communication module, temperature humidity sensor module and power module. Figure 2 shows the structure of node.
Fig. 2. The node is composed of CC2430, SHT15 and power modules
618
S. Gong et al.
3.1 Processor Module Microprocessor is the central processing unit of the node. As the data that the microprocessor handled is large, the microprocessor is one of the most energy consuming components of the node. In the low power design, the microprocessor that the node used should be low power consumption and support sleep mode. As wireless communication account for most of the power consumption, the selection of the wireless module is very important. The following advantages of CC2430 make it become the ideal solution to solve the problem. CC2430 SoC integrates the CC2420 RF transceiver and enhanced 8051MCU. Its current consumption is 0.9uA on sleep mode and can be waked up by external interrupt or RTC wake-up system. Its current consumption is 0.6uA on dormancy mode and can be waked up by external interrupt. CCC2430 requires a large voltage supply between 2.0V and 3.6V [2]. 3.2 Sensor Module Sensor module uses temperature humidity sensor module SHT15. SHT15 integrates temperature humidity sensor, the conditioning and amplifying, A/D converting and I2C bus in one chip. The serial interface of SHT15 has a definite advantage both on the reading of sensor signor and power consumption. Current consumption is 550uA in measuring, 28uA in average, 3uA during sleep.
Fig. 3. This circuit diagram shows the interface of SHT15 and CC2430
3.3 Power Module Power module is mainly for CC2430 and SHT15. Voltage range of CC2430 is 2.03.6V; the SHT15 is 2.4-5.5V.The power of the system uses 2 #5 battery (3V), we can get 3.3V for the system after using DC-DC L6920 step-up chip. The following advantages make L6920 become the best choice: output voltage is 3.3V or 5V, even when the input voltage as low as 0.7V the system still can work (This can make sure that after a period of time of battery usage and the input voltage is falling the output voltage is still 3.3V ). This can enhance the stability of the system. The power consumption of L6920 is low. The working current is 18uA and only 5uA on sleep mode. Figure 4 shows Power circuit.
Low-Power ZigBee Wireless Temperature Humidity Sensor Network
619
L1 VIN
LX 10uH SHDN LBI
47uF
C3 REF GND
7
8
5
1
2
3
4
6
VCC
OUT FB C2 LBO 10uF
C1 47pF
GND
L6920
GND
Fig. 4. Power circuit diagram shows that the 3.3V battery
3.4 Processing of Spare Pins The system uses the CMOS chips. The charges cumulated by spare pins make the potential between “0” and “1”. The current complementary consist of an enhanced NMOS and an enhanced PMOS, the jumping of the potential from “0” to “1” or “1” to “0” will make the two pins instant short [3]. The intermittent short will bring short consumption. So connecting spare pins to the ground can reduce power consumption.
4 Low-Power Consumption of Software With the integrated circuit technological progress, power consumption of processor module and sensor module become very low. Figure 5 can explain the sensor node energy consumption, most of the energy consumption on wireless communication module; communication module has four states: transmitting, receiving, leisure and sleep, and energy consumption on transmitting, receiving and idle state is large, but the energy consumption on sleeping state is small [4]. So how to reduce power consumption on wireless communication has become an important problem.
Fig. 5. This shows all the consumption of sensor node
4.1 Working Mode of CC2430 In general CC2430 has low power consumption, it offers four power modes to choose: PM0/ PM1 /PM2/ PM3.
620
S. Gong et al.
CC2430 works on full-function PM0 mode, the frequency is 32MHz and 32.768 kHz, and the power consumption is less than 1mW. When it works on sleep mode PM1/PM2, Only the 32.768 kHz low-speed crystal run, the power consumption is less than 0.9uA, when it works on Hibernate mode PM3, there is no crystal run, so the power consumption is less than 0.6uA. After sending information, the node will be run into PM2-sleep mode. Because in this mode, it can run into full-function mode through internal sleep clock wake-up, and it does not require manual operation to wake-up. When the system works on SET_POWER_MODE (2) mode, sleep time can be set. the frequency of PM2 mode is 32.768kHz,timer of sleep is a 24-bit counter, so the longest sleep time of system is: T=2*24/32768=512s,the shortest sleep time is: T=1/32768 and it is about 30.5us. The setting function of sleep time is Set_ST_Period (uint16 sec), where sec is set according to user needs. 4.2 Low-Power Software Design The system wakes up nodes at regular time. Temperature and humidity is a gradual process, so it is no need to monitor it all the time. Monitoring it at regular time not only can monitor the change of temp and humidity, but also can reduce the working time of nodes. it is not necessary for the sensor nodes to communicate with coordinator all the time, this can avoid data collision and can reduce the power consumption.
Fig. 6. This shows the flow chart of software design and the sleep mode of CC2430
Low-Power ZigBee Wireless Temperature Humidity Sensor Network
621
After the establishment of ZigBee networks all sensor nodes send temp and humidity information to the coordinator. After receiving the confirmed information of coordinator the node gets into sleep mode. The coordinator connects to PC through RS-232 serial port. The temp and humidity information will display in upper machine. When the wake-up time arrives, the node will rouse from sleep and continue sending messages to coordinator. This method can make all nodes in sleep state in most of the time, which will significantly reduce the power consumption of the entire network. Figure 6 shows the flow chart of software design.
5 Implement of ZigBee Sensor Network The system uses a tree topology. The advantages of tree topology are as follows: it can extend transmission distance between coordinator and sensor nodes advance the ability to carry the load and enhance the area of the network.
Fig. 7. This shows the results that the PC collects. The information is temperature and humidity of greenhouse.
The connection of ZigBee coordinator and sensor nodes is bonding. Bonding is a mechanism from one application layer to another [5]. 2, 7, 8 are the labels of source nodes in the network, which can be determined when the program is downloaded. Each network ID is the only one in the network and each network ID correspond one sensor node [6]. We can monitor the temperature and humidity of the greenhouse through network ID. The power consumption of the nodes is 32.5mA in average.
6 Conclusion ZigBee wireless sensor network overcome many shortcomings that wired media bring in. This paper implements a ZigBee wireless sensor network that can transmit and
622
S. Gong et al.
monitor temperature and humidity of the greenhouse. This wireless sensor network resistant to stable work, tight structure, large loads and low power consumption. After low-power design, the power consumption of the network reduces, the effective working time of the sensor nodes extend and the stability of the system enhance. Each node can carry many sensors and controlled equipment, so the node can be added more sensor such as light sensor and CO2 sensor to collect more information of greenhouse. This ZigBee wireless sensor network is part of Internet of Things, and can be used in precision agriculture and industry and many other fields. Acknowledgments. Funding for this research was provided by Harbin Special Funds for Research on Technological Innovation Projects (NO.2008RFXXN003).
References 1. Bogena, H.R., Huisman, J.A., OberdÊrster, C., et al.: Evaluation of a low cost soil water content sensor for wireless network applications. Journal of Hydrology, 32–42 (2007) 2. Zhang, X., Zhang, C., Fang, J.: Smart Sensor Nodes for Wireless Soil Temperature Monitoring Systems in Precision Agriculture, pp. 237–240 (2009) 3. Li, W., Duan, C.: ZigBee2007/PRO experiment and practice of stack. Press of Beihang University, Beijing (2007) 4. Gao, J.: Study of energy consumption of ZigBee wireless sensor network node. Electronic Test, 1–4 (2008) 5. Sun, Y., Liu, Z.C., Cai, C.: Design of Low-power wireless sensor networks nodes. Computer Applications, 21–26 (2009) 6. Li, J., Zhang, C., Fang, J.: Design On The Monitoring System Of Physical Characteristics Of Dairy Cattle Based On Zigbee Technology. In: IEEE Proceedings of the 2009 International Conference on Computer and Computing Technology Applications in Agriculture (2010)
Land Evaluation Supported by MDS* Fengchang Xue School of Remote Sensing, Nanjing University of Information Science & Technology, Nanjing, Jiang Su Province, China. 210044
[email protected] Abstract. GIS-MCE is the main method in land evaluation,but it is a linear method and neglects multidimensional complexity of factors used in land evaluation, which leads to information loss. Multidimensional scaling (MDS) originates from psychoanalysis, which is used to describe multidimensional data in higher dimensions by transforming data in higher dimensions into geometry structure in Lower dimensions. In the Land evaluation model supported by MDS, the data is transformed into a similar space and land evaluation is completed according to the spatial clustering based on the data’s spatial similarity ,which is a method drived by data and not dependenting on others priori assumption. Taking expropriation division in Xuzhou city as an example, it shows that the land evaluation based on MDS meets the requirements of land classification. Keywords: Land Evaluation, MDS Spatial Cluster model.
1 Introduction Land evaluation involves factors of soil, climate, vegetation, topographic and hydrology, which is an analysis integrating spatial information. In analysis of integrating spatial information, spatial data of different types and different sources can be taken as attribute of spatial cells, which have different spatial resolution and different spatial scale. So land evaluation can be taken as integrating spatial information of different sources and different spatial scales in specific information space. GIS-MCE is the main method in land evaluation, but it is a linear method and neglects Multidimensional complexity of factors used in land evaluation, which leads to information loss. Multidimensional scaling(MDS) originated from psychoanalysis, and it is used to describe multidimensional data in higher dimensions by transforming data of higher dimensions into geometry structure in Lower Dimensions. Based on this,Land evaluation supported by MDS is a method drived by data and not dependenting on others priori assumption. *
Foundation: Science and Technology support project of Nanjing University of Information Science and Technology(Project Number: S8108232001, S8109008001).
D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part IV, IFIP AICT 347, pp. 623–628, 2011. © IFIP International Federation for Information Processing 2011
624
F. Xue
2 The Principles of MDS
( )
Set the Attribute measurement matrix C(C= dij n×n) of n objects has been given in r-dimensional space,said C is the similarity matrix for the n objects. MDS using C to obtain p (p r)-dimensional vector Z = (z (1), z (2), ..., z (n)), and n objects can be exˆ is Attribute measpressed by vector Z, said that Z is a Mimetic Structure of C. Set D ˆ urement matrix Obtained from the Z,and said D is Mimetic distance matrix of C. MDS's target is both to make the n objects to achieve p (p r)-dimensional expression, ˆ as close as possible to C, and the process achieves the classificabut also to make D tion of n objects ,where only n-matrix C is available. Young- Household theorem approachs creating p-dimensional vector Z from the matrix C. For the matrix C = (dij) n × n, set B = (bij) n × n, where
≦
≦
(- + ∑ dij + ∑ dij - ∑∑ dij ) (1) Set λ ≧λ ≧…≧λ is positive characteristic roots of B, e ,e …,e is eigenvector corresponding to λ ≧λ ≧…≧λ take Z = (e , e , ..., e ) = (x ) , each row of Z matrix 1 bij= 2
1
2
dij2
n
1 n
n
1 n
2
j =1
2
i =1
r
1
2
r
1
2
r
n
n
1 n2
i =1 j =1
1
2
2
r
ij n × r
corresponds to a mimetic spatial coordinate of object. Take Z dimension p 800 nm can minimize interferences caused by the presence of dissolved light-absorbing substances(Dag Hongve and Gunvor Åkesson, 1998); (3) Water samples colour absorbs light of certain wavelength, and affect measurement precision, near infrared light can minimize interferences caused by water samples colour. Consequently, the LED used in this work is designed for maximum emittance at 940 nm. The detector is a matching, high sensitivity Silicon Photoelectric Generator with peak sensitivity also at 940 nm. Both the emitter and detector provide a narrow and closely matched spectrum. The electric circuit used to drive the sensor and to treat the signal is shown in Fig. 6. Conversion of the Silicon Photoelectric Generator current and amplification of the voltage is achieved by using one wide range of input offset, low offset voltage drift, high input impedance, extremely low power, and high gain quad operational amplifiers (TLC27L4). Power requirements for the sensor are +3.3V for the operational amplifiers.
Fig. 6. Electric circuit for the Turbidity sensor signal treatment
A Simple Temperature Compensation Method for Turbidity Sensor
655
3.3 Methods To measure the response of the sensor to different temperature, Turbidity standard solutions of Formazin were adjusted with filtered water into water samples of 0-100NTU. 11 different concentrations were measured, ranging from 0 to 100NTU. Take Bunsen beaker filled with Formazin standard solution and cooling to 1 , then immerse the sensor in the solution. The Bunsen beaker and sensor were then placed in a water bath at room temperature. Solution in the Bunsen beaker was gradually heated to 40 and the output recorded at 5s intervals.
℃
℃
4 Results and Discussion 4.1 Calibration of Sensor Table1 shows the relationship between the turbidity of water samples prepared by diluting turbidity standard solutions of Formazin and the values measured by the sensor under environment temperature 20 . As can be seen, the solutions measured gave nearly linear relations between the concentration of the standard solutions and the measured turbidity. Calibration equation of the sensor fitted according to the principle of least square is y = 1.0003 x − 0.0181 , and the correlation coefficient is R 2 = 0.9993 .
℃
Table 1. Relationship of measured turbidity and concentration of turbidity standard Water samples/NTU
measured values /NTU
5 10 20 30 40 50 60 70 80 90 100
5.50 10.55 19.42 29.39 39.37 49.35 59.88 71.52 80.38 90.92 98.68
correlation coefficient R2 0.9993
4.2 Temperature Effect Fig. 7 shows the relationship between temperature and the output signal of sensor. As the figure shows, the experimental curves seem to be similar in shape to each other, but had larger values as solution concentrations increased. The output signal of sensor
656
H. Tai et al.
increased according to the rise of temperature found from the experiment. But the output signal of sensor decreased according to the rise of temperature found from the experiment at initial stage, this result probably arose from the temperature between the sensor and solution failed to come up to equilibrium, the temperature of sensor is higher then solution.
550
2
y = 0.2131x - 6.3821x + 432.32 2
R = 0.9973
500
80NTU
2
y = 0.1303x - 0.5072x + 255.9
450 ) V m ( l400 a n g i s t350 u p t u O300
2
R = 0.9923
40NTU
2
y = 0.142x - 2.3413x + 254.9 2
R = 0.998
10NTU
2
y = 0.1478x - 2.5972x + 237.4 2
R = 0.993
0NTU
250 200 0
5
10
15
20
25
30
35
40
Temperature(℃)
Fig. 7. Relationship of temperature and output signal of sensor
5 Temperature Compensation As Fig. 7 shows, when the temperature was in the range of 1
~40℃, the relationship
between the output signal of sensor and temperature can be given by
y = mx2 + nx + p
(1)
Where y , is the output signal of turbidimeter, x is temperature, m , n and p is coefficient. Consequently, a general curve can be obtained as Fig. 8 shows. Three points A(x 0 , y 0 ) B(x1 , y1 ) and C(x 2 , y 2 ) on the curve in Fig.8 can be used
,
to solve the quadratic equation (1). But it is difficult to solve the quadratic equation (1) directly. Another form of the quadratic equation (1) as follow can be used.
y = a( x − x0 )( x − x1 ) + b( x − x2 ) + c Where
a
, b and c is coefficient.
(2)
A Simple Temperature Compensation Method for Turbidity Sensor
657
y
y2 y1 y0
C
B A
x0
x1
x2
20
Fig. 8. General curve of the relationship between temperature and output signal of sensor
Substitute the quantitative data of A(x0 , y 0 ) , B(x1 , y1 ) and C(x 2 , y 2 ) into equation (2), the three coefficients can be obtained, and the results as follow:
y2 − y0 y1 − y0 y 2 − y0 − −b x2 − x0 x1 − x0 x2 − x0 a= = x2 − x1 x2 − x1 b=
y1 − y0 x1 − x0
c = y0 + ( x2 − x0 ) When the temperature is 20
y1 − y0 = y0 + ( x2 − x0 )b x1 − x0
℃, output signal of sensor is given by
y = a(20 − x0 )(20 − x1 ) + b(20 − x2 ) + c
(3)
Then substitute the equation (3) into calibration equation of the sensor which obtained at temperature of 20 . Reference value of the solution can be acquired finally. The above-mentioned method is designed to program, and be carried by MCU which is used to compensate the measurement errors by soft programming.
℃
6 Conclusion In the detection system, the high measurement accuracy is required, and decrease or eliminate temperature error is very important. By the experiment, the following results were obtained:
658
H. Tai et al.
(1) The output signal of sensor increased according to the rise of temperature found from the experiment, and a quadratic equation give the best fit. (2) The method carried by MCU which is used to compensate the measurement errors by soft programming is simple and effective; it can improve the error compensation precision on sensors greatly. Acknowledgements. This work was supported by 863 programs “Research and Application of Digital Integrated System for Intensive Aquaculture” (2007AA10Z238), 948 Project of China Agriculture Ministry (2010-Z13). And the programs “Development and Applications of sensor network applied to monitor bloom of blue-green algae in Taihu Lake” (2010ZX03006-006).
References 1. Gong, D.-t., Li, F.-h., Fan, Z.-g., Tao, J., Liu, S.-l.: Measurement of CL− and SO42 − in boric acid by turbidimeter. J. Inorganic Chemicals Industry 40, 56–58 (2008) 2. Brasington, J., Richards, K.: Turbidity and suspended sediment dynamics in small catchments in the Nepal Middle Hills. Hydrol. Process. 14, 2559–2574 (2000) 3. Gippel, C.J.: Potential of turbidity monitoring for measuring the transport of suspended solids in streams. Hydrol. Process. 9, 83–97 (1995) 4. Lewis, J.: Turbidity-controlled suspended sediment sampling for runoff-event load estimation. J. Water Resource 32, 2299–2310 (1996) 5. Lawler, D.M., Brown, R.M.: A simple and inexpensive turbidity meter for the estimation of suspended sediment concentrations. Hydrol. Process. 6, 159–168 (1992) 6. de Lacy Costello, B.P.J., Ewen, R.J., Ratcliffe, N.M., Richards, M.: Highly sensitive room temperature sensors based on the UV-LED activation of zinc oxide nanoparticles. J. Sensors and Actuators B: Chemical 134, 945–952 (2008) 7. Sheng, Q., He, X.-g.: The Far-infrared LED and the Detection of Water’s Turbidity. J. Sci-Tech Information Development & Economy 17, 274–275 (2007) (in Chinese) 8. Zhang, W., Yang, J.-f., Yan, Q.-g.: Experiment research on the property of silicon solar cell. J. Experimental Technology and Management 26, 42–46 (2009) (in Chinese) 9. Siqueira Dias, J.A., Leite, R.L., Ferreira, E.C.: Electronic technique for temperature compensation of fibre Bragg gratings sensors. J. AEU - International Journal of Electronics and Communications 62, 72–76 (2008) 10. Du, Y.-p., He, X.-y.: Brief discussion on temperature compensation technology of sensor. J. Electronic Design Engineering 17, 63–64 (2009) (in Chinese) 11. Xu, J.: Using S-C Processor Software to Achieve the Sensor’s Temperature Error Compensation. J. Modern Electronic Technique 10, 97–99 (2002) (in Chinese) 12. Možek, M., Vrtačnik, D., Resnik, D., Aljančič, U., Penič, S., Amon, S.: Digital Self-Learning Calibration System for Smart Sensors. J. Sensors and Actuators A: Physical 141, 101–108 (2008) 13. Hongve, D., Åkesson, G.: Comparison of nephelometric turbidity measurements using wavelengths 400–600 and 860 nm. J. Water Research 32, 3143–3145 (1998)
A Wireless Intelligent Valve Controller for Agriculture Integrated Irrigation System Nannan Wen1, Daoliang Li1,*, Daokun Ma1, and Qisheng Ding1,2 1
College of Information and Electrical Engineering, China Agricultural University, Beijing 100083, P.R. China
[email protected] 2 Xuzhou Normal University, Xuzhou, Jiangsu, P.R. China
Abstract. Based on the problems caused by current automatic irrigation system with a large number of cables, such as installation, maintenance, and difficult expansion, this paper proposed a wireless integrated irrigation control system and developed a wireless intelligent valve controller which can satisfy the requirements for agricultural irrigation in our country. The controller consisted of the control unit, power unit, wireless communication unit, relay boost driver unit, state feedback switch, etc. It received the control command from the remote control center by the wireless communication unit, actuated the relay boost driver unit to turn on/off irrigation solenoid valves. The controller includes the functions of remote control, parameter setting, feedback status detecting, etc. Since the rational design, simple command, and low cost, the controllers have been pilot in Beijing and Xinjiang. The results indicate that the controller is effective in practice and can be used a normal irrigation season. Keywords: Wireless, Valve controller, Irrigation system.
1 Introduction With the popularization and improvement of water-saving irrigation technology, agricultural irrigation control technology is applied gradually. However, irrigation control equipments in application at this stage are from abroad and also very expensive, which are only for demonstration and lack of significance for promotion. Current automated irrigation system with a large number of cables brings many problems such as installation, maintenance, expansion and other issues. Therefore, it is hot to develop a low-cost irrigation valve controller independently for agriculture irrigation system. For example, a low cost solar-powered feedback controller for DIC of fixed irrigation systems was developed. The controller uses soil water potential (SWP) measurements to control the amount of water applied to each specific management area of a field, and measured system hydraulic pressure to communicate with other controllers. The results indicate that the controller was *
Corresponding author.
D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part IV, IFIP AICT 347, pp. 659–671, 2011. © IFIP International Federation for Information Processing 2011
660
N. Wen et al.
effective in maintaining the SWP in the root zone close to a predetermined management allowed deficit (MAD). (F.R. Miranda et al, 2005) A wireless broadcast irrigation control system and a low power wireless DC solenoid valve based on OOK modulation and powered by battery were design and applied throughout the greenhouse irrigation in a plantation. (Shen Changjun et al, 2009) With advanced electronic computer control and wireless transmission technology, a precision irrigation control system was designed for water-saving irrigation. (Kuang Qiuming et al, 2007) A new automatic smart controller for irrigation based on GSM network has been developed, it can receive GSM message from PC and mobile, then control the valves by wireless radio communication. (Yang Genghuang et al, 2005) Many irrigation-scheduling methods have been developed over the years, but adoption by producers has been limited by cost, installation time, maintenance, and complexity of the decisions involved. (F.R. Miranda et al, 2005) The above-mentioned irrigation controllers which used OOK modulation, infrared, GSM and other means of communication bring with high price and high maintenance costs. Besides, most commercially available sensors and actuators assembled for irrigation system networks are too complex and/or costly to be feasible for site-specific management of integrated irrigation systems. A wireless intelligent valve controller for site-specific management and/or operation of a wireless integrated irrigation control system is easier to install and maintain as compared to cable irrigation control. As a result, a potential solution to these problems is a total wireless automation of irrigation control system. Wireless automation irrigation controllers can be implemented using spatially variable irrigation systems to optimize yields and maximize water use efficiency for fields with variation in water availability due to different soil characteristics or crop water needs. (Zhang Wei et al, 2009) The objectives of this research were to develop and test an autonomous, low cost, wireless intelligent irrigation valve controller for wireless integrated irrigation control system for the actual demands in agricultural irrigation.
2 Architecture of Wireless Integrated Irrigation Control System Integrated agricultural production demands and learning from foreign experience in research, the architecture of wireless integrated irrigation control system is shown in Fig. 1. The system was composed of two parts, wireless sensors monitoring network and remote control center. Wireless sensors monitoring network consists of wireless information acquisition nodes and irrigation control nodes distributed in the field, the remote control center sends the commands of data acquisition and valve control though the routing nodes. Wireless sensors monitoring network uses star topology, end nodes route by way of the routing nodes to the sink nodes, then the sink nodes transmit information to the wireless access point by GPRS and control computer receives
A Wireless Intelligent Valve Controller for Agriculture Integrated Irrigation System
661
information from the wireless access point by RS232. Information acquisition, remote control, parameter setting, intelligent alarm functions are achieved by the remote control center. The system passed soil information and meteorological parameters collected by sensors in real-time monitoring to the remote control center though wireless network. Then the remote control center automatic analysis and makes irrigation management decision to open or close valves. With timely scientific guidance amount of irrigation, the system achieved the sustainable use of water resources and automatic and intelligent irrigation. The remote control center can also share information across users through the Internet.
Fig. 1. Architecture of wireless integrated irrigation control system
3 Materials and Methods 3.1 Irrigation Valve Controller Description For irrigation implementation, a field is typically divided into irrigation management units based on soil characteristics, crop water requirements, and/or economic factors prior to the installation of the control system. A wireless integrated irrigation control system is installed in each irrigation management unit to monitor soil temperature, humidity and meteorological parameters and ensure water-saving irrigation. The valve controller developed in this study is designed to work autonomously without hard-wire connections between individual control units. Each controller is
662
N. Wen et al.
designed to operate on battery power. Soil moisture determinations and irrigation decisions occur at fixed regular intervals set by the remote control center. Real-time clocks for all controllers are synchronized for measurement and irrigation decisions. After measurements and irrigation decisions are made, remote control centre sends signal to valve controllers. Each controller installed in system is programmed to process the control commands received from remote control centre to autonomously open an irrigation solenoid valve, triggering relay boost driver unit. Finally, each controller stores the corresponding valve status feedback information from state feedback switch and transmits to remote control center. 3.2 Controller Hardware Wireless Intelligent Valve Controller shall be the core of the wireless irrigation control node, hardware modularization was adopted. It consisted of the control unit, power unit, wireless communication unit, relay boost driver unit, state feedback switch, etc. A block diagram of Wireless Intelligent Valve Controller hardware developed in this study is shown in Fig. 2. Electronic devices, chips, MCU were selected to meet the low power and low cost required for the controller. The sub-components of the control unit are a microcontroller, reset circuit, real-time clock, UART, IO expansion, analog-to-digital (A/D) acquisition, flash memory, and Interface circuit. Each irrigation controller controls four solenoid valves.
Fig. 2. Block diagram of Wireless Intelligent Valve Controller developed in this study
A Wireless Intelligent Valve Controller for Agriculture Integrated Irrigation System
663
3.2.1 Control Unit The microcontroller (MCU) consists of the Jennic5139, is the master device of control unit that is programmed to keep time and control the actuators. The device integrates a 32-bit RISC processor, with a fully compliant 2.4GHz IEEE802.15.4 transceiver, 192kB of ROM, 96kB of RAM, and a rich mixture of analogue and digital peripherals. The JN5139 is a low power, low cost wireless microcontroller suitable for IEEE802.15.4 and ZigBee applications. It has 96kB RAM for share program and 192kB ROM for program code. It’s peripherals include a 56-pin DIP package with 21 digital I/O pins, 4-input 12-bit ADC, 2 11-bit DACs, 2 comparators, 2 Application timer/counters, 3 system timers, 2 UARTs, SPI port with 5 selects, 2 wire serial interface, etc. These features all make for a highly efficient, low power, single chip wireless microcontroller for battery-powered applications. This MCU was selected based on low-cost, processor speed, low power requirements, rapid software development, and ease of system integration with custom circuits. The reset circuit and clock circuit ensure equipments work reliability and stability. The JN5139 has two independent Universal Asynchronous Receiver/Transmitter (UART) serial communication interfaces. The UART connected to a level shifter connector to provide the RS232/RS485 line voltage compatible with a PC. The software developer kit uses such an interface as the debugger interface between the JN5139 and a PC. The information can be observed directly in local with RS485/RS232 communication. There are 21 Digital I/O (DIO) pins, which can be configured as either an input or an output, and each has a selectable internal pull-up resistor. Most DIO pins are multiplexed with alternate peripheral features of the device. The Serial Peripheral Interface (SPI) allows high-speed synchronous data transfer between the JN5139 and peripheral devices. The JN5139 operates as a master on the SPI bus and all other devices connected to the SPI are expected to be slave devices under the control of the JN5139 CPU. The Intelligent Peripheral (IP) Interface is provided for systems that are more complex, where there is a processor that requires a wireless peripheral. The intelligent peripheral interface is a SPI slave interface and uses pins shared with other DIO signals. The interface is designed to allow message passing and data transfer. The controller with above DIO pins and interface circuitry can support 4-way analog inputs, 4-way control power output. A controller can support 2-way pulse solenoid valves and state switch, 2-way soil moisture sensor. Controller extends Flash memory. Data and program are stored in non-volatile memory to prevent loss if a power failure occurs. The user can download recorded data to analyze system performance.
664
N. Wen et al.
3.2.2 Power Unit Two 1.5V batteries as the input source through the power unit provide a stable output voltage for other units. Power unit contains polarity protection circuit and voltage regulator circuit. Polarity protection circuit includes Schottky diode and the decoupling capacitors. Voltage regulator circuit utilizes voltage regulator chip to ensure output voltage stably. After fully considering energy in application field, controller utilizes ordinary battery as power supply to ensure that the system works over an irrigation quarter. 3.2.3 Wireless Communication Unit ZigBee is a two-way wireless communications technology in short distance communication, with low complexity, low power consumption, low data rate. Its complete protocol stack is only 32 KB, can be embedded in a variety of devices, while support geographic targeting. It is a protocol specification developed for small device wireless network. It is a part of the family of IEEE wireless network protocol. It has a very complete hierarchy. Wireless communication unit supported by low-power radio frequency chip with the typical external circuitry, communicated with the control unit by MCU interface and completed the wireless transceiver function. It works in the global free public band (2.4GHz), reducing the traditional wireless operating cost. 3.2.4 Relay Boost Driver Unit In sleep mode, relay boost driver unit is off. Once the controller receives the valve-driven control signal, it enables relay boost driver unit. Low-power boost chips installed in the relay boost driver unit circuitry will boost the low input voltage to 5 ~ 24V working voltage suitable for solenoid valve, send the driver pulse to control valve. Finally, each controller stores the corresponding valve status feedback information from state feedback switch and transmits to remote control center. The valve controller uses 3V battery as power, based on ZigBee protocol, waiting for remote control command. It can actuate relay boost driver unit to control valve. The MCU collects the state feedback information in real-time and transmits to remote control center. The controller supports a number of pulse solenoid valves from various valve manufacturers and can be easy to installation and operation. 3.3 Controller Software There are 4 kinds of the operation states during the controller working: the initial state, the task query state, the task execution state, sleep state. Diagram of specific state transition is shown in Fig. 3. After the wireless valve controller was powered into the initial state, the microcontroller completed hardware initialization, including the port initialization, watchdog initialization, the system clock initialization, and variable initialization, etc. Then the
A Wireless Intelligent Valve Controller for Agriculture Integrated Irrigation System
Fig. 3. Diagram of specific state transition Star
Initialization No Register to routing node? Yes No
Receive the control order? Yes
valves control
status detecting
parameter setting
Task execution
Idle?
No
Yes Sleep
Wake up
Fig. 4. Flow diagram of controller program
665
666
N. Wen et al.
controller starts testing a higher-level control signal which drives it into the task query state. If the wireless communication unit detected execution signal, the microcontroller will further determine the type of tasks, for instance, valve control, status detecting, parameter setting, and then enter the task execution state. If no task is detected, the controller enters sleep state. It will be wake up automatically by software timing, then checks control signal again and enters into the task query state. The irrigation controller program was written in C. The program uses the memory capacity of the microcontroller (MCU). The program flow diagram is shown in Fig. 4. It contains a main loop and several subroutines such as valve control, status detecting, and parameter setting that enable the MCU to perform the following tasks: - initialize hardware, including the port initialization, watchdog initialization, the system clock initialization, variable initialization; - register to routing node in the same network, report device information; - request tasks from superior; - query tasks, receive remote control commands; - process the remote commands to determine the type of tasks, execute the task - store control information, time, and valve status in the controller memory; - receive control commands from remote control center to drive irrigation valve; 3.4 Irrigation Valve Controller Testing To measure the valve controller performance, a resistor circuit is used to series connect with the controller (Fig. 5). The circuit was powered by battery. Using a small resistor, resistance on the circuit can be neglected in the energy. Measured the voltage across the resistor by oscilloscope, circuit current value can be obtained in the series circuit to calculate the energy consumption of valve controller. battery
oscilloscope
small resistor valve controller
Fig. 5. Resistor circuit used to calculate the valve controller energy consumption
The valve controller during operation including the initial state, the task query state, the task execution state and sleep state. To facilitate the analysis, the above conditions are divided into the state of active and sleep. Active Consumption Pw can be calculated using the equation:
A Wireless Intelligent Valve Controller for Agriculture Integrated Irrigation System
Pw = I w × Tw
667
(1)
I w is the Active Current, Tw is the Active Time for the valve controller. The Average active current I wa can be obtained by the following equation: Where
I wa =
∑ Iw × T ∑ T + Ts
w
(2)
w
Where Ts is the Sleep Time. Similarly, Average sleep current using the equation:
I sa =
I s × Ts
Isa can be calculated (3)
∑ Tw + Ts
Is is the Sleep Current. The relationship among the Average active current I wa , Average sleep current Isa and Average current Iav is as follows: Where
I av = I wa + I sa
(4)
Consequently, Consumption in one day Pwa is the average current consumption for each hour multiplied by 24 hours per day, the formula can be given by
Pwa = I av × 24
(5)
Through analysis of valve device energy consumption in different states on one day in laboratory, we estimate the actual duration in work as follows. Days:
D=
1200 I av
(6)
D 30
(7)
Months:
M=
4 Results and Discussion The controller performance was evaluated at demonstrations in Beijing and Xinjiang. To verify adequate operation of the hardware and software, and the controller in the root zone on a real-time basis, some controllers were tested to simulate a site-specific irrigation system in laboratory. We chose one of them to analyze its consumption.
668
N. Wen et al.
4.1 System Hardware and Software Performance
The controller hardware and software performed all tasks as designed. Data downloaded from the controllers showed that the irrigation control system continually measured apparatus parameters and status information, and opened or closed the solenoid valve when needed without failure. User interactions with the controllers using a notebook computer were also successful. 4.2 Consumption Current in Active and Sleep State
Using two ordinary batteries to power the valve controller, about 1200mAh, specific test results are as follows. Table 1. Valve controller power consumption current in active and sleep state
I w (mA)
Tw (ms)
pw (mAys)
I s (mA)
10.000
2.5
0.025
0.300
20.000
2.0
0.040
0.300
40.000
1.5
0.060
0.300
70.000
1.5
0.105
0.300
40.000
2.5
0.100
0.300
140.000
2.0
0.280
0.300
60.000
3.5
0.210
0.300
140.000
1.0
0.140
0.300
50.000
100.0
5.000
0.300
Valve controller power consumption current in active and sleep state is shown in Tab.1. The range of active current I w is between 20 ~ 140mA, the operating current is 50mA in most of the time and the sleep current is 0.3mA. Total active time is 114.0 ms. 4.3 Average Energy Consumption and the Working Duration
As the sleep time is set by software, furthermore the different sleep time preprogrammed will affect the energy consumption, we test the different average current under the different circumstance. According to the different average current, we can further compute the number of days that the irrigation controller could operate without external charging at the study.
A Wireless Intelligent Valve Controller for Agriculture Integrated Irrigation System
669
Table 2. Controller average energy consumption and the working duration
Ts (ms)
I wa (mA)
I sa (mA)
I av (mA)
10000
0.586810
0.297
0.883
20000
0.295068
0.298
30000
0.197084
40000
P
D(d)
M(m)
21.202
56.598
1.887
0.593
14.241
84.265
2.809
0.299
0.496
11.903
100.817
3.361
0.147953
0.299
0.447
10.730
111.832
3.728
50000
0.118430
0.299
0.418
10.026
119.690
3.990
60000
0.098729
0.299
0.398
9.556
125.578
4.186
120000
0.049411
0.300
0.349
8.379
143.214
4.774
180000
0.032951
0.300
0.333
7.986
150.258
5.009
600000
0.009890
0.300
0.310
7.436
161.377
5.379
av
Tab.2 shows controller average energy consumption and the working duration. The sleep time Ts was default in the program. The table lists 10s to 10 minutes of sleep time. The average energy consumption can be calculated by different time. Can be seen from Tab.2, the longer the sleep time, the longer controller can work. Moreover, the result showed that the 1200mAh battery was sufficient to maintain an irrigation quarter. The performance of the irrigation controller was analyzed in terms of satisfactory operation of the controller hardware and software. The irrigation controller performance was considered satisfactory and two batteries would be sufficient to power the irrigation controller during the normal irrigation season. If the sleep time was set at 10min or more, the controller proved to be effective for the actual application in the root zone.
5 Conclusion A Wireless Intelligent Valve Controller for wireless integrated irrigation control system was developed and tested. The controller proved to be reliable, affordable, and effective in practice without hard-wire connections. The wireless intelligent valve controller with control unit, combined with low-power microprocessor chip and a sound external circuit, powered by battery, received control command from remote control center, executed the tasks, such as information acquisition, remote control, parameter setting, state feedback, while the task of valve control must be driven though the relay boost driver unit. The wireless transceiver capabilities reflected in the wireless communication unit. The valve control can be widely used in different irrigation regions.
670
N. Wen et al.
Wireless intelligent valve controller fully considers the energy in application field, setting the collection/control frequency in accordance with the user demands. Increasing solar power charge unit to controller to ensure the power supply is the next research work. Wireless intelligent valve controller is simple construction, easy maintenance, low cost and has good application prospects. Acknowledgements. This research is supported by National Science and Technology Support Programmed (2008BAB38B03-03), Land Consolidation and Rehabilitation Center of Ministry of Land Resources and Beijing UNISM Technology Co., Ltd for their Technical Support. The authors would like to thank all the members of EU-China Center for Information and Communication Technologies for their assistance.
References 1. Miranda, F.R., Yoder, R.E., Wilkerson, J.B., Odhiambo, L.O.: An autonomous controller for site-specific management of fixed irrigation systems. Computers and Electronics in Agriculture 48, 183–197 (2005) 2. Mareels, I., Weyer, E., Ooi, S.K., Cantoni, M., Li, Y., Nair, G.: Systems engineering for irrigation systems successes and challenges. Annual Reviews in Control 29, 191–204 (2005) 3. Cardenas-Lailhacar, B., Dukes, M.D.: Precision of soil moisture sensor irrigation controllers under field conditions. Agricultural Water Management 97, 666–672 (2010) 4. Davis, S.L., Dukes, M.D., Miller, G.L.: Landscape irrigation by evapotranspiration-based irrigation controllers under dry conditions in Southwest Florida. Agricultural Water Management 96, 1828–1836 (2009) 5. McCready, M.S., Dukes, M.D., Miller, G.L.: Water conservation potential of smart irrigation controllers on St. Augustinegrass. Agricultural Water Management 96, 1623–1632 (2009) 6. Blonquist Jr., J.M., Jones, S.B., Robinson, D.A.: Precise irrigation scheduling for turf grass using a subsurface electromagnetic soil moisture sensor. Agricultural Water Management 84, 153–165 (2006) 7. Wei, Z., He, Y., Qiu, Z., et al.: Design of precision irrigation system based on wireless sensor network and fuzzy control. Transactions of the CSAE 25(Supp. 2), 7–12 (2009) (in Chinese) 8. Zhang, J., Chen, J., Hu, J., Zhao, Y.: Control system of urban green land precision irrigation based on GPRS / SMS and μC / OS embedded technology. Agricultural Engineering (09), 1–6 (2009) (in Chinese) 9. Gao, F., Yu, L., Zhang, W., Xu, Q., Yu, L.: Research and design of crop water status monitoring system based on wireless sensor networks. Agricultural Engineering (02), 107–112 (2009) (in Chinese) 10. Shen, C., Zheng, W., Sun, G., Yan, H., Xing, Z.: Development and Application of Low Power Wireless DC Solenoid Valve and Controlling Module. Agricultural Machinery 40(z1) (2009) (in Chinese) 11. Liu, H., Wang, M., Wang, Y., Ma, D., Li, H.: Development of farmland soil moisture and temperature monitoring system based on wireless sensor network. Jilin University (Engineering Science) (03), 604–608 (2008) (in Chinese)
A Wireless Intelligent Valve Controller for Agriculture Integrated Irrigation System
671
12. Kuang, Q., Zhao, Y., Bai, C.: Automatic monitor and control system of water saving irrigation. Agricultural Engineering (06), 136–139 (2007) (in Chinese) 13. Yang Genghuang, Guo Kairong, Ya-Weili, A new automation device for irrigation based on GSM network. Journal of Shenyang Agricultural University, 2005 - 12, 36 (6): 753 - 755, (in Chinese)
Evaluation of the Rural Informatization Level in Central China Based on Catastrophe Progression Method Lingxian Zhang, Xue Liu, Zetian Fu, and Daoliang Li* College of Information & Electrical Engineering, China Agricultural University, P.O. Box 209, 17 Qinghua East Road, Haidian District, Beijing, China
[email protected],
[email protected] Abstract. This paper developed a methodology based on the catastrophe theory for estimating the rural informatization level in central China, and took evaluation of the rural informatization level among the six provinces of central China as an example to test the effectiveness of the method. Taking data from reference and constructing the hierarchy based on catastrophe progression method, it was calculated the scores of rural informatization level among the six provinces of central China using normalizing formula. The results are found to be coincident with practical situation, so it proves the catastrophe progression method works well. Keywords: rural informatization, level, evaluation, Catastrophe Progression Method, the central China.
1 Introduction As rural informatization is an important component of the national informatization development process, it is the basis for decision-making to make an objective assessment and analysis to national and regional level of development of rural informatization. There have many methods for multiple object comprehensive evaluation [1-4], such as factor analysis, RITE’s Index of Information, Information Society Index(ISI), UN’s Information Utilization Potentials, fuzzy evaluation, analytic hierarchy process [5] and so on. But some methods have defects with more subjectivity on weight decision or too complex processes in calculation. The paper introduces the main steps and idea of a method using catastrophe theory (CT), namely, called catastrophe progression method (CPM). Its characteristic is combining catastrophe theory with fuzzy mathematics, and only considers the relative importance of the indices, so the method avoids the subjectivity for weight decision. Catastrophe progression method also has advantages in solving problems of fuzzy multiple object decision because the catastrophe progression is a multidimensional fuzzy membership function. With these characteristics, the method is easier and the results are more precise [6]. *
Corresponding author. Tel.: +86 010 62736764; Fax: +86 010 62737741.
D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part IV, IFIP AICT 347, pp. 672–679, 2011. © IFIP International Federation for Information Processing 2011
Evaluation of the Rural Informatization Level in Central China Based on CPM
673
2 The Evaluation Methodology for the Rural Informatization Level in Central China 2.1 Evaluation Index System of the Rural Informatization Level According as the composition project of national informatization index issued by Information & Industry Ministry of China on July 29th, 2001, taking the literature [7] from reference, this paper presents the evaluation index system for the rural informatization level in the central China, which includes the 5 categories (secondary indices) of the Economic strength, Information infrastructure, Information terminal equipment, Human resource and Information utilization, and 14 tertiary indices (Table 1). Table 1. Evaluation index system for the rural informatization level in the central China Evaluation objective
Category
Indices GDP per capita
Economic strength
informatization level of rural areas
Revenue per capita Per capita income of rural households Long-distance telephone exchange capacity Mobile switching capacity Information Length of long-distance optical infrastructure cable Internet broadband access ports Information terminal equipment
Index abbreviation GDP RP PI TC MC OC IB
Color television ownership
TO
Computer ownership
CO
Mobile telephone ownership Percentage of literate population to Human total aged 15 and over Number of rural information resource service employees Proportion of per capita information consumption Information expenditure of rural households utilization Proportion of Internet users in rural areas
MO LP IEm IEx IU
2.2 Assessment Model of the Rural Informatization Level Establishment of the hierarchical structural model. According to the regulation of indices grouping aforementioned, we constructed the catastrophe progression model for informatization level of rural areas in central China by four hierarchies, which was broken down in a manner as shown in Fig.1.
674
L. Zhang et al.
GDP ( x11 ) Economic strength (B1)
RP ( x12 )
PI ( x13 )
Cusp
Information infrastructure (C1) Information terminal equipment (C2)
Information hardware (B2)
TC ( X21 )
MC ( X22 ) Butterfly OC ( X23 )
IB ( X24 ) TO ( X31 ) Swallowtail CO ( X32 )
Human resource (B4)
Membership degree of rural informatization level in central China
Butterfly
Swallowtail
MO ( X33 )
Cusp
LP ( X41 )
Information utilization (B5)
IEm ( X42 )
Cusp
IEx ( X51 )
IU ( X52 )
Fig. 1. Catastrophe progression model for rural informatization level in the central China
Confirming the catastrophe type in every catastrophe subordinate system. Catastrophe theory analyses critical degenerate points of the potential function, where not just the first derivative, but one or more higher derivatives of the potential function are
Evaluation of the Rural Informatization Level in Central China Based on CPM
675
also zero. There are seven fundamental types of Catastrophe models presented, three types of which, such as elliptic umbilic catastrophe, hyperbolic umbilic catastrophe, and parabolic umbilic catastrophe, are under behavior variation of two dimension. Among seven fundamental types only four common catastrophe types are cusp catastrophe, swallowtail catastrophe and butterfly catastrophe, respectively, which potential functions are as follows [8]:
,
Fold catastrophe: f ( x) = x3 + ax . Cusp catastrophe: f ( x) = x 4 + ax 2 + bx . Swallowtail catastrophe: f (x ) =
1 5 1 3 1 2 x + ax + bx + cx . 5 3 2
1 1 1 1 Butterfly catastrophe: f (x) = x 6 + ax4 + bx3 + cx2 + dx . 6 4 3 2 Where f(x) is the potential function of the state variable x, and a, b, c and d are control variables of the state variable. In the catastrophe progression model for informatization level assessment of rural areas, the catastrophe type was confirmed in every evaluation index layer (see Fig.1). There were three catastrophe type, such as cusp catastrophe (one index divided into two sub-indices), swallowtail catastrophe (into three sub-indices) and butterfly catastrophe (into four sub-indices). Normalizing the control variables of the catastrophe model Standardizing data. The standardized data are in the value range of from 0 to1. Equation (1) is adopted if the index is positive, equation (2) is applied if negative. a) For The-Larger-The-Better indices, let
yij =
xij − xmin ( j ) xmax ( j) − xmin ( j)
xmin ( j ) < xij < xmax ( j )
(1)
xmin ( j ) < xij < xmax ( j )
(2)
b) For The-Smaller-The-Better indices, let
yij =
xmax ( j ) − xij xmax ( j ) − xmin ( j )
Where j is a sequence number for catastrophe subordinate systems, i is a serial number of control variables in a catastrophe subordinate system;
xij is merely the values
of x corresponding to control variables in a hierarchy of i and j, respectively;
yij is
676
L. Zhang et al.
only the standard transformation values of xij ; xmin ( j ) , xmax ( j ) are lower limit or upper limit values of x corresponding to control variables in a catastrophe subordinate system, respectively. Formulas for normalization of control variables. For the cusp, fold, swallowtail, and butterfly catastrophe system, the decomposition forms of the equation of the bifurcation point set and the normalization formulas are showed in Table 2. Table 2. Catastrophe models under behavior variation of one dimension Catastrophe type
Dimensions of control variation
Potential function
Bifurcation set
Fold
1
f x x 3 ax
a 3x 2
Cusp
2
f x x ax bx
a 6 x , b 8x
Swallowtail
Butterfly
4
f x
3
f x
4
2
2
1 5 1 x ax 3 5 3
1 bx 2 cx 2 1 6 1 1 x ax 4 bx 3 6 4 3
1 cx 2 dx 2
a
6 x , b
c
3x 4
2
Normalization formula
xa xa
3
3
8x ,
10 x 2 , b
20 x 3 ,
c
15 x , d
5
4
a , xb
xa
a , xb 4
xc
a
4x
xa xc
a
4
3
b 3
b,
c
a , xb
3
b,
c , xd
5
d
Principles of finding values of catastrophe subordinate functions. During the process of computation, we must follow two principles, i.e., a complementary and a non-complementary principle. The non-complementary principle means that the smallest of the state variable values corresponding to the control variables (xa, xb, xc and xd) is chosen as the state variable value of the whole system; However, the complementary principle implies x=(xa+xb+xc+xd)/4. Hierarchically doing the calculation in the same way, the value of the overall catastrophe subordinate function can be found [9].
3 Results and Discussion In the light of the evaluation methodology of rural informatization level aforementioned, the paper take evaluation of rural informatization level among the six provinces of central China as an example to test the effectiveness of the Catastrophe Progression Method and referred to interrelated data on Beijing and nationally average in the mainland of China. The central China is a central region, including six provinces, such as Shanxi, Anhui, Jiangxi, Henan, Hubei, and Hunan (see Fig.2), with total population of 360 million people in the end of 2008. It plays an important role on the pattern of economic and social development in China, which land area is 1.03 million square kilometers, Gross Regional Product of which reached 6.3188 trillion RMB Yuan in 2008, accounting for 19.3% of GDP in China.
,
Evaluation of the Rural Informatization Level in Central China Based on CPM
677
Fig. 2. Space diagram of the six provinces of central China
Taking data from reference [7], the region′s overall number of communication facilities was in place of four indices of information infrastructure in rural areas, telecommunications and other information transfer service employees in place of indices about the number of rural information service personnel, the proportion of local comprehensive internet users in place of the proportion of internet users in rural areas. Table 3 showed some practical index values of the rural informatization level among the six provinces of central in 2006, Table 4 showed the standardized data of practical index values for the rural informatization level from the six provinces of central China transformed by equations (1) and (2). At the same time, the scores of rural informatization level among the six provinces of central China were calculated using normalizing formula (see Table 5). Table 3. Collected data from the six provinces of central China [7] Region x11
x12
x13 X21 X22 X23 X24
X31
X32
X33
X41
X42
X51
X52
Beijing 4.98 7066.08275 365 129 1788 21
133.2 36.1 104.3 95.5
37.6 1542.2 30.4
Shanxi 1.41 1729 3181 86.2 43.2 1445 4.4
98.4 1.38 72.1 95.6
9.8
564
11.3
Anhui 1.01 701 2969 78.5 23.7 1873 2.2
91.8 1.16 79.2 83.7
4.4
527.9
5.5
Jiangxi 1.08 704 3460108.0 40.3 1052 2.8
90.0 1.10 64.7 90.8
5.9
538.4
6.6
Henan 1.33 723 3261 62.4 31.3 1764 2.5
88.8 1.02 53.4 91.4
4.3
419.4
5.5
Hubei 1.33 836 3419 82.1 44.7 1234 3.9
92.1 2.33 57.3 90.2
8.6
538.4
9.3
Hunan 1.19 754 3390 73.4 31.7 1393 3.3
80.9 0.95 60.2 93.5
5.3
591.4
6.4
89.4
7.4
593.9 10.5
Nationally average 1.79 1417 3587106.8 47.3 752 5.02
Data source: Huang, 2009.
2.7
62.5 90.69
678
L. Zhang et al. Table 4. Standardized data from the six provinces of central China
Region
x11
x12
x13
X21
X22
X23
X24
X31
X32
X33
X41
X42
X51
X52
Beijing 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 Shanxi
0.101 0.162 0.040 0.079 0.185 0.669 0.117 0.335 0.012 0.367 1.008 0.165 0.129 0.233
Anhui
0.000 0.000 0.000 0.053 0.000 1.082 0.000 0.208 0.006 0.507 0.000 0.003 0.097 0.000
Jiangxi 0.018 0.000 0.093 0.151 0.158 0.290 0.032 0.174 0.004 0.222 0.602 0.048 0.106 0.044 Henan
0.081 0.003 0.055 0.000 0.072 0.977 0.016 0.151 0.002 0.000 0.653 0.000 0.000 0.000
Hubei
0.081 0.021 0.085 0.065 0.199 0.465 0.090 0.214 0.039 0.077 0.551 0.129 0.106 0.153
Hunan 0.045 0.008 0.079 0.036 0.076 0.619 0.059 0.000 0.000 0.134 0.831 0.030 0.153 0.036 Nationally 0.196 0.112 0.116 0.147 0.224 0.000 0.150 0.163 0.050 0.179 0.592 0.093 0.155 0.201 average
Table 5 showed catastrophe progression values of the rural informatization level among the six provinces in central China. From Table 5, catastrophe progression values (CPV) of the rural informatization level averaged 0.793 in the mainland of China; CPV is bigger in only one province of the six provinces in the central China than nationally average in the mainland of China. CPV is larger in Beijing Municipality than in each region of the six provinces in the central China. Among the six provinces in the central China, CPV of Shanxi province is first with 0.800, Hubei province second with 0.773, Jiangxi province third with 0.722, Hunan province fourth with 0.714, Henan province and Anhui province least with 0.354 and 0.222, respectively. The results are found to be coincident with practical situation. Table 5. The evaluation result by CPM
Beijing
Catastrophe progression values 1.000
Shanxi Anhui
0.800 0.222
1 6
Jiangxi
0.722
3
Region
No.
Region Henan Hubei Hunan Nationally average
Catastrophe progression values 0.354 0.773 0.714
No. 5 2 4
0.793
4 Conclusion The paper take evaluation of rural informatization level among the six provinces of central China as an example to verify the effectiveness of the method. Based on catastrophe progression method, it calculated the scores of rural informatization level among the six provinces of central China using normalizing formula. The results are found to be coincident with practical situation, so it proves the catastrophe progression method works well.
Evaluation of the Rural Informatization Level in Central China Based on CPM
679
From the calculation process, it was found that catastrophe progression method works simply and quickly. So the method was proved feasible in practice and provides us a new way in solving the problems of multiple objects comprehensive evaluation [6]. Acknowledgments. Thanks are due to National Science & Technology Support Program (2008BAB38B06) and Project from the Ministry of Agriculture of China (The twelfth five-year plan of agriculture and rural informatization development in China (2011-2015)) for financial support.
References 1. Bazewicz, M.: Information systems paradigms for design, engineering and education, Cybernetics and Systems 1994. In: Proceedings of the Twelfth European Meeting on Cybernetics and Systems Research, vol. 1, pp. 391–398 (1994) 2. Yu, E.J., Leem, C.S., Park, S.-K., Kim, B.W.: An integrated evaluation system for personal informatization levels and their maturity measuement: Korean motors company case. In: Gervasi, O., Gavrilova, M.L., Kumar, V., Laganá, A., Lee, H.P., Mun, Y., Taniar, D., Tan, C.J.K. (eds.) ICCSA 2005. LNCS, vol. 3482, pp. 1306–1315. Springer, Heidelberg (2005) 3. Chang, M.S., Guo, W., Yang, L.: Research on integrated information system of regional enterprise informationization market. Computer Integrated Manufacturing Systems 9(1), 6– 10 (2003) 4. Hu, J., Yan, Y., Lu, J.P., Zhao, C.H.: A study on the informatization evaluation index system of manufacturing enterprises and evaluation standard. Modular Machine Tool & Automatic Manufacturing Technique 12, 97–99 (2005) 5. Zhang, L.X., Wen, H.J., Li, D.L., Fu, Z.T., Cui, S.: E-learning adoption intention and its key influence factors based on innovation adoption theory. Mathematical and Computer Modelling 51(11-12), 1428–1432 (2010) 6. Li, Y., Chen, X.H., Zhang, P.F.: Application of catastrophe progression method to evaluation of regional ecosystem health. China Population Resources and Environment 17(3), 50– 54 (2007) (in Chinese) 7. Huang, Z.W.: Study on analysis evaluation for China’s rural informatization in six provinces of the central. Modern Agricultural Science and Technology 15, 370–373 (2009) (in Chinese) 8. Huang, Y.L.: Application of catastrophe progression method to sustainable usage of water resource. Arid Environmental Monitoring 15(3), 167–170 (2001) (in Chinese) 9. Zhang, T.J., Ren, S.X., Li, S.G., Zhang, T.C., Xu, H.J.: Application of the catastrophe progression method in predicting coal and gas outburst. Mining Science and Technology 19(4), 430–434 (2009)
GIS-Based Evaluation on the Eco-Demonstration Construction in China Lingxian Zhang, Juncheng Ma, Daoliang Li, and Zetian Fu* College of Information & Electrical Engineering, China Agricultural University, P.O. Box 209, 17 Qinghua East Road, Haidian District, Beijing, China
[email protected],
[email protected] Abstract. The ecological construction for demonstration area is an idealized model for the establishment of rural sustainable development in the county-scale. This paper presents a GIS-based evaluation system of ecological construction for demonstration area. It includes an index system with the 4 secondary indices and 24 factor indices, and a method of ecological construction for demonstration area utilizing multi-attribute decision models of AHPCI method for comprehensive performance assessment. Taking Xinyang city of Henan province in China as a case study, the comprehensive assessment result were found to be coincident with practical situation, so it proves that the GIS-based evaluation system was fit for county-level ecological construction assessment for demonstration area as a beneficial reference framework elsewhere. Keywords: Evaluation system, Eco-demonstration construction, AHPCI, Sustainable development.
1 Introduction Since United Nations Conference on Environment and Development (UNCED) in 1990s, the sustainable development concept has gradually been accepted all over the world with respect to almost all aspects of human development [1]. The ecodemonstration area is regarded as an important way to maintain regional sustainable development in many countries. The ecological construction for demonstration area is an idealized model for the establishment of rural sustainable development in the county-scale. Eco-demonstration area is a compound ecosystem of society, economy and nature, which is relatively independent and exoteric [2]. The eco-demonstration concept lies close to the UN Millennium Project [3] and aims at making a contribution to this programme. Eco-demonstration construction refers to the application of ecological principles to the development of human ecosystems in order to achieve sustainability. It consists of three components: ecological engineering, ecological institutional reestablishment and ecological cultural remolding. Since 1980, a vigorous campaign for the ecological construction for demonstration area has appeared in China, and many eco-polis, eco-counties, eco-villages as well as eco-families had sprung up all over the country [4]. In 1996, the Ministry of *
Corresponding author. Tel.: +86 010 62736323; Fax: +86 010 62737741.
D. Li, Y. Liu, and Y. Chen (Eds.): CCTA 2010, Part IV, IFIP AICT 347, pp. 680–690, 2011. © IFIP International Federation for Information Processing 2011
GIS-Based Evaluation on the Eco-Demonstration Construction in China
681
Environmental Protection of China(MEPC) carried out the activities of the national creation of eco-demonstration area, and a lot of eco-demonstration areas had been created in our country, as not only promoted coordination of economy, society and environmental protection in pilot region but also have positive effect on surrounding areas [5]. The evaluation of sustainable development for eco-demonstration area is the base of perfecting the sustainable development theory for eco-demonstration area and guiding its construction [6]. To promote the benign development of eco-demonstration area construction, this paper promoted an evaluation index system of eco-demonstration area construction based on criterion for eco-demonstration area construction issued by MEPC, and developed a GIS-based assessment model utilizing the Analytical Hierarchy Process (AHP) and Competitiveness Index (AHPCI) method.
2 Evaluation Methodology for Intra-county Eco-Demonstration Construction 2.1 Evaluation Index System of Intra-county Eco-Demonstration Construction Taking the criterion of the national creation of eco-demonstration area and goal of eco-demonstration construction into account, the paper presented a set of quantitative and hierarchical index system with the 4 secondary indices, and 24 factor indices by integrating the assessment goal and the standards of eco-demonstration construction announced by MEPC according to the design principles and requirements of the index system (Table 1). Table 1. Index systems and its standard for the intra-county eco-demonstration construction Primary Second index Third index index
Name
Comprehensive Level of sustainable development for Eco-demonstration
GDP per capita Disposable income of urban residents Per capita annual net income of peasantry Unit GDP energy consumption(ton/ten thousand) Unit GDP water consumption meter/ten Economic (cubic thousand) structure Proportion of investment in environment protection in GDP (%) Water productivity (kilogram/cubic meter) National income
Economic development
Fourth index
Reference value Code Upper Lower limit limit Q11 Q12 Q13
>4000
1600
Q21
1.3-1.4 1.5-1.6
Q22
90
>35
Q34
>70
>35
Urbanization (%)
Q41
GDP per capita
Q42
Disposable income of urban residents
Q51 National criterion
Resource Per capita annual net utilization income of peasantry Unit GDP energy consumption(ton/ten thousand) Unit GDP water3 consumption(m /ten Ecological thousand) construction Proportion of investment in environment protection Environment in GDP (%) protection Water productivity (kg/m3) The natural population growth rate (%) Comprehensive utilization of straw Manure disposal rate (%) Environment protection Amount of fertilizer application (equivalent pure kg/ha) Amount of pesticide application (equivalent pure, kg/ha)
,
Q52
>10
>7
Q53
100
>80
Q61
>10
>10
Q62
>90
>80
Q63
>80
>60
Q64
>50
>30
Q71
>90
>70
Q72
100
80
Q73