2004 New and Renewable Energy Technologies for Sustainable Development: Evora, Protugal, 28 June-1 July 2004

2004 New and Renewable Energy Technologies for Sustainable Development This page intentionally left blank 2004 New...

Author: Maria da Graca Carvalho | Naim Hamdia Afgan (eds.)

15 downloads 1040 Views 23MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

2004 New and Renewable

Energy Technologies for Sustainable Development

This page intentionally left blank

2004 New and Renewable

Energy Technologies for Sustainable Development 28 June - 1 July 2004

Evora, Portugal

Editsrs

Maria da Graca Carvalho Naim Hamdia Afgan Institute superior Tecnico Portugal

world scientific tific NEW JERSEY * L O N D O N * S I N G A P O R E

BElJlNG * S H A N G H A I * HONG K O N G * TAIPEI

*

CIiEMNAI

Published by World Scientific Publishing Co. Pte. Ltd. 5 Toh Tuck Link, Singapore 596224 USA office; 27 Warren Street, Suite 401-402, Hackensack, NJ 07601

U K ofice: 57 Shelton Street, Covent Garden, London WC2H 9HE

British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library.

2004 NEW AND RENEWABLE ENERGY TECHNOLOGIES FOR SUSTAINABLE DEVELOPMENT Copyright 0 2007 by World Scientific Publishing Co. Re. Ltd All rights reserved. This book, orparts thereof. may not be reproduced in anyform or by any means, electronic or mechanical, including photocopying, recording or any information storage and retrieval system now known or to be invented, without written permission from the Publisher.

For photocopying of material in this volume, please pay a copying fee through the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA. In this case permission to photocopy is not required from the publisher.

ISBN-13 978-981-270-505-1 ISBN40 98 1-270-505-8

Printed in Singapore by World Scientific Printers (S) Pte Ltd

CONTENTS Foreword ....................................................................................

ix

Chapter 1 - Renewable Energy Sources Promoting Renewable Energies and Energy Efficiency through the CDM Funding Options K.D. Patlitzianas, A . Flamos, H. Doukas, A.G. Kagiannas and J. Psarras ...........

1

H2RES,Energy Planning Tool for Increasing the Penetration of Renewable Energy Sources in Island Energy Supply M. Lerer, N. Duic, L.M. Alves and M. G. Cawalho.. .....................................

15

The Shape of Complete Renewable Energy Systems in the World H. Yamamoto and K. Yamaji ...............................................................

31

Experimental Investigation and Modelling of Thermoelectric Generators for Use in Hydrogen Based Energy Systems J. Eriksen, R. Glocknev, V.A. Yartys, B. T. Hafsteinsson and T.I. Sidusson ...........

41

Intelligent Operation Management of Fuel Cells and Micro-Turbines Using Genetic Algorithms and Neural Networks A. M Azmy and I. Erlich ...................................................................

53

Chapter 2 - Biomass Energy Municipal Solid Waste Valorisation as Energy for Isolated Communities G. Tavares, Z. Zsigraiova, V. Semiao and M G . Cawalho ..............................

67

Production of Sustainable Hydrogen Using Thermochemical Gasification of Biomass J. Andries, W. de Jong and H. Spliethoff.. ................................................

77

Characterization of Kenaf Potential in Portugal as an Industrial and Energy Feedstock A.L. Fernando, M P . Duarte, J. Morais, A . Catroga, G. Serras, B. Mendes andJ.F.S. Oliveira.. ........................................................................

87

Implementation of Anaerobic Digestion Plants in Breweries - Difficulties and Benefits G. Pesta and R. Meyer-PittrofJ:............................................................

95

V

vi

Contents

Application of Biofuels to Compression Ignition Engines A. Kowalewicz ...............................................................................

105

Kinetic Study of Froth Flotation for PET-PVC Separation E. Agante, T. Cawalho, F. Durao, A. Pinto and T. Mariano.. ..........................

117

Use of Biomass in Small Direct Fired Systems C. Syred, W. Fick, N. Syred andA.J. GrifJhs ............................................

127

Emissions Reduction by Co-Firing Biomass or Waste with Coal in a Pressurized Fluidised Bed Combustion Combined Cycle Power Plant Y. Huang, J. T. McMullan, D. McIlveen- Wright and S. McCahey .......................

153

Production of Biocoal from Cassava Stalk T. Puthikitakawiwong, R. Boonsu and 0.Joompha ......................................

165

A Comparison of Power Generation from Biomass in a Small CFBC Plant with Biomass Co-fired with Coal in a Large CFBC D.R. McIlveen- Wright, B. C. Williams and J. T. McMullan. ..............................

171

Characterization of Sweet, Fibre and Biomass Sorghum Potential in Portugal as an Industrial and Energy Feedstock A.L. Fernando, M.P. Duarte, J. Morais and J.F.S. Oliveira.. ...........................

183

Chapter 3 - Solar, Wind, Wave and Geothermal Energy Process Dynamics of Fossil Steam Power Plants Induced by the Integration of Transient Solar Heat V. Scherer, K. Roth and M Eck.. ..........................................................

193

Directions for Renewable Energy in Canada’s Smallest Province A. Tvivett.....................................................................................

205

Externalities Affecting the Viability of Wind Power for Hydrogen Production N. Kassem ...................................................................................

21 1

Non-Technical Barriers to Large-Scale Wave Energy Utilisation A.J.N.A. Sarmento, F. Neumann andA. Brito-Melo.. ....................................

225

Solar Power Agriculture: A New Paradigm for Energy Production U. Bardi .....................................................................................

235

Contents

Possibility Assessment of Wind Energy Utilization in Bosnia and Herzegovina F, Begic, A. Kuzagic and N.H. Afgun.. ....................................................

vii

249

Chapter 4 - Greenhouse Gas Emissions Earth Tube Ventilation System: A Project Pre-Feasibility Analysis Tool MA.A. Costa, A4S.A. Oliveira and N. Martins ...........................................

259

Greenhouse Gases Emission Reduction in an Urban Bus Fleet: Application to a Case Study in Funchal, Madeira Island A.M Simoes, P. Lages, T.L. Furius, C.M Silvu and M N . Aguas .......................

273

Environmental Impact of Hydrogen in Urban Transports K.R. Calhau, G.A. Goncalves and T.L. Furius.. ..........................................

285

Chapter 5 - Gas Production and Utilization Devolatilisation of Coal-Biomass Blends and Combustion Behaviour of their Chars P. Ciambelli, V. Palma, P. Russo, S. Vuccaro and V. Vaiano ...........................

295

Dual Fuel Combustion as a Way of Increasing Sustainability of Power Production D.R. Schneider and Z. Bogdan. ............................................................

3 11

Hydrogen Production by Allothermal Reforming of Ethanol for Fuel Cell Applications. Prototype Development A.J. Marin Neto, E.P. Silva, J.C. Cumargo, N.P. Neves Jr. and C.S. Pinto ............ 325 Hydrogen Production from Steam or Autothermal Reforming of Ethanol for Low- and High-Temperature Fuel Cell V. Galvita ....................................................................................

335

Hydrogen Fuel Cell Urban Buses Operating in the City of Porto G.A. Gonculves, T.L. Furias, R. Teixeiru undA. Silva.. .................................

343

Chapter 6 - Sustainability Development Energy in Slovenia and Croatia: Challenges and Possibilities for Sustainability A . Zidunsek, R. Blinc, I. Sluus andD. Nujdovski .........................................

353

Sustainability Assessment as a Basis for the Decision Making in Selection of Energy System F. Begic and N.H. Afgan ...................................................................

363


FOREWORD “Since 1990, the EU has been engaged in an ambitious and successful plan to become world leader in renewable energy. To take one example, the EU has now installed wind energy capacity equivalent to 50 coal fired power stations, with costs halved in the past 15 years. The EU’s renewable energy market has an annual turnover of 15 billion Euros (half the world market), employs some 300,000 people and is a major exporter. Renewable energy is now starting to compete on price with fossil fuels. In 2001, the EU agreed that the share of electricity from renewable energy sources in the EU consumption should reach 21% by 2010. In 2003, it agreed that at least 5.75 % of all petrol and diesel should be bio-fuels by 2010” GREEN PAPER European Strategy for Sustainable, Competitive and Secure Energy, 2006. The United Nations World Summit on Sustainable Development (UNWSSD) 2002 (Rio+lO) produced an Action Plan to guide increased efforts towards meeting Sustainable Development targets as set out in the Millenium Development Goals. In this document scientific and engineering communities are invited to take the leading role in the promotion and organisation of actions that are designed to deliver sustainable development. Modem technologies are a key part of achieving this objective. In this respect the utilisation of new and renewable energy technologies provide milestones for the improvement of the economic, social and environmental quality of our life. Due to the heavy reliance of modem civilisation on intensive energy consumption, the accelerated development of renewable technologies will play a crucial role in the realisation of the shared vision of sustainable development. It will impose a new dimension in the new and renewable energy technology development. Increasing efficiency is an ever-lasting demand of modem technologies. The miniaturisation of the systems for energy conversion has opened a new sphere of activity which is of particular importance to renewable energy technologies. The Evora International Conference on New and Renewable Energy Technologies for Sustainable Development marks a new development in modem technologies. It reflects on the progress towards meeting the goals set out in the Johannesburg Conference Action Plan. This Conference followed the objectives of the Lisbon (1998), Madeira (2000) and Azores (2002) Conferences “to bring together engineering practitioners, product developers and researchers with economists, political scientist and government administrators to explore the multifaceted relationship between renewable technologies and sustainable development”. Key lectures frame the technical and policy issues confronting the sustainable development movement and enrich the dialog between the various segments of the community. This dialog provided the context for more detailed technical presentations and panel discussions on energy systems, renewable resource exploitation, and the engineering design and optimisation for minimum consumption of water, with sanitation management, including agriculture productivity and biodiversity and ecosystem management. ix

X

Foreword

The Proceedings are organised in the following chapters: Renewable Energy Sources; Biomass Energy; Solar, Wind, Wave and Geothermal Energy; Greenhouse Gas Emissions; Gas Production and Utilisation; and Sustainable Development. The Editors would like to express their appreciation to Maria Fernanda Afonso for her excellent work on the technical editing of the manuscript. It is our duty to recognise also the contribution of Ana Mascarenhas for typing and Jorge Coelho for the preparation of the figures.

The Editors May, 2006

PROMOTING RENEWABLE ENERGIES AND ENERGY EFFICIENCY THROUGH THE CDM FUNDING OPTIONS K.D. PATLITZIANAS, A. FLAMOS, H. DOUKAS, A.G. KAGIANNAS, J. PSARRAS Management & Decision Support Systems Laboratory, Department ofElectrica1 and Computer Engineering National Technical University of Athens, Heroon Polytechniou 9, 157 73, Athens, Greece The promotion of Renewable Energy Sources (RES) and Energy Efficiency (EE) is a central aim of the world energy policy in order to contribute in reducing greenhouse emissions. Moreover, Clean Development Mechanism (CDM) allows Annex I countries to achieve their emission targets. Financial assistance to developing countries is needed in order to avoid high increase of their greenhouse emissions associated with their development. Today, a large number of Annex I countries are also setting up programmes to fund RES and EE investment through CDM. These include Denmark, Great Britain, Italy, Japan, Norway, Canada and Austria. These programmes are being applied by international organisations and numerous non-governmental organisations and they facilitate projects throughout the developing world. Initial experiences have also been gained from the “World Bank Prototype Carbon Fund - PCF” and the “Dutch CDM tender-programme CERUPT”. The most important result of these initial experiences is the creation of appropriate organisational structures and the use of qualified personnel by both the host and investor countries and the respective participants in the project, thus, building the capacity to facilitate the CDM. This paper describes the main role of CDM funding programmes and their potential role in promoting RES and EE projects.

1. Introduction Energy is essential for economic and social development. About ninety per cent of the world energy supplies are provided by fossil fuels, with the associated emissions causing local, regional and global environmental problems. Most energy projections show that current and expected hture global energy demand patterns are not sustainable. Even when assuming massive improvements in energy intensity, long term projections indicate that world energy demand may increase dramatically, with most of this increase, taking place in developing countries. These trends indicate that, in order to comply with the necessary conditions for the three dimensions of sustainability (economic, environmental and social) with respect to energy production and consumption, a decoupling of economic activity from fossil primary energy consumption should be achieved. Increasing the share of Renewable Energy Sources (RES) in the energy balance enhances sustainability and helps to improve the security of energy supply by reducing dependence on imported energy sources. In addition, the development of Energy Efficiency (EE) is a central aim of the world energy policy in order to contribute in reducing greenhouse emissions [ 11. The interest in RES and EE increased when the oil crises of the 1970’s made everyone aware of the fact that fossil resources would run out one day - but since there is some uncertainty about when that will actually happen the efforts made in this area remained rather tentative. 1

2

K.D. Patlitzianas et al.

Nowadays, growing environmental concerns and limitations in the exploitation of conventional energy resources have given new impulse in modem RES technologies. Beyond 2020, new technologies, such as hydrogen-based fuel cells and carbon sequestration, hold out promising prospects of plentiful, clean energy supplies for the world. So, RES projects and EE will need to play a greater role in the future energy mix in order to achieve low-carbon intensive energy systems [2]. Despite the fact that RES and EE provide a series of benefits in national and private scale, the development and dissemination process of such technologies has been slower than expected. The state cannot always place sufficient amounts for the modernisation of the energy sector through the promotion and penetration of these technologies. In addition, the users of energy often face the implementation of RES and EE projects hesitantly. A pool of inhibits that justify this attitude, includes [2]: The high initial cost of RES and EE projects in relation with the long time period of depreciation of the investment; The lack of available funds of the enterprises for the implementation of these projects. In most cases the enterprises cannot allocate sufficient amounts of their budget in such projects, since they have to overcome more demanding priorities, such as the improvement of their competitiveness and the identification of new markets; The financial, technological and performance risks of these projects are often high for an enterprise related to the expected results; The lack of awareness regarding the performance of modern and innovative renewable technologies. Besides the barriers mentioned above, a project investment in GHG abatement technologies might encounter several other barriers, especially when the investment takes place in developing countries. For instance, one barrier may be the flow credit worthiness that often developing countries have. The Kyoto Protocol through the Clean Development Mechanism (CDM) can exert a leverage effect to increase the attractiveness of new and renewable technologies [3]. The purpose of the CDM, as defined in Article 12 of the Kyoto Protocol, is to assist Parties not included in Annex I in achieving sustainable development and to assist Parties included in Annex I in achieving compliance with their quantified emission limitation and reduction commitments [4-61. The additional funding channeled through the CDM could assist countries in reaching some of their environmental and sustainable development objectives, such as C 0 2 emission reductions, cleaner air and water and reduced dependence on imported fossil fuels. CDM projects are comprised of two parallel income flows [7-91. The first flow refers to the base project and is typical of any traditional investment in an energy project; The second flow is the “carbon flow”. The products of this flow are tonnes of carbon dioxide avoided or reduced in the case of energy projects. Through the CDM project cycle, these reductions are converted into certified emission reductions (CERs).

Enera EfJiciency through the CDM Funding Options

3

These CERs can be purchased by greenhouse gas emitting sources, such as thermal power generation companies in industrialized countries, to meet domestic emissions reductions requirements. The market price of CERs will fall in between the cost of production (including all transaction costs incurred in the CDM project cycle) and the maximum purchaser price. This price is not anticipated to exceed the cost of domestic reduction measures in industrialized countries. The sale of CERs can significantly increase the internal rate of return (IRR) of an entire project. In the case of projects with a healthy IRR on the traditional flow, the CER is an incentive to implement a more greenhouse gas efficient project. In the case of projects that are not quite commercially viable, the added input of financing for the resulting CERs could make them viable. In this context, the objective of this paper is to present the main role of CDM funding programmes and their potential role in promoting RES and EE projects. The paper is structured along 3 parts, as follows: The first part is the introduction of the paper; The second part is devoted to the review of the objectives and the procedures of the main CDM - funding programmes (PCF, CERUPT, CDM Programme of Canada, Rabobank and Finland Programme) presenting the progress so far. This part is deovoted to the interpretation of the results of CDM funding programmes and concerns actions and measures to be undertaken for maximum RES and EE penetration; The last part presents the conclusions of this study. 2. CDM - Funding Options

Governments and private companies from Annex I parties are the main end-users of CERs. Funds to implement CDM project in a developing country can flow from three different trading models: Multi-lateral model - financial resources from Annex I investors flow through a centralized investment fund and channeled toward project activities in host developing countries; Bilateral model - this involves collaboration among investors from Annex I countries, project developers, and host country governments. These parties agree on project selection, funding and credit sharing arrangements; Unilateral model - the host country invests in a project, and sells or banks CERs. The carbon funds of the World Bank (Prototype Carbon Fund, Community Development Fund, BioCarbon Fund) are examples of multi-lateral funds using the World Bank as fund manager. The Dutch Government adopts several means of producing emission reduction credits, through multi-lateral organizations such as World Bank and European Bank for Reconstruction and Development (EBRD), through banks (e.g. Rabobank) or through bilateral contracts and via its own tender (CERUPT). Similarly, the Government of Japan uses the Japanese Bank of Industrial Development to manage its CDM funds. More recently, the Canadian and Netherlands governments have initiated

4


bilateral transactions with several Latin American countries while the Danish government has recently come to bilateral agreements with Malaysia and China [lo]. The World Bank’s PCF and the Dutch Government’s CERUPT tender are the current main buyers of CERs through direct purchase transactions. As of 2002, the PCF Fund Management Committee and the Participants Committee approved a total of 16 CDM projects with emission reduction purchase agreement of almost 24 Mtn COz. The CERUPT tender, on the other hand, approved 18 projects in 2003, to generate emissions reductions of 16, 7 Mtn COZ. A number of PCF projects have been commissioned in 2002. Most of the PCF and CERUPT’s projects would be commissioned between 20032007. Furthermore, the Finnish and Swedish government launched CDM tenders in 2002 while the Austrian government is currently in the process of preparing CDM/ JI tenders. Today, CDM funds and investment approaches are growing and expanding. The Prototype Carbon Fund (1 80 million $), the Community Development Carbon Fund (100 million $) and the World Bank Bio-Carbon Fund (100 million $) are the main multilateral funds. The main government funds are categorized as follows [lo]: Own Tender: Dutch Government C-ERUPT Program; /Finnish CDWJI Pilot Program (20 million $); J Sweden International Climate Investment Program - CDM; J Austria J K D M Procurement Program; CommerciaVDevelopmentBanks: JRaboBank (Dutch Government); Japanese Bank of Industrial Cooperation (Japan CDM Fund - 4 billion yen); JDevelopment Bank of Japan (Japan CDM Fund - 3 billion yen); Multilateral Institutions: World Bank (The Netherlands Clean Development Facility - € 70 million); JIFC (IFC-Netherlands Carbon Facility - € 44 million); Bilateral Transactions: Canadian Government with Colombia and Chile; J Dutch Government with Bolivia, Colombia, Uruguay and Ecuador; /Danish CDM Portfolio -bilateral agreements with Malaysia and China. In the above framework, the most known CDM funding schemes are analysed as follows:

2.1. Prototype Carbon Fund (PCF) 2.1.I . Objective

Recognizing that climate change will have a significant impact on the world, on July 20th, 1999, the Executive Directors of the World Bank approved the establishment of the Prototype Carbon Fund (PCF). The PCF, with the operational objective of combating climate change, aspires to promote the Bank’s tenet of sustainable development,

Energy Eficiency through the CDM Funding Options

5

demonstrate the possibilities of public/ private partnerships, and offer a “learning-bydoing” opportunity to its stakeholders [ 1I]. The PCF has three primary strategic objectives [12]: Show how project-based greenhouse gas emission reduction (ER) investment can contribute to sustainable development; Provide the Parties of the UNFCCC, the private sector, and other interested parties with an opportunity to “learn by doing” in the development of policies, rules, and business processes for the achievement of emission reductions under JI and the CDM. 2.1.2. Progress so far

PCF has prepared and successfully negotiated an Emission Reduction Purchase Agreement with 12 Host Governments. The Prototype Carbon Fund signed emission reductions purchase agreements for most of its projects in 2005. More specifically, the following RES developing projects have been selected [12]: Brazil: Plantar Sequestration and Biomass Use; Bulgaria: 2 projects: J 13,4 MW biomass-based boiler to utilize wood waste produced at the Svilosa pulp and cellulose plant to replace coal; /District heating system upgrades for the city of Sofia; Chile: Mid size run-of river hydropower plant with capacity of 25 MW and 175 GWh displacing the dispatch of coal thermal power generation; Colombia: Wind farm in the northern part of Colombia to displace a mix of coal-and gas-based power generation; Costa Rica: 2 projects: Wind farm to displace thermal power capacity addition; J Small hydro to replace thermal power generation; Czech Republic: EE measures and RES through the Czech Energy Agency (CEA); Guatemala: Peaking run-of-river hydro-electric plant in the west coast of Guatemala to displace thermal power plants; Hungary: Conversion of Pecs Power plant’s existing coal-fired boilers to biomass, with annual generation 162 TJ heat and 334,3 GWh electricity; Indonesia: Implement energy efficiency measures including reducing clinker contents in the produced cement, burning alternative fuels for clinker formation, utilizing heat power generation; Latvia: Methane capture from waste management and carbon dioxide reduction from power generation; Moldova: Soil conservation afforestation of 14,394 hectares of degraded and eroded state-owned and communal agricultural lands throughout Moldova; Romania: Afforestation of 6,852 hectares of public land; South Africa: Durban Municipal Solid Waste 10 MW landfill gas-fired generator to produce electricity from landfill-collected methane;


Uganda: Off-grid hydropower development in the West Nile region of Uganda. In addition several projects are under negotiations as of September 30, 2004, such Brazil: 3 projects: JAlta Mogiana Bagasse Cogeneration; JGuarani Cruz Alta Bagasse Cogeneration; JLages Wood Waste Cogeneration Facility; China: 2 projects: J Jincheng Coal Mine Methane Recovery; 4Xiaogushan Run-of River Hydropower; Costa Rica: 2 projects: J Rio General; JVara Blanca Wind Farm; India: 2 projects: JMunicipal Solid Wastes (ABIL); JNitrous Oxide Removal Project; Mexico: 2 projects: J INELEC Hydros; /Umbrella Waste Management; Philippines: North Wind Bangui Bay Wind Farm; Poland: Stargard Geothermal Heating; Vietnam: Grontmij Landfill in Ho Chi Minh City. A broad balance has been achieved in the number of projects undertaken in economies in transition and in developing countries. Furthermore, major emphasis has been laid on the development of projects in the area of RES technology such as, but not limited to, geothermal, wind, solar and small-scale hydro energy projects.

2.2. Certified Emission Reduction Unit Procurement Tender (CERUPT) 2.2. I . Objective

Through CERUPT (Certified Emission Reduction Unit Procurement Tender), the Netherlands is aiming at attracting CDM projects by providing funds for the acquisition of CERs. The responsibility for CDM in the Netherlands is with the Minister of Housing, Spatial Planning and the Environment (VROM). The Minister has appointed Senter International as tendering authority for CERUPT [ 131. The main objective of CERUPT is to buy CERs at a low price and acceptable risk. Senter International, the organisation that manages the CERUPT for the Netherlands’ Government, therefore assesses primarily price and project risk and does not have a preference for specific project types, which means that the submitted project proposals most likely reflect the actual supply on the market. CERs are generated and delivered as follows: CERs are generated in CDM projects. To generate CERs a CDM project has to realise a GHG emissions reduction compared to the baseline scenario. The CDM


7

project’s feasibility should be proven at the moment of submission to Senter. The CDM project should be operational in the contract period. To ensure that all these requirements have been met, a business plan for the project must be included in the proposal. CDM projects have to be part of one of the following project categories: JRES (e.g. solar, wind, biomass, hydro); JEE (e.g. CHP, lighting, insulation, process optimisation); /The replacement of C 0 2- intensive fuels (e.g. oil to gas, coal to gas); J Waste processing (e.g. land fill gas extraction, waste incineration); CDM projects have to assist host countries in achieving sustainable development goals. The host country will have to judge impacts on the environment of the CDM project. This may require an Environmental Impact Assessment on request of the host country; Public participation is essential in CDM projects. Local stakeholders are to be invited to comment the draft project design. The final project design should take these comments into account. Further general requirements are that: Greenhouse gase (GHG) emission reductions should be additional to any that would occur in the absence of the CDM project. This must be demonstrated through in a baseline study; CDM projects are validated by Designated Operational Entities (DOE) and registered by the Executive Board (EB); Emission reductions are verified and certified by DOES, CERs are issued by the EB; CDM projects are subject to levies by the EB. The EB will withhold part of the CERs (2%) to cover administrative expenses and to assist least developed countries in meeting costs of adaptation to the adverse effects of climate. CDM projects in one of the least developed countries are exempt from the adaptation levy [14]. Furthermore, the EB will charge an administration fee on each registered CDM project. The volume of this fee has not been specified yet. 2.2.2. Progress so far The program has been launched on 1 November 2001 and resulted in 78 expressions of interests covering all relevant regions and technologies. Among these proposals, 26 qualified for the awarding stage and 18 proposals have been already awarded (13 March 2003) [ 141: 11 projects from Central and Latin America: JTotal emission reductions achieved: 7,7 Mton; /Type of projects: Biomass, geothermal, hydro, landfill, wind and EE projects; 0 7 projects from South-East Asia: /Total emission reductions achieved: 8,8 Mton; /Type of projects: Biomass, geothermal and wind projects. It is unclear what the future of CERUPT will be. There has only been one tender period which officially closed on January 3 1, 2002 and is therefore no longer open for

8


submission of projects. Since CDM projects can often take a long time to process, it is likely that the Dutch Government is hoping to see first if the CDM Executive Board approves the 18 projects under CERUPT and then evaluate. As of yet none of the 18 projects have passed through the CDM approval process, though the Wigton Wind Farm project in Jamaica which CDM Watch describes as 'business as usual' has been approved by the CDM Executive Board 'subject to required changes'.

2.3. CDM Programme of Canada's CDM and JI Office 2.3.1. Objective

Canada's Clean Development Mechanism and Joint Implementation (CDM & JI) Office was established within the Climate Change and Energy Division of the Department of Foreign Affairs and International Trade (DFAIT) in 1998. The Office is the federal government's focal point for CDM and JI activities. It was created to enhance Canada's capacity to take advantage of the opportunities offered by the CDM and JI and is guided by an Interdepartmental Steering Committee comprised of representatives from Industry Canada, Natural Resources Canada, Agriculture Canada, Environment Canada, the Canadian International Development Agency and Canada's Climate Change Secretariat. Under Action Plan 2000, the CDM & JI Office received funds for the period 200 1-2005 in order to pursue the following three objectives: To strengthen Canada's capacity to take maximum advantage of the Kyoto Mechanisms. These include the Clean Development Mechanism, Joint Implementation and Emission Trades backed up by emission reduction projects; To encourage and facilitate Canadian participation in the Kyoto Mechanisms by building awareness, promoting cost-effective opportunities and lowering transactions costs, while also engaging developing countries and countries-in-transition in such activities; To assist Canadian entities in obtaining emissions reductions credits from CDM and JI type projects, this can assist Canada in meeting its Kyoto target, according to international rules and guidelines. With the elaboration of international rules and guidelines at Marrakech in November 200 1, the Office is primarily focused on project facilitation including: Guiding companies on the technical requirements for CDM activities and issues related to ownership of emission reduction credits; Providing financial support for market identification studies, feasibility assessments, baselines and monitoring plans, risk assessments and environmental impact studies; Preparing memoranda of understanding and project-specific agreements with host country governments. Project proposals submitted to Canada's CDM & JI Office will be assessed in reference to the following key criteria:

Energy Eflciency through the CDM Funding Options

9

Promotion of the objectives of the UNFCCC. The UNFCC seeks to stabilize GHG concentrations in the atmosphere at a level that does not create dangerous anthropogenic interference with the climate system; Contribution to the objectives of Canada's CDM & JI Office; Consistency with Canada's International Strategy on Climate Change Proposals must respond to at least one of the Strategy's objectives: JMaximise Canada's ability to meet its UNFCCC commitments and Kyoto Protocol targets at the lowest cost; J Contribute to the achievement of global climate change objectives; JMaximize business opportunities for Canadian business interests in international projects and initiatives on climate change [15-161.

2.3.2. Progress so far Indicative recent events of the Canada's CDM & JI Office in order to strengthen Canada's capacity for CDM are the following: Introductory Workshops on the Kyoto Mechanisms, SPRING 2006; Trade Team Canada Environment (TTCE) Mission to China, February 27 to March 10, 2006; CDM Seminars in India, March 1-11,2006; Canada's CDM & JI Office. DNA and Focal Point Workshop, March 27th-28th, 2006, Vancouver, BC; Carbon Expo, Cologne, Germany, (May 10-12,2006); Trade Mission to Carbon Expo 2005. May 11-13. Cologne, Germany; Business Development Mission to India Partnering Together. February 2005; AMERICANA 2005, Pan-American Environmental. Technology Trade Show and Conference; COP- 10, Buenos Aires, Argentina, (December 6th-l7th, 2004); Green Power Conference. Metro Toronto Convention Centre, (November 24 & 25, 2004); DNA and Focal Point Conference Ottawa, (September 30th- October lst, 2004); Canada's CDM & JI Office at the Emissions Marketing Association (EMA), Fall Meeting. Toronto, (September 2004); Canada's CDM & JI Office at Globe 2004.Vancouver, British Columbia, (March 3 1April 2); Globe 2004 Conference Program; CDM & JI Workshop at Globe 2004 (March 29,2004); Salon des technologies environnementales du QuCbec 2004 (March 17-19,2004) ; CDM & JI National Technical Workshop Ottawa (March 16-17,2004); Canada-Cuba Seminar on Clean Development Mechanism - Havana, (February 2324,2004).

10


In addition, 65 CDM and JI proposals have been awarded from the following regions: Africa and Middle East (14 projects); Latin and South America (19 projects); Central Eastern and Europe (I 5 projects); Asia (17 projects). 2.4. Rabobank Fund 2.4.1. Objective

In January 2003, Rabobank lnternational has established the Carbon Procurement Department for the sale and purchase of project-related emission reductions from projects in developing countries More specifically, the Rabobank International concluded a contract with the Ministry of Housing, Spatial Planning and the Environment for the Dutch government to purchase ten million tons of certified greenhouse gas reductions from projects in developing countries in Asia, Latin America and Africa. These projects should have a sustainable character. In these projects, Rabobank may participate as financier via its international network.

2.4.2. Progress so far During the next eighteen months Rabobank will conclude contracts for the purchase of emission reductions with the owners of CDM projects. These will involve bilateral contracts and project-related reductions such as: A wind farm and a hydroelectric power station in China; A hydroelectric power station in Chile; A wind farm and a biomass project in India; A manure management project in Brazil. In some cases the extra proceeds generated by the sale of emission reductions will enable projects to be carried out. Otherwise, these projects would be difficult to be financed. In addition to reducing harmful emissions, these projects often also serve to strengthen the local economy, in part by improving the infrastructure and creating employment [ 171.

2.5. Finland Programme 2.5.I . Objective Finland has a commitment of 8% reduction of greenhouse gases according to the burden sharing agreement under the EU as a result of the Kyoto Protocol. The Ministry of Foreign Affairs (Development Cooperation) is currently exploring the possibilities of


11

purchasing certified emission reductions via small-scale CDM projects. The guidelines for small-scale projects are simplified, reducing transaction costs, and small-scale projects are in general more beneficial for sustainable development than large-scale CDM projects. The government of Finland has published an invitation to submit project proposals [15-161.

2.5.2. Progress so far Currently, the Pilot Programme has seven on-going CDM and five on-going JI projects. The CDM projects are located in Viet Nam, Honduras, El Salvador, Costa Rica, India and Zambia. The JI projects include four projects in Estonia (Tamsalu, Kadrina, Paide and Pakri). The projects are at different stages. Some are already validated or determined and some are only in their early stages. Most of them are RES projects and include among others Hydro Power; Biogas; Landfill; Gas Recovery; Wind Power. 3. Conclusions

RES and EE constitute really important factors to sustainable development. However, the development and dissemination process of such technologies has been slower than expected because the use and promotion of RES and EE is inhibited by a number of factors. The most important factors are the high initial costs, the financial, technological and performance risks, the scarcity of investment capital and the investing priorities of private companies. The new climate change regime also offers an opportunity for RES and EE as they meet the two basic conditions to be eligible for assistance under the UNFCCC implementing mechanisms: They contribute to global sustainability through GHG mitigation; They conform to national priorities by leading to the development of local capacities and infrastructure. The application of CDM funding programmes could facilitate the development of RES and EE projects both in national and private level. RES, as indigenous sources can enable local development since they can secure the environmental integrity through the minimisation of C02 emissions. In addition, RES projects can contribute to the technology transfer through the CDM of the Kyoto Protocol [ 18-19]. CDM projects require upfront investments that are normally obtained from different sources such as loans, equity and grants. As in conventional projects, the funding of CDM projects can be arranged either through corporate or project funding. Additional project revenues (i.e. CERs) could leverage debt financing.

12


In addition, the net financial gain derived from the sale of CERs is the difference between the project CER value and the transaction costs. There are three elements that influence the net impact of CERs on project profitability [20]: Value of CERs (low CER value implies low net benefits); Overall transaction costs (high transaction costs yield low net benefits); Up-front transaction costs (high upfront payments could also result in low benefits). Project developers generally expect up-front transaction costs within the range of 5 to 7% of the net present value of the revenue or total transaction costs around 10 to 12% of the net present value of revenue. A positive net financial gain means that CER revenues improve the financial viability of the project [lo]. Nowadays, the decisions adopted at the COP 9 and the recent adaptation of the directive linking the EU Emissions Trading Scheme to RES and EE projects, made clear that CDM will play a significant role as an effective mechanism for achieving the binding targets of the Kyoto Protocol. Within this framework it is expected that the EU Governments as well as the developing countries will make further use of the CDM funding programmes, which will play a more and more important role in funding RES and EE projects [21,22]. References

[ 13 European Commission (2003) Renewable Energy Technologies and Kyoto Protocol Mechanisms - Joint Implementation in Central and Eastern Europe - Clean Development in the Mediterranean Area.

[2] Second European Climate Change Programme (ECCP) Progress Report. (2003) Can we meet our Kyoto targets? [3] Geres R, Michaelowa A. (2002) A qualitative method to consider leakage effects from CDM and JI projects. Energy Policy 30 (6): 461-463. [4] The Pembina Institute for Appropriate Development. (2002) A User’s Guide to the Clean Development Mechanism (CDM). Second Edition. Canada. [5] Wees M. (2002) Current developments in CDM implementation: from theory to th th practice. National Policy Seminar on Clean Development Mechanism 5 -6 August, Kuala Lumpur. [6] Friedman S . (2002) The use of benchmarks to determine emission additionality in the clean development mechanism. US Environmental Protection Agency. [7] Anagnostopoulos K, Flamos A, Kagiannas AG, Psarras J. (2003) The impact of clean development mechanism in achieving sustainable development. International Journal of Environment and Pollution 21 (1): 1-23. [S] Flamos A, Anagnostopoulos K, Askounis D, Psarras J, Butzengeiger S, Gaast W. (2003) E-serem - A web-based manual for the estimation of emission reductions from JI and CDMprojects. Special Issue Mitigation and Adaptation Strategies for Global Change. Kluwer Academic Publishers. [9] Flamos A, Anagnostopoulos K, Askounis D, Psarras J. (2003) The multiple benchmark system application to Indonesia, Russia and Panama. Special Issue

Energy Eflciency through the CDM Funding Options

13

Mitigation and Adaptation Strategies for Global Change. Kluwer Academic Publishers. I101 Myung-Kyoon L. (2004) CDM Information and Guidebook. United Nations Environment Programme (UNEP) - Netherlands Ministry of Foreign Affairs. Unep Riso Centre on Energy, Climate and Sustainable Development Riso National Laboratory Roskilde, Denmark. [ 113 Word Bank (2006) Carbon Finance at the World Bank (http://carbonfinance.org). [12] Prototype Carbon Fund (PCF) - The World Bank. (2005) Annual Report - 2004. Washington. [13] Ministry of Housing, Spatial Planning and Environment of the Netherlands International Environmental Affairs Directorate - Clean Development Mechanism Division. (2006) CERUPT Guideline, (http://www.senter.nV). [14] Ministry of Housing, Spatial Planning and Environment of the Netherlands International Environmental Affairs Directorate - Clean Development Mechanism Division. (2006) CERUPT Tender Document. (http://www.senter.n1). [15] Coninck HC, Linden NH. (2003) An overview of carbon transactions - General characteristics and specific peculiarities, ECN, Peten ECN-C-03-022, March 2003. [16] Davidson 0. & Sparks D. (Eds) (2002) Developing energy solutions for climate change: South African research at EDRC - Energy & Development Research Centre, University of Cape Town, pp. 83-103. [17] Maruyama A. - UNEP, Aalders E. - IETA, Lee MK - UNEP (2005) Carbon Market Update for CDM Host Countries. Newsletter 1. [ 181 World Bank Carbon Finance Business, International Energy Agency, International Emissions Trading Association - PCFplus. (2004) Estimating the market potential for the clean development mechanism: Review of models and lessons learned. Washington DC. [19] Barcena A. (1994) What is capacity in environment? A developing country perspective in capacity development in environment. Proceedings of a Workshop held in Costa Rica, 9-1 1 November 1993, Paris: OECD: 46-53. [20] Wohlgemuth N. (2001) Directing investment to cleaner energy technologies: The role of financial institutions. In: Sustainable banking, the greening of finance. Greenleaf Publishing, Sheffield, 401-4 11. [21] Patlitzianas K, Kagiannas A, Askounis D, Psarras J. (2003) The policy perspective for RES development in the new member states of the EU. Renewable Energy 30 (4): 477-492. [22] Doukas H, Patlitzianas K, Kagiannas A, Psarras, J. (2005) Renewable energy sources and rationale use of energy development in the gcc region: Myth or reality? Renewable Energy 31: 755-770.


HJWS, ENERGY PLANNING TOOL FOR INCREASING THE PENETRATION OF RENEWABLE ENERGY SOURCES IN ISLAND ENERGY SUPPLY M. LERER, N. DUIC, L.M. ALVES, M.G. CARVALHO Instituto Superior Tecnico, Av. Rovisco Pais, 1049-001, Lisbon, Portugal The islands’ high energy costs make them excellent test beds for introduction of new technologies. Some islands are trying to become so called renewable islands, to satisfy their energy demand mainly or entirely from indigenous and renewable sources, thus increasing the security of supply, and employment opportunities, without necessarily increasing the costs. Advanced energy planning must consider different intermittent and constant resources, to satisfy the demand for energy, water and fuel for transport, taking into account security of supply on different time scales and necessary supply storage. Such studies cannot be done with traditional methods (LDC and Weibul curves) therefore time series analysis methodology is a requirement. The most promising energy storing technologies are reversible hydro (where geography allows) and storing hydrogen, making possible in the first case integration of the water system, and in the second case integration of the future hydrogen based power and transport systems. This paper describes the new developments in H2RES model, and its application on the example of an isolated island in Madeira archipelago, Porto Santo. HzRES is based on time series analysis that includes wind, precipitation and solar as renewable resources and presents reversible hydro, batteries and hydrogen as energy storages. It also allows for deferrable and hydrogen loads. Considering the present situation and the expected energy needs in Porto Santo by 2010, 7 possible cases of evolution of renewable energies in the island are presented, considering the power and road transport systems. The renewable sources available are wind and solar. In some of the cases, storage is introduced: the excess renewable energy can be used to electrolyze water and the produced hydrogen is stored for later use by fuel cells, either for supplying the power system or for transport needs. Renewable sources can be limited to 30% or to 100% of load, on hourly basis. It is seen that it is possible to significantly increase the penetration of renewable energy sources, albeit at a relatively high cost, with hydrogen storage technology. The H21ZESmodel can serve as a valuable tool for island energy planning.

1. Introduction

As discussed in the European Commission’s White Paper on Renewable Energy Sources [l], the United Nations Conference on Islands and Small Island States (Barbados 94, [ 2 ] and the lStEuropean Conference on Island Sustainable Development, there is a need to create conditions for developing the renewable energies, specifically regarding islands. The European Island Agenda [3] considers “the non-renewable energy sources as provisional solutions, inadequate to solve in the long term the energy problems of the islands”. Porto Santo, in Madeira archipelago, Portugal, is one of the European islands that aim at becoming 100% renewable. Environmental reasons are not the only good reasons to turn to renewable islands. Insularity, in general, means isolation andor dispersion, with the associated supply related obstacles. For these small local markets, the costs of transports, communications and energy are high when compared to the closest continental regions. Regarding electricity generation, most of islands depend on fossil fuel imports, whose costs are increased by transportation and that may suffer from security of supply problems. From the economic point of view, the higher the costs of conventional fossil solutions, the more attractive the renewable energy technologies become. 15

16

M. Lerer et al.

In some regions of Europe, intermittent renewable energies supply already about 10% of the demand (one example is Denmark’s wind electricity). In special cases of very good resources (for instance, steady wind) this fraction can be increased to about 20%, but for higher penetrations energy storage is necessary. Islands arise as excellent small scale systems to access the technical fitness of such schemes. As common tourist destinations, islands shall take special care of their environmental conditions, so use of fossil fuels should be decreased as possible. The increasing awareness of the public relating environmental issues makes it care about the impact of tourism in the local environment and “green” turns out to be appealing for marketing. As a consequence, Eco-tourism is a new market for islands, and allows highlight the sustainability of the community and tourist project. Dimensioning the renewable capacity of an island’s power system is a complex task. In some cases, the insular electrical grid may be connected to the mainland, but if the island is remote the grid will be isolated. If the island is connected, the excess of renewable energy can usually be taken by the larger system of the mainland without disturbing it. In case of isolated grids with important intermittent renewable energies supplies, such as solar or wind, the stability of the grid imposes limiting the amount of power taken and also requests control of frequency and voltage. The maximum intermittent power that enters the grid must be limited, and the rest of it should be stored or rejected. A very effective way of storing surplus energy is reversible hydro, but it may be inadequate for some geographical and climatic conditions. A new and very promising storage technology is based on hydrogen. The hydrogen is obtained by electrolysing water with the excess renewable energy, then it is stored, and its energy can be used later by a fuel cell, for stationary or mobile uses. For conventional energy planning, it is usual to use the statistic approach based on load duration curves (LDC), where the load is sorted by magnitude instead of time, and Weibull curves. This statistical approach cannot be used well when there are intermittent renewable sources, especially if the sources show patterns different from the intermittence of the load and when both are of the same order of magnitude, as desired for the renewable islands. In such case it is necessary to model the system on an hour per hour basis, for a representative year regarding resources. On islands that show a relevant tourism activity the load is likely to show not only the usual daily intermittence pattern but also an important seasonal fluctuation, and the same applies to water demand. For energy planning should be considered the expected evolution for the convenient time horizon (10-30 years). The conventional planning tools, like ENPEP, cannot be used in such situations, and several new energy-planning tools are being developed. EnergyPlan [4]for example is well adjusted for decentralised power generation, and it also integrates heat demand into the model, enabling the optimisation of combined heat and power generation, which already delivers nearly 50% of power to Danish system. It also integrates other intermittent resources, and optimises different strategies to treat the power excess. On the other hand, it does not treat hydro resource, water demand, hydrogen demand, reversible hydro, hydrogen storage, batteries and other specifics of the isolated power systems.

Energy Planning for the Penetration of RES

17

HOMER on the other hand is made particularly for small isolated power systems, although allows for grid connection, and has some of the needed technologies, but still lacks reversible hydro and water demand treatment, that is the cheapest way to store energy in those islands where there is potential [5].There are also models that contains more precise physical models of some technologies, like Hydrogems [6], but it also lacks hydro resource, reversible hydro storage and water demand, among others, and for the purpose of energy planning it is not necessary to go into conversion detail, beside the necessary level, although when it comes to component dimensioning, it might be beneficial. 2. H2RESModel

The H2RES [7] model is based on hourly time series analysis of demand (water, electricity and hydrogen), and of resources (wind speed, solar radiation and precipitation). It is composed of several modules that consider specific user data on system components (power, nominal efficiency, output curves, storage capacities, etc.) and control characteristics (maximum renewable penetration on hourly basis, security of supply, etc.). The wind module uses the wind velocity data, typically from the meteorological station, measured at 10 m height and, for a given choice of wind turbines, adjusts the velocity data to the wind turbines hub level and considers the efficiency curve to convert the velocities to the output. The solar module converts the total radiation on the horizontal surface, obtained typically from the meteorological station, to the optimum inclined surface considering geographical location, and then to electrical output. The hydro module takes into account precipitation data, typically from the nearest meteorological station, water collection area, and evaporation data, based on the reservoir free surface, to predict the water net inflow into the reservoir. Load module, based on a given criteria for the maximum acceptable renewable electricity in the power system, puts a part or all of wind and solar output into the system and discards the rest of the renewable output. The excess renewable electricity is then either stored as hydrogen, pumped water or electricity in batteries, consumed for some non-time critical use, or rejected. The energy that is stored can be retrieved later, and supplied to the system as electricity. The rest of demand is covered from Diesel blocks. The model can also optimise the supply of water and hydrogen demand. There are two abilities of H2RES that make it specifically suited for islands: it can manage the water supply and consider hydrogen load for other than power supply. The hydro module allows managing potable water demand and excess energy storage at the same time, using pumping as a deferrable load and considering pumped potable water as a storage for reversible hydro power. Hydrogen has been seen as a new energy vector and can be specially suited for islands and remote regions, when produced in a renewable way, because it increases the security of supply and opens the renewable market for other areas, as for instance transport. For islands lacking fresh water, there is the possibility of developing a new module based on desalination to be associated to the hydro module. In the current version

18

M. Lerer ef al

~esa~ination c m be treated as deferrable load, but there is no link between it and water storage module.

3. Port0 Santo Popto Santo i s a 42 Ian2island in Madeira Archipelago, situated at about 16"W 32% on p h e rover a ~ 800 km away from any cont~ne~tal the Atlantic Ocean? an ~ ~ ~ ~ ~ p e r iregion land. The main economic activity is towism, with a clear predom~ancein summer, which is ~ ~ s ~ o nfor s ~anbincrease ~e from about 5500 occupants in winter to about 15000 in summer. The island has a mild temperature all year (average of 18.4"C), showing a temperate, oceanic, semi-arid climate. Regarding final energy use, transport arises as the most important sector (Fig. I), also indirectly related to tourism activity.

'industry

Hotel Sector

Agriculture, Cattleraising, Fishing -.

I

Fig. 1. Port0 Santo final energy distribution in 2000, by sector

3.2. Resou*ces

The solar resources of Port0 Santo are quite good, presenting about 1700 k W m 2 year. Tbe annual distribution of solar radiation is shown in Fig. 2. Currently there are some localked ~ns~a~lations of thermal solar panels for heating water at ~ w e ~ ~ but ~ ~ g § , photovoltaic technology has not been used except for eventual stand alone systems and do not provide electricity to the grid.

Energy Planningfor the Penetration ofRES

I

19

1,200,

I

0

1000

2000

3000

5000

4000

6000

7000

8000

hours

Fig. 2. Hourly average total solar radiation on horizontal surface, meteorological station, Porto Santo, 2000

The 30 years averaged wind velocity, as measured at the meteorological station on Porto Santo airport, at 10 m height, is only 4.2 d s , what is not particularly high. The lowest monthly average is 3.4 m/s corresponding to September, while the highest average corresponds to April, 4.5 mh, which makes it fairly constant throughout the year. The results from exploitation of installed wind turbines are giving 35% higher average wind velocity at 10 m, at the location. The wind velocity at the airport meteorological station is given for year 2000 in Fig. 3:

0

1000

2000

3000

4000

5000

6000

7000

8000

hours

Fig. 3. Hourly average wind velocities at 10 m, meteorological station, Porto Santo, 2000.

The island lacks water resources, both from springs, wells and precipitation, so hydro energy is not considered an option, unless the water system was re-designed and water storage could be integrated into a reversible hydro facility. The biomass is scarce. The wave conversion technology is not yet feasible. It was not considered technically practical usage of any other renewable resources. Among the fossil fuels, only oil derivatives are imported to Porto Santo: petrol, Diesel, fuel oil and LPG. In Madeira Archipelago there has been discussion on the potential for LNG. The technology of sea transport of liquefied natural gas is well established on large scale. It is expected that with maturing of the technology, smaller scale applications will become

M Lerer et al.

20

viable, which will show a new way for islands, starting from bigger ones, like Cyprus or Madeira, but if the scaling down does not reach some unsurpassable obstacle, it is probable that it will also be applicable to small scale islands, like Port0 Santo, in the same way LPG distribution is made today. 3.2. Electricity Demand and Power System

The annual load profile for electricity is given in Fig. 4.

1000

'

I

0 , 0

1000

2000

3000

4000

7000

6000

5000

8000

hours

Fig. 4. Hourly average electricity system load, 2000.

The system had in 2000 a peak load of about 5.5 MW and a low load of about 1.1 MW. The seasonal fluctuation is considerable and mainly related to tourism activity. The increase in electricity peak and annual demand for recent years is shown in Figs. 5 and 6:

1.:

-

..

..

..

. " " .... .

,.

. .. . .

...

.,

,

__

.

1

2,

1

Fig. 5. Peak Demand, 1996-2000 [8].

..

,

.

......

.

..

Energy Planningfov the Penetration ofRES

Annual Electricity Production

1

1991

1992

1993

1994

1995

1996

1997

21

IHThermal .Wind)

1998

1999

2000

2001

2002

Fig. 6. Annual electricity production 1991-2002 [8].

Regarding the production of electricity, there is a wind park with two 225 kW Vestas and one 660 kW Vestas, and a thermal plant that has 3 fuel-fired units (installed from 1998 on) of 3.41 MW each and 2 old diesel-fired groups (3.5 MW each) that are used mainly for backup and rare peak times. Since the installation of the 660 kW Vestas in the end of 2000, the wind has been providing about 8% of the yearly electricity demand. For 2000 the total demand was 25.2 GWh, increasing to 29.8 GWh in 2002: Table 1 , Electricity generation history, Porto Santo, 1991-2002.

The lack of water, mainly during summer, led to the installation of a reverse osmosis desalination plant in 1990, which is currently the bigger electricity client of the island. As the desalination plant doesn’t have storage for fresh water, the plant functions mainly during the day, contributing to increase the load at peak times. Installing a water tank to store water produced during the night would allow a better management of electricity consume, both reducing the load at peak time and increasing the load at night, thus admitting the penetration of intermittent sources to increase. 3.3. Road Transport Sector

The number of vehicles has been growing fast in the last years in the whole archipelago, especially concerning private owned vehicles, thanks to the increase in wealth of local communities and the great development in road accesses. Porto Santo has 5000 yearlong inhabitants and the penetration rate of private owned vehicles is 522%0(the greatest of all

22

M. Lerer et al.

the Municipalities in the archipelago) or about 2250 vehicles [9]. In 2000, Porto Santo road transport consumed 2274 toe of fuel, which corresponds to 75% of the total spent in transport in the island (3028 toe), and almost 40% of all primary energy consumed (5761 toe). The land transport has doubled its fuel consumption between 1991 and 2000.

3.4. Energy Demand Expected Evolutionfor 2000-2010 Based on hypothesis of regional development, national and European orientations, international markets of energy products, technological progress and environmental aspects, three scenaria (high, low and alternative) were constructed for the energetic sector development in the Madeira Region, as described in [9]. The values presented here for the demand are based on the high development scenarium forecasts, and it is assumed that Porto Santo Island will show the same evolution as Madeira Region as a whole, The forecasts can then be used for medium-range planning on the energy sector. The yearly demand growth rate is expected to be 8% during 2000-2005 period and 7% during 2005-2010 period, both concerning electricity and fuel for road transport. This means Porto Santo’s electricity production will increase from 25.2 GWh in 2000 to 37.0 GWh in 2005 and to 51.9 GWh in 2010. If the yearly peak follows the same growth rate, it will increase from 5.4 MW in 2000 to 8 MW in 2005 and to 1 1 . 1 MW in 2010. The fuel consumption for land transport will grow to 3341 toe in 2005 and 4686 toe in 2010.

4. Becoming Renewable Small power systems usually have their power frequency controlled by a single block. Small amounts of power coming from other sources will easily adjust to synchronous operation. It is safe to say that, at any single hour, the maximum power that can come from sources without frequency control is around 30%. That allows for even higher values during smaller periods of time, but at any moment will in principle stay under 50% of the load. With the present wind turbines technology, it would not be possible to increase the wind capacity in Porto Santo because it is already higher than 50% of the low load value (1.1 MW). This means that, if a low load situation occurs in a time when wind availability is good, the momentary penetration of wind may be higher than 50%, and the quality of the electricity delivered will be unacceptably low. Such a limit placed on intermittent sources will, typically for wind, allow only 10-15% of the total yearly electricity produced to be renewable. According to [lo] for a 5 MW system, as Porto Santo, one could possibly expect, at the current level of technology, to get less than 20% of electricity from wind, on yearly basis. That would mean either accepting more than 30% of wind electricity in some intervals, with consequences to the quality of electricity delivered, or installing active power controlled variable pitch wind turbines that can easily adjust the output to the load, andfor would condition installation of frequency and voltage control for all wind turbines and other renewable sources, and some kind of energy storage [ 1 1 - 161.

Energy Planning for the Penetration of RES

23

In this work we consider that all new wind turbines to be installed should be of variable pitch and be indirectly connected to the grid, with asynchronous AC-DC-AC conversion and frequency and voltage active control (Enercon 300). For the photovoltaic and the fuel cell systems, the connection is DC-AC with frequency and voltage active control. We consider 7 different cases that allow increasing the renewable share, especially Concerning the electrical demand:

4.1. Case I : Wind 30% Baseline Scenarium This scenarium is used as baseline scenarium, the one that could be considered a best conservative economically viable scenarium, while in the same maximising the renewable penetration. Wind energy now can be considered economically viable on islands, when not surpassing certain limit of penetration. For the purpose of this scenarium that limit is established at 30% renewable penetration on the hourly basis, meaning that no more than 30% of load during one hour can be covered by power coming from wind turbines. The system is optimised so that yearly wind penetration was maximised while keeping the rejected wind electricity close to 10%.

4.2. Case 2: Wind 100% Scenarium This scenarium allows the hourly penetration rate to reach loo%, meaning that when there is enough wind, Diesel generators will be completely shut down. The number of wind turbines was optimised so that yearly wind penetration was maximised while keeping the rejected wind electricity close to 30%.

4.3. Case 3: FCWind 30% Scenarium This scenarium is similar to case 1 in limiting the wind penetration to 30% on hourly basis, and optimising it so that less then 10% of wind potential is rejected. The difference is that a peak shaving fuel cell is added, together with electrolyser and hydrogen storage, in order to store the excess of wind generated electricity. Water is electrolysed to produce Hydrogen using the renewable excess that cannot be taken directly by the power system. The case defines the peak periods when load is more than 80% of the weekly moving hourly peak. The system is optimised so that fuel cell covers 1% of the yearly load.

4.4. Case 4: FCWind 100% Scenarium FCWind 100% scenarium enables that all the electricity in Port0 Santo is supplied from wind turbines, either directly, or through storage oE the excess production into hydrogen. Direct wind power is maximised (up to 100% hourly load), so that no more than 30% is rejected, and the rest is supplied via fuel cell. This is a 100% renewable scenarium concerning electricity.

24

M Lerer et al.

4.5. Case 5: FCSolar Wind 30% Scenarium This scenarium is similar to case 3, just that wind turbines are kept exactly as in case 1 and solar PV panels supply the rest of renewable electricity, with penetration from intermittent sources limited to 30% on hourly basis, and optimised so that less then 10% of renewable potential is rejected. The fuel cell is used for peak shaving, supplying approximately 1% of yearly load.

4.6. Case 6: FCSolar Wind 100% Scenarium This scenarium is similar to case 4, just that instead of only wind, such a combination of solar and wind is chosen, in which for the same results (30% rejected, 0% Diesel) one try to minimise the sum of installed wind + solar power. This is a 100% renewable scenarium for electricity generation.

4.7. Case 7: FCWind Transport 30% Scenarium This is very similar to case 3, just there is additional hydrogen load, represented by three shuttle vans operating on fuel cell, and doing 40,000 km per year per shuttle. With an average car economy of 0.05 kgH2h-1, that makes 2000 kgH2 per shuttle per year. For simplicity we can consider the hourly load of Hydrogen for the shuttles as constant trough the year, which does not cause much error as the load is quite small. The hydrogen for stationery fuel cell as well as for shuttle vans is produced by electrolyser using the excess wind electricity and stored, In order to have reasonable security of supply of hydrogen for shuttle, there has to be one month transport operation reserve (since the stationery use can be backed up with a Diesel engine). When the hydrogen falls under that level, stationery fuel cell will not be used. For these simulations we assumed the following efficiencies for the equipment: PV: 5.78% total (8.5% nominal for PV panels, 80% for converter, 15% losses); Electrolyser: 60%; Fuel Cell: 50%; Diesel blocks: 30%. The losses associated with compressing and storing the Hydrogen are included in the electrolyser efficiency. The wind turbines efficiency curve is given by the producer. Next we present the results for year 2010.

4.8. Results for 30% Hourly Penetration: Cases 1 , 3 , 5 and 7 These cases refer to a limit of 30% on hourly penetration of intermittent renewable sources but differ in technologies installed, regarding both generation and possibility of storage of surplus energy. Fig. 7 shows the equipment power and storage capacity that should be installed for each scenarium, in 2010:

Energy Planningfor the Penetration of RES

60

t

I -

50

40 P

30 0

a! m

E

20

f

10

0 Case 1 Wind 30%

Case 3 FCWind 30% Case 5 FCSolarWind 30%

Case 7 FCWindTransport

Fig. 7. The comparison of components needed to be installed in different cases of 30% intermittent limit, year 2010.

The fraction of electricity supplied by renewable sources (Fig. 8) is about 11% in Case 1, 16% in Case 3, 17% in Case 5 and 17% in Case 7. This means that, by applying the peak shaving Fuel Cell and the Hydrogen based storage cycle (Cases 3, 5 and 7) we could increase renewable penetration in about 5-6%: I

Case 1 Wind 30%

Case 3 FCWlnd 30%

Case 5 FCSolarWind 30%

Case 7 FCWindTransport

Fig. 8. The demand supply per technology, in different cases of 30% intermittent limit, year 2010.

On comparing the generation technology chosen, we see that by installing a convenient power of photovoltaic panels we can reduce the Hydrogen storage to 115 (electrolyser

M Lever et al.

26

power also decreased, although not as significantly), while keeping the fuel cell at the same value, and still increase a little the overall renewable penetration. This is because the intermittence of the solar potential is smaller than the intermittence of wind potential, and so the storage is filled up more regularly and can cover the same supply by the end of the year. The increase in storage and electrolyser from Case 3 to Case 7 is related to the additional Hydrogen load for shuttles. The significant increase in hydrogen storage comes not from a great increase in hydrogen load, but from the increase in security of supply condition for transport (a reserve of 30 days of hydrogen for transport is considered as a minimum for allowing the hydrogen to be used for power system). From the point of view of power system, those cases are identical. As result from the introduction of the peak shaving fuel cell, the use and wear of the Diesel engines will be reduced, as can be seen (Fig. 9) from the Load Duration Curves for cases 1, 3 , 5 and 7. The curve for case 3 is very similar for those of cases 5 and 7, showing that there is not much difference when we introduce the solar panels and confirming that case 5 and 7 are approximately equal concerning electricity.

0

1000

2000

3000

4000

5000

6000

Diesel output [kWhlh]

Fig. 9. Load duration curves for diesel generators.

4.9. Results for 100% Hourly Penetration: Cases 2 , 4 and 6 Cases 2 , 4 and 6 differ in technologies installed, regarding both generation and possibility of storage of surplus energy, but they all allow for 100% penetration of intermittent renewable sources. Also, Cases 4 and 6 represent 100% Renewable Scenaria for electricity generation. Fig. 10 shows the equipment power and storage capacity that should be installed for each scenarium, in 2010:

Energy Planning for the Penetration ofRES

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 .~~~~~

1

~~.~~ . . . . . . . . . . . . . . . . . . . . . . . .

21

~ ~ . . . . . ~ . ~ 3000. . ~

2500

Case 2 Wind 100%

Case 4 FCWlnd 100%

2000

F

1000

t


Fig. 10. The comparison of components needed to be installed in different cases of 100% intermittent limit, year 2010.

It is clear, for 100% renewable scenaria, the storage size reduction thanks to PV introduction is not very important. This happens because, as the Hydrogen cycle must cover all unsupplied demand, the storage is not allowed to be empty at any time and the relevance of the smaller intermittence of solar resources is not as important as in the peak shaving cases. As seen in Fig. 11, in Case 2, when 100% intermittent limit is accepted but no storage is used, the electricity supplied by wind is about 50% on yearly basis. With storage based in Hydrogen Cycle (Cases 4 and 6) one manage to increase the renewable penetration to loo%, and virtually eliminate the use of Diesel blocks, that are kept as backup.

Care 2 Wind 100%

Case 4 FCWind 100%


Fig. 1 1. The demand supply per technology, in different cases of 100% intermittent limit, year 2010.

M Lerer et al.

28

It was shown how using hydrogen loop the renewable penetration could be increased to 100% in Porto Santo (cases 4 and 6). Since that would currently be too expensive a possible first step would be to install peak shaving systems (cases 3 and 5). Also, using hydrogen for transport would be straightforward, as seen in case 7. In order to have a reference case, a conservative scenarium for renewable energies penetration, as described in case 1, is based on the proven technology of wind farms, with up to 30% penetration on hourly base. In this case, and keeping the rejected renewable energy lower than lo%, more 5x300 kW wind turbines would have to be installed until 2010. These turbines could provide about 1I% of the electricity used in Porto Santo. Adding a peak shaving FC, fuelled on renewable Hydrogen, even if the annual load is of about 1% only, can increase the renewable penetration to about 16% of the supply, but this scenarium needs an infrastructure with electrolysers and H2 storage. The decision on including the still expensive PV panels together with the wind turbines should be weighted considering the reduction in cost for Hydrogen storage. Although the diversification of energy sources can be an advantage, because the overall intermittence is reduced and security of supply is increased, the decision must consider geographical and economical concerns as well as strategy arguments. In case we allow for up to 100% penetration of intermittent renewable energy on hourly basis, with no energy storage (case 2), about 50% of electricity could be provided by wind energy. The switch to 100% renewable (cases 4 and 6) corresponds to a high investment in generation (PV and wind turbines on one hand, electrolysers and FC on the other) and energy storage. Acknowledgements The authors would like to thank the European Comission for supporting the project in which this paper was based: “Renewable Energy Solutions for Islands RENEWISLANDS” Accompanying Measure of the European Commission’s Framework 5 ‘Energy, Environment and Sustainable Development’ Research Programme. Contract NO. NNE5-2002-00073. References [ l ] European Commission, Communication from the Commission - Energy for the Future: Renewable Sources of Energy, White Paper for a Community Strategy and ActionPlan, COM (97)599 final (2611 1/1997),

http://europa.eu.intfcommlenergy/library/599fi~en.pdf. [2] United Nations, Report of the Global Conference on the Sustainable Development of Small Island Developing States, NCONF. 16719, Bridgetown, Barbados, 1994, http://www.unep.ch/islands/dsidscnf.htm. [3] INSULA - UNESCO - European Commission - Consello Insular de Menorca, First European Conference on Sustainable Island Development, The Minorca Commitments, European Island Agenda, Ciutadella Declaration, Minorca, 1997.

Energy Planningfor the Penetration of RES

29

[4] Lund H, Munster E. (2003) Modelling of energy systems with a high percentage of CHP and wind power. Renewable Energy 28: 2179-2193. [51 HOMER, The optimisation model for distributed power. http://www.nrel.gov/homer/. [6] HYDROGEMS, HYDROGen Energy Models. http://www.hydrogems.no/. [7] Duic N, Carvalho MG. (2003) Increasing the penetration of intermittent renewable energy sources in island energy supply. CD, Proc. of the 2nd Dubrovnik

Conference on Sustainable Development of Energy, Water and Environment Systems, Dubrovnik. [S] Empresa de Electricidade da Madeira. http://www.eem.pt/. [9] Melim Mendes JM, Oliveira F, Freitas D, Henriques C , Olival E, Branco S. (2002) Plano de politica energetica da Regiao Autonoma da Madeira, Funchal. [lo] Lundsager P, Binder H, Clause NE, Frandsen S, Hansen LH, Hansen JC. (2001) Isolated systems with wind power. Main Report, Riso-R- 1256(EN), Riso National Laboratory, Roskilde, http://www.risoe.dklrispubWEA/veapdf/ris-r-1256.pdf. [ll] Milborrow DJ. (2001) Wind and storage - and a look at Regenesys. "Windpower on Islands" Conference, Gotland, 12-14 September. Gotland University. [ 121 Milborrow DJ. (2002) Assimilating wind. IEE Review 48. [ 131 Milborrow DJ. Penalties for intermittent sources of energy. http://www.cabinetoffice.gov.uWinnovationl;!002/energy/report/working%2Opapers/Milborrow .pdf. [ 141 Stavrakakis G. (2000) Conclusions, Workshop: Dissemination of the advanced control technologies and SCADA systems for the isolated power networks with increased use of renewable energies. Ajaccio, Corsica, http://power.inescn.ptljaneca/pdfs/CorsicaConclusions.pdf. [15] Altmann M, Niebauer P, Pschorr-Schoberer E, Zittel W. (2000) WHySE WindHydrogen Supply of Electricity, Markets - Technology - Economics, Wind Power for the 2 1st Century, Kassel, Germany. [ 161 Putnam R. (1996) How difficult is it to integrate wind turbines with utilities? Wind Energy Weekly 680.


THE SHAPE OF COMPLETE RENEWABLE ENERGY SYSTEMS

IN THE WORLD H. YAMAMOTO (I), K. YAMAJI(*) (I)

Socioeconomic Research Center, Central Research Institute of Electric Power Industry (CRIEPI), Tokyo, Japan. "School of Engineering, The University of Tokyo, Tokyo, Japan

The uses of fossil fuels cause not only the resources exhaustion but also global warming. Before they exhaust all the fossil fuels or fall into catastrophic climate change, they need to develop the new energy systems that are fossil-fuel-free and complete renewable. In the study, the authors developed a global land use and energy model including new technological concepts such as intermittent renewables (photovoltaic and wind) directly connected with electricity storage of small capacity and large capacity. Using the model, the authors conducted simulations and obtained the following results. In the sustainable scenario in 2060 (when they achieve zero fossils and completely renewable energy systems) the dominant resource of primary energy supply will be photovoltaic directly connected with electrolysis at the amount of 455 EJ.yeai' in the world. The second and the third largest resources are biomass and photovoltaic directly connected with large capacity storage at 322 EJ.yeai' and 118 EJ.yeai', respectively. At that time, the dominant energy resource of the power generation connected with the electricity grid will be the photovoltaic directly connected with large capacity storage at 118 EJ.year-l in the world. BGCC (biomass gasifier combined cycle), hydropower, and photovoltaic directly connected with small capacity storage are above 40 EJ.yeai' each in the world. However, the energy system costs in the sustainable energy scenario will be at 18 trillion US$.year-l and be almost three times as expensive as the costs in the base case (without constraints of fossil fuels or C 0 2 emissions). The sustainable energy systems are the expensive energy systems.

1. Introduction

The uses of fossil fuels cause not only the resources exhaustion but also environmental problems such as global warming. Before they exhaust all the fossil fuels or fall into catastrophic climate change, they need to reach the sustainable energy systems that are fossil-fuel-free and completely renewable. However, in the past studies the clear shape of the sustainable energy systems was not found. Even in the cases of high carbon taxes the authors could not find the sustainable energy path in the past study [ 11. In order to find the shape of the sustainable energy systems, the authors developed a global land use and energy model (GLUE) [2-31. The model figures the global energy supply systems in the future considering the cost minimization. The model includes overall energy resources including fossil fuels and renewables and overall energy conversion technologies including a variety of technologies concerning power generation, gasifier, and liquefaction. In the study, the authors developed the new version of the model (GLUE 3.0). It is based on a new calculation method of foresighted dynamic recursive. In addition it includes new technological concepts of intermittent renewables (photovoltaic and wind) directly connected with electricity storage of small capacity and large capacity. The authors assume a sustainable energy scenario that they will set a target in 2030 when they will reach no fossil fuel and complete renewable energy systems by 2060 (in 30 years after 2030). The authors conduct simulations under not only the sustainable energy scenario but also a base scenario (that is without constraints on fossil fuel uses or 31

32

H. Yamamoto & K. Yamaji

C 0 2 emissions). Then the authors discuss the shape and the cost of the sustainable energy systems.

2. The Model of Glue 3.0 2.1. Outline of the Model In this section the authors outline the structure of Global Land Use and Energy model version 3.0 (GLUE3.0) model. The world is divided into 11 regions in the model (Table 1). The model can conduct simulations from 2010 to 2200 with ten-year time step. The authors show the simulation results until 2060 in the study. The model consists of two parts: an energy systems part and a land use part. The energy systems part is based on a global energy systems model named New Earth 21 (NE21) and the land use part is base on a global land use and energy model (GLUE-11) (Fig. 1) [2][3]. The land use part covers a wide range of land uses and biomass flow including food chains, material recycling, and discharge of biomass residues. Those two parts are connected through common variables concerning supply potential of bioenergy and photovoltaic. The model minimizes the summation of the energy system costs. The authors prepare data set for the model using the data of FAO, IPCC, World Bank, DOE, and so on. For example, the final energy demand will increase following IPCC SRES-B2 scenario. Most input data are explained in the reference [2]. 2.2. Renewables in the Model

In the model the authors considers the following renewables: bioenergy (including fourteen kinds of bioenergy), hydropower and geothermal power (hydropower is the representative and includes geothermal power in the study), wind power, photovoltaic, and innovative renewables (such as space power system). Intermittent renewables (such as wind power and photovoltaic) are divided into four kinds such as type 1 (without electricity storage), type 2 (directly connected with electricity storage of small capacity), Table 1. Regions in the model.

CO2 balance in land use part (wood biomass

-.

.

I

CO2 balance in ene

I

Energy part

Land use part

pasture, other land, addtiional arable

each primary energy (resources and wsts; except bioenergy)

wnversion

slash-and-burning, (biomass pmduction per

(mundwocd to timber, cerealto meat, etc.)

(population,

technologies (ind efficiency and

secondary

demands per

TI

lntrcduction

energy. (including bioenergy)

of each sewndary energy

r

'

t Determination of biomass flow

~upp~ycurve of each bioenergy -+

Discharge of chemical product scrap Determination of energy flow

34

H. Yumumoto & K. Yumaji

type 3 (directly connected with electricity storage of large capacity), and type 4 (directly connected with electrolysis and not connected with electricity grids). It is assumed that type 2 improves stability of the outputs of the renewables and eases the upper limits of introduction of the intermittent renewables in the electricity grid compared with type 1. Type 2 is being developed in Japan funded by NED0 (New Energy and Industrial Technology Development Organization) [2]. It is assumed that type 3 is similar to dammed hydropower that can supply electricity to the grid following the demands. Therefore type 3 is a kind of stable energy and is not suffered the constraint of the electricity grid stability. The authors consider that type 3 will be realized when the price of the capacity (in kWh) is reasonable. Type 4 is not connected with electricity grids and is not suffered the constraint of the electricity grid stability. 2.3. Sustainable Energy Scenario and Base Scenario The authors assume a sustainable energy scenario that they will set a target in 2030 when they will use no fossil fuel by 2060 (in 30 years after 2030). In the other words, they can use only renewables (such as hydropower, wind power, photovoltaic, and innovative renewables (that the authors assume at Space Power System) in 2060. Under the assumption of resource estimation of WEC and an energy demand scenario of IPCC SRES B2, they will exhaust reserves of conventional oil conventional natural gas until 2060 (Figs. 2 and 3)[2]. In 2060 there will be plenty of reserves of the other energy resources (such as unconventional oil, unconventional natural gas, and coal). However the depletion of conventional oil and conventional natural gas will have human reconsider the relationship between human and exhaustible energy resources. It is assumed that they will be able to use intermittent renewables with electricity storage or electrolysis (see the previous subsection). It is assumed that they will adopt nuclear (LWR) phase-out where they will introduce LWR following the LWR scenario at IAEA by 2020 and will phase out the capacity after 2020 and abolish it in 2060 [2]. In the study the authors conducts not only the sustainable scenario but also a base scenario. The base scenario is the scenario without constraints about fossil energy uses and C 0 2 emissions. The authors discuss the results such as the structures and the costs of the energy systems concerning the two scenarios. 3. Simulation Results

Using the model of GLUE, the author conducted simulations of two scenarios (that are the sustainable energy scenario and the base scenario). The authors found feasible solutions even in the sustainable energy scenario. The authors explain the simulation results of primary energy supply, electric power supply, and energy trade.

The Shape of Complete Renewable Energy Systems in the World

35

3.1. Primary Energy Supply In the base scenario the dominant resources of primary energy and power generation are both coal in the world.

-+Cumulative consumption 20000 18000 16000 14000 12000 10000 8000 6000 4000

,

,

40

35 30 25 % 0 20 .= 15 10 2 h

-

5 0

2000

I

45

1970 2000 2030 2060 Fig. 2. Proven reserve and RIP ratios of conventional oil.

+Cumulative consumption FUP ratio(left-hand axis)

a,

2 c

-

g %.z 0 0

k

20000 18000 16000 4000 2000 0000 8000 6000 4000

70

I

60 50

2

30

'z

20

$

m a, 40 2 0 L

10

2000 0

0 l J l U

LUUU

LUJU

LUUU

Fig. 3. Proven reserve and RIP ratios of conventional gas.

36

H. Yamamoto & K. Yamaji

In the world in 2060 in the sustainable energy scenario the largest primary energy resource will be photovoltaic type 4 (that is photovoltaic directly connected with electrolysis and supplies hydrogen) at 455 EJ.year-'. The second and the third largest resources are bioenergy at 322 EJ.year-' and photovoltaic type 3 (that is photovoltaic directly connected with electricity storage of large capacity) at 118 EJ.year-', respectively (Fig. 4).

I2Oo

1 w Chemical products scrap w Bioenergy w SPS w PV and electrolysis w PV(with large capacity storage)

1000

i i 5

3

800

rn PV(with small capacity storage)

=-a. -

Y

d>0.125 mm 3. S: Commercial Swedish Wood powder: 2>d>0.063 111111,dmean=OSmm Mean size distributions of the fuels were determined using sieves of different sizes. Each fuel has somewhat different size distribution, resulting in slightly different operation of the gasifier. For instance this gives a slightly unstable start-up condition for Fuel A when the fuel is first introduced to the rig and the gas is turned off. This is most likely due to the finer particles are initially needed to quickly provide the combustion heat for stable start-up operation. With the absence of finer particles more time is needed to change the wall temperature patterns of the cyclone from preheat to gasification processes. Experimentally a wide range of wood powder feed rates and equivalence ratios were used, resulting in a range of exhaust gas temperatures. All tests resulted in stable steady conditions. Detailed investigations of four conditions for typical conditions are shown in the following section. Table 4 below shows the gasification conditions investigated. Table 4. Gasification conditions.

Trial No Fuel Type Total Air Flow Rate [l/min]

Wood Fuel Feed Rate [kg/h] Equivalence Ratio [-3

a

b

C

d

Fuel A

Fuel A

Fuel S

Fuel S

550

660

550

660

26

31

26

40

0.243

0.205

0.243

0.190

A large number of trials were carried out with fuel A and S at different equivalence ratios to characterise the cyclone gasifier, exhaust temperatures are shown in Fig. 6 as a lowest exhaust gas temperature achieved at the smallest equivalence ratio). Exhaust gas temperatures above 1OOO°C do not represent realistic gasification conditions, regardless of the mixture ratio. Thus the lowest equivalence ratio at any feed rate is selected for further investigation. Exhaust gas temperatures were generally slightly higher for fuel A but fluctuations were higher with fuel S. This was due to fuel S being composed of smaller particles that are more reactive and hence release heat for the gasification reaction more quickly. The lowest equivalence ratios were obtained for higher fuel feed rates (Fig. 6). Wall temperatures increased as the equivalence ratio was increased. This effect was more pronounced at lower fuel feed rates and not present at all for fuel S at 40 kgh. The

139

Use of Biomass in Small Direct Fired Systems

inlet and exhaust sidewall temperature distribution are different with cyclic variations of about 50°C. However, local fluctuations of wall temperatures were very low. 1300 1250

E

I2O0

@ 1150 J

E

1100 1050 1000

f

950

f

900

U 26 kg/h, Fuel S

850 800 500

550

650

600

750

700

800

850

900

mr Flow Rate [llrninl 1300

oe-

1250 1200

@ 1150

=

1100

n 1050 u)

2

1000

c

3 2

950 900

t 4 0 kg/h, Fuel S

+26 kglh. Fuel A

850

t37 kg/h. Fuel A 800 0.15

0.17

0.19

0.21

0.23

0.25

0.27

Equivalence Ratio

0.29

0.31

0.33

0.35

[-1

Fig. 6 . Exhaust gas temperatures as a function of airflow and equivalence ratio.

Detailed experimental characterisation of the flow pattern within the cyclone gasification chamber was undertaken by Fick et al. [19-211. Two dimensional LDA measuring techniques were used. The isothermal and combusting flow patterns were shown to be generally independent of Reynolds and swirl number, however, negative or positive peak values were affected. It was found that regardless of the inlet flow rate, minimum losses of tangential velocity at the cyclone inlet and the best conservation of tangential velocity in the main cyclone gasification chamber were obtained for a geometrical swirl number of 5.1. The LDA data for axial and tangential velocity distributions measured for isothermal and combusting (natural gas) conditions was used as guides for the velocity distribution to be expected for the relevant gasification cases, Figs. 7 a) and b). Fig. 7 c) shows the measured temperature distribution under combustion conditions (the system was probed with fine wire thermocouples). An air flow rate of 1900 I/min and a gas flow rate of 100 l/min (mixture ratio-2) were chosen to simulate the

140

C. SyTed et al.

conditions for the gasification trials at realistic wall temperatures of around 850OC.With a Reynolds n ~ m b of e ~30055 in the cornbusting case (Fig. 12 c)) and an inlet velocity of 6.17 ads (for a total inlet Row rate of2000 Vhh-i),the Reynolds number for the relevant ~ a s ~ ~ c aconditions ~ i o n at about 650 llxnin decreases to 12755 based QPI an inlet velocity of 2 m/s for Fig. 4 a) and b). Fig. 7 a) and b) highlight the axial and tangential velocity d i s ~ ~ ~fort these ~ Q ~~ ~s ~ d i ~ i Q ~ s .

Fig. 7. Derived mean a) axial, b) tangential velocities for gasification, c) tempera^^.

141


3.3. Energy Balance An energy balance was performed for the four test cases, shown in Table 5 below. Table 5 . Summary of energy balance.

Exhaust gas velocity hot [ d s ]

7.80

9.44

7.88

9.60

Cal. Value exhaust gas [MJ/m3]

0.78

2.99

4.26

5.91

36

59

70

76

Efficiency [%]

Highest calorific value was achieved with trial d), including (by volume). 52% of nitrogen, 13% of carbon dioxide and remaining products, methane, carbon monoxide and hydrogen (35%). Losses for trial d) are lowest at 24%, followed by c) at 30%, then b) at 41% and a) at 64%. The higher calorific values and lower equivalence ratios for the fuel S result in lower losses. At a lower equivalence ratio condition b) could certainly reach higher efficiencies. However conditions a) and c) are approaching the lower operating limit of the gasifier wall heat loses are excessive. Trial c) represents the highest efficiency possible for such a low load condition. Table 6 shows the exhaust gas composition for the four cases.

3.4. ElementaVProximate Analysis: Separation Performance The separation of ash from the flow is very important to minimise damage to the turbine and additional hot gas clean up for use of the gas in the gas turbine. The VCP and CCP were designed to remove these particles from the flow as described earlier. A whole range of different tests were undertaken to investigate the overall separation performance of the gasifier as well as the collection efficiency for individual elements. To determine separatiodretention performance of the VCP and CCP a reference source to the ash was produced by forming 100% burnout ash from each fuel. This was done by burning a set amount of fuel in a stainless steel pot. The size of the 100% burnout ash for fuel A was typically 3 times smaller than the fuel and 5 times smaller for fuel S. The ash collected in the VCP and CCP was generally larger than the 100% burnout ash. For case d the CCP residual is of very similar size to the fuel with some 10% of the particles bigger than

142

C. Syred et al. Table 6 . Exhaust gas composition.

2mm. As fuel S contains no particles larger than 2 mm there is clear indication that agglomeration of the particles is occurring. The axial and tangential velocity distribution (Figs. 7 a) and b)) can explain this occurrence. Fuel particles move upstream and gasify close to the cyclone wall from r/De=0.7 to r/De=l.O. The particles not being separated by the VCP situated at r/De=l.O and z/De=3.33 join the downstream flow which leads into a reverse floe region from r/De=O to r/De=0.5 and z/De=1.53 to z/De=1.00. With low tangential velocities also in this region, low shear stresses are present assisting agglomeration in this region. This effect is however beneficial to the separation performance since the larger particles are more likely to be separated by the CCP collector pocket situated further downstream. Although particles collected in the VCP are generally smaller than the CCP ones, they are typically 30% to 70% heavier. This is because the centrifugal force field in the VCP region of the flow had tangential velocities identical to the inlet air velocity; the heavier particles are more likely to be separated. The remaining lighter particles in the flow are then separated in the CCP although heavier particles not captured by the VCP are again more likely to be separated. The VCP and CCP collected different percentage of the retained ash for each separate trial, highlighting the importance of both VCP and CCP collector pockets to provide good separation performance for a wide range of fuels and operating conditions. The overall gasification performance of the rig was characterised, with carbon burnout of 98/99% burnout. The collection efficiency for Al, Ca, K, Mg, Mn, Na, P and Si with respect to the input of those elements evaluated from the elemental analysis of 100%

143


burnout ash for fuel A and fuel S is shown in Fig. 8. The results are presented independent of the fuel feed content, i.e. 100% burnout ash content. Fig. 8 shows that, regardless of the air flow rate and the fuel load, there are trends only dependent on the fuel. For fuel S, trials c & d, typically 50% of all elements are separated with the VCP and CCP ash. For fuel A, trial a & b, the trend is also very consistent but with significant differences in separation performance between different elements. Only 10% of A1 was removed compared to well above 90% of Mn, doubtless due to volatilisation of Al. Large differences in separation performances were seen between the two fuels for some of the elements and it is clear that the separation rate is much more dependent on fuel type than any other operating condition. The retention rates of Alkali’s, Na and K, however were independent of the fuel type and identical for the moIre relevani: cases, trials b & d, with around 50/60% retention of Na and K. 100 90

F-

80 70

0

$

60

s 2 so I

5 ._

i 5 2-

s

40 30

20 10

0

A1

Ca

K

Mg

Mn

Na

P

Si

Element

Fig. 8. Collection efficiency for Al, Ca, K, Mg, Mn, Na, P and Si in VCP & CCP (%wt. of fuel feed)

As carbon burnout and the ash separation performance are generally very high, low particulate emissions in the exhaust stream can be expected. This was only the case however for the more relevant operating conditions (at lower equivalence ratios) trials b & d (here exhaust particulate emissions were down to 0.044 g/s, 0.4% of the inlet fuel, with between 70 to 80% of ash and unburnt char being collected). High particulate emissions were obtained for trials a & c, presumably due to the lower flow rates, hence velocities and cyclonic separation. The tar content of the particulate emissions in the exhaust gas was very low, less than 5% for trial b & d, indicating very good carbon burnout throughout the system. Very good separation and gasification performance was obtained with the inverted cyclone gasifier under atmospheric conditions at higher fuel feed rates. The lower fuel load 26 kg/h was found to be far too low for this test rig. The inlet velocities of 2 d s or

144

C. Syred et al.

less for the proposed conditions were on the lower operating limit of the test rig. Higher fuel feed rates and subsequently air flow rates/ inlet velocities are certainly possible and higher turn down ratios of 5 or even higher can certainly be achieved. This highlights the robustness of the design with clear benefits. 4. Cyclone Combustor

A secondary combustor was developed for integration into the small-scale biomass cogeneration process describe above. This combustor should be capable of running on the gasifiers’ exhaust gas whilst still be able to utilise natural gas or oil for start up and shut down procedures or as a pilot fuel during certain operational conditions. To minimise pressure loss across the system and maximise efficiency the use of complex hot gas clean up system should be unnecessary, due to the separation capability of the gasifier. The combustor will be indirectly fired to the gas turbine. Present designs of combustor that can run on multiple fuels are either direct fired or involve complex gas clean up systems. All these designs are essentially derived from conventional gas turbine combustor systems fired on conventional liquid fuels or natural gas. They are all designed to be tired on cleaned bio-gas. This arises because the turbines have sophisticated blades incorporating numerous fine cooling passages susceptible to blockage. Conversely the present combustor address a different problem involved with small scale power systems. Here gas turbine systems are generally of simpler construction with un-cooled turbine blades and can sustain modest levels of fine particulates less than 5 microns in size. Turbine inlet temperatures are up to 900°C. The combustor for this process can be more robust and is indirectly fired. The combustor should be capable of removing any particles remaining above 5 pm. 4.1. CombustorDesign

The cyclone type combustor is designed with several tangential inlets as shown in Fig. 9, an air inlet, a high CV fuel inlet for oil or natural gas and a low CV gas inlet. The tangential inlets create a strongly swirling flow that gives good mixing and burn out rates. The combustor is to be operated at a maximum thermal input of 550kW. The combustor is mounted vertically and has a cone section at the base to collect larger particles in the flow. The combustor is designed with a long chamber to allow the flame to move up and down with varying thermal input and quality whilst giving sufficient residence time for fuel burnout and thus low emissions of CO and volatile hydro carbons. The central section of the combustor is refractory lined allowing substantial heat storage capacity helping to create stable flames. There is a tangential off take on the combustor that recovers energy from the flow and also forces the exhausting flow tangentially across the VCP aperture, hence increasing separation capability. The combustor is to be used to fire a small gas turbine operating at an inlet temperature of 800°C. The exhaust gas of the combustor has higher temperatures than this, and is diluted by a co-flowing air stream that is passed through a jacket surrounding the combustor.

145

IJse of Biomass in Small Direct Fired Systems

This co-flow air lowers the temperawe of the exhaust gas and acts as a difhser to the Blow, lowering the pressure drop across the combustor. The VCP is placed just before the tan~ent~ off-take, a~ which removes any fme particle above 5 microns that could damage the mbine. Tbe VCP also removes the need for a cyclone separator to remove the particles, which would increase the pressure drop across the system further.

Tangential exit

rc’

Oil lniet

\e.

Air inlet 2 Air Inlet1

LCV vrood gas inlet

Fig. 9. Schematic ofthc cyclone combustor with 4 inlets at the bottom, a VCP to collect particles at the top and a tangential offtake to reduce pressure drop.

4.2 Combustor Flow Characteristics A cyclone type combustor was built with several tangential inlets, for ak, oil and gasifier LCV gas. The tangential inlets produced a swirling flow with good mixing, creating a stable flame. Fig. 18 shows the swirling flame inside the combustion chamber, (gas oil as fuel) with the top plate removed.

Fig. 10. Photo of swirling flame inside combustion chamber.

C. Syred et aL

146

4 3

cm ~0~~~~~~~

The omb bust or was initially modelled using the CFD package Fluent 4 , under atmospheric conditions but with velocities representative of those occurring under pressurised gas turbine operating conditions.‘ An unstructured t ~ e ~ ” d i m e n s i ogrid n a ~was applied for this analysis. The Reynolds Stress turbulence model was used as the combustor has strong swirling flow. Experience has shown that this model predicts swirling flow with s ~ ~ i f ~ c a nmore t l y accuracy thm other models. The LC’V gas species and gas oil were modelled using the mixture fractiodpdf approach thus a s s ~ i n g chemical equilibrium. The non-adia~dticpdf model was used as radiative heat transfer to wall boundaries would have to be considered. The local ~ e ~ o - c h e m i cstate a l is also related to entbalpy as well as the mixture fraction. The D T W radiation model was used t ~ o u g ~ o the u t cal~u~ation. Particles (assumed to be spherical) were i ~ ~ o d ~ into c e dthe LCV gas stream using the discrete phase function. This enables the particles to interact with the flow md predict particle trajectories, hence separation. The c o ~ ~ u s t was or modelled using the gas composition from the cyclone gasifier, trial b, Tables 5 and 6 . Two ~ a n ~ ~ ninlets ~ i atol the combustor were used, the LCV gas inlet and an air inlet. The combustor was designed with multiple inlets to allow high to low CV gas to be tested with two inlets used for the modelling described below, a LCV gas inlet and an air inlets. The combustor was modelled at 3Q% excess air to allow good mixing, produce a clean exhaust gas with low emissions and to allow wall temperatures to stay within accepkable limits. CFD results show using a typical LCV wood gas the combustor produced a strongly swirling flame. The vortex structure of the flow field caused an extended residence time producing complete combustion of the &el. Experimental tests demonstrated the swirling flame in Fig. 10. Velocity contours across the combustor are shown in Fig. 11. The LCV gas and air enters the combustor ~ ~ e n tthrough i a ~their ~ ~ respective inlets. Velocity contours taken at a plane through the inlets, Fig. 12 a), show high tangentiai velocity at the inlets and close to the wall.

Fig. 1 1 . Velocity contours across combustor d s .

Use qfRiomass in Small Direct Fired Systems

147

Tangential velocities decrease towards the centre md thus the flow is o f forced vortex form. Areas of high tangential velocity near the wall correspond to regions of high t e ~ p e r a ~ ~Fig. e s ,12 b). The flow has good mixing and produces a stable swirling flame. The majority of the swirling flow stays near the chamber walls. As the flow moves downstream. the velocity of the flow increases in the centre, indicating the presence of complex r e c i r c ~ ~ apatterns ~ i o ~ and secondary flows. The LCV inlet is designed with a relatively small area to increase mixing with the air. The tangential exit produces a near ~ i velocity ~ profile o thus ~creating more stable inlet c o n d ~ t ~ ofor n ~the gas turbine.

Fig. 12. a) Velocities and b) temperatwe contours thro~ghplane o f combustor inlet.

~ e ~ p e r contours, a ~ e Fig. 13, show that combustion is initiated in the wall region near the inlets and then spreads across the whole top 50% of the unit, p r e s ~ a b l ydue to the secondary flow effects. The exit temperature profile is quite uniform and when mixed with dilution air will form a good flow to fire into the turbine. Average exit temperatures are around 1350 K. Minimising pressure drop across a system is important, especially in small-scale plants, a level not exceeding 0.2 bar is needed here. Fig. 14 shows the pressure drop across the combustor is typicaliy less than 0.025 bar. This could M e r be reduced by placing a 4" difkser on the tangential off-take as would d i ~ ~ the ~ nhaus g us^ gas with co-flowing air, to lower temperatures suitable for the turbine inlet. Particles were introduced into the flow through the LCV gas inlet. Fig. 15 shows the tracks of 5 microns particles through the combustor. The VCP removes the particles -firom the flow before they exit in the exhaust. Thus the p e ~ o ~ a n of c e the VCP is adequate to remove particles &om the flow without the addition use of cyclone separators d o ~ s t r of e ~the combustor. The VCP acts by collecting particles from the wall b o ~ layer d ~where they are concentrated by the centrifugal force field, they pass though the aperture between the VCP and the main cyclone chamber into the VCP where they are slowed by fiction and drop to the bottom for collection.

148

.........

Fig. 13. Temperature profiles across combustor inlet and outlet.

"500 diameter (ma) Fig. 14. Pressure drop WFOSS the combustor


149

Fig. 15. Particle tracks of 5 micron particles through combustor and into VCP

5. Conclusions

The inverted cyclone gasifier developed by Cardiff University is suitable for gasifying biomass under a range of conditions with varying equivalence ratio between 0.19 and 0.24. The design is robust with good mixing and 98-99% burnout rate. The VCP and CCP removed particles from the flow, with high separationhetention rates. Up to 50% of the Alkali’s Na and K were removed from the flow. A good quality LCV gas was produced with maximum calorific value of 5.91 MJ/m3. The LCV gas produced is suitable for firing directly into the secondary combustor. The combustor produced a strongly swirling flame with good mixing and burnout. An area of secondary flow and recirculation occurred where combustion was enhanced. The combustion patterns showed good mixing with near full burnout and hence low CO and hydrocarbon emissions. Velocity and temperature profiles at the combustor exhaust were nearly uniform, forming stable conditions for the turbine inlet. The VCP positioned just before the combustor exit effectively removed particles above 5pm from the flow. The system produced gas suitable for indirect firing of the gas turbine. The pressure drop across the system was relatively low at a predicted maximum level of 0.2 bar.

Acknowledgements The authors gratefully acknowledge the support of the European Union in this work, both through the Euroflam Scheme and most directly through contract number JORC398081-

150

C. Syred et al.

start date 01/06/98. The work would not have been possible without this support. The support of the UK EPSRC is also gratefully acknowledged. Professor Syred gratefully acknowledges the award of a Royal Academy of Engineering Global Research Award.

References [ 11 Lees I. (2001) Idris Jones Lecture. Cardiff Castle, Wales, UK. [2] Loram G. (2001) Turning straw into power at Ely. Energy World286: 14-15. [3] Pilavachi PA. (2000) Power generation with gas turbine systems and combined heat and power. Applied Thermal Energy 20 (15-16): 1421-1429. [4] Reed TB. (1985) Principles and technology of gasification. Advances in Solar Energy: An annual review of research and development 2: 125-174. [5] Engstrom F. (1998) Hot gas clean-up bioflow ceramic filter experience. Biomass and Bioenevgy 15 (3): 259-262. 161 Wilkes C. (1999) Sulphur deposition in a gas turbine natural gas fuel control system. Joint Power Generation Conference FACT 23 (1). [7] Leyens C, Wright IG, Pint BA. (2000) Hot corrosion of an EB-PVD thermal-barrier coating system at 950°C. Oxidation ofMetals 54 (516). 181 Simms NJ, Oakey JE, Nicholls JR. (2000) Development and application of a methodology for the measurement of corrosion and erosion damage in laboratory, burner rig and plant environments. Materials at High Temperatures 17 (2): 355362. 191 Neilson C. (1998) Gas turbine modifications for biomass fuel operation. Biomass and Bioenergy 15 (3): 269-273. [lo] Williams RH, Larson ED. (1996) Biomass gasifier gas turbine power generating technology. Biomass and Bioenergy 10 (2-3): 149-166. [ l I] Salo K, Mojtahedi W. (1998) Fate of alkali and trace metals in biomass gasification. Biomass and Bioenergy 15 (3): 263-267. 1121 Fredriksson C. (1999) Exploratory experimental and theoretical studies of cyclone gasification of wood powder. PhD Thesis, Lulea University of Technology, Lulea, Sweden. [13] Gabra M, Nordin A, Ohman M, Kjellstrom B. (2001) Alkali retentiodseparation during bagasse gasification: A comparison between a fluidised bed and a cyclone gasifier. Biomass and Bioenergy 21 : 46 1-476. [14] Hamrick JT. (1991) Development of wood as an alternative fuel for gas turbines. Battelle Memorial Institute. Report PNL-7673iUC-245. [15] Biffin M, Syred N. (1985) Vortex collector pockets to enhance dust separation in a gas cyclone. Filtration and Separation 22 (6): 365-372. [16] Biffin M. (1984) Improved cyclone dust separators for hot gas clean up. PhD Thesis, University of Wales, Cardiff, UK. [17] Syred N, Biffin M, Dolbear S, Wright M, Sage P. (1986) Evolution of new concept for compact cyclone dust separators. Proc. Con$ on Gas Cleaning at High Temperatures EFCE, Event No 340, Inst. of Chemical Engineering: 17-30.


151

[ 181 Morgan DJ. (1990) Characteristics of non slagging cyclone combustors for solid fuels. PhD Thesis, University of Wales, Cardiff, UK. [19] Fick W, Griffiths AJ, Syred N. (1999) Cyclone gasification of pulverised biomass for operation of gas turbines in cogeneration plants. Isothermal LDA Velocity Measurements, Report no. 2435, Cardiff School of Engineering, Cardiff University, UK. [20] Fick W, Griffiths AJ, Syred N. (1999 b), Cyclone gasification of pulverised biomass for operation of gas turbines in cogeneration plants. LDA, Temperature, Pressure and Isothermal PIV, Report no. 2528, Cardiff School of Engineering, Cardiff University, UK. [21] Fraser T. (2004) CFD modelling of an inverted cyclone gasifier. PhD Thesis, Cardiff University, UK.


EMISSIONS REDUCTION BY CO-FIRING BIOMASS OR WASTE WITH COAL IN A PRESSURIZED FLUIDISED BED COMBUSTION COMBINED CYCLE POWER PLANT Y. HUANG, J.T. MCMULLAN, D. MCILVEEN-WRIGHT, S. MCCAHEY NICERT, University of Ulster at Jordanstown, Newtownabbey BT37 OQB, Northern Ireland, UK One potentially attractive power generation option to reduce harmful emissions and to reduce the waste disposal problem is co-firing biomass or waste with coal in the pressurised fluidised bed combustion (PFBC) system. In this study, mixtures of coal and biomass or waste, such as straw, wood chips and sewage sludge are fired in the PFBC, and the resulting gases and sensible heat are applied in a combined cycle. In order to investigate the effects of cofiring biomass or waste with coal on the performance of the PFBC system from operational, technical, economical and environmental considerations, detailed simulations were carried out with biomass or waste additions of up to 40% of the total thermal input. The results showed that the main parameters affecting the overall efficiency were the co-firing ratio and the biomass and waste properties. It was also found that the steam cycle output was more sensitive to the fuel used than the gas turbine cycle. As expected, the increased fraction of biomass or waste significantly reduced net COz emissions, and had a beneficial influence on SO, emissions. Co-firing wood chips resulted in lower NO, emissions and co-firing straw or sewage sludge slightly increased NO, emissions.

1. Introduction

Combustion of fossil fuels will remain the primary energy source in the world for the foreseeable future. It is also recognised that the conversion of fossil fuels into energy for electricity generation unavoidably leads to many ecological problems. The major pollutants formed in the combustion process are considered to be COz, SO, and NO,. One of the more promising methods for reducing the environmentalimpact of power generation in the future is the co-firing of biomass or wastes with coal in a PFBC [l], in a combined cycle configuration. The benefits of using this technology are related to both the reduction of environmental emissions and possibly the reduction of a waste disposal problem [2]. Unlike coal, biomass and wastes such as sewage sludge, wood chips and agricultural wastes in general contain a very small amount of sulphur. Hence, adding biomass or waste to the fuel stream in a coal-fired power plant can result in a proportional reduction in SO, emissions. In addition, replacing fossil fuel by carbon dioxide neutral biomass (that is absorbed during plant growth) is also an approach to decrease C02 emissions. Although the environmental advantages of biomass wastes are well known, the use of biomass wastes as a fuel source is still primarily restricted to small scale heat and power generation [3]. The problems associatedwith biomass or waste application in large capacity power plant are due to low quality feedstock, i.e., high moisture content, low energy density, impurities and low calorific value, resulting in handling and operating difficulties. However, as a mean of reducing coal consumption and improving environmentalperformance,co-firing of biomass or waste with coal can reduce the impact of these problems. The main purpose of this paper is to investigate the impact of variations in the cofiring ratio of biomass or waste with coal on the technical and environmentalperformance of an advanced PFBC system and to determine whether the biomass or waste addition had any 153

154

Y. Huangetal.

noticeable influence on combined cycle operations. To achieve these objectives, the approach to conducting case studies is as follows. A reference coal and three biomass or waste fuels were selected. The fundamental process modules making up the PFBC combined cycles were modeled and the necessary information for the process was collected including the properties of coal and biomass or waste. Following the modeling of the PFBC combined cycles and definition of the design parameters, numerical simulation of the process and a range of sensitivity studies were carried out, as described later, using the ECLIPSE process simulator [4-51. 2. The Pfbc Power Plant

The PFBC system selected here is shown in Fig. 1. It is based on a commercially available P800 module, which is the most recent version developed by ABB Carbon [6-71. Accordingly, the boiler system is designed to match the required steam conditions of 241 bar and 565 "C at a full load flow of 720 tonnes per hour with steam reheating to 48 bar, 593 "C. The steam turbine condenser pressure is 0.041 bar. The overall electric power output of the plant is expected to be about 360 MWe. Raw coal blended with limestone is fed to the preparation plant and crushed to a correct size distribution. Limestone sorbent is used if necessary, to allow for the relatively high sulphur fuels. The mixture is converted to a paste by mixing with water, normally maintaining 25 percent water by weight [8]. This paste is pumped into the pressurized fluidised bed combustor. Air is compressed to approximately 16 bar by a two stage compressor which is powered by the gas turbine driven by the hot flue gas. The outlet from the low pressure stage passes through an intercooler, to limit the outlet temperature from the high pressure compressor to 300 "C. The air from the compressor passes through the outer annulus of a coaxial aidflue gas pipe before it enters the combustor. This arrangement cools the wall of the inner pipe, which carries the flue gas to the gas turbine. Inside the combustor vessel, the air flow provides the oxygen for the co-firing of coal and biomass and enhances thermal contact between the gas and the solids. The fuel paste is burnt directly under pressure in the fluidised bed. The operating temperature of the bed is around 855°C. This temperature is well above the minimum coal and biomass combustion temperature and provides sufficient margin to prevent melting of the ash. It also minimises the products of another potential pollutant, nitrogen oxide (NO,) generated from either the combustion air or the fuel-bound nitrogen in the fuel. The sulphur compounds which are produced from the relatively high sulphur content of the coal during the combustion process are captured by direct sulphation of the calcium carbonate in the sorbent material, which is mixed with the coal in the fluidized bed and thus eliminates the need for additional flue gas desulphurisation (FGD) scrubbing system. The steam which is generated and superheated in tubes immersed in the fluidised bed is sent to the turbine stop valve. It is then expanded through the high pressure turbine which also drives a generator. The steam turbines have facilities for steam extraclion and allow for

Emissions Reduction by Co-Firing Biomass or Waste with Coal

155

transfer of steam to the feedwater heaters. The steam from the high pressure turbine is reheated before passing through an intermediate pressure and a double flow low pressure turbine. The hot flue gases are expanded in a gas turbine, which powers air compressors and provides further electricity generation. Bed ash is removed from the bottom of the bed to a bed ash cooler, where some sensible heat is recovered by low pressure feedwater. Particulate removal is achieved using parallel sets of two-stage cyclones [9] before the gas turbine, and final cleaning using a cold-side electrostatic precipitator. From the gas turbine exit, the flue gas passes to a waste heat boiler, where sensible heat is absorbed by the feedwater via a set of economizers. The plant includes all of the necessary power, water and waste treatment systems. Natural draught cooling towers are used to provide cooling water for the turbine condenser.

Fig. 1. Schematic of the pressurized fluidized bed combustion combined cycle.

3. Fuels The coal chosen in this study is the American Federal coal which serves as a reference coal. Federal coal has a medium calorific value of 35.64 MJ kg-' (HHV, daf), a low moisture content of 6.3% (ar) and a low ash content of 6.6% (db), but a higher sulphur content of 2.6% (daf). The biomass wastes selected during the case studies are straw, wood chips and sewage sludge, of which the calorific values range from 18.73 MJ kg-' to 22.93 MJ kg-' (hhv, daf), the moisture contents vary from 5.4% to 33.3 % (ar), the ash contents change from 0.9% to 21.8% (db), the sulphur contents are between 0.1% and 0.9% (daf), chlorine content (ranging from 0% to 0.4%) and nitrogen content (ranging from 0.1% to 5.1%). What should also be

Y. Huang et al.

156

considered in fluidised bed applicationsis the effect of mixing the biomass derived ash with the coal ash. In particular the potentially high alkali metal content of biomass ash can reduce the ash softening temperature and lead to agglomeration in the fluidised bed. Table 1 contains basic information about the feedstocks used in the co-firing simulation. Table 1. Fuel analysis. Description

Base case

Case 1

Case 2

Case 3

Fuel name

Federal coal

Straw

Wood chips

Sewage

Water % (ar)

6.3

10.6

33.3

5.4

Ash % (db)

6.6

4.6

0.9

21.8

HHV MJkg (daf)

35.6

19.9

18.7

22.9

Carbon

84.0

48.8

51.0

53.9

Hydrogen

5.7

7.1

6.0

7.9

Nitrogen

1.5

1.3

0.1

5.1

Sulphur

2.6

0.2

0.1

0.9

Chlorine

0.1

0.3

0.0

0.4

Oxygen

6.1

42.4

42.9

31.9

4. Process Simulation

To ensure that the investigation and comparisonswere performed on a consistent and reliable basis, all the cases were simulated using the ECLIPSE personal computer based process simulation package. ECLIPSE was developed for the European Commission over the period since 1986 by the Energy research Centre of the University of Ulster [4-51. A technical assessment study is performed in a logical sequence, in which the first stages involve the preparation of a process flow diagram for the system to be analysed, the addition of the technical informationused for the models and the completion of a converged mass and energy balance. When the mass and energy balance has been completed the next stages involve the economic analysis. These analyses provide all the data required to complete the assessment study.

5. Results and Discussion Table 2 summarises the major technical and environmentalperformance results for the PFBC combined cycles with the range of co-firing ratios and feedstocks given in Table 1. The sensitivity ofthe efficiencyto changes in the biomass types and co-firing ratios, as presented in Fig. 2.


157

Table 2. Mass and energy balance results.

Biomass name

Straw

Co-firingratio (thermalbasis) Fuel flow, kg s-' (ar) Thermal input, MW (HHV) Gas turbine output, MWe Steam turbine output, MWe GT share of gross output, %

I 0% 1 10% I 20% I 30% I I 26.95 I 29.20 31.44 33.69 I 840.10 I 840.4 840.00 840.00 I 78.10 I 80.27 I 82.47 I 84.67 I I 297.20 I 292.64 I 287.72 I 282.54 I I 20.81 I 21.53 I 22.30 I 23.05 I

40% 35.94 840.00 86.87 276.74 23.89

13.61

13.58

13.53

13.46

13.44

361.69

359.33

356.66

353.75

350.17

Net efficiency, % (HHV)

43.1

42.8

42.5

42.1

41.7

C 0 2 g kWh-' (total)

730

738

746

754

762

COz g kWh-' (excluding biomass)

730

661

592

523

453

185

216

249

Total auxiliary usages, MWe Net plant output, MWe

SO, mg (Nm3)-'

(02

6%)

NO, mg (Nm3)-' (026%)

-' (026%)

HCI mg (Nm3)

Main steam mass flow, kg s-'

187.7

I

181.2

184.6

I I

177.6

I I

173.6

334.1

326.3

Heat duty (SH & RH), MW

=-I-= I 182.0

178.8

Ash cooler heat duty, MW

6.0

5.8

5.7

5.5

5.4

Heat duty (HRSG), MW

105.0

106.3

107.0

107.9

108.2

Exhaust flue gas flow, m3 s-'

401.9

405.2

410.4

414.8

422.1

Exhaust flue gas heat loss, MW

102.2

108.6

115.0

121.9

131.5

Exhaust temperature, OC

122.0

122.3

123.5

124.5

127.5

Hot gas flow (GT inlet), m3 sC1 Heat duty (Boiler & Econ.), MW

1 Biomass name

1) Co-firing ratio (thermal basis)

I

I

0%

I I

78.8

10%

79.5

80.2

168.0

I

Wood chips 20%

I

30%

I

I 40%

Fuel flow, kg s-' (ar)

26.95

31.06

35.13

39.22

43.31

Thermal input, MW (HHV)

840.1

840.00

840.0

840.00

840.00

1

158

Y. Huang et al. Table 2. (Continued)

Biomass name

Wood chips

Gas turbine output, MWe

78.10

82.20

86.91

91.33

95.73

Steam turbine output, MWe

297.20

287.93

277.75

268.05

258.17

GT share of gross output, %

I

20.81

I

22.21

I

23.83

I

25.41

I

27.05

13.61

13.68

13.72

13.81

13.85

361.69

356.45

350.93

345.57

340.05


43.1

42.4

41.8

41.1

40.5


730

75 1

773

795

816

C 0 2 g kWh-' (excluding biomass)

730

663

596

528

46 1

SO, mg (Nm3)-' (026%)

175

157

140

123

106

Total auxiliary usages, MWe Net plant output, MWe

1 NO, mg (Nm3)-' (026%)

I

-' (026%)

126

I

114

I

102

I

90

I

78

121

108

97

84

72

Main steam mass flow, kg s-l

187.7

181.1

174.4

167.9

161.7

Hot gas flow (GT inlet), m3 s-'

77.4

79.1

80.9

82.6

84.3

Heat duty (Boiler & Econ.), MW

334.1

320.2

305.4

291.1

276.6


182.0

175.6

168.8

162.5

156.5

HCl mg (Nm3)

1 Ash cooler heat duty, MW 1 Heat duty (HRSG), MW (1 Exhaust flue gas flow, m3 s-'

I I 1

6.0 105.0 401.9

I I

I

5.6 107.8 411.6

I I

I

I

I 108.3 I 108.4 I 426.3 I 443.2 1

453.3

5.3

5.2

5.0 110.8


102.2

116.4

131.5

146.9

162.6

Exhaust temperature, "C

122.0

124.7

129.9

135.5

141.6

Biomass name Co-firing ratio (thermal basis)

Sewage sludge 0%

10%

20%

30%

40%

Fuel flow, kg s-' (ar)

26.95

29.21

31.46

33.72

35.97

Thermal input, MW (HHV)

840.10

840.20

840.20

840.30

840.10

Gas turbine output, MWe

78.10

80.16

82.18

84.24

86.29

Steam turbine output, MWe

297.20

292.85

288.59

283.54

277.85


159

Table 2. (Continued)

Biomass name

Sewage sludge

GT share of gross output, %

20.81

21.49

22.16

22.90

23.70

Total auxiliary usages, MWe

13.61

13.60

13.59

13.54

13.51

361.69

359.41

357.18

354.24

350.63


43.1

42.8

42.5

42.2

41.7


730

734

738

743

750

C 0 2 g kWh-' (excluding biomass)

730

661

590

520

450

SO, mg (Nm3)-' (026%)

175

167

158

150

141

NO, mg (Nm3)

(026%)

126

138

149

160

172

HCI mg (Nm3)

(026%)

121

160

199

237

276

Main steam mass flow, kg s-'

187.7

184.7

181.6

178.0

174.9

Hot gas flow (GT inlet), m3 s-'

77.4

78.4

79.2

80.2

81.1

Net plant output, MWe

~

Heatduty(Boiler&Econ.),MW


Ash cooler heat duty, MW

I 1

334.1 182.0

I

I I

326.2 178.7

I

I I

319.0 175.8

I

I 311.5 I I 172.3 I

I

305.4 168.9

I

6.0

6.4

6.7

7.0

7.2

Heat duty (HRSG), MW

105.0

106.3

107.5

108.2

108.7

Exhaust flue gas flow, m3 s-'

401.9

406.9

411.7

418.3

425.9


102.2

108.6

115.0

121.9

129.1

Exhaust temperature, "C

122.0

122.2

122.8

124.5

126.9

In general, as shown in Fig. 2, the overall efficiency, as a function of co-firing ratio, decreases with an increase in the biomass or waste proportion. In comparison with the base case (100% coal firing), case 1 (co-firing of straw and coal) displays a 1.4% (HHV)reduction of net efficiency when the co-firing ratio varies from 0% to 40% on thermal basis. Also similar trends were seen in case 2 (co-firing of wood chips and coal) and case 3 (co-firing of sewage sludge and coal), where the overall efficiencies were lowered by 2.6% (HHV)and 1.3% (HHV)points respectively. It can be seen that there is a notably different effect on the overall efficiency depending on the type of biomass or waste. The average rate of efficiency reduction from wood chips was about twice of the rate obtained from straw and sewage sludge.

Y Huang et al.

160

In general, the main reason for the efficiency reduction by the biomasses and wastes is the low energy density of the feedstocks, their high moisture (straw and wood chips) and the ash content (sewage sludge). The magnitude of the efficiency reduction reflects the relative significance and quantities ofwater and ash present in the feedstock. Since the PFBC system uses a wet fuel feeding, an even greater quantity of water is fed to the combustor in the case of the low energy density feedstocks than with coal. As a result, it is not surprising that more sensible heat is lost in the flue gases. Also, the higher ash content of the sewage sludge results in more sensible heat leaving the combustion chamber. Therefore, there is an increased amount of unrecovered energy from the flue gases and the ash removal system. However, the moisture content has a much more significant effect on the overall efficiency than the ash content.

g 5 5 .

’%

62 2

0

43,5 43,O 42,5

-

42,O 41,5 41,O 40,5 40,O 0

-

1 1

10

20

30

40

Biomass and waste fraction % (thermal input)

- - 0 - .Straw

--O- Wood chips +sewage

sludge

Fig. 2. Overall efficiency.

The effect of co-firing ratios on the combined cycle operating characteristics has been found and the results are shown in Table 2 and Fig. 3. Firstly, because the biomass or waste has a lower energy density than coal, when the co-firing ratio is increased up to 40%, the mass flow of fuel feed stream (including limestone) is enlarged by 33% for straw, 60% for wood chips and 33.5% for sewage sludge, compared to coal only firing. This is a disadvantage which requires more fuel storage area and fuel processing and handling equipment. Secondly, increasing the co-firing ratio increases the flue gas massholume flow, resulting in a larger fraction of gas turbine output. At the same time, the waste heat flow is enlarged. Although the electrical output from the expander section of the gas turbine is increased, in this case, less heat is available from the PFBC for the steam cycle, which decreases the steam output quite markedly. Compared to coal only firing, case study 2, when the co-firing ratio is set to 40%, the gas turbine output is increased by 17.6 MWe and the steam turbine output is decreased by 39.0 MWe. Consequently, the steam mass flow and the heat duty of the evaporating unit and superheating and reheating units are reduced by 14%, 17% and 14% points, respectively. There is about 10.7MW net heat duty reduction in the evaporating unit,


161

which is compensated for by increasing the feedwater temperature. In other words, it is necessary to increase the heat duty of heat recovery steam generator (HRSG). On the other hand, since the amount of waste heat in the flue gas path is increased, more feedwater is needed to reduce the flue gas temperature. As a result, if there is not sufficient heat exchange in the flue gas path, an additional increase of the exhaust temperature is unavoidable. Lastly, as mentioned earlier, since the co-firing ratio affects the gas turbine output, the gas turbine selected should have enough margins in capacity to cope with the changes in power output. With regard to the total emissions of COz from the stack, an increase in the co-firing ratio reduces the overall efficiency and therefore gives a higher emission compared with 100% coal firing (730 g kWh-'). For example, emissions are approximately 762 g kWh-' for straw, 816 g kWh-' for wood chips and 750 g kWh-' for sewage sludge when the co-firing ratio is 40%. However, if the net emissions to the environment are calculated, ie. the straw and wood chips are considered to be neutral due to the absorption of CO2 from the atmosphere during their growth, then the co-firing processes, as shown in Fig. 4, demonstrate a much lower level of CO2 emissions. Referring to case study 1, in total gaseous CO2 emissions, of the 762 g kWh-' was emitted by coal and straw at 40% co-firing ratio, about 309 g kWh-' was contributed by the straw. The result is a net avoidance of 38% COzemissions compared to pure coal combustion.

'

28,O

-

27,O

-

1

25,O 24,O 23,O 22,o 260

3 @ 0 g

2

-

21,o

%

6

20,o 0

20

I0

30

40

Biomass and wastefraction % (thermal input)

- - 0 - - Straw +Wood chips ~

~~

Fig. 3. Gas turbine output share.

+sewage

sludge

Y. Huang et al.

162

800 700 600

500 nnn 7""

1

0

20

10

30

40

Biomass and wastefraction 96 (thermal input)

- - 0 - . Coal+straw - - ik - .Excluding straw C02

-D-Coal+wood chips

+Excluding

wood chips C02

I

Fig. 4. Net C02 emissions.

As shown in Fig. 5 , since biomass or waste mostly contains considerably less sulphur than coal, an increase the co-firing ratio always makes the SO, emissions decrease proportionally. However, the sulphur rich sewage sludge gave less reduction of SO, emissions compared with straw and wood chips. If the limestone addition is adjusted to give the same SO, emission level, the option of co-firing of straw or wood chips and coal gives an improvement in efficiency of about 0.2% points at 40% co-firing ratio.

3

.3

120

8 3 2

100

I

0

10

20

30

Biomass and waste fraction % (thermal input)

- - 0 - .Straw

+Wood chips

Fig. 5 . SOX emissions.

-d-Sewage sludge

40


6 2

0

I0

20

30

163

40

Biomass and waste fraction % (thermal input) - - - 0-- - Straw

+Wood chips

+Sewage

sludge

Fig. 6 . NOx emissions.

Since the NO, emissions are highly dependent on the fuel nitrogen content and boiler operating conditions, increasing biomass or waste fraction in the fuel blends shows different trends, as shown in Fig. 6, but they remain at low levels. This is the contribution partly from the low flame temperatures associated with the pressurised combustion and partly from the lower fuel nitrogen levels in terms of nitrogen per unit energy fed. The emissions of carbon monoxide in the flue gas of pressurized combustors were low, which was less than Smg/MJ [9-101 due to the operating pressures [ll]. It should be noted that there is a higher chlorine content present in straw (0.3%) and sewage sludge (0.4%), which potentially gives a higher HCl emissions, 249 mg (Nm3)-' for straw and 276 mg (Nm3)-' for sewage sludge at 40% co-firing ratio. 6. Conclusions

The ECLIPSE process simulator was used to successfully perform technical and environmental studies for the co-firing of coal and biomass with different ratios in the PFBC system, where straw, wood chips and sewage sludge were chosen to blend with Federal coal (a reference coal). The process simulation and analysis of the co-firing in the PFBC power plant led to the following conclusion: The technical and environmental impact of co-firing of coal with straw, wood chips or sewage sludge in the PFBC system has been assessed theoretically. Increasing the co-firing ratio will, in all cases reduce the overall efficiency. However, the co-firing of wood chips with coal gives the lowest efficiency, due mainly to its higher moisture content.

164

Y. Huang et al.

The co-firing of biomass or waste with coal has a reasonable effect on the PFBC combined cycle. As expected, the influence on the gas turbine cycle operation is smaller than on the steam turbine cycle operation, although the gas turbine output does change significantly. Also, this change links with the feedwater preheaters and the economisers. In order to maintain the PFBC plant output, it is suggested that when blending with high moisture content biomass wastes, more heat transfer surface in the HRGS is needed. Proportional effects are shown on the reduction of both net C 0 2emissions (straw and wood chips) and SO, emissions with increasing co-firing ratio. NO, emissions from the cofiring of wood chips and coal are much less than the coal firing process, due to the lower he1 nitrogen levels in terms of nitrogen per unit energy fed. However, the straw and sewage sludge additions give a slight increase of NO, emissions. References [I] Lindberg T, Anderson J. (1997), Co-firing of biomass and coal in a pressurised fluidised bed combined cycle. Results of pilot plant studies. The 14th ASME International FBC Conference, Vancouver, May. [2] Van De Kamp WL, Morgan DJ. (1996) The co-firing of pulverised bituminous coals with straw, waste paper and municipal sewage sludge. Combust. Sci. and Tech 121: 317-332. [3] Pedersen LS et al. (1996) Full-scale co-firing of straw and coal. Fuel 75 (13): 1584-1590. [4] Williams BC, McMullan JT. (1988) Development of computer models for the simulation of coal liquefaction processes. In Imariso and Bemtgen (eds), Progress in Synthetic Fuels, pp. 183- 189. Graham and Trotman, London. [5] ECLIPSE Process Simulator. Copyright 1992. Energy Research Centre, University of Ulster, Coleraine, BTS2 1SA. [6] Veenhuaizen D. (1999) Private Communication, S.A. Jansson & J. Anderson. Progress of ABB’s, PFBC Projects, the 15th International Conference on Fluidised Bed Combustion, Savannah, Georgia, USA. [7] Gustasson 0. (1993) PFBC for combined heat and power generation, Fernwurme International-FWI 22: 210-217. [8] Anderson J. and L. (1997) PFBC: Competitive power with minimal environmental impact. Practical experience. IChE Conference November 3-4, London, UK. [9] Advanced Electric Power Generation Fluidized Bed Combustion, http :Ilwww .lanl.govlproj ects/cctc/factsheets/tidd/tidddemo. [ lo] Podolski WF et al. (1983) Pressurised fluidised bed combustion technology, NOYES Data Corporation, New Jersey, USA.

PRODUCTION OF BIOCOAL FROM CASSAVA STALK T. PUTHIKITAKAWIWONG, R. BOONSU, 0. JOOMPHA

Faculty of Science, Mahasarakham University, Thailand Thailand is an agricultural country. Each year the country produced not only agricultural product but also more than 50 million tons of agricultural residues. Cassava stalk is the fourth largest agricultural residues which accounted more than 4 million tons per year. Cassava stalk consider as useless agricultural residues. Production of biocoal from cassava stalk is the process to convert the useless residues into useful h e l . Production of biocoal from cassava stalk consisted of two main processes: production of charcoal from cassava stalk and briquetting of cassava stalk charcoal. The charcoal yield was 22.18 % and the suitable binder to charcoal ratio was 1 to 7. Biocoal from cassava stalk has more ash than Eucalyptus charcoal but fixed carbon and calorific value are lower than Eucalyptus charcoal. However, the calorific value of this biocoal is still higher than calorific value of firewood. So, biocoal from cassava stalk can be used as household cooking fuel instead of firewood and charcoal.

1. Background and Objective

Thailand is an agricultural country. The majority of the populations are live in the rural area. Agriculture is the main academic activity for the rural people. Which consider as country’s lifestyle. The residues from the agricultural products in the rural area are abundant. Each year the country produced more than 54 million tons of agricultural residues. The ten largest agricultural residues in Thailand are, paddy straws, bagasse, rice husk, cassava stalk, corn cob, coconut husk, coconut shell, cotton stalk, saw dust and peanut shell. Cassava stalk is the fourth largest agricultural residue in the country. The total amount of cassava stalk is around 4 million tons per year. This amount of cassava stalk is considered as useless waste. The aim of this study is to convert this amount of useless waste into useful fuel. Biocoal is one of the conversion process that suitable for this purpose. Biocoal is the combine process of carbonization and briquetting.

2. Biocoal Production The carbonization process converted cassava stalk into cassava stalk charcoal. The briquetting process converted the cassava stalk charcoal into cassava stalk charcoal briquetted, which is called biocoal. Binder is use during the briquetting process. The detail processes of biocoal production are separated as follows: carbonization process in Fig. 1 and briquetting process in Fig. 2.

3. Result of Study Result of study showed that the biocoal production from cassava stalk has two main processes. The result of each processes are as follow.

3.1. Carbonization Process This process converted cassava stalk into cassava stalk charcoal. Cassava stalk and property are showed in Fig. 3 and Table 1. 165

166

T. Puthikitakawiwong, R. Boonsu & 0.Joompha

Cassava Stalk

+ Drying

1 1

Sizing

Carbonization

+ Cassava Stalk Charcoal Fig. 1. Carbonization process

Cassava Stalk Charcoal

1 1

Grinding

Binder Mixing

+ Briquetting

Biocoal Fig. 2. Briquetting process

Production of Biocoalporn Cassava Stalk

167

Fig. 3. Cassava stalk. Table 1. Propemes of cassava stalk.

-

The charcoal kiln, cassava stalk charcoal and charcoal yield are showed in Figs. 4 and 5 and Table 2.

Fig. 4. Charcoal kiln.

168

Fig. 5. Cassava stalk charcoal. Table 2. Carbonization of cassava stalk.

I- - - -

Average

32.

~

~ Ppscess ~

~

U

@

~

~

This process briquetted the charcoal fine (mixed with binder) into the final product whish call briquetted charcoal or biocoal. The briquettirag machine sand biocoal product we showed ir! Figs. 6 and 7.

Fig. 6. Briquetting machine

~

I69

Fig. 7. Biocoai firom cassava stalk.

3.3. Binder to ratio Result o f study showed that the suitable binder to charcoal ratio was 1 to 7. The details o€ the study are showed in Table 3 . Table 3. Binder to ~ h a ~ cratio. ~al

The suitable binder to charcoal ratio are 1 to 6 and B to 7. Because of the costly binder, the least use of binder is the best choice. So that the best binder to charcoal ratio is 1 to 7.

3.4 ~

0~

~

~~ w~~ ~ Charcoal o ~ c

o

~~

~

~

~

s

The biocoal from these processes has the calorific value lower than Eucalyptus charcoal. Eucalyptus wood charcoal has calorific value around 29 MJkg. The calorific value of biocoal fram cassava stalk is around 23.5 MJkg. The calorific value of firewood is around 14 MJkg. So that the calorific value of biocoal fi-om cassava stalk is in between firewood and charcoal. This biocod can be used instead of firewood or charcoal in the nurd c 0 ~ ~ ~ ~ ~

~

170

T. Puthikitakawiwong, R. Boonsu & 0.Joompha

4. Conclusions

Production of biocoal from cassava stalk has two main processes, carbonization and briquetting. The yield of carbonization of cassava stalk was 20.8% dry basis. Oil drum kiln is use as charcoal kiln; it is still cheap, efficient and easy to operate. The briquetting process by screw type briquetting machine. The optimum ratio between binder and charcoal was found to be 1 to 7 respectively. The calorific value of biocoal is lower than the calorific value of Eucalyptus charcoal but is still higher than the calorific value of firewood. So, this biocoal can be use as cooking fuel instead of charcoal or firewood. The disadvantage of this study is the briquetting and grinding machines, which is very expensive for the rural community. References [ 11 Bhattacharya SC, Shrestha Rh4. (1990) Biocoal technology and economics. Regional

Energy Resources Information Center (RERIC), Asian Institute of Technology, Bangkok, Thailand. [2] Stassen HEM. (1997) Biomass energy. Biomass Technology Group (BTG), Enschede, The Netherlands. [3] Grover PD. (1981) Development of briquetted fuel (PARU) from agricultural and other biomass residues. Producer Gas System and Agricultural Application. [4] Panya Thanya W. (1998) Charcoal making technology for rural forest product. Royal Forest Department, Thailand.

A COMPARISON OF POWER GENERATION FROM BIOMASS IN A SMALL CFBC PLANT WITH BIOMASS CO-FIRED WITH COAL IN A LARGE CFBC D.R. MCILVEEN-WRIGHT"), B.C. WILLIAMS(*),J.T. MCMULLAN'~) (')Northern Ireland Centrefor Energy Research and Technology (NICERT), University of Ulster, Coleraine, BT52 ISA, Northern Ireland, United Kingdom. "Exus Energy, Templemore Business Park, Derry, BT48 OLD, Northern Ireland Biomass is one of the renewable energy sources which is not intermittent, location-dependent or very difficult to store. If grown sustainably, biomass can be considered to be C 0 2 neutral. The use of biomass for power generation is also considered to be important in increasing the electricity output from renewable energy sources. However, power plants dedicated to the use of biomass fuel are not in widespread use and the acceptance of this he1 and development of the infrastructure for biomass production and transportation remain in their infancy. If small ratios of biomass can be co-fired with coal in large-scale conventional power plants, without significant technical, environmental or economic penalties, it could lead to a greater demand for biomass and stimulate the industry. In this study a 80 MWth CFBC, fuelled by biomass only, and a large-scale 1000 MWth CFBC, cofired with coal and 8% biomass, and the same large CFBC system, fired only with coal, are modelled using the ECLIPSE process simulation package and their technical, environmental and economic properties analysed and compared. The co-firing of biomass with coal was found to have little effect on the large-scale CFBC system, when a small ratio of biomass is used. The large scale system was found to have higher efficiency, lower C02 emissions and lower break-even electricity selling price than the small biomass-helied CFBC. Co-firing of biomass with coal could be a promising way of promoting the production, use and acceptance of biomass as a fuel in electricity generation.

1. Introduction

The co-combustion of coal and biomass has received widespread interest for some time as a means of conserving coal reserves and reducing net CO2 emissions [ 11. Several other environmental advantages have been reported e.g. co-firing high-sulphur bituminous coal with 20% straw gave a net reduction in NO and SO2 emissions [2]; lower NO, emissions may be found during co-combustion, since there is high volatile content in biomass and biomass nitrogen preferentially forms NH3 to HCN which is formed preferentially by nitrogen from coal [3]; and the primary reactions of thermal decomposition of biomass fuels are not significantly affected by the presence of coal, which itself does not seem to be influenced by the release of volatile matter from biomass [4]. The use of biomass, which is considered to produce no net COz emissions in its life cycle, can reduce the effective C 0 2 emissions of a coal-fired power generation system, when co-fired with the coal, but may also reduce system efficienay and increase electricity selling price. An analysis of several power generation technologies, using 100% coal, 100% biomass and coal-biomass mixtures has been made to identify the economic cost of reducing C 0 2 emissions through the replacement of coal with different amounts and types of biomass [ 5 ] . More recently there have been further financial incentives for co-firing, such as the requirements for increasing the percentage of electricity generated from renewable sources, carbon taxes, the increasing cost of gate fees at landfill sites and the ban on putrescible wastes going to landfill. In this study co-firing of a high ash coal with 8% (by thermal input) biomass in a large scale power plant is compared with a smaller plant fired only by the same quantity of biomass, both using fluidised bed technology. Experiments on the co-firing of these 171

172

D.R. Mcllveen- Wright, B.C. Williams & J.T. McMullan

mixtures were carried out and the results used in process simulation software to model power plants employing fluidised bed technology and perform technical, environmental and economic analyses of such systems.

2. Circulating Fluidised Bed Combustion Plants In this paper a computer simulation, using the ECLIPSE process simulation software package [6], was made of a large scale 1000 MWth CFBC power generation system. The process flow diagram of the modelled system was based on the power plant at Gardanne, France, which has been tested for co-firing with coal and biomass. Simulations were also made of a smaller 80 MWth CFBC power plant, which was fuelled completely by biomass in this instance.

3. Typical 1000 MWth Coal-Fired Circulating Fluidised Combustion (CFBC) Power Station In a typical CFBC plant (see Fig.1) coal would first be transferred from the normal coal storage facilities where it is then pulverised in mills, before being pneumatically transferred, together with limestone, using preheated primary air to a balanced draught, circulating fluidised bed boiler. Secondary air is injected through a set of nozzles higher up the chamber walls. The high fluidising velocity forms an expanded bed with material carried out of the combustor. Cyclones separate the majority of the solids from the flue gas. These solids are returned either directly to the combustor or through a set of external heat exchangers which receive preheated fluidising air. The low operating temperature (850 "C) and the staged combustion of the coal helps to reduce NO, formation. Sulphur retention is achieved by adding limestone, so no additional flue gas desulphurisation is required. In the combustor, the walls are lined with tubes which remove the radiant heat, and maintain the furnace temperature at 850 "C. Approximately 40% of the bed material is removed periodically from the base of the combustor and heat is extracted for lowpressure boiler feedwater heating. The rest of the solids are carried forward with the hot gases and removed by bag filters. High ash resistivity makes cold side electrostatic precipitators unsuitable and bag filters have the added advantage of promoting further sulphur retention. Before reaching the bag filter the gases are cooled by transferring heat first to steam in the superheater and reheater tubes, then to condensate in the economizer, and finally by passing through air preheaters at the back of the convective pass section. Superheating is achieved in both the external heat exchangers and the convective pass section. The reheater tubes are also located in these external heat exchangers and final economising also occurs in the convective pass section. The cooled gases are exhausted to the atmosphere via the induced draught fan and stack. The steam from the superheater goes to the turbine stop valve and is expanded in the high-pressure turbine. The steam turbines have facilities for steam extraction and allow for transfer of steam to the regenerative feedwater heaters. Drains from the three high-pressure feedwater heaters are fed to the deaerator. The steam from the highpressure turbine is then reheated before passing through intermediate pressure and double flow low-pressure turbines. At the crossover from the intermediate to the low-pressure turbines steam is extracted for the deaerator. Drains from the three low-pressure feedwater heaters are fed to the condenser.

A Comparison of Power Generation?om Biomass

173

Fig 1. Diagram of 1000 MWth CFBC.

The steam from the low-pressure turbine is condensed and the condensate is pumped by the extraction pump through three low-pressure surface-type heaters and a parallel ash cooler to the deaerator. Here the incoming water is heated by direct contact with the bleed steam. The boiler feed pump forces the condensate through three high-pressure feedwater heaters and the economizer before reaching the boiler and completing the steam cycle. Recent trials have shown that, when around 543% of the feedstock is not coal i.e. consists of biomass andor certain wastes, no modifications of the coal-fired plants are necessary. In the large scale system proposed here, there probably would need to be some additional reception, size reduction, handling and storage facilities for the biomass, which has been taken into account in the desigdmodification of the power plant. 4. 80 MWth CFBC System

A typical CFBC system of this size (see Fig. 2) would have standard biomass feed preparation, storage and handling facilities from which the fuel would be transferred, together with limestone absorbent when coal is involved, to an atmospheric circulating fluidised bed combustor. Air is first heated in an air preheater and external heat exchangers and then fed to the base of the combustor. The high fluidising velocity of this air causes an expanded bed to form and which carries material out of the combustor into the recirculation cyclones. These cyclones separate the majority of the solids, which are then returned to the base of the combustor via external heat exchangers. Most of the ash is removed from the base of the combustor and the highly efficient cyclone filters remove

174

D.R.Mcllveen- Wright,B.C. Williams & J. T. McMullan

the rest. The low operating temperature helps to reduce NOx formation, and sulphur retention (when coal is used) is achieved by the addition of limestone.

!

Fig. 2. Diagram of 80 MWth CFBC

The hot gases from the recirculation cyclones are cooled initially in a heat recovery steam generator and finally in combustion air preheaters. The cooled gases are exhausted to the atmosphere via the induced draught fan and stack. The steam from the heat recovery steam generator goes to the turbine control valve and is expanded in the steam turbine. The steam turbine has facilities for steam extraction to allow for transfer of steam to the feedwater heaters and to the deaerator tank. The low pressure steam from the steam turbine is condensed and the condensate is pumped by the low pressure (LP) pump through the LP heater to the deaerator tank and the high pressure (HP) pump through the HP heater, before reaching the heat recovery steam generator and completing the steam cycle. 5. Simulation Results

The two power plants were modelled using the ECLIPSE process simulation software and technical, environmental and economic analyses made. The larger scale 1000 MWth CFBC was assessed when fuelled by a high-ash, medium sulphur Puertollano coal, and also when co-fired with 8% (by thermal input) with biomass. The 80 MWth CFBC was analysed when fuelled by the same amount (and type) of biomass as in the 1000 MWth plant.

A Comparison of Power Generation from Biomass

175

5.1. Fuel Analysis The calorific values, proximate and ultimate analyses of these fuels are shown in Table 1. Table 1. Fuel properties.

I

I

I

Volatiles Fixed Carbon

Coal Feedstock

Biomass

Puertollano Coal

Pine Residues

Proximate Analysis % wiw 24.9 37.3

I

I

16.2

Moisture

5.5

Ash

32.3

0.5

Total

100

100

I

I

72.6 10.7

Ultimate Analysis % daf C H

77.33 5.31

4.94

N

1.93

0.90

I

S

I

I

0 Total

I

5 1.57

1.29

I

14.15

I

0.00 42.58

100.00 19.06

LHV (ar) MJikg HHV (daf) MJikg LHV (daf) MJ/kg

I

I

100.00

I

I

20.20

18.21

19.01

30.64

22.75

29.28

21.41

6. 1000 MWth System, Coal Only and Co-fired with 8% Biomass

6.1. Technical Results The technical and emissions results for the simulations of the 1000 MWth power plant are shown in Table 2. Table 2. Technical results for the 1000 MWth CFBC.

I I I

coaltype

1

Reheat?

excess air

Steamcycle Fuel Mix

FD fans (3) ID fan

I I I I

Peurtollano 20% 160bari538C Yes 100% coal

I I I I

Peurtollano 20% 160bad538C Yes 92% coal/ 8% wood

11086.2

10854.6

3882.5

3829.4

I I I I

D.R. McIlveen- Wright, B. C. Williams & J. T. McMullan

176

Table 2. (Continued) Slag Outlet

148.6

Bag Filter

7.8

7.4

Coal Crusher

22 1.2

135.5

Conveyors (4)

2042.9

2132.8

HP & LP Pump

I

9774.3

I

Elect. Utilities

4899.21

9765.6

I

4894.91

Total Usages

32062.8

31760.3

HP Turbine

108079.5

107985.1

LP Turbines (4)

179475.3

I793 18.5

IP Turbines (3)

149386.2

149255.7

I I

I I I

Net Electricity

I

Thermal Inuut HHV

I

140.1

Gross Electricity

Thermal Input LHV Efficiency. LHV

I I I

43694 1

I

436559.3

404878.2

I

404799.0

1000.00 1046.7 40.49

I I I

1000.00 1048.0 40.48

Efficiency, HHV

38.68

38.63

C02 g/kWb

873.7

866.3

C02 mgfNm3 at 6% 0 2

239935

241577

SO2 mg/Nm3 at 6% 0 2

243.4

250.8

NOx mg/Nm3 at 6% 0 2

340.38

352.95

CO mgfNm3 at 6% 0 2

59.82

62.01

4.04

4.04

0 2 (dry)vol %

I

I I I I I

6.2. Economic Results The economic results for the ECLIPSE simulations of the large CFBC system are shown in Table 3. Table 3. Economic results for 1000 MWth CFBC.

A Comparison of Power Generationkom Biomass

177

Table 3. (Continued) 20 1.29

CFBC

37.82

37.45

253.15

250.32

CFBC HRSG Sub Total

113.37

Steam turbine

I

I

199.35

Steam System & Cond

I

37.03

Cooling Water

I

15.03

Water Treatment Chimney

15.02

10.62

10.62

4.82

4.77

67.50

67.40 494.05 12

1208.2

1220.7

48.79

48.51

SI ($/kWe) BESP ($IMWh)

I I

36.99

489.1904

Sub Total Total

113.29

I 1

7. 80 MWth System, 100% Wood

7.1. Technical Results The technical and environmental results for the CFBC using 100% biomass are shown in Table 4. Table 4. Technical results for the 80 MWth CFBC.

I I

Fuel Mix

I

FD-Fan

I

ID-Fan

I 1 I

8% Biomass 739.5 118.7

Gas Cleaning

I

0.0

WoodConvey

I

205.8

AbsConvey

I

0.0

Coal Convey

0.0

Ash Convey

5.5

Plastic Convey

474.8

LP Pump

2.9

Size Reduction Total

I I

LPTurbine

I

Electric Process

IPTurbine

Electric Utility

I I I

0.0

HP Pump

I

I I

800

I I I I

2347.2 13457.9 14337.3 25448.0 450.6

I I I I

178


Steam Cycle bar/"C

921495

Thermal Input LHV

80

Thermal Input HHV Efficiency LHV %

31.25

Efficiency HHV %

29.40

C02 glkWh

1019

SO2 mgMm3 at 6% 0 2

0.0

Nox mgmm3 at 6% 0 2

348.3

CO mgMm3 at 6% 0 2

61.2

0 2 (dry) Vol %

5.5

7.2. Economic Results The economic results for the 80 MWth CFBC using 100% biomass fuel are shown in Table 5. Table 5. Economic results for the 80 MWth CFBC.

Total ($M) Specific Investment ($/kWe) BESP ($/MWh)

49.233 1970 59.83

A Comparison of Power Generationfrom Biomass

179

8. Comparisons 8.1. Efficiency The efficiency of the larger CFBC was found to change little from the case where it is fuelled solely by coal (40.49%, LHV) to the case where it is co-fired with 8% biomass (40.47%, LHV), as shown in Fig.3. The 80 MWth CFBC system was found to be much lower (3 1.29%) in efficiency, probably because the steam cycle conditions result in an intrinsically less efficient system (538°C with reheat and 160 bar for the larger system compared with 495OC and 92 bar for the smaller system). Electrical Efficiency 45

Fig. 3. A comparison of electrical efficiencies.

8.2. Specific Investment

The capital cost per unit of electricity generated, or specific investment, SI, of the larger CFBC increases slightly from 1208 $/kWe to 1220 $/kWe if the cost of additional equipment for biomass reception, storage, size reduction and handling is taken into account. The SI for the 80 MWth CFBC fuelled by 100% biomass is much higher, around 1970 $Awe, as shown in Fig. 4. Specific Investment 2000 180

160 1400

1000 MWth

00 MWth

Fig. 4. A comparison of specific investments.

180

D.R. Mcllveen- Wright, B. C. Williams & J. T. McMullan

8.3. Emissions The C 0 2 emissions of the larger CFBC (1000 MWth) were found to be around 874 glkWh for 100% coal, dropping to about 866 gkWh when it is co-fired with 8% biomass. The 80 MWth CFBC was found to emit around 1029 glkWh C02 when fuelled with 100% biomass, as shown in Fig. 5. The smaller CFBC emits more CO2 than the larger one due to its lower efficiency. If the biomass is grown sustainably, it can be considered to be carbon-neutral. The net C 0 2 emissions from the 80 MWth CFBC could be considered to be zero, when using 100% biomass fuel, and the 1000 MWth CFBC would have net C 0 2 emissions of around 803 g/kWh, when co-fired with 8% biomass, which is an 8% emission reduction over the 100% coal case.

Fig. 5 . C 0 2 emissions for both CFBC systems.

9. Electricity Generation Costs The break-even electricity selling price (BESP) for the larger CFBC system was found to be 48.79 $/MWh using 100% coal and 48.51 $/MWh when co-fired with 8% biomass. The coal and biomass costs were taken to be $52.251 daf tonne and $26.061 daf tonne respectively. The sensitivity of the BESP to variations in the cost of coal or biomass is shown in Fig. 6. The coal and biomass costs are varied by k 100% of their nominal values.

Fig. 6 . Variation of BESP with fuel price.

A Comparisonof Power Generationfrom Biomass

181

10. Conclusions 10.1. Efficiency

The efficiency of the larger 1000 MWth CFBC was only negligibly affected by changing the fuel from 100% coal to co-firing with 8% biomass. The smaller 80 MWth CFBC was much less efficient than the 1000 MWth CFBC (3 1.2% compared with 40.5%, LHV). Co-firing biomass in the larger CFBC is the more efficient method for using biomass to generate electricity. 10.2. C 0 2 Emissions

The less efficient 80 MWth CFBC emits more C 0 2 than the larger 1000 MWth CFBC per unit of electricity generated, but no SOX, when fuelled by 100% biomass. However, the net C 0 2 emissions for the 80 MWth CFBC system may be considered to be zero, if the biomass is sustainably managed. Co-firing with biomass lowers C 0 2 emissions and net C02 emissions of the larger CFBC. 10.3. Specific Investment

The 80 MWth CFBC has an SI (about 1970 $/kWe) more than 50% higher than that of the 1000 MWth CFBC system (around 1200 $/kWe), since the larger system is more efficient.

10.4. BESP The 1000 MWth CFBC system co-fired with 8% biomass has a lower BESP value than the 80 MWth CFBC using the same amount and type of.biomass. In addition, the BESP for the 1000 MWth system is only negligibly affected by variations in the cost of biomass, whereas the 80 MWth BESP is significantly affected, as shown in Fig. 4. Financial incentives for the use of biomass are currently available in some countries for co-firing, such as the requirements for increasing the percentage of electricity generated from renewable sources, carbon taxes, the increasing cost of gate fees at landfill sites and the ban on putrescible wastes going to landfill. These incentives have not been taken into account in this paper, as their long-term availability for co-firing applications is currently unconfirmed.

10.5. Summary In summary, the ECLIPSE simulations of the 1000 MWth CFBC co-fired with 8% biomass showed this system had a higher efficiency, lower C02 emissions, lower SI and BESP than the 80 MWth CFBC fuelled solely by biomass. The larger system was also much less sensitive to variations in the cost of biomass, which would have a significant effect on the economic viability of the small biomass-fuelled CFBC.

182


Co-firing small ratios of biomass with coal offers a promising way of promoting the production, trade and infrastructure of a biomass wastes or energy crops industry and the wider acceptance of biomass as a long term fuel for electricity generation. References

[I] Hein KRG, Bemtgen JM. (1998) EU clean coal technology - co-combustion of coal and biomass. Fuel Processing Technology 54: 159-169. [2] Pedersen LS, Nielsen HP, Kiil S, Hansen LA, Dam-Johansen K, Kildsig F, Chrstiensen J, Jespersen, P. (1996) Full scale co-firing of straw and coal. Fuel 75: 1584-1590. [3] Spliethoff H, Scheuer W, Hein KRG. (2000) Effect of co-combustion of sewage sludge and biomass on emissions and heavy metal behaviour. Proc. SaJ Environ. Prot. 78: 33-39. [4] Biagnini E, Lippi F, Petarca L, Tiognotti L. (2002) Devolatilization rate of biomasses and coal-biomass blends: an experimental investigation. Fuel 81: 1041-1050. [5] McIlveen-Wright DR, McMullan JT, Williams BC. (2003) The economics of reducing carbon dioxide emissions by the use of biomass co-combustion, Proc. Seventh International Conference on Energy for a Clean Environment ‘CleanAir 2003 ’, Calouste Gulbenkian Foundation, Lisbon, Paper 16.1. [6] Williams BC, McMullan JT. (1996) Techno-economic analysis of fuel conversion and power generation systems - the development of a portable chemical process simulator with capital cost and economic performance analysis capabilities. Int. J. Energy Research 20: 125-142.

CHARACTERIZATION OF SWEET, FIBRE AND BIOMASS SORGHUM POTENTIAL IN PORTUGAL AS AN INDUSTRIAL AND ENERGY FEEDSTOCK A.L. FERNANDO, M.P. DUARTE, J. MORAIS, J.F.S. OLIVEIRA

Grupo de Disciplinas de Ecologia da Hidrosfera / Unidade de Biotecnologia Ambiental, Faculdade de Cidncias e Tecnologia da Universidade Nova de Lisboa, Quinta da Torre, 2829-51 6 Caparica, Portugal: e-mail: [email protected] orjjio@ct. unl.pt The purpose of this work was to investigate the influence of crop management on the sorghum biomass quality and productivity, in Portugal, in order to access its potential as an industrial and energy feedstock. To do so, the effects of different fertilization levels, the sowing date and the harvest date in the biomass quality and productivity, were studied. Productivity was affected by the level of nitrogen and by the sowing date. Highest productivities were obtained in the most N-fertilised fields and in fields sowed between 251hMarch and 1’‘ May. Among subspecies, Sweet and Biomass Sorghum presented, globally, higher productivities than Fibre Sorghum. In all the experimental years, highest productivities, in all the fields, were achieved at 193 5 9 days after sowing, regardless of all the different sowing dates. Biomass quality was affected by the crop productivities. Fields with lowest productivities resulted in a biomass with a higher nitrogen and ash contents, namely the fields sowed at 2nd June and the fields without N-fertiliser. In the experimental conditions at Caparica, sweet sorghum crop showed a better potential as an industrial and energy feedstock, and to get better results, economically and environmentally, fields should he sowed between 25’h March and 1’‘ May, with a high nitrogen input (120 kg N/ha).

1. Introduction Sweet, Fibre and Biomass Sorghum (Sorghum bicolour L. Moench) are annual C4 plant species with high water efficiency, high biomass yield and high dry matter accumulation rates on a daily basis [1,2]. Although sorghum is of tropical origin, the plant is well adapted in subtropical and temperate regions [2,3]. The greatest interest in Sweet, Fibre and Biomass sorghum is biomass production. Fibre sorghum’s stalks are fibrous, not juicy [4]. This involves using Fibre sorghum as an energy plant or as pulp for paper industry. Due to its high sugar content Sweet sorghum is gaining attention as a potential alternative crop for the production of bioethanol or sugar-sirup for industrial productions [4]. Stalks are crushed for juice extraction to separate the sugar from the bagasse. After extraction the sugar can be concentrated for storage or it can be converted into fuel ethanol [ 11. Bagasse can be dried for long term storage or can be processed immediately into bio-crude oil, gasoline, electricity, pulp for paper, compost or charcoal pellets [l]. The study of sorghum as an energy crop started in Portugal in 1991, with UBiA, in the scope of the project “Activity in the Sector of Biomass Production Sweet Sorghum Network” (JOULE Programme, JOUB 0036) supported by the European Union. Two others projects followed this one: “Sweet Sorghum, a sustainable crop for energy production in Europe: Agricultural and Industrial improvement, optimization and implementation” (AIR 1 CT92-004 1) and “Environmental studies on sweet and fibre sorghum sustainable crops for biomass and energy” (FAIR-CT96-1913), both supported also by the European Union, with the participation of UbiA. These three projects 183

184

A.L. Fernando et al.

permitted the study of sweet, 'fibre and biomass sorghum in Portugal, namely the morphologic and morphometric characteristics of the plants, their productivity and their biomass quality. 2. Materials and Methods

The experimental fields are situated in the Peninsula of S e ~ b a lin , the south border of the river Tejo, near the estuary and the Atlantic coast (latitude 38'40' N, longitude 9" W, altitude of 50 m). The experimental plots were established from 1991 to 1994 and from 1997 to 2000, in a clayey and alkaline soil, using a row spacing of 0.70 m with a distance within row of 0.10-0.12 m. Three different subspecies of sorghum were studied, sweet sorghum (Keller) (1991-1994 and 1998-2000), fibre sorghum (H128) (1997-2000) and biomass sorghum (MN 1500) (1997). During the experimental years, the fields were sowed between the end of March and the beginning of June: 1991 - 2Sth April; 1992 - 25'h March; 1993 - gth May; 1994 - 9" April; 1997 - 2"d June; 1998 - lstMay; 1999 - loth April. Top fertilization was applied after the emergence of the plants. For the sweet sorghum fields, from 1991 to 1994, two levels of N-fertiliser were applied: 75 and 100 kg N.ha-'. For the biomass and sweet sorghum, from 1997 to 2000, three levels of Nfertiliser were studied: O, 60 and 120 kg N.ha-'. For fibre sorghum, from 1997 to 2000, only one level of N-fertiliser was studied: 60 kg N.ha-'. P-fertilizer (120 kg P.ha-') and Kfertilizer (120 kg K.ha-') were applied in all the fields. All the fields were fully irrigated in order to compensate the water deficit of the soil, and to prevent water stress. At the end of the growing season the vegetable material was harvested and the total aerial dry weight, ash, nitrogen and phosphorus contents, the fibre content, the sucrose content and the gross heat of combustion, were determined in order to evaluate the productivity and the quality of the biomass. The chemical analyses were performed according to the following procedures: a) ash content: calcination at 550 50°C for two hours in a muffler furnace; b) nitrogen content: by the Kjeldahl method; c) phosphorus content: by the ascorbic acid method, after digestion of the sample; d) fibre: by the Belluci method; e) sucrose: sugar extraction by hot distilled water, sucrose determination by the UV method for determination of sucrose and D-glucose in food stuffs and other materials (Boehringer-Mannheim); f) gross heat of combustion: using an adiabatic calorimeter.

*

3. Results and Discussion 3.1. Biomass Productivity Fig. 1 shows the differences, in terms of the dry matter productivities, for sorghum, among fields sowed at different dates. According to these results there were differences among fields sowed at different dates. Better results were obtained in the fields sowed between 25" March and lst May.

185

Sweet, Fibre and Biomass Sorghum Potential in Portugal

Sowing as late as gth May resulted in lower productivities, but still, higher than 20 t/ha, not compromising the economical viability of the crop production. But, sowing as late as

I

45

I

I

40

f

35

b 30 U m

25

z

20

c

p

15

!10 U

E

5 0

25th

9th April

March

~

10th April

28th

1stMay 9th May

2nd June

April

Fig. 1. Sorghum productivities obtained in fields sowed at different dates.

2"d June resulted into extremely low productivities, and this can compromise the net economical and environmental gain of this crop. But, these significant lower productivities obtained in fields sowed at 2nd June, can also be a consequence of the climatic conditions observed in 1997, namely, a colder summer than usual with heavy showers. Fig. 2 shows the differences, in terms of the dry matter productivities, among fields with different nitrogen fertiliser levels. According to these results there were significant differences among productivities obtained in fields with different levels of nitrogen: productivity increased with the increase in nitrogen input. Fig. 3 shows a comparison, in percentage terms, of the dry matter productivities, between biomass sorghum (MN 1500) and fibre sorghum (H128) and between fibre sorghum (H128) and sweet sorghum (Keller).

I

35

.

30

25

P m 20 c 0

15

TE i

10

0

Low

High

Fig. 2. Sorghum productivities obtained in fields with different nitrogen application levels (0 = 0 kg Nha; Low = 60 or 75 kg N/ha; High = 100 or 120 kg Nha).


186

100 90 80 70 60 S 50 40

--

30 20

10 0

Biomass Sorghum

Fibre Sorghum

Fibre Sorghum

Sweet Sorghum

Fibre vs Sweet

Biomass vs Fibre

Fig. 3. Comparison of dry matter productivities between Biomass Sorghum (MN 1500) and Fibre Sorghum (H128) and between Fibre Sorghum (H128) and Sweet Sorghum (Keller) (in percentage terms).

According to these results, Sweet and Biomass Sorghum presented, globally, higher productivities than Fibre Sorghum. In all the experimental years, highest productivities, in all the fields, were achieved at 193 f 9 days after sowing, regardless of all the different sowing dates. Before 193 days, biomass is still in the process of growing and productivities were lower. After 193 days, the productivities lowered significantly mainly due to the loss of the leaves and to the degradation of the non-fibre components of the stems, specially the sugars. Depending on the sowing date, the 193 f 9 days after sowing corresponds to a period that range from 14" October to 31d December.

3.2. Biomass Quality Table 1 shows the results obtained concerning the moisture content, the ash, nitrogen and phosphorus contents, of the sorghum biomass obtained in the experimental fields, along the experimental years. Table 1. Moisture content, ash content, nitrogen content and phosphorus content of Sorghum biomass obtained in the experimental fields at harvest (0 = 0 kg Niha; Low = 60 or 75 kg N/ha; High = 100 or 120 kg Niha). Sowing dates Moisture (%)

Nitrogen (% dry matter) Phosphorus (% dry matter)

Low

High

64i7

~

23rdMarch 9" May

6.2k0.8

4.1k0.8

3.610.9

2ndJune

8.2i0.3

6.2+0.2

4.8i0.2

~

Ash (% dry matter)

0

23rdMarch 2"d June

231d March

~

91hMay

0.3 f 0.2

Znd June

0.9 f 0.2

231d March 2"d June

0.2 i 0.1

~

Moisture content decreased along the growing season, ranging from 57% to 71% at the time of harvesting. Moisture content at harvest was not influenced by the sowing dates


187

and by the level of N-fertiliser and there were no differences among the subspecies of sorghum. However, moisture content, at harvest, was influenced by the climatic conditions: drought autumns leaded to lower moisture contents and rainy autumns to higher moisture contents, in the biomass. Ash content decreased along the growing season attaining the lowest values 193 f 9 days after sowing. No significant differences were observed among ash contents of sweet, fibre and biomass sorghum. However, significant differences were observed among plants obtained from fields with different levels of N-fertiliser. In fact, plants from fields with no application of nitrogen showed significantly higher ash content than plants from fields with low (60 or 75 kg Nka) and high N-application (100 or 120 kg N/ha). In the fields with no N-fertiliser application, plants can only accumulate and absorb nitrogen from the soil reserves. As a result, the production of proteins, nucleic acids, chlorophyll and other biomass molecules, by the plants, is more limited, and, a lower synthesis of glucides through photosynthesis can be observed. Consequently, lower productivities and lower organic matter contents might take place, as it did in the experimental fields at Caparica. Fields sowed between 23rdMarch and gthMay, didn't show significant differences in terms of the ash content of the biomass material, at harvest. But, plants from fields sowed at 2"d June, showed a significant higher ash content than fields sowed earlier. Probably, due to the late sowing date, but also to the disadvantageous climatic conditions observed in this year, the percentage of glucides synthesised was lower than in other experimental years, thus explaining the lower organic matter content and also the lower productivities. Nitrogen content decreased along the growing season getting lowest values 193 f 9 days after sowing. No significant differences were observed among nitrogen contents of sweet, fibre and biomass sorghum and among plants obtained from fields with different levels of N-fertiliser. Fields sowed between 231d March and gth May, didn't show significant differences in terms of the nitrogen content of the biomass material, at harvest. But, plants from fields sowed at 2"d June, showed significant higher nitrogen content than fields sowed earlier. Again, the lowest development of the plants from these fields resulted in a higher accumulation of minerals. As for nitrogen and for ash contents, phosphorus content decreased along the growing season getting lowest values at harvest. No significant differences were observed, in terms of phosphorus content, among sweet, fibre and biomass sorghum and among plants obtained from fields with different levels of N-fertiliser. No significant differences were also observed, among plants obtained from different sowing dates. Sorghum is characterized by low nitrogen and phosphorus requirements leading to less nitrogen and phosphorus fertilization inputs. Table 2 shows nitrogen and phosphorus offtakes observed in the experimental.


188

Table 2. Nitrogen and phosphorus offtakes observed in the experimental fields (0 75 kg Niha; High = 100 or 120 kg Niha). Sowing dates 231dMarch

Nitrogen offtake (kg/ha)

~

91hMay

2ndJune 231dMarch

Phosphorus offtake (kg/ha)

~

9" May

2"d June

=

0 kg N/ha; Low

=

60 or

0

Low

High

27

98

115

50

40

83

46

57

63

15

13

17

According to these results, sorghum depletes the soil nitrogen reserves in ca. 27 kglha, when no nitrogen is applied to the soil. Application of low levels of nitrogen fertilizer into the soil (60 or 75 kg Nha) can also contribute to a depletion of the soil N-reserves. In the fields were a high level of N-fertilizer was applied, 100 or 120 kg N/ha, the Nexported by the plants was approximately the same what was applied, not causing a depletion of N-resources but also, not contributing to the contamination of soils and waters with an excess of N-compounds. P-applied (120 kg P.ha-') was considerably higher than what was uptaken by the plants. Although phosphorus presents low mobility in soils, in the perspective of a sustainable development, P-fertilizer should be applied in balanced amounts. This means that, in the experimental fields at Caparica half of the Papplied would be satisfactory for the crop growth. Nitrogen fertilizer use efficiency was obtained as the slope of a linear regression between the aerial dry matter obtained at harvest and the nitrogen fertilization applied (Fig. 4). I

35000 30000 25000

20000 15000 10000

0

20

40

60

80

100

120

Nitrogen fertilizer applied (kglha)

Fig. 4. Nitrogen fertilizer use efficiency of Sorghum plants obtained in Caparica.

According to Fig. 4, the nitrogen fertilizer use efficiency was very high, 173 kg dry biomass per kg N-fertilizer. Fig. 4 also shows what was previously observed in Fig. 2: that there were significant differences among productivities obtained in fields with different levels of nitrogen fertilizer.

189


3.3. Potential Energy Production The potential energy produced during combustion of the harvested biomass was evaluated as the productivity (t/ha) x gross heat of combustion (GJ/t) (Table 3). Table 3. Gross Heat of combustion of Sweet, Biomass and Fibre Sorghum and the Potential energy production (GJiha) of Sorghum biomass obtained in fields with different nitrogen application levels. Gross Heat of combustion (GJit)

0 kg N h a

60 kg Niha

120 kg N h a

Sweet Sorghum Biomass Sorehum

12.3 12.9

176 60

326 93

484 118

Fibre Sorghum

11.1

Potential Energy Production (GJiha)

210

In terms of gross heat of combustion, there were no statistical significant differences among subspecies of sorghum, although fibre sorghum presented a lower value. There were also no significant differences among plants obtained in fields with different levels of nitrogen and among different sowing dates. Then, in terms of the potential energy production, differences among the fields were observed only due to the differences in the productivities. According to this, highest values were obtained in the most fertilized fields (120 kg Nha) and with sweet sorghum. Biomass sorghum presented the lowest values but these results were obtained only in one experimental year, in 1997, when the sowing date was at 2”dJune, and when productivities were remarkably low. 1997 year results were not considered for the calculation of the potential energy produced by fibre sorghum. Considering the results obtained for sweet sorghum with a N-fertilizer application of 120 kg Nha, 9000 1 of gasoline, 10000 1 of diesel oil or 15 t of coal, could be saved. Table 4 shows the potential energy produced during conversion of sorghum sugars into fuel ethanol. This potential energy was determined considering that the heating value of ethanol is 26.9 MJ/1 and that the theoretical ethanol yield (ma) can be calculated from the formula [I]: Total sugar content (“3 dry matter) x 6.5 (conversion factor) x 0.85 (process efficiency) x total biomass (tha dry matter). Table 4. Saccharose + Glucose content of Sweet, Biomass and Fibre Sorghum and the Potential energy production (GJha) from Sorghum biomass sugars, that can be obtained from fields with different nitrogen application levels. Saccharose + Glucose (“hdry matter)

0 kg N h a

60 kg Niha

120 kg Niha

Sweet Sorghum

31

91

129

21 1

Biomass Sorghum

36

20

36

60

Fibre Sorehum

35

Potential Energy Production (GJiha)

102

190


As for the gross heat of combustion, in terms of the saccharose + glucose content, there were no statistical significant differences among subspecies of sorghum and there were no significant differences among plants obtained in fields with different levels of nitrogen and among different sowing dates. Then, in terms of the potential energy production, differences among the fields were observed only due to the differences in the productivities. According to this, highest values were obtained in the most fertilized fields (120 kg N/ha) and with sweet sorghum. Biomass sorghum presented the lowest values but these results were obtained only in one experimental year, in 1997, when the sowing date was at 2"dJune, and when productivities were remarkably low. 1997 year results were not considered for the calculation of the potential energy produced during conversion of fibre sorghum sugars into fuel ethanol. Considering the results obtained for sweet sorghum with a N-fertilizer application of 120 kg Niha, 4600 1 of gasoline, 5000 1 of diesel oil or 7.4 t of coal, could be saved. It should be mentioned that, beyond this high potential energy production from sorghum sugars, a significant amount of lignocellulosic material is left as bagasse. This bagasse can be used for pulp production or the cellulosic fraction can be fermented, thereby increasing ethanol yields, and consequently energy yields, by at least 40%.

3.4. Potential Pulp Production

The potential pulp produced after the harvest of the biomass was evaluated as the productivity ( a a ) x fiber content ("3 dry matter)/l00 (Table 5). Table 5. Potential pulp production of Sorghum biomass obtained in fields with different nitrogen application levels. Potential Pulp Production (tiha)

Fibre (%

matter)

0 kg N/ha

60 kg N/ha

120 kg Nlhb

Sweet Sorghum

44

5.7

10.8

15.6

Biomass Sorghum

38

1.7

2.1

3.1

Fibre Sorghum

51

9.4

In terms of the fibre content, fibre sorghum presented a higher value than biomass and sweet sorghum, although differences among subspecies were not statistically significant. There were, also, no significant differences among plants obtained in fields with different levels of nitrogen and among different sowing dates. Then, in terms of the potential pulp production, differences among the fields were observed only due to the differences in the productivities. According to this, highest values were obtained in the most fertilized fields (120 kg N/ha) and with sweet sorghum. Biomass sorghum presented the lowest values but these results were obtained only in one experimental year, in 1997, when the sowing date was at 2ndJune, and when productivities were remarkably low. 1997 year results were not considered for the calculation of the potential pulp produced by fibre sorghum.


191

4. Conclusions

Productivity was affected by the sowing date and by the level of N-fertilization. Better results were obtained in the fields sowed between 25'h March and lst May and productivity increased with the increase in nitrogen input. Among subspecies, Sweet and Biomass Sorghum presented, globally, higher productivities than Fibre Sorghum. In all the experimental years, highest productivities, in all the fields, were achieved at 193 9 days after sowing, regardless of all the different sowing dates. In terms of moisture, ash, nitrogen and phosphorus contents, there were no significant differences among sweet, fibre and biomass sorghum. Moisture and phosphorus contents were, also, not affected by the sowing date and by the Nfertilization. In terms of ash content, fields with lowest productivities, namely fields where no N was applied and fields sowed after 2"d June, resulted in a higher accumulation of minerals. For the nitrogen content of the plants, no significant differences were obtained among plants cropped from fields with different N-fertilizer. In relation to this parameter, there were also no significant differences among plants obtained from fields sowed at different dates, except for the plants obtained from fields sowed at 2"d June that presented a significant higher value. In the perspective of a sustainable growth and development, nitrogen fertilizer was applied in balanced amounts not contributing to the contamination of water resources with an excess of N-compounds, like, for example, nitrates. But fields where no nitrogen was applied or where 60 kg N/ha were applied, led to the depletion of the soil N-reserves. Calculus of the nitrogen use efficiency, with the experimental results obtained at Caparica, showed that 173 kg dry biomass can be produced per kg N-fertilizer. Compared with other crops, this value is higher, characterizing Sorghum as a low N-input crop. More essays with a higher N-fertilizer application, 180 and 240 kg N/ha, for example, should be experimented, in order to access, if after 120 Kg N/ha, the N-use efficiency show a decay. P-fertilizer application was higher than the amount uptaken by the plants, showing that sorghum can be characterized by low phosphorus requirements, thus leading to less P-fertilizer inputs. Considering the potential energy production and the potential pulp production, highest values were obtained in the most fertilized fields (120 kg Nha) and with sweet sorghum, due to the higher productivities obtained in these fields. So, in the experimental conditions at Caparica, sweet sorghum crop showed a better potential as an industrial and energy feedstock, and to get better results, economically and environmentally, fields should be sowed between 25th March and 1st May, with a high nitrogen input (120 kg Nha).

*

192


References [ 11 El Bassam N. (1998) Energy Plant Species. James & James (Science Publishers) Ltd. [2] Duarte MP, Fernando AL, Guimarfies H, Amparo V, Alves L, Oliveira JFS. (2001) Study of sweet and fibre sorghum crops in Portugal. Effect of climatic conditions and sowing date on the final productivity and on the quality of biomass. In S. Kyritsis, A.A.C. Beenackers, P. Helm, A. Grassi, D. Chiaramonti (eds) Proceedings of the 1st World Conference on Biomass for Energy and Industvy, Seville, Spain, 5-9 June 2000, James & James (Science Publishers) Ltd, London, United Kingdom, 361-364. [3] Duarte P, Fernando A, Oliveira JFS. (1996) Analysis of correlation between productivity and biochemical parameters for the sweet sorghum in Portugal. In Proceedings of the First European Seminar on Sorghum for Energy and Industry. Toulouse, France, 1 - 3 April, 243-247. [4] Duarte P, Fernando AL, Alves L, Amparo V, Silva S, Guimaraes H, Oliveira JS (1998) Characterization of Sorghum potential as an industrial and energy feedstock - the influence of crop management. In H. Kopetz, T. Weber, W. Palz, P. Chartier, G. L. Ferrero (Eds) Proceedings of the 1Oth European Conference and Technology Exhibition Biomass for Energy and Industry, Wiirzburg, Germany, 811 June, C.A.R.M.E.N., Rimpar, Germany, 1046-1049.

PROCESS DYNAMICS OF FOSSIL STEAM POWER PLANTS INDUCED BY THE INTEGRATION OF TRANSIENT SOLAR HEAT

v. SCHERER(’), K. ROTH(’), M. ECK(*) (I)

Department of Energy Plant Technology, Ruhr-Universitaet Bochum, Bochum, Germany (2)GermanAerospace Center (DLR), Stuttgart, Germany

In the current paper a concept to integrate solar steam generated by parabolic trough collectors into a fossil fueled power plant is being investigated. Feed water preheaters are one favourable location to integrate the external heat into the water steam cycle. Since the addition of the external energy causes less steam consumption in the high pressure and low pressure preheaters, less steam is extracted from the steam turbine. This leads to an increasing output of electrical power, which is a profitable product at the energy stock markets. The occurring time dependent processes are studied on a numerical basis, using the dynamic process simulation software APROS from VTT. With APROS a model of the parabolic trough collectors (62 MW maximum thermal output) was built-up, adapted and parameterized with typical lay-out data. Using meteorological data from southern Europe the time dependent steam generation was simulated. The collector model has been connected with a comprehensive APROS-model of an existing fossil steam power plant with a nominal output of 393 MW. The results show, as expected, that the steam generated by the solar collectors varies directly with the irradiation present. In July the power plant output increases at noon by 17.5 MW from 393 MW to 410.5 MW with a maximum gradient of 0.3 MW/min.

1. Introduction and Purpose In order to reduce the emissions of greenhouse gases like COz and to promote renewable energy a specific percentage of the electricity has to be generated by renewable energy sources in some European countries. The legislation in Italy, for example, requires 2 % nowadays, rising to 4.4 % in 2012. A promising technique to meet these goals is to employ a hybrid power plant, where a renewable energy source is added to a conventional and fossil fueled power plant cycle. The great advantage of this concept is, that it can be realized as a retrofit to existing power plants. Solar thermal plants are favourable to be located in southern European countries, that are offering a high insolation. Such kind of power plants, for example with parabolic trough collectors heating a synthetic oil which transfers the heat via heat exchangers to a conventional steam turbine cycle, are operating since many years and have proven their technological reliability. Direct solar steam generation is a possible improvement of the parabolic trough power plants vaporizing water directly in the absorber pipelines and eliminating costly equipment like the heat exchanger and the oil pump. This solar generated steam can be integrated into the water steam cycle of a fossil fueled power plant [ 11. A schematic process diagram is shown in Fig. 1. By this means less steam is consumed by the high pressure and low pressure preheaters. One or more steam extraction lines can be closed and the steam can be used for a very fast and dynamic additional power generation in the steam turbine. The power plant’s efficiency is improved simultaneously. Since the integration of the transient heat sources will lead to substantial dynamic energy shifts within the power plant’s water-steam-cycle, one objective of the developed automation concept is to keep the boiler inlet feed-water temperature constant to avoid any effects on the complex boiler control. The description and analysis of these shifts together with a cycle evaluation using an unsteady process simulation tool are the content of the current paper. 193

V. Scherer. K. Roth & M Eck

194

External (transient) Heat

Fig. 1. Schematic steam power plant with external heat source.

2. Process Simulation Software APROS

The time dependent processes are studied on a numerical basis, using an unsteady process simulation software. Such software tools are based on physical principles and empirical correlations and solve the one-dimensional unsteady conservation equations for mass, energy and momentum [2]: mass balance:

dAp -+-

aApv

az

at

momentum balance:

dApv ~

dAph 0

energybalance:

at

at

=o

+

dApv2 ~

aZ

dApvh

-I--

az

=

dAp += s, az

s,

(3)

On the basis of a detailed software evaluation [2,3,4], reviewing more than 15 simulation tools, the unsteady simulation software APROS from the finish supplier VTT was selected. The Advanced PROcess Simulation environment APROS is one of the very few software tools, that provides a 6-equation model for calculating the conservation equations for water and steam separately. This is of certain interest for the calculation of so called zero flow conditions appearing during start-up and shut-down when the flow velocities decrease to zero. Most of the other tools provide 5-equation models only, neglecting the momentum differences between the liquid and gaseous phase. Further advantages are the various interfaces providing several alternatives for connecting external software to the simulation programme like the IAPWS-IF97 for the thermodynamic properties of water and steam [6]. A long list of reference users working with APROS especially in Europe is another advantage. The number of available power

Process Dynamics of Fossil Steam Power Plants

195

plant components is very large and nearly every plant type from fluidized bed boilers to even coal gasification combined cycles including automation and electrical systems can be modelled.

3. Parabolic Trough Collectors and Solar Irradiation In the current paper parabolic trough collectors with direct solar steam generation are employed to supply steam to the preheating section of a conventional power plant. Although other solar thermal concepts feature higher concentration factors and higher system efficiencies the parabolic trough technology has been chosen because it is a proven technology which has demonstrated its reliability in large scale applications (80100MW) [l]. Tracking the sun from sunrise to sunset the cylindrical parabolic mirrors concentrate the sun’s radiation on the black absorber tubes along their focal line transforming radiation into heat. In these absorber tubes water gets vaporized and superheated to temperatures of more than 400°C.

Fig. 2. Parabolic trough collector LS-3 (schematic) [7].

Fig. 2 shows a typical parabolic trough collector. The aperture area is about 545 m2 with a length of 99 m and the reflectivity is more than 97% [7]. Favourable locations within Europe to operate solar power plants are southern countries like Spain, Italy or Greece. Because of the Italian legislation having a special interest in promoting renewable energy sources the location Greater Bari in southern Italy was chosen. The yearly global irradiance is about 1600-1750 kWh/m2a as it can be seen in Fig. 3. [8]. The yearly direct normal irradiance, that is used by the parabolic trough collectors, is up to 2200 kWh/m2a. In order to design and simulate a parabolic trough collector the monthly average values of irradiation were chosen at first.

196

V. Scherer, X Roth & M Eck

In Fig. 4 the average dkect normal irradiance of the last ten years in greater Baai can be seen for the months January to December. The highest maximum imadiance o c ~ w s in July at noon (884 W/m*) and the lowest is observed in January (349 W/ma).The parabolic trough collector plant which is the basis BOP the current study is not an existing one but is an ~ p ~ r o to a csimilar ~ existing plants and ~nvesti~a~ions in the field of direct solar steam generation [I].

Fig. 3. Global imdiance in Europe per year [S].

Fig. 4. Average direct normal itradiance in Greater Bari.


197

4. Reference Plant and Parabolic Trough Model In order to examine the feasibility of integrating external solar heat into the water steam cycle of a power plant, a model based on an existing reference plant was defined. This reference plant is a typical fossil fuelled conventional steam power plant with an output of 393 MW and 7 .feed-water preheating stages. Using APROS a comprehensive model of this power plant was built-up, adapted and parameterized. Especially the parameterizing is a very time consuming step. A great amount of lay-out data like geometries, isometries, geodetic elevations, thermodynamic data, valve characteristics and automation concepts have to be integrated into the model. In Fig. 5 a more detailed APROS view of the plant’s high-pressure preheater section is illustrated as an example. The automation and measuring concept is not shown. The power plant has been subdivided into more than 500 single components. For each component the one dimensional unsteady differential equations for the conservation of mass, momentum and energy are solved. Heat transfer, heat capacity of solid walls and two phase flow phenomena are taken into account. On the basis of this model several different kinds of plant configurations with or without an external heat source have been simulated. Careful calibration of the model has been carried out to meet the steady-state design and guarantee values. A parabolic trough collector was modelled in APROS. Design values calculated by steady state simulations and taken from the literature were used [ 1-91. The generation of steam with a heat capacity of 60 MW at noon in July was set as boundary condition for the collector design. The simulation is done with parabolic collectors of the type LS-3 (see Fig. 2) with a length of 100 m. Ten of them are added up to a 1000 m collector line. According to this there are 15 lines in parallel necessary to provide a peak load of 60 MW. The collector efficiency is assumed to be constant at 67 %. This simplification is of sufficient accuracy for the simulations carried out in this paper but will be corrected in future work. The collector feedwater pump control guaranties that the water pumped through the absorber tubes gets vaporized and superheated to a constant temperature of 380°C.

5. Integration of the Parabolic Trough Collectors and Power Plant Simulation The concept of integrating the parabolic collectors into the cycle is illustrated in Fig. 6. The temperature level of the solar steam is high enough to be integrated into the highpressure preheater section and to preheat a part of the boiler feedwater. Thus the preheater’s steam consumption is reduced and the steam remaining in the turbine can be used for an additional power generation. The condensate leaving solar preheater 1 is still at a temperature level to preheat a part of the feedwater in the low pressure preheating section (solar preheater 2). The main part of the feedwater is heated in the preheaters 1 to 4 by extraction steam conventionally.

198

V. Scherer, K. Roth & A4 Eck

Bypasses with solar preheaters are integrated into the water steam cycle of the power plant as shown in Fig. 6. In the solar preheater 1 the superheated steam gets condensed via heat transfer to the feedwater. The feedwater is switched into the bypass by the main control valve MCV. The solar heat and correspondingly the feedwater mass flow through solar preheater 1 is permanently varying with a maximum at noon caused by the changing irradiation in the course of the day.

Fig. 5. High pressure preheating section in APROS.

Fig. 7 illustrates that the water at the outlet of the parabolic trough (PT) gets warmed up in the early morning (t=5h), is then vaporized (t=6h) and finally superheated to a constant temperature of 380°C. It is obvious that the PT outlet temperature follows quite closely the irradiation curve. In parallel to the irradiance the feedwater mass flow through the parabolic trough increases till noon and then decreases. Because of mass release effects during the vaporization phase in the morning the mass flow characteristic through the absorber tubes shows a peak. Effects like these in combination with long dead times resulting from long absorber tubes and low flow velocities complicate the development of reliable automation concepts for the water mass flow through the absorber tubes in the parabolic trough collectors [9]. Therefore a pilot control concept based on measured irradiance data to pre-adjust the water mass flow has been applied. The steam generated in the PT is to be integrated into the preheating section of a conventional steam power plant. A control strategy which is illustrated in Fig. 8 has been developed that allows the addition of the external heat into the preheating cycle without affecting the feed-water temperature at the boiler inlet (25 1"C) and therefore without affecting the complex boiler control.


I

199

Time

Fig. 6. Integration of the parabolic trough collectors into the cycle.

This control strategy consists of 2 control loops employing the feedwater temperatures as controlled variables: In the first loop the feedwater temperature at solar preheater 1 (SP1) exit is measured by the temperature sensor TC1. The PID-controller opens the main control valve (MCV) and the feedwater is directed into the bypass successively in order to keep the given temperature set point. Because of the lower irradiance in the morning the parabolic trough supplies hot water instead of steam which leads to lower feedwater temperatures. In order to avoid any mixing of “cold” bypass feedwater with the mainstream, it gets discarded by the control valve CV as long as it is below the set point temperature. In the evening the irradiance decreases and the bypass feedwater is discarded again.

Solar Preheater 1

Parabolic Trough

\

/

J

Solar P r e h e a t e r 2

PT Pump

\ /J

1

1

1

Fig. 7. Irradiance, temperature and mass flow characteristics in July.

V. Scherer, K. Roth & M. Eck

200

,

y....................... W

.......

cv

~

0

.......................,, 8

P A ,$ .

.

. . .

~

Fig. 8. Bypass control concept.

Another temperature sensor TC2 can be found in the feedwater line downstream of heat exchanger PH7 and is part of the second control loop. This temperature measurement is used to control the steam extraction valves V1 and V2 to keep the mainstream feedwater temperature constant. The control concept in the low pressure preheating section is working similarly. Fig. 9 illustrates the mass flow characteristics in the high pressure preheating section during the course of a day in July. As it can be seen in Fig. 9, up to 179 kgls of feedwater can be guided into the bypass to be preheated by the solar generated steam. In consequence the mainstream feedwater mass flow decreases from 320 kgls to 141 kgls. Since the reduced feedwater mass flow needs less steam the controller closes down the extraction steam valves V1 and V2 as it is shown in Fig. 10. The processes appearing in the low pressure preheating section are very similar.

+Mass

5

7

9

11

13

Row through HP bypass

15

Time

Fig. 9. Mass flow characteristics.

17

19

21

20 1


20

1-

5

7

11

9

13

15

17

I9

21

Time

Fig. 10. Extraction steam mass flows into PH6 and PH7.

Closing the steam turbine extraction lines lead to a significant increase of the steam power plant’s power output as shown in Fig. 11. The power increase is following the solar irradiance characteristic nearly without any time delay - this is an essential result of the investigations carried out. The combination of integrating an external heat source into the cycle and closing the extraction steam lines leads to a rise of the steam power plant output from 393 MW up to 410.5 MW at noon in July.

1200

1000

$

800

.-

C

600 C

m

s

-

400

200

0 5

7

9

I1

13

15

17

19

21

Time

Fig. 11. Irradiation and power output characteristics.

This corresponds to a power increase of approximately 4.5 percent (17,5MW). Compared to the results in July, the potential power increases in the other months are lower, for example in January the maximum additional power rises about 6.7 MW. On the basis of

K Scherer, K. Roth & M. Eck

202

this simulation data an efficiency factor can be calculated, defined as a ratio of the parabolic trough thermal power output integrated into the cycle and the achievable electrical power output increase at the steam turbine's generator. This efficiency factor is about 28 % at noon. In addition to the use of average irradiance data for the simulation a single day in July showing strongly varying irradiation caused by clouds was selected as a test case. The simulation results can be seen in Fig. 12. Because the irradiance is temporarily too low to preheat the feedwater in the solar preheater 1 to 25 1"C, the feedwater is not led back to the mainstream but discarded. In these periods no additional power is generated in the steam turbine. Hence significant output losses are to be noticed in the example presented. However, the results favourably show that even fast changes in solar irradiation can be managed with simple control strategies. 404

400

-

.E

P

c

0 L

396

g -8 m c

I-

* 6

8

10

12

14

16

18

20

22

Time

Fig. 12. Simulation of a single day in July.

6 . Conclusion The integration of external solar generated steam into the water steam cycle of conventional steam power plants is a promising concept to promote renewable energy and as a consequence to reduce the COz emissions. The simulation results based on meteorological data of a location in Italy show a power increase following the solar irradiance characteristic nearly without any time delay. This can be managed even with simple control strategies. Since the peak of electricity consumption is usually to be found at noon, the additionally produced power matches very well the electricity demand of industrial countries. The efficiency factor defined as the ratio of the parabolic trough thermal power output integrated into the cycle and the achievable electrical power increase at the steam turbine's generator amounts 28% maximum.

Process Dynamics ojFossil Steam Power Plants

203

Acknowledgements

Building up the reference plant model in APROS was carried out within the Arbeitsgemeinschaft Turbomaschinen 2 (AG Turbo 2) and was financially supported by the BMWiiBMWA Germany (Ministry of Labour and Economics) and ALSTOM Power Support Mannheim. The authors would like to acknowledge the provision of detailed power plant data by the project partner ALSTOM. References [ l ] Eck M. (2001) Die Dynamik der solaren Direktverdampfung und iiberhitzung in

Parabolrinnenkollektoren.ed. VDI-Verlag, Dusseldorf. [2] Roth K, Scherer V, Behnke K. (2002) Wirtschaftliche Nutzung von transienten Waermequellen im Wasser-Dampf-Kreislauf. Proc 8. Statusseminar der AG Turbo, Koeln. [3] Roth K, Scherer V, Reinig G, Gebhardt B, Behnke K. (2003) Steigerung der Leistungsdynamik von Dampkaftwerken durch Einbindung externer transienter Warmequellen in die Speisewasservonvarmung. Proc VDI Fachtagung Fort schrittliche Energiewandlung und - anwendung, Stuttgart, 1746: 303-3 16. [4] Scherer V, Roth K, Behnke K. (2003) Enhancing the dynamic performance of electricity production in steam power plants: Integration of transient waste heat sources into the water steam cycle, Proc 7th International Conference on Energy for a Clean Environment, Lisbon. [ 5 ] Silvennoinen E, Juslin K, Hanninen M, Tiihonen 0, Kurki J, Porkholm K. (1989) The APROS software for process simulation and model development. Espoo, Technical Research Centre of Finland, Finland. [6] Accuracy of APROS’ (1998) Fast steam tables for real time applications. VTT, Technical Research Centre of Finland, Finland. [7] Deutsches Zentrum f i r Luft- und Raumfahrt; Institut fiir technische Thermodynamik; Germany, www.dlr.de/tt. [ 8 ] Suri M, Huld TA, Dunlop ED. (2005) PVGIS: a web-based solar radiation database for the calculation of PV potential in Europe. International Journal of Sustainable Energy 24 (2): 55-67. [9] Eck M, Steinmann WD. (2000) Dynamic behaviour of the direct solar steam generator in parabolic through collectors: A simulation study. Proc SolarPACES, Internat. Symp. on Solar Thermal Concentrating Technologies 10: 101-106.


DIRECTIONS FOR RENEWABLE ENERGY IN CANADA’S SMALLEST

PROVINCE A. TRIVETT

Department of Engineering, University of Prince Edward Island, Charlottetown, PEI, Canada Prince Edward Island’s economy relies almost entirely upon energy that is imported. Up to 200MW of electrical power is imported through an underwater cable from the province of New Brunswick, while the bulk of our energy is imported by tanker truck and ship in the form of oil, gasoline and propane. Political, economic and environmentalpressures are forcing us to re-think the sources and supply routes of our energy. A recent provincial government policy draft has suggested that Prince Edward Island, Canada, should be generating 10%-15% of its electrical energy from wind power by 2010. Since we are already generating close to 10% of our average power from wind, several industry representatives have proposed that this target is too low. It can be argued that we have sufficient wind resources to be planning the generation of 10-20% of the energy for the ALL the MARITIME PROVINCES from PEI wind resources, which would mean more than 200MW of electrical capacity from wind turbines. Doing so could make PEI a net energy exporter, but concern over the variable nature of wind power makes the local utility very concerned over the viability of this option. At the same time as we debate the large-scale renewable energy strategy, there is a new demonstration project to produce hydrogen fuel from wind generated electricity being planned. The hydrogen will be used in a small, remote community to provide energy storage in order to replace wind-derived energy at times of low wind speeds, as well as to supply wind-derived fuels for vehicles. This paper will review some of the energy realities of a small Canadian island, outline the PEI Wind-Hydrogen Village project, and discuss the practicalities of making major energy supply changes in a rural setting.

1. Energy Supply in Canada’s Smallest Province

As a nation, Canada is a net exporter of energy. We are the largest trading partner of the United States, and a major commodity which we supply to our neighbour is energy. In 2002, Canada exported 6% of its total electricity generation to the United States, while reciprocally importing only 3%. 55% of this electricity was produced from hydro electric dams. The remainder came from nuclear power or conventional coal and oil combustion

HI. While as a whole, Canada has a wealth of energy, and this is being utilized from a wide variety of sources, the picture is quite different for some of the regions and provinces which make up the Canadian Federation. Prince Edward Island is the smallest province in Canada, consisting of no more than 140,000 souls [l]. We are situated on combustionturbine

3%

(,

receipts

deliveries

3%

6%

7

steam nuclear 11%

r-

conventionalsteam

55%

22%

Fig. 1. Total electrical energy generation in Canada for 2002 [ 11.

205

A . Trivett

206

what is fondly thought of as the "garden in the Gulf ', an island in the Southern portion of the broad Gulf of St. Lawrence. The island is dominated by fishing and farming industries, with a small, but growing manufacturing sector. There is no commercial energy resource industry, despite recent high-profile exploratory wells drilled in hopes of finding natural gas or oil. At the present time, the prospect of a home-grown oil and gas supply is very slim. The electrical supply of our energy is mostly from imports through an underwater cable to the power grid in New Bmnswick. Fig. 2 shows the breakdown of electrical energy generation for the province of PEI. The Statistics Canada source does not include wind generation in its database, so the wind component of our power base is estimated using data from the number and size of wind turbines at the PEI Energy Corp site in North Cape. Steam generation is a very small component of the total due to the electrical utility's policy of maintaining a 60MW generation plant in Charlottetown as a backup facility. This plant is operated primarily on extremely cold winter days when the supply from New Bmnswick is in highest demand locally in New Bmnswick. A 38MW gas turbine facility is also maintained for backup power generation. PEI 8 NB Energy Generation

'%

Total conventional steam ge"Wall0" 2%

f

Total hydro generation 11 4%

)

Total rkceipls wind generation 9 1% 0 4%

Fig. 2. Electrical energy generation on PEI: the vast majority of power is purchased from off-island sources in New Bmnswick. The graph to the right shows the break-down of the combined New Bmnswick and PEI electrical generation [ 11.

It is clear from Fig. 2 that the vast majority of our energy is not under control of our governmental or corporate entities. When looking at the source of this power from New Brunswick, it is apparent that the mix of energy comes from primarily oil burned in one of several stationary combustion plants. The fuel for the generation plants is from oil imported by ship into the port of St. John, New Brunswick. 2. The Total Energy Demand

Electrical power generation is not the complete picture, however. The use of electrical energy from the power grid is not as great as the demand we have for fuels used separate from the electrical grid. The energy from electrical grids is only a minor part of the total.

Directions for Renewable Energy in Canada's Smallest Province

207

Fig. 3 shows the total use of energy from all sources, for all uses. The consumption of transport fuels dwarfs our use of electricity. Energy Consumption on PEI 30000

25000

Fig. 3. Total energy consumption on PEI. The component of electrical consumption includes all grid power imported directly from New Brunswick, which, from the previous figures, 60-70% originates from burning

fossil fuels [l].

A closer look at our use of petroleum products is important to understand what actions can be taken in the future. The Island has no public transport system that is island-wide. In recent months, a very limited bus system has been offered within the capital city, Charlottetown. Ridership has not been very high in the early stages. The only means for travelling for the vast majority of residents is by private automobile. Fig. 4 shows the seasonal changes in the use of different fuels. Gasoline and Diesel fuel are in much higher demand in the summer months as a result of our very large influx of tourists between June-September. Tourists to the island travel by automobile almost without exception. The businesses on PEI are oriented towards the tourist trade to a large degree, and thus the demand for fuel from the businesses coincides with the peak tourist season, and in some businesses, energy intensive activity leads the tourist season by a few months as businesses prepare for the summer. In contrast to the high summer demand for transport fuel and business-related demands, we have a very cold winter climate and this creates a high demand for home heating fuel from November-March. At the same time, fewer residents on the island and

208

A . Trivett

less travel reduces the consumption of gasoline and diesel fuel. The combined effect of these two factors tends to smooth some of the seasonal variations. The peak season for total energy demand is, as a result, in the spring months, March-May.

Mar-97

Mar-98

Mar-99

Mar-00

Mar-01

Date

Fig. 4.The seasonal variation in usage of different fuels on PEI. The cold climate results in a very high demand for home heating oil in the winter. Our large tourism industry results in high demand for gasoline and diesel fuel from May-September [ 11.

3. Alternatives for the Future

3.1. Wind Power At the same time as liquid fuels are very expensive, the public perception of wind power is positive. Maritime Electric, the private utility which distributes all of the electricity to PEI, offered customers a chance to purchase blocks of “green” power from the PEI Energy Corporation’s wind turbines at North Cape in 2001. 389 Residents and 32 business customers voluntarily bought all of the 1097 5OkWmonth blocks for an additional 35% premium to support the use of wind power. This was widely seen as a strong indication of support. The Island has a proven wind resource, with capacity factors of up to 42% at North Cape, and potentially similar factors at other sites across PEI. This resource can be a benefit through electricity exports. There are presently 16 Vestas V47 turbines installed by the PEI Energy Corporation at North Cape. A prototype V90 was installed by Vestas Canada in the fall of 2003. Future expansion of this site is expected. Shallow waters surrounding PEI may make future expansion of ocean wind farms a practical alternative if land use issues become a concern.

Directionsfor Renewable Energy in Canada's Smallest Province

209

3.2, Wind and Hydrogen The greatest challenge for the adoption of wind power is the inherent variability of the supply. Numerous alternatives exist for complementary renewable energy sources that can balance the variable nature of wind energy. The alternatives have not been researched, and this must be considered to be a high priority for the engineering community. To this end, many groups worldwide are investigating the use of hydrogen (H2) as a component in new energy strategies. Implicit in environmentally-mindedH2 policies is that the fuel, Hydrogen, must be produced from zero-emission renewable primary energy sources. Demonstrating potential for generating significant amounts of H2 from renewable resources such as wind or solar power is an important aspect of these programs. We see a future potential for H2 not only as a replacement for battery storage of wind energy, but as a means to use wind power for existing transport fuels. We have proposed to develop a small coastal community of 50-100 people demand in the range of 75-150 kW of electrical power (using PEI average estimates). If we factor in the fuels used for transport and home heating, this may amount to approximately 300-500kW of power to supply all their needs. Such a community would benefit from the installation of 3-5 lOOkW wind turbines. The turbines and associated H2 equipment capital cost must be compared with the alternative cost to the residents for continuing to purchase all their fuel and energy from outside of the community. In this light, 100 residents who may currently spend 1500-3000$per year, per person on all fuels (a figure which is on the low extreme of what is currently spent) could, in theory, finance installation of 1.5-5M$ of infrastructure for self-generated electricity and fuels. To be justified, the equipment would operate over a 10-15 year lifetime. If such a system is shown to be technically practical, the economics may also prove to be attractive to local residents. Such a community will serve to address not only the supply of electricity from renewable sources, but also the stabilization of supply from wind generators in addition to a means of replacing fossil fuels in transportation. Since, as was shown previously, the transportation and home heating use of fuels outweighs the use of direct electricity, a solution which serves all three needs is one worth studying. 4. Conclusions

The fuels we use are supplied to Prince Edward Island via a tanker which docks in Charlottetown approximately three times per month. Recently, while speaking to 10 and 11 year old school children at a local public school, I asked them if they knew where their energy came from. After some interesting guesses, I asked how many had noticed the particular tanker which docks regularly within sight of the school. All the children recognized the ship, having seen it numerous times, but none connected its regular routine with our driving habits and our energy use.

210

A . Trivett

A significant challenge for energy policy in Canada is making a direct connection between the source of supply and our consumption in the public. Being part of a seemingly infinite supply of electrical power and liquid fuel, regardless of price, can delay public action towards sustainable practices. For the first time in two generations, the price of gasoline has reached l.OO$/litre, a psychological price barrier never before crossed. All other petroleum based fuels are at record-setting high prices. Despite the opportunities for a shift in direction towards more sustainable energy future, there is no concerted effort from all parties to develop a coherent plan. There has been no ongoing research into energy alternatives since the late 1980’s. Projects, such as the North Cape wind farm, and a district heating plant in Charlottetown burning municipal waste are isolated projects. This is slowly changing. We have seen in the past year recognition from the provincial government of the value from R&D in renewable energy. The challenge for our future is to integrate the most appropriate energy solutions for a small Northern Island. Acknowledgements The author wishes to acknowledge support form the University of Prince Edward island office of Research, and well as the Dean of Science. References [l] Statistics Canada: Table 128-0002 - Supply and demand of primary and secondary energy, http://cansim2.statcan.ca/cgi-win/cnsmcgi.exe. [2] Statistics Canada: Table 127-0001 - Electric power statistics, monthly, http://cansim2.statcan.ca/cgi-win/cnsmcgi.exe. [ 3 ] State of the Environment, Published by the Prince Edward Island Department of Fisheries, Aquaculture and Environment, June 2003, available online at http://www.gov.pe.ca/photos/original/fae-soe-report.pdJ:

EXTERNALITIES AFFECTING THE VIABILITY OF WIND POWER FOR HYDROGEN PRODUCTION N.KASSEM Royal Institute of Technology, Dept. of Energy Technology, S-100 44 Stockholm, Sweden Off-peak electricity from wind power farms is a valuable and cheap source of energy, which is hlly competitive, with fossil fuel and nuclear power generation. Wind power is one of the lowest-priced renewable energy technologies available today depending on the wind site, the type of wind turbines, and project financing. Coupling of wind turbines to high temperature steam electrolyzers would have the potential of providing low cost hydrogen source. For offshore wind farms to be economic, they have to be of sizes greater than 150 MW using large turbines (> 1.5 MW). The study uses methods of risk analysis to assess the economic viability of wind power and novel electrolytic processses to evaluate the simultaneous effect of multiple input uncertainties on system output. The results show that stochastic variations in wind speed, electricity prices, and interest rate have direct impact on investment capital requirements, and hence on the cost of hydrogen.

1. Electrolytic Hydrogen from Renewables

Hydrogen is an alternative energy carrier, which would significantly contribute to the growth of sustainable energy systems. It is considered a renewable fuel if produced from renewable energy resources such as hydropower, solar-and wind-energy. The opportunities for innovation and economic growth in hydrogen energy utilization are virtually untapped, and many industrial nations are dedicating their efforts to establish a leading position in this expanding field. A great deal of R&D work is conducted worldwide to utilize hydrogen as a new energy source in district heating, power utilities, and transportation sectors. Hydrogen can be produced in virtually unlimited quantities from renewable resources. Thus it is a primary element to sustainable growth in power generation and transportation systems. Conventional technologies for hydrogen productions [ 1-31 include primarily partial oxidation of heavy oils, steam reforming of light hydrocarbons, coal gasification, and water electrolysis. Electrolysis is an energy intensive process, where the cost of electricity has a direct impact on hydrogen economy. Although not economically competitive, electrolytic hydrogen provides high product purity and flexible operation modes. Electrolysis plants may operate over a wide range of production capacities, thus they are more convenient for electricity load leveling either at the generation site or at any location directly supplied from the grid. Wind power is an emerging technology with great potentials for electricity generation and has been grown steadily, over the last 10 years. The main growth, however, has been in Europe, where government policies and high conventional energy costs favor the use of wind power. Almost 3 1 GW, of wind power were already installed worldwide by the end of 2002 [4]. The major challenge in the process of wind energy assessment is how accurately the airflow and hence the error in wind energy can be predicted and how the total capital investments are estimated. Failure to account for uncertainties would result in deterministic estimates that tend to overstate performance and underestimate costs. Accounting for uncertainties in emerging technologies, requires applying methods of uncertainty analysis in any life cost assessment. This study provides an assessment of the risks and pay-offs associated with electrolytic hydrogen production using electricity from offshore wind farms and heat from the combustion of bio-fuels. 21 1

N. Kassem

212

2. Offshore Wind Power

Wind power is a potential renewable energy resource in many countries with substantial installed capacities, Germany (12GWe), Denmark (2.88GWe), Spain (4.8GWe), and India (1,2GWe). The U.S. Department of Energy (DOE) recently launched the Wind Powering America initiative with goals to increase federal use of wind energy to 5 % by 2010, and to satisfy at least 5% of the nation's electricity demand using wind power by year 2020. As part of its strategy, the EU-Commission has launched a target to increase the Union's installed capacity of wind power from 7 GWe in 1998 to 75 GWe by year 2010 [4]. Wind power is a source of clean, nonpolluting electricity, and at favorable wind sites, it is fully competitive with fossil fuel and nuclear power generation. The available wind resources at any given site are crucial in wind energy implementation. The minimum economic value for the average wind speed at a given location is about 6-8 d s , a condition that is satisfied in many places in the United States, Europe, and Asia. In Antarctic, however, the average speed for 75 YOof the year is about 3-4 folds of this value, which make wind power more cost-competitive. Finally, good wind sites are often located in remote locations far from areas of electric power demand. Most of these problems have been greatly reduced by properly sited wind farm or through technical development. Main drawbacks of wind power are attributed to the fact that it is an intermittent electricity generator and hence it does not provide power on an "as needed" basis. The potential power yield, expressed in Watt varies with the cubic power of the wind speed as given by Eq. 1. E, = 0.125 zr Ow2p V 3

Watt

(1)

In this study, however, the uncertainty in wind speed is accounted for through including a

1

XI0

random number generator in the velocity component of Eq. (1). The wind variation for a typical site is usually described by a Weibull distribution which has a probability density fimction given by Eq. (2). The statistical distribution of wind speeds varies from place to place around the globe, depending on local climate conditions, the landscape, and its surface. The Weibull distribution may thus vary in its shape factor k, and the scale factor c. If the shape parameter is exactly 2, the distribution is known as a Rayleigh distribution. Wind turbine manufacturers often use this distribution to present the standard performance figures for their machines The economics of wind power are compelling and the cost of installing and operating wind turbines has been fallen dramatically by almost two thirds between 1981 and 1995. The typical capital costs of large onshore, small onshore, and offshore wind farms in year 2000 is estimated between 1000 to 1700 $/kW. The annual investment requirement for achieving 10% of the world's electricity from wind energy is about $3 billion in 1999. It is expected however, to reach a peak of $78 billion in 2020 (green peace, October, 1999). These are only a fraction of overall global energy investments of $170-200 billion per year in 1999. Wind energy is a domestic, reliable resource that

Wind Power for Hydrogen Production

213

provides more jobs per dollar invested than any other energy technology including fossil and nuclear fuels. Employment implications of the 10% target are quite significant; and would create 1.7 million jobs in manufacturing, installation, and operation worldwide. 3. High Temperature Steam Electrolysis

Well-developed and novel processes for hydrogen production, based on both fossil and non-fossil fuels, are frequently reviewed in the literature, Winter and Nitsch [ 13. In this study, however, the main focus is on the High Temperature Steam Electrolysis (HTE), detailed in [5], using bio-fuels supplementary thermal source. The total energy requirement (AH) for the electrolysis process consists of an electrical energy part (AG) and a thermal energy part (Q=T.ds) to provide the heat demand of the endothermic electrolytic reactions. The basic idea of HTE-technology is that a substantial saving in electric power requirement can be achieved by increasing the operating temperature. Up to 38% of the total energy demand can be satisfied by a cheaper energy source than electricity, thus resulting in a significant improvement in the overall efficiency of electrolytic hydrogen production. The standard free energy for water dissociation is A c O 2 9 8 =237.4 kJ/mol., equivalent to the standard potential E'29.9 ,1.230 V, while the standard enthalpy of the electrolytic reaction is -AH298= -286.4 kJ/mol, equivalent to standard potential of 1.483 V. The difference has to be supplied by other energy sources. The amount of heat needed (T m) for the endothermic electrolytic reactions can either be supplied by internal resistivity losses of the cell itself due to high current densities (300-500 mA/cm2) or by direct heat input from high temperature heat source. Compared with conventional electrolysis, the electrolytic decomposition of steam requires quite different technologies, whereby the liquid electrolyte is replaced by a special solid electrolyte having good ionic conduction properties. The solid electrolyte is usually in the form of a porous metal oxide ceramics, such as zirconium oxide stabilized with yttrium oxide (Zr02 - Y2 O,), which allow the oxygen ions to migrate by ionic conduction to the anode compartment. The operating temperature is governed by the electrolyte resistance and electrode polarization. However, the present achievements in thin-film technology allow the fabrication of electrolysis elements with electrolyte film thickness in the range of 10-50 pm. This is expected to lower the operating temperature of the HTE process by 100-200 "C. Under these conditions, the operating temperature is more likely to be determined by electrode polarization rather than electrolyte resistance. A basic requirement for the solid electrolyte ceramics is that its structure and composition would remain unchanged during electrolysis processes. The conditions for reliable HTE operations is discussed by Perfiliev [6] The electrode polarization, due to electric current flow, has significant effect on the electrode/electrolyte interface, where the oxygen activity decreases at the cathode while it increases at the anode. A schematic flow diagram of the HTE process is shown in Fig. 1. Saturated steam from renewable resources such as thermal solar power or waste heat, after being superheated in a heat recovery unit, is mixed with hydrogen recycle stream and preheated to 1100 "C (2012 OF). This takes place in ceramic conductor tubes in a bio-fuel -fired furnace before being fed to the electrolyser. At this temperature, only ionic conduction can cause oxygen ions to move across an ionic conductor. Single cells, produced from the ionic conductors arrayed in various geometrical forms, are connected

N. Kassem

214

in series or parallel to form a battery of cells. A stack of these batteries are hooked together to form a large-scale electrolyser. 4. Performance and Economic Models

Engineering models include both performance and economic models based on thermodynamic relationships, conservation equations, cost scaling, thermo-economic

Offshore Wind farm

Rectifiers

Seperator

0, Storage

0, compressor

Fig. 1. Schematic flow diagram for hydrogen production using renewable resources

relationships, and published cost curves for generic process areas. These models are mostly empirical in nature with the parameters being fitted using real or statistical data. The cost models must be compatible with performance model simulation results, which include key parameters needed to determine the total capital investment, running costs, and cost of hydrogen. In order to account for design and economic uncertainties, and to apply methods of risk analysis, a stochastic modeling capability is developed, which allows input uncertainties to be represented using probability distributions. The modeling capability provides an automatic random sampling of model inputs/outputs, and a systematic method to evaluate the simultaneous effect of multiple input uncertainties on key plant outputs. It is also possible to apply methods of statistical and sensitivity analyses to model outputs, and to illustrate the results as cumulative distribution functions (cdf). Like any other modeling tool, the outcomes of the stochastic methods are very much dependent on the type and quality of the data and judgments provided by the user. A summary of performance and cost models used in this study is given below.

A. Electrolvser Model [7]

+ C, + CR

(Cinv),

= CA

C,

=

C,

=

c, + 6 I

=

pru

c, c,

A


215

B. Hydrogen liquefaction. & Compression [S] Fj, = 3 . 3 elf, F&

kW

= R T l n ( p , ) Q l q c I3600 kW

(G")lLq,c,p

= 25422(p)0.61

C. Hydrogen Storage [S].

(Ci,,)slo

= 9.17 1 exp ( f ( v > )

v = volume of H , storage

f ( v ) = nonlinear function of volume Models for estimating the cost of electrolytic hydrogen are available with varying degrees of detail and sophistication in the literature [7, 81. Generally, electrolysis plants are designed in modular form, which are not sensitive to the economy of scale. Design data for water electrolysis are readily available in the literature and can be adapted for the given model. Economic input data, however, are rather rare especially for large-scale operation of advanced technologies. 5. Economic Assessment

For power utilities and chemical industries, there are generally accepted guidelines regarding the methodology of developing cost estimates. The levelized cost of electrolytic hydrogen is determined based on the methods and assumption described in the TAG report of EPRI [9], which include typical cost estimate parameters. The cost of hydrogen includes the cost of compression, liquefaction, and storage, and the allowance for byproduct credits. The main cost items considered in this assessment are: 1. Facility Invest. Costs (FIC), i.e. equipment and machinery. 2. Indirect Costs (IDC) including engineering, installation, contingency, and administration costs. 3. Direct Costs (DC) including Feedstock, operation and maintenance (O&M) costs, and utilities. 4. General Administration: (GA) including insurance, taxes and inflation costs, royalties, and depreciation cost. All cost estimates are evaluated by applying a GNP price deflector factor of 1.32, for 1990198, obtained from economic indicators published in IEA 1999 edition. The inflation rate is taken as 2% and all costs are expressed in 1999 US$. With limited cost data, cost scale factors ranging between 0.5-0.8, are used to determine the approximate capital cost of the new plant. Indirect cost (IDC), which includes installation, piping, building instrumentation, auxiliaries, and structures are estimated at 2.0-2.5 of delivered equipment cost. Engineering, start-up cost, plant contingencies, and interest during construction were taken as 35 % of FIC whereas the working capital is estimated at 3% of FIC. The outside battery limits was estimated at 35 % of the inside battery limits. The contingency represents additional costs that are expected to occur and yet it is not explicitly included in main parts of cost estimates. It is a single large and least

216

N. Kassern

documented expense item. Contingency factors, estimated by conventional approach, are inadequate to quantify the risk of cost increases in new technologies and tend to be underestimated as discussed by Milanese [ 101. By-product credits, mainly oxygen, were taken into account and being estimated at $45 ton-' oxygen. The power generated by the wind farm is based on a standard 1.5 MW turbines, which have a capital investment cost of $1000-1400/kW installed.

6. Uncertainty Analysis Prediction of the future commercial-scale performance and costs of a new alternative technology are often based on limited data from small-scale version of the plant. These predictions are generally expressed in terms of single-point deterministic estimates based on the best guessed values of key performance, design and economic variables, while ignoring whatever uncertainty associated with them. Such estimates are inherently uncertain and generally provide misleading estimates, which can have serious implications on project planning and resources allocation Thus, the ability to conduct uncertainty (risk) analysis is of special importance in the context of ongoing assessment of emerging technologies, where technical and economic figures-of-merit are not wellestablished. The extent of uncertainty will vary from one parameter to another depending on the state of development of the selected technology, the level of detail of relevant estimates, and future market values. The extent and severity of uncertainties reflects the degree of confidence the designers might have towards input assumptions of the applied model. Uncertainty in key performance and economic parameters e.g. wind energy potential, feedstock prices, contingency factors, and interest rates would lead to uncertainties in key output variables such as overall efficiency, plant capital cost, cost of electricity, or cost of hydrogen. Studies of advanced power generation, prepared for EPRI by Hager and Heaven [ l I], include uncertainty analysis, which consider only uncertainties in cost related input parameters, in particular capital costs. The type of distribution chosen to characterize uncertainty reflects the amount of information available on a given variable. Probability distribution defines the range of values, which the process variables can assume, and the likelihood of occurrence of each value within the defined range. If data are not available, uncertainties can be expressed in terms of probability distributions based on the judgment of technical and economical experts. Compared to conventional sensitivity analysis, the development of estimate of uncertainties requires detailed thinking about the possible outcomes and their relative likelihood. This eventually would lead to anticipating a broader range of problems or system payoffs, that otherwise might have been overlooked. For practical problems with many variables subject to uncertainty, the combinatorial explosion of possible sensitivity scenarios becomes unmanageable. This type of analysis is inadequate and provides no insight into the likelihood of obtaining any particular results that might be of use in the decision making process. The main objective in using such tools is to assess the risks and the payoffs of the new concepts compared to conventional ones. To account for uncertainties in wind power based hydrogen technologies, a stochastic modeling tool is required. Two types of such tools are frequently used in the literature, namely the Latin Hypercube Sampling (LHS), where the distribution of a given input is divided into intervals of equal probability, and one sample


217

is taken from within each interval at a random manner. The second tool is the MonteCarlo Sampling (MCS) method, where samples from the input parameter distribution are taken randomly from the whole range. Detailed descriptions of both procedures are given by Ang and Tang [12]. Considering the advantages and disadvantages of both methods, the MCS method has been selected as a suitable stochastic modeling tool for conducting the required uncertainty analysis in this study. The, the stochastic modeling tool provides explicit and quantitative measures of uncertainty in key figures-of-merit, thus helping the analyst to focus the efforts on reducing their effect.

7. Results and Discussion The main focus of the techno-economic assessment carried-out in this study is to predict the uncertainties associated with the cost of electrolytic hydrogen from renewable resources. The main parameter values used in this analysis are given in Table 1. Table 1. Main parameter values ~

500 $/mz fm = 0.17 k = 2.0 p = 100 $/kW qc= 0.75 c,

=

c , = 5000 D,=64m c = 0.125 F=0.05 $/A Q = 10-70 metric ton H2/hr

A stochastic modeling capability using Monte-Carlo sampling technique has been applied to monitor the effect of input uncertainties on a number of performance and economic figures-of-merit of the combined offshore wind farm and high temperature electrolytic hydrogen plant. Production rates in the range of 10-70 metric ton/ hour of hydrogen ware considered suitable for the study, while the wind turbine capacity factor (CF) is assumed to be 0.4. The main input uncertainties considered are: wind speed, cost of electricity, and the interest rate. The uncertainty in wind speed is described by a Weibull distribution, Eq. (l), with parameters k and c being given in Table 1. The uncertainty in electricity and biofuel costs, and the interest rate are represented by normal distributions having mean and standard deviations as shown on the diagrams The probability density (pdf) and cumulative distribution functions (cdf) for the selected key input uncertainties are shown in Fig. 2 . The engineering models were run repeatedly, using values, generated from those distributions using Monte-Carlo sampling method. For multiple input parameters, the values of all input are sampled simultaneously during each computational cycle. The number of repetitions is equivalent to the sample size, which is selected based on the desired precision of the output distribution. The results are in the form of numerical values that can be treated statistically as sets of experimental values and are given in the form of cumulative distribution functions (cdf), as shown in the following figures. 7.1. Figures-of-Merits

Evaluation of the electrolytic hydrogen technology requires the access to quantified measures of the projected performance and cost parameters. These measures include

218

N. Kassem

k = 2.0

f

0.4 0.05

0.2

0

10

5

15

20

Wind s p e e d m/s

cs =

5 10 Wind s p e e d

m/s

0.01

0.03

15

20

0.00

0.4

ft

0.4

0.2

0.2 0

0

0.02

0.01

0.03

0.04

0

0 0 '

2

4

6

I

8

. 0

lntrest rate %

ft

0.2 0 0.8

0.4

= 2%

0.2

1

1.2

W p invest K$/kW

0.04

z 2

1

V 4

6

6

Interest rate %

p = 6% -3

0.02

cost of elect. COE $/kWh

Cost of elect. COE W k W h

1.4

2

4

6

8

W p invest. $/kW

Fig. 2. Probability functions of input uncertainties.

performance and economic figures-of-merits (FOM), which can be determined using the engineering model outputs. For each set of randomly selected input values, a set of FOM evolves at each Monte-Carlo simulation cycle. This procedure is iterated repeatedly until sets of sufficient size emerge with statistical properties that are meaningful to the decision-maker. Probability distribution functions of some relevant figures- of-merits are shown in Fig. 3. Here, the top left plot represents the cumulative probability of the generated power by a single wind turbine having CF= 0.4, and reveals that there is about 80% probability to get a power output up to 2 MW. This applies to the entire range of production rates. The plot on the top rights of Fig. 3 shows, however, the amount of wind energy generated per square meter of the swapped blade area of the turbine. It indicates an 80% probability to generate up to 800 W/m2 using 64 m-blade diameters. The next four diagrams represent the probability of capital investments for the main facility items namely, the electrolyser, wind turbine, liquefier, and bio-fuel (BF)-furnace. The diagrams, however, reveal that there is a great deal of uncertainty (big variance)) associated with the last three items, while the electrolyser shows a reduced degree of uncertainty (small variance). They also show that the electrolyser has the highest capital investment cost of about 4$/GJ H2 with a probability greater than 95%, while the BF-furnace has the lowest investment cost of about 0.06 $/GJ H2 with 50% probability. The capital investment of

219


both the wind turbine and liquefier are quit close to 0.6 $/GJ H2 with a probability of 75% and 50% respectively. In case of wind power investment, the probability distribution is almost linear indicating a higher degree of uncertainty. Of all the four production items, the capital investment of the liquefier showed some degree of sensitivity towards the production rates of hydrogen. The bottom diagrams show the total capital requirement TCR and the return-on-investment (ROI) expressed in $/GJ H2 produced. The diagram on the left reveals that the certainty in TCR, which is a measure of the cost of hydrogen (COH) lies in the range of 12-18$/GJ H2 with 90% probability that the expected value is about 15 $/ GJ H2. The ROI on the right has a certainty range of 8-12 $/GJ H2 with an expected value of 10$/GJ H2 at 90% probability. Both diagrams indicate that the effect of production rate on TCR and ROI is insignificant. 1

I

500

Wind power M W

1000

1500

Wind Energy swaped W/m2

0.9,

0.7

0.5

Wind power invest. $/GJ H 2

8

1 5

10

Electrol. invst. $/GJ H 2

0.2 0

0

0.2 0.4 0.6 Liquef. invest. $/GJ H 2

0.8

Tot. capit. requir. T C R $/GJ H2

0 0.02 0.04 0.06 0.08 B F Furnace invest. $/GJ H2

Ret. on invest. ROI $/GJ H 2

Fig. 3. Probability functions of some figures-of-merits.

7.2. Multiple Sensitivity Analysis

To assess the technical and economic viability of new process concepts, the designer need to get more information on how input uncertainties would affect some key figures-ofmerit. To achieve this aim, the designer should conduct some sort of sensitivity analysis

N. Kassem

220

to identify input uncertainties and to characterize their effects on process outputs. In this study, however, multiple sensitivity analysis was conducted by allowing process inputs to change randomly in accordance with their probability distributions, while observing their impact on a number of key economic figures-of-merit. The results of this study are illustrated in Fig. 4. The two plots at the top represent the effect of interest rate and the cost of electricity (COE) on the cost of hydrogen (COH) expressed in $/GJ at different production rates. The COH showed almost the same behavior with respect to both input parameters, but showed no distinct sensitivity towards the economy of scale. A gradual increase in the interest rate in the range of 2-6% resulted in a linear increase of hydrogen cost from about $15 to $20 /GJ H2. Beyond the selected range of interest rate, there is a dramatic decrease or increase of the expected hydrogen cost. A similar behavior is resulted, when the cost of electricity is allowed to change in the range of $10-$30/ MW, which is illustrated by the plot on the top right of Fig. 4.

-- I

2

4

6

5'

8

10

20

Cost of elect.

Interest rate %

30

I

0.1

'

I

2

4

6

8

I

-

Lu

2

Interest rate %

Interest rate %

z: 1

0.8,

0.1

n

I

prod. rate 10-70 ton/hr H2

A I

2

4

6

Interest rate %

0

C O E $1 MWh

8

0.01

1

2

4

6

Interest rate %

Fig. 4. Sensitivity analysis of some key figures-of-merits

1

a

22 1

Wind Powerjor Hydrogen Production

The next four plots illustrate the impact of the intere te on the facility investment costs . The diagrams show a gradual (FirC), in $/GJ H2, of the main fom production fac linear increase of about 25% in FIC as the interest rate increases in the range between 2In a ~ ~ i t i othe n ~facility cost (FIC) o f all items, except for the elecrolyser, showed ct sensitivity towards the economy of scale. ~ncreas~ng the production rates in the range between 10-70 metric tonshour, indicated by the direction of the arrow, would result in reducing the i n v e s ~ e n cost t by almost 50%. The electrolytic unit, however9is n o ~ a c~o ~n sy~ c t in e ~modules, thus i s not sensitive to the econorny of scale.

20

1s 10 S

0

7 6

8. a 4

a 3

8 2 1

10 20

30 40 SO

60 70

0

10

M e t r i c tonlhr H2

a

30

40

50

60 70

M e t r i c tonlhr H2 10

10 ‘

20

E

6 4 2

0 5 0

20

30 40

50

60 70

Metric ton/hr H2

10

20

30 40 SO

60 70

M e t r i c ton/hr H2

Fig. 5. Relative magnitudes of various investment costs.

The relative order of ~ a ~ of ~various ~ economic d e ~ ~ e s - o € - m eexpressed ~t, in $/GEE, for the p r o d ~ c ~ ~ofo hydrogen n in liquid (LEI21 and gaseous (GH2) forms are i ~ ~ ~ sin~Fig. a t 5e. ~Itemized capital such as facility investments (FIC), indirect versus the production rates in the first f v ~ o costs (BDC), and direct costs (DC) are stack plots at the top. Here the dominant item is the direct costs, since it include the cost of e ~ e c and ~ other ~ c feedstock. ~ ~ The capital costs o f the faci~~ties are shown in r n i d ~ ~ e plots and clearly indicate that the electrolytic facility has the most dQminantcost f o ~ ~ ~ by wind farm and ~ i ~ u e f a c t ~costs. o n Finally the two bottom plots compare the ~ e e ~ s t o c ~

~

N. Kassem

222

(electricity, bio-fuel and steam) costs indicating that the cost of electricity is the dominant cost, especially for liquefied hydrogen.

8. Conclusions 0

Wind power has a great potential for electricity generation and is fully competitive with fossil fuel and nuclear power generation; Improved design of wind turbines would enhance the economic viability of offshore wind farm as a source of cheap electricity; Characterization of uncertainty by their probability distribution provides more insights into performance and economic estimates compared to deterministic ones; Interactions among uncertainties in key performance and cost parameters are propagated through the model to yield an explicit indication of output variable uncertainties; The capital investment cost of the electrolytic hydrogen plant, and electricity cost requirement are the most dominant cost items in electrolytic hydrogen production.

Nomenclature A

AF CA Cinv COE COH CR CS Dw Ew

DC FIC GRR IDC I P Pr

Q RC ROI TCR

Total Electrode area Annuity factor Area related costs $/MWh Capital investment costs $ Cost of Electricity Cost of Hydrogen $IGJ Rectifier related costs $ Auxiliary related costs $ Wind turbine blade diameter m Wind power Watt Direct costs Facility investment costs $ Gross revenue required MM$/yr Indirect costs Total Current Amper Power kW Pressure ratio Flow rate kg H 2 h Running costs Return on investment MM$/yr Total capital requirement MM$/yr

U V WE WP c k cdf Ca > Cs

fm V

X

Cell voltage V Windspeed m!s Wind Energy Wh Wind Power kW Weibull factors Cumulative DF Constants Figure-of-merit storage volume m3 random number

Greek symbols p Constant parameter $lkW 6 Constant parameter $/A p air density qc Compressor efficiency Subscript Ele liq CmP

Electrolyser H2 Liquefier H2 compressor


223

References

[ 11 Winter CJ, Nitsch J. (1989) Hydrogen as an Energy Carrier. Springer-Verlag, Berlin. [2] Hammerli M. (1984) When will electrolytic Hydrogen become competitive? Int. J. Hydrogen Energy (IJHE) 9: 25. [3] Basye L, Swaminathan S. (1997) Hydrogen Production Costs - A Survey. Report Sentech, Inc Bethesda, MD. [4] The European Wind Energy Association, EWEA, Briefing report, October 2003. [5] Doenitz W. et al. (1980) Hydrogen production by high temp. electrolysis of water vapor. Int. J. Hydrogen Energy (IJHE) 5 : 55-63. [6] Perfiliev, M. (1994) Problems of high temperature electrolysis of water vapour. Int. J. Hydrogen Energy (IJHE) 19 (3): 227-230. [7] Stuki S. (1991) The cost of electrolytic hydrogen from off-peak power. Int. J. Hydrogen Energy (IJHE) 16 (7): 461-467. [8] Syed M. et al. (1998) An economic analysis of three hydrogen liquefaction systems. Int.J. of Hydrogen Energy (IJHE) 23 (7): 565-576. [9] EPRI TAGTM(1986) Technical assessment guide, 1: Electricity supply, EPRI rep. P-4463 SRI, Palo Alto. [ 101 Milanese JJ. (1987) Process industry contingency estimation. Publ. N-2386-PSSP, Rand Corp., St. Monica, CA. [ 111 Hager RL, Heaven DL. (1990) EPRI GS-6904, Palo Alto, CA. [12] Ang AH-S, Tang WH. (1984) Probability Concepts in Engineering Planning and Design. 2, Decision, Risk and Reliability. Wiley, New York, NY.


NON-TECHNICAL BARRIERS TO LARGE-SCALE WAVE ENERGY UTILISATION A.J.N.A.SARMENTO'',~),F. NEUMANN'~), A. BRITO-MELO'~) (I)

Instituto Superior Tecnico, Lisbon, Portugal,

(21

Wave Energy Centre, Lisbon, Portugal

In order to enable a large-scale implementation without violating the principles of sustainability, it is of high importance to avoid conflicts that partly emerged from the context of the large-scale implementation of other RE technologies, mainly with respect to environmental and societal sensitivities. In addition to the lessons learnt from other technologies, there is a number of specific aspects for the utilisation of wave energy that have to be taken into account with sufficient care in the early phases of development. In the present paper the approach of the WEC (Wave Energy Centre) to seize these aspects, is presented, based on results of a case study on these non-technical barriers to large-scale development of wave energy basically arising from environmental sensitivity aspects and competing uses of the ocean. The utilisation of wave energy is generally deemed environmentally benign and only results of several years of operation at significant scale may confirm the validity of this evaluation. Therefore, emphasis in this phase is given to the aspects of conflict of use, which can easily be underestimated in the implementation of this renewable energy. Although the study has been elaborated for the situation in Portugal, several findings are expected to have generic character.

1. Introduction

The present energetic situation world-wide and in particular in EU countries demands a high priority to intensify and diversify the exploration of endogenous energy resources in the near future. According to geographic and geologic characteristics, the resulting renewable energy (RE) mix can assume very different compositions, among which hydropower and wind energy already cover notable shares in a number of countries. Among the RE resources that have not yet contributed in a significant scale, ocean wave energy is likely to play a major role in the f i h r e for a number of regions, such as countries at the European Atlantic coasts including the UK, Ireland, France, Spain and Portugal. Despite the vast energy potential of ocean waves and various attempts to tap this for the last 30 years, technical drawbacks and high development and installation costs were the main obstacles for the implementation of this RE technology. Meanwhile, a number of systems have reached the prototype stage and some promising achievements indicate that it is likely to overcome technical difficulties and yield sufficient maturity for large-scale utilisation in the medium term. On the background of the priority that is presently given to RE implementation on the community political level, also the prospects for economic viability of wave energy implementation are encouraging. In order to enable a large-scale implementation of wave energy compatible with the principles of sustainability, it is of high importance to avoid mistakes that were partly committed in the context of the large-scale implementation of other RE technologies, mainly with respect to environmental and societal sensitivities. In addition to these lessons learnt over the last two decades, there is a number of specific aspects concerning the utilisation of wave energy that have to be taken into account with sufficient care in the early phases of development. It is expected that large-scale implementation will take place by deploying offshore submerged or floating devices at depths around 50 m, grouped in wave farms with specific configurations and dimensions depending on the type of technology. In contrast to wind energy, wave energy has not converged yet to a unique technology. Many different concepts of devices have been proposed so far and presently some have 225

226

A.J.N.A. Sarmento. F. Neumann & A. Brito-Melo

achieved the prototype status and are under tests at real sea conditions. An emerging industry in wave energy domain has accompanied this development with strong enthusiasm, and from the viewpoint of technological development and economical considerations, large-scale implementation in Portugal and other countries may be realistic by 200712008. In Portugal there is a substantial “know-how“ in the field of wave energy. R&D on wave energy utilization was initiated in 1978 at Instituto Superior Tecnico (IST), joined by INETI in 1983. The work conducted by this team in the numerical and experimental fields lead to the construction of the 400 kW onshore OWC pilot plant at the island of Pico, Azores [ 11. This construction was funded by the EC, EDA (Azores utility) and EDP (Portuguese mainland utility), and was technically supported by a number of Portuguese companies. In 1997, IST established collaboration with the Dutch company Teamwork Technology, developer of the offshore submerged device “Archimedes Wave Swing (AWS)”, whose 2MW prototype is presently being deployed offshore Povoa de Varzim (www.waveswing.com). The Portuguese %now-how“ in the field of wave energy, together with the existence of a good natural resource at the Atlantic Portuguese coast and good infrastructures, make Portugal a favoured country for playing an important role in the industrial development of wave energy conversion, compared to its European cogeneris. This poses a number of subsequent aspects that have to be considered in view of such development, based on a planning that weighs the eventual impacts or conflict of uses with other sectors of activity. There are several aspects to have in mind that may constitute potential barriers to large-scale implementation of wave energy. Among these are certainly still a number of “know-how” barriers related to technological experience or availability of specific expertise and processes. However in the present study priority is given to the aspects not directly related to technological development, which are assumed to be overcome in the next years.

2. Competing Uses of the Maritime Space Having in mind the increasing maturity of wave energy technology, it is opportune to investigate the main concerns and issues over the Portuguese coast to large-scale development of wave energy basically arising from competing uses of the ocean space. In the maritime domain there are several players and entities with different interests and competences, so that it is important to try to understand and, if possible, to conciliate the different viewpoints. The experience from offshore wind energy has demonstrated that this is a difficult task, which has to be approached as early as possible. The lack of sensitivity of the entities involved in a large-scale wave energy project may in some cases generate conflicts and some concern by the communities more directly involved in the traditional activity sectors of the maritime domain. This may constitute a barrier to the realisation of a project, and to the acceptance of the technology. In countries with significant coastline, the near-shore area has been a traditional domain of the fishermen communities and the interference with these communities can turn out to be problematic for the development and implementation of the project. This study identifies the main problems, provides a perception of relevant issues by interested parties and gives recommendations for preventive mitigation measures. On the other hand also potential benefits and opportunities of large-scale wave-energy development are briefly discussed. Successfully overcoming the non-technological barriers enables to

Non-Technical Barriers to Large-Scale Wave Energy Utilisation

227

provide a framework to facilitate the licensing process of offshore wave energy increasing investor confidence and market penetration. The methodology applied in the study has consisted in defining the different areas in the maritime domain subject to restrictions and thus identifying a set of uses susceptible of conflict with the implementation of wave energy projects, focussing on the activities around the bathymetric line of 50 m where it is expected that large-scale development will concentrate. To achieve this objective, a number of entities with activities or jurisdiction in the maritime field has been consulted, namely those related with the environment, fisheries, authorities, submarine cables and prospecting1 exploitation of petroleum. The restricted areas were compiled in a map of the coastline of continental Portugal, including main bathymetric lines and port locations (see Fig. 1). 2.1. Fishermen’s Society

In Portugal fishery represents in economic terms an important sector with special social relevance. The implementation of large-scale wave energy projects in an area, which is traditionally regarded as domain of the fishery community requires a careful approach of the involved fishermen community. The barriers imposed to the development of wave energy may be mainly a result of the fishermen acceptance to its installation. In order to identify eventual conflict of uses with the fishery activity, the maritime authorities along the West coast of Portugal have been contacted. The opinions collected in these meetings indicated that the eventual conflict is related mainly with the area of activity of the different types of fishery: the traditional fishery done by small local vessels close to the shoreline in shallow waters (in a distance less then 3 miles from the coast) do not achieve in general the 50 m water depth except in the Southwest of Alentejo and in the Vicentine coast where the bathymetric of 50 m is much more closer to the shoreline. Meanwhile in this area the devices may be placed more distant from the coast as the slope of the continental platform is very soft and the 60 m water depth is about 4 miles from the coast. The industrial fishery is only allowed at a distance beyond 6 miles from the shoreline, which in the Portuguese coast occurs in depths of more than 50m, except for the area about 50km south of Oporto. According to the maritime authorities, the wave energy farms installed in water depths of 50 m will generally not interfere significantly with the main types of fishery and thus should not represent any serious conflict with the activity of the fishermen. However it is necessary to consider navigation corridors between the farms for the ships to go round the area delimited by the wave energy farms, and to be sure that they are neither placed in hot-spots from the fisheries point of view, nor in navigational corridors between hot-spots. Fishermen assume in general an attitude of apprehension and doubt to the introduction of a new activity in the sea. In regard to wave energy they assume a defensive opinion and consider the wave energy farms as a barrier to the fishery activity, recognising however the benefits of wave energy utilisation. The general attitude is not against the installation of large-scale wave energy projects, as long as adequate lengths between the farms are maintained and security and insurance aspects are taken into account by the promoters of the project. The question of attribution of responsibilities in case of an accident is a relevant aspect for the fishery communities. Finally they show clear interest in being informed of the plans for future projects and to be involved in the licensing process.

228

A.J.N.A. Sarmento, F. Neumann d; A. Brito-Melo

R

.EGEND Coastline 50mdepth 100 rn depth 200 m depth 6 miles distance

~

iESTRlCTEDAREAS I-,-"

;,g&j Nature reserves ZdCd

,g Submarine cables >TKT.,

' %_* 2

Militar use

~~

Port authorities

ZOMPETING USES Traditional fishery Industrial fishery

Fig. 1. Wave energy resource conditioning map for Portuguese nearshore.


229

2.2. Port Authorities The maritime traffic is done in a large distance from the coast (typically more than 5-10 miles), where wave energy devices are not expected to be installed. The only conflict may be the entrance channels to the harbours. According to the maritime authorities, a sector of 135" at the entrance of the main harbours (Sines, Setubal, Lisbon, Figueira da Foz, Aveiro, Leixoes and Viana do Castelo) is recommended to be reserved just for navigational use.

2.3. Submarine Cables In the Portuguese coast there are submarine telecommunication cables protected areas in the near-shore of Lisbon and Sesimbra. Within these areas it is forbidden to perform any activity susceptible of damaging the cables. Therefore wave energy devices can not be installed in these areas. However they coincide with zones where it is not expected the utilisation of wave energy due to the fact that the shore is protected from the main incident waves and on the other hand due to the presence of the two main ports - Lisbon and Setubal (nearby Sesimbra) -with an intense traffic. 2.4. Military Use Some coastal areas are periodically reserved for military exercises. On the Portuguese coast these are announced by radio navigational warnings. These areas are delimited and represented in nautical maps. Although there are no restrictions to maritime use outside the exercise periods, it is apparent that wave energy devices should not be installed in these areas. However this does not constitute a relevant conflict of use with wave energy as one of the zones is located in the vicinity of the area in Sesimbra designated as submarine cable protected area; and the second one is a small area in front of the Vicentine coast (Southwest of Portugal).

2.5. Marine Resource Exploration There is some activity Concerning hydrocarbon exploration in Portuguese waters, although to date there are no prospection and extraction activities. The discovery of petroleum in Cadis Sea (Southern Spain) has valorised also the Portuguese coast and an increase in the exploration activities is expected. These activities may affect the installation of wave energy devices just in specific localisations. The prospection and extraction processes would not be significantly affected by wave energy installations, as it is possible to do inclined perforations. Further the device locations may be deviated if an important petroleum field is discovered. It is recommended that plans for large-scale offshore wave energy implementation is synchronised with the activities of the petroleum industry, in order to avoid potential conflicts. It should also be noted that there can arise a number of technological synergies concerning the use of offshore installations that may be explored by such an approach. 2.6. Nature Reserves

In the maritime domain there are some areas delimited as natural parks and reserves. At the bathymetric line of 50m the following areas were identified: 0 Berlengas-rectangular area including Berlengas archipel;

230

A.J.N.A. Sarmento, F. Neumann & A . Brito-Melo

Southwest of Alentejo and Vicentine coast-2 km wide corridor along the entire extension of the coastline. In these areas the installation of wave devices it is not recommended. The limit of the 2km corridor on the Southwest of Alentejo and Vicentine coast reaches the 50m depth contour, but as referred before when discussing the fishery activities, the slope of the continental platform is very soft and the devices can be installed in a depth of 60m (3 to 5 miles offshore), neither interfering with the natural reserve, nor with the traditional fishery. Although other natural reserves in the Portuguese shoreline do not reach the 50m water depth contour, they need to be taken into account in the underwater crossing and installation of electric cables from the wave energy farms through land. However a large scale utilisation of wave energy will possibly require an underwater transmission electrical line, extended from North to South along significant parts of the coast, thus reducing the number of connections to the shore.

3. Environmental Concerns Beyond the conflict of use with nature reserves, which indeed originate from environmental concerns, the general consequences of large scale utilisation of wave energy are of highest importance to the technology’s future potential. Lessons have been learnt from related technologies, such as wind energy, and it is essential to give importance to a number of potential concerns, such as visual impact and noise. 3.1. Visual Impact

The experience from offshore wind energy clearly identifies the visual impact as a main concern. The coastal areas are usually important for recreational uses and so the visual impact of shoreline projects should be considered. Offshore devices will have certainly less visual impact. The visual impact may be relevant in areas where the water depth required is attained at small distance visible from the coast. The visibility of the offshore devices from the coast will depend much of the signalisation with lights and therefore its effects must be considered. Some recommendations in order to improve the public acceptance have been proposed from wind offshore such as to avoid recreational areas, to consider a planning phase well open to the public with eventual involvement of the public and to discuss and analyse the configuration of the farms, number of devices and dimensions previously to any final decision. It was verified that many times the turbines are assumed as an intrusive element on the landscape and tests have been done by different suppliers in order to investigate the most adequate colour for the towers. However, the comparison with wind energy appears very conservative when taking into account the structural dimensions of wave energy devices. These are either submerged or surface-piercing but with substantially lower heights than wind turbines. It can therefore be expected that the visual impact of wave energy parks will be negligible for offshore wave energy and very reduced for near-shore exploration.

3.2. Noise Noise is mainly caused by the mechanical components or air-flow in case of OWC devices. The noise may affect the habitat of marine species and if audible by the


23 1

population on the coast may reduce the public acceptance. Care needs to be taken that the noise emitted from devices does not cause interruptions to migratory pathways or breeding grounds [2]. In the onshore plant at the Islay, Scotland it was verified that the turbine was responsible for unpleasant noise which was audible above the background of wind and waves and a solution was adopted to reduce the level of noise. To demonstrate that the noise resulting from wave energy devices is an impact that can be mitigated to nonsignificant levels is a very important task for the future of wave energy and that is why the experience gained in individual projects is very important. It shall be noted that such impacts are considered to be negligible for offshore wave energy, as the noise levels 4-5 miles away from the shore will not seriously affect population on shore. 3.3. Emissions

Wave energy generation does not bring along Greenhouse Gases emissions or any other atmospheric pollutant. However the emissions may result from other stages of the life cycle of wave energy technologies (Resource extraction, resource transportation, materials processing, manufacture, transportation, construction, operation, maintenance decommissioning and product disposal). Several detailed studies with other renewables energies demonstrated that in general the emissions released during the manufacture of the materials the most important, depending on the industrial practices and regime of pollution control in the country [3], but similar to what happen with the conventional processes of energy production. 4. Further Barriers

Although the competing uses and the environmental impacts are considered to be the most important potential restraints to large-scale wave energy implementation in the future, there is a number of other factors that presently limit a more efficient development and a faster implementation. These are mainly related to the fact that the development of utilising the ocean as major energy resource is a relatively recent challenge to society, resulting in the following main barriers:

4.1. Marketing Deficits Marketing barriers require the characterisation of the potential market for shoreline and offshore systems at the Iberian Peninsula scale, but also, whilst at a less detailed levels, at the European and word scales. For such procedure, different wave energy technologies should be grouped by similar requirements concerning minimum or maximum water depth, ocean bottom and subsoil characteristics, maximum allowable tidal amplitude and currents, minimum wave resource for economical competitiveness and technological support and infrastructure. 4.2. Administrative Procedures and Legal Aspects

Administrative barriers are mainly related to permits procurement. The experience in Portugal, both with the Pic0 and the 4 W S plants shows that this is a very important area.

232

A .J.N.A. Sarmento, F. Neumann & A . Brito-Melo

In Portugal, and possibly in many countries, the regulations for the sea are less well established, no experience exists in licensing wave energy plants and a large number of public departments are called to participate in the permit procurement. This turns the permit procurement a very complicated and time-consuming process. In the case of the AWS permit about 10 different public departments were involved (the Defence, Environmental and Public Works ministries were directly involved with more than one department from each). Regarding this aspect, a similar approach like the new Strategic Environmental Assessment (SEA), as practiced in the United Kingdom, could straighten the procedure. In this methodology, the National waters are divided in several zones, for which the necessary permission and licensing procedures and the involved competent entities are clearly defined. Concerning the permissions and licensing to feed electricity into the electrical grid, the restrictions to power quality are quite clearly defined by Portuguese law, however the quantity and the timing for large-scale feed-in is not easily sizable. The work to be done includes the identification and comparison of legal procedures and constraints related to the access to the electrical grid, with the objective of removing the constraints and simplifying and standardising the procedures required for wave energy. The identification of the electrical grid connection points availability and characteristics in the areas of interest for wave energy is an important step towards enabling large-scale implementation. 4.3. Financial Barriers

Financial barriers are related to the feed-in tariff of the produced energy, the financial schemes to support the technology development (funds or special interest loans) and their evolution with the development of the technology. The situation in Portugal concerning the feed-in tariff is extremely favourable compared to other European countries, which is why a strong development can be expected during the next years. In particular in the United Kingdom, various players of the wave energy community stated that this is the most desired kind of governmental support, as it is calculable and rewards the actual cost efficiency of devices. However, in the present phase of pre-commercial development, additional measures to finance the development are essential. In a following phase, the potential involvement of large industrial players or risk capital is much more likely to occur than at present, which is why strong incentives are particularly important before reaching that point. They are also require to assist suppliers of services and components to target their expertise to wave energy, since the development of wave energy and the potential benefits of exportation of the technology can not be achieved without a competitive and competent cluster of suppliers. 5. Conclusion

The main non-technical barriers are related with competing uses of navigation activities, fishing resources and, to a small extent, petroleum prospectiodexploitation. Restricted areas such natural reserves, military exercises areas and submarine crossing cables need to be taken into consideration as well but are not critical in Portugal. Another aspect that may constitute a barrier to the development of wave energy projects is the public acceptance. It can play an important role for the critical path of large-scale projects and may impose serious conditioning to the time-schedule and realisation of a project. Public


233

acceptance barriers can have various origins; often they are a mixture of societal aspects and competing uses, as identified in this paper. A major barrier could be the licensing process that although recently adoptions in the law were made, could play an important role in a country, in which this process takes several years for wind parks and mini-hydro plants, technologies with substantial operational experience. Generally, the environmental impact of the wave energy technologies presently relevant is considered to be much reduced, provided that the site selection is done prudently and a controlled planning underlies for sensitive locations. In spite of having a small to moderate expected level of environmental impact, the penetration of wave energy technology will highly depend on this aspect. Thus, it is important to assess the environmental impact of different wave energy technologies and the limitations that may result from it, both in what concerns the market and alternative technological developments. A substantial baseline for this will be provided by the pilot plants presently operational or in the deployment phase. Measurements on these plants are foreseen and will allow updating the present understanding and providing means for a adequate legislation, before passing from the demonstration to an industrial phase. On the other hand, there is to date no substantial operational experience that can confirm this evaluation. It is mandatory to investigate and monitor the first systems of significant scale with respect to potential impacts, in order to identify their extent and to adopt mitigation measures where necessary. To conclude it is perceived that the non-technical barriers resulting from conflicts of interest do not constitute major barriers for the large-scale development of wave energy; however they need to be taken into account in a strategic plan for large-scale wave energy development. The competing uses are indicated in a map of the coastline of continental Portugal (see Fig. I), including main bathymetric lines and port locations. From this figure can be seen that about 300 km of the Portuguese west coastline are potentially available to wave energy utilization.

Acknowledgements This work was partly sponsored by the Marie Curie Fellowship Scheme (contract EVK32000-55 135), the Portuguese Science and Technology Foundation (FCT; contract SFRH/BGCT/11789/2003) and Direccao Geral de Energia, Portugal (Portuguese Directorate General Energy).

References

[I] Falcao AFO. (2000) The shoreline OWC wavc power plant at the Azores. Proceedings of the 4thEuropean Wave Energy Conference, Aalborg, Denmark, pp. 42-48. [2] Thorpe TW. (1999) A Brief Review of Wave Energy. DTI, United Kingdom. [3] Wavenet - Results from the work of the European Thematic Network on Wave Energy, ERK-CT- 1999-2001 2000-2003, European Community, March 2003.


SOLAR POWER AGRICULTURE: A NEW PARADIGM FOR ENERGY PRODUCTION U. BARD1

Dipartimento di Chimica, Universita di Firenze. Polo ScientiJico di Sesto Fiorentino, 50019 Sesto Fiorentino (Fi) Italy, [email protected] ~

The present paper argues that solar power technologies, such as photovoltaic panels and wind turbines, should be regarded as a form of agriculture in the paradigm named here “solar power agriculture”. These technologies share several features with traditional agriculture, for instance the basic process of energy conversion by collecting solar light over relatively large areas. Solar power technologies and agriculture also share a dynamic resource growth curve which is s-shaped in contrast with that of non-renewable fossil resources which is bell shaped. Conventional agriculture uses solar light to produce food, textiles and other products, whereas solar power technologies mainly use light to generate electric power. Framing renewable energy within a well known and accepted paradigm, that of agriculture, leads to major advantages in terms of public image, with the consequent unlocking of the financial resources needed to increase the use of renewable resources. The concept of solar power agriculture provides a bridge between the present situation in which renewable play a marginal role in the world’s energy production to a future situation when renewables will instead play a major role.

1. Introduction

The world is not (yet) running out of fossil fuels, but progressive depletion is making these fuels expensive as the result of factors which include increasing direct costs (e.g. the need to exploit smaller oil fields) and increasing external costs related, for instance, to the greenhouse effect generated by the products of combustion. Rising costs will cause the production of fossil fuels to reach a peak at some point in time and subsequently to decline as extraction becomes progressively less and less convenient in economic terms. Some projections [l-41 indicate that the peak for crude oil production could take place within the first decade of the 2 1st century, to be followed by the other fossil fuels peaking in later decades. The production curve of a mineral resource is often described as “bell shaped” [5]. In contrast, the curve for agricultural production is, in principle, s-shaped, i.e. it reaches a plateau when all productive land is exploited. Agriculture is sustainable, at least in theory. Historically, both soil and water may turn out to be a non-renewable commodity when misused [6-71, however it is at least a physical possibility for agricultural societies to reach a stable state. The subject that the present paper addresses is how fossil fuels can be replaced as a source of energy using renewable solar technologies in the “agricultural paradigm”, that is in order to abandon the “bell shaped’’ curve and reach, instead, a plateau of production at levels comparable to the present values. In this approach, called here ”Solar Power Agriculture” or, also “agri-energy” electric power produced on agricultural lands is considered as just another form of agricultural product, to be approached with the same social and economic approaches which are commonplace for conventional agriculture. Several previous reports (e.g. 8-16) have discussed the ultimate limits of solar technologies in terms of land requirement. These studies arrived to the conclusion that a fraction of the earth’s equatorial deserts would be sufficient to provide abundant energy 235

236

U. Bardi

for humankind's needs. However, large scale renewable energy plants in deserts do not appear to be on the verge of materializing. The main problem appears to be the need to attract the huge investments needed, both for the plants and for the related energy vectoring system. Here, a different approach is considered on the basis of the idea that renewable energy can to make significant inroads in power production only if its introduction is gradual and it starts from a relatively small scale. This approach leads to the idea of embedding solar plants within areas used for conventional agriculture. It is an approach, in fact, that has already been tested for wind energy in countries such as Denmark and Germany. Obviously, the possibility of expanding this strategy to obtain a significant fraction of the worldwide energy needs depends on a number of factors: 1. Technological factors: can renewables produce enough energy for the needs of humankind without competing with conventional agriculture for land needed for food production? 2. Cost: Even when embedded in conventional agricultural production, won't renewable energies remain too expensive? 3. Public opinion. Can the relatively large extension of land needed for solar energy be made acceptable to the public? The present paper will show that the answer to these questions is, in principle, positive and that the concept of solar power agiculture has a strong potential in order to speed up and favor the introduction of solar renewable technologies in the world.

2. Technological Factors Human beings have been exploiting solar energy for their needs since the time when agriculture was first developed, about 10,000 years ago. As a technology, agriculture was hugely successful and in time it diversified and expanded until, at present, most of the land "potentially suitable for agriculture" on the planet, about 130 million km2, [ 17-181 is exploited for human needs. The fraction of this land which is exploited for food production, i.e. the sum of pasture land and arable land corresponds to approximately 50 million km2that is about 38% of the total [ 17-181. Agriculture uses solar energy to carry out chemical synthesis, in turn based on photosynthesis, a biochemical process which uses solar light as the source of energy necessary to synthetize carbohydrates from water and carbon dioxide. Further stages of the biochemical process can produce a variety of organic substances. Mainly, agriculture is used to produce foodstuff and fiber for textiles. Traditionally, agricultural biomass has been also used as energy source for residential heating but in recent times it is increasingly exploited to produce electric power and, in some cases, fuel for traction. Other physical and chemical processes to exploit solar light for human use have been known for a long time but the large scale development of what we call "renewable energy technologies", started only during the first decades of the 20th century with hydropower plants. The last decades of the 20th century saw the extensive development of the technologies which go under the name of "new renewables". Among these, we can define

Solar Power Agriculture: A New Paradigm for Energy Production

237

as "solar direct" methods those which collect and exploit solar light: mainly photovoltaic (PV) and solar concentration. We may define as "solar indirect" methods those which exploit the effect of the solar light in heating the atmosphere or the ocean: wind power, small hydropower (SHP), wave, ocean currents, and others. There exist also renewable technologies which are not based on solar energy, i.e. geothermal and tidal which may play an important role in the future but which will not be considered in detail here as they can't be framed within the "agricultural" paradigm. The energy needs of humankind will vary in the future depending on population and on the human lifestyle. The present work does not attempt to address projections for the future growth of population, nor the need of equilibrating the energy use by different fraction of the population. However, considering the slowdown of population growth observed during the past decades, most authors indicate that a stabilization of the world population might take place before reaching mid 21st century for numbers not much higher than the present ones. Considering also the ample possibilities for rich countries to reduce the amount of energy utilized by measures of efficiency and conservation, an "order of magnitude" estimation of the energy needed for humankind can be obtained examining the amount of energy used nowadays. The parameter to be examined here is the "total final consumption" (TFC). This is a better parameter to consider in the present context than the other common one, "Total Primary Energy Supply" (TPES), since renewables normally provide immediately exploitable energy in the form of electric power. The world TFC is reported to be 8 . 4 ~ 1 0 ~ G W y e a r [20] of which electric power is about lx107 G W y e a r . The OECD countries have access to 53.5% of the world TFC [20] for a population of 17% of the world total. The amount of solar energy that reaches the surface of the earth is commonly given as 1x1012 GWWyear (see. e.g. ref. 16): a value more than ten thousand times larger that the present TFC. In addition, we can roughly estimate the wind energy generated by solar irradiation as ca. 2% of the total solar energy arriving on the earth surface [ 161 that is about 2x10'' GWhJyear, again very large in comparison to the target. The energy associated with the global water flow generated by thermal effects is very difficult to estimate, but it is another very large amount. These data are listed in Table 1, together with two more data for comparison: the amount of the total biomass produced every year on the planet, some 2x10" tondyear, or lx109 G W y e a r [21] and the total human metabolic requirement [2 11. Table 1. Comparison of energy flows and energy requirements.

Human total metabolic energy requirement (present) World totai electric power generation (present) World total Anal consumDtion. T K Inresent) World total biomass produced Solar energy arriving on the earth's land surface I Total solar enerov arrivino on the earth's surface

5x106 1x10~ sx1a7

11~10~

I

3x1Ol1

I

I l Y l n ~ ~I

238

U. Bardi

The actual possibility of exploiting the large amount of solar energy arriving on the earth surface depends on the efficiency of the technology used. Furthermore, we have to consider the quality of the energy produced by solar technologies and not just its amount. Passive solar collectors, for instance, are not suitable for power production. Most renewable technologies (e.g. photovoltaics and wind turbines) generate high quality energy in the form of electric power and can therefore be used for the direct replacement of fossil fuels in power production. It is also always possible to use high grade energy to efficiently produce low grade energy, such as domestic heating. Regarding vehicle fuels, at present hydrogen, which can produced from electricity and water, appears to be a possible choice as replacement for fossil fuels. A different situation exists with biomass, which can be used to fuel a thermal engine in order to generate electricity or simply burned to obtain heat. In practice, for the purpose of comparison, in the present discussion it can be simply assumed that the energy output of renewables in the form of electric power can be used to replace all the forms of energy that compose the world TFC of today. This approach has been used, among others by Pimentel [ 151. This said, we can calculate the fraction of land area (or “footprint”) needed to generate the amount of energy needed by humankind by renewable technologies. In general, solar irradiation on populated lands ranges from a minimum of ca. 900 kWh/m2/year (e.g. northern Europe) to values of the order of 2200 kWhIm2lyear in subtropical regions, with even higher values in equatorial areas such as the Sahara. An approximate worldwide average value for populated areas could be taken as ca. 1500 kWWm2Iyear. The OECD countries tend to be located somewhat farther away from the equator, so that an average value in this case should be lower. Commercial photovoltaic panels have an efficiency that today can routinely reach 14%. Experimental PV cells may do significantly better. The efficiency of the power delivered to end users by this kind of plants can be reasonably considered as 10%. In terms of land use, with such efficiencies, solar direct plants can bring to the end user about 120 kWh/m2/year in medium latitude areas (1200 kWh/m2/year). As an order of magnitude estimation, it is possible to consider an average value of 100 kWh/m*/year for the power delivered to end users. In the case of solar indirect technologies, it is not possible to define a land requirement in the same way as for photovoltaics. Wind turbines must be spaced at a certain distance from each other, but most of the land occupied remains available for agricultural use. The physical footprint of a wind turbine may be estimated to correspond to a land use of about a factor 100 smaller than PV panels for the same delivered power. Of course, offshore wind turbines do not occupy any land area, and the same is true for SHP plants and for all the schemes which use wave power or oceanic currents. Finally, as a direct solar technology, biomass is relatively inefficient. The upper limit of photosynthesis efficiency at the molecular level is estimated as 6% by Tiezzi [21] and as 4.5% by Patzek [22]. In practice, the efficiency of biomass production in plant organisms is much lower. From the data of table 1, it can be calculated that the “planetary efficiency” for biomass production is about 0.1%. In specific cases, the efficiency can be higher. Patzek [22] reports a calculation on the average potato yield in England where the photosynthesis efficiency turns out to be


239

0.4%. According to Tiezzi [21] crops may have yields of the order of 1%, but only at the cost of extensive use of irrigation and of fertilizers derived from fossil fuels. Another source [23] reports that some ecosystems, such as tropical forests, can reach efficiencies of conversion higher than 1% whereas crops cultivation has an average efficiency of ca. 0.3%. Taking into account the need of using a thermal engine to produce power, the sustainable efficiency of biomass as a source of electric power cannot be considered as higher than ca. 0.1%, again as an order of magnitude estimation. Obviously, the generation of electric power is not the best energetic use for biomass, but this value will be retained as a means of comparing biomass conversion with other energy sources. A further important parameter affecting land use is the energy return on energy invested (EROEI). This parameter can be defined as the total amount of output energy produced by the plant over its lifetime divided by the input energy, that is the energy required to build, maintain, fuel (if needed) and eventually dismantle the plant, as well as energy needed for vectoring the energy produced to end users (see e.g. ref. 15). The concept of EROEI can be expanded to include also the external costs (e.g. greenhouse gases produced) of the plant, measured again - in terms of energy. This parameter defines the “ecological footprint” (EF) of an energy producing plant, which is larger (and may be much larger) than the actual footprint of the plant. For instance, for fossil fueled energy plants, the EF is large because it is calculated taking into account the area of forest needed to remove the amount of C02 generated in the combustion of the fuels [24]. On the basis of the concept of EROEI, the Ecological Footprint (EF) of a renewable technology can expressed in units of area needed to provide a certain yearly output of energy, taking into account the area needed for the input energy, obtained using the same technology and averaged for the same period. Simple considerations lead to express the EF as a function of the actual footprint (AF) as EF=AF x EROEI/(EROEI-1). The EROEI is a difficult parameter to calculate and the estimates reported in the literature are widely variable. Some significant values for the technologies considered here are listed in the following table. For comparison, a giant oil field in its prime is reported to have an EROEI of ca. 30-50, whereas the EROEI for partially depleted and/or smaller oil fields may be around 10 or lower [25]. ~

Table 2. Comparison of different renewable technologies in terms of energy return for energy invested (EROEI) and ratio of “ecological foot print” (EF) to “actual foot print” (AF)

The values of Table 2 are to be considered as estimations only, however they do indicate that for the “new renewables” (mainly PVand wind) the EF is of the order of 10% or less larger than AF and therefore can be considered to lie within the uncertainty of the estimation of the needed land area. For biomass, the area correction for a value EF/AF ratio of 1.5 cannot be neglected, but it must also be taken into account that the EROEI

U. Bardi

240

calculated by Pimentel [ 151 includes the loss due to the transformation of heat to electric power. Therefore, the approximate value of "0.1%' calculated before can be retained since it already took into account this inefficiency factor. It is to be stressed that this biomass is intended to be mainly harvested from natural forests. Crops or other agricultural products make little sense as sustainable energy sources because of the extensive need for fertilizers, irrigation, etc which makes their EROEI close to one or even lower than one as reported by Pimentel [29] for bio-ethanol. The last important parameter to be considered here is the need of energy storage for intermittent sources such as photovoltaic and wind power plants. It is known that when the amount of energy produced by intermittent sources exceeds ca. 20% of the total grid power, it becomes difficult to control the energy input to the grid without some kind of energy storage. There exist a large number of methods for electrical energy storage, several still in the development stage (e.g. hydrogen coupled with fuel cells) others are possible only as large scale solution (e.g. pumped storage). For the present land area estimation, it will suffice to note that storage does not involve large areas and so it does not need to be considered here. Once these data are available, the simplest approach is to examine the land use for each kind of technology and from that calculate the extent of planetary land needed to generate, for instance, an amount of energy in the form of electric power equal to the present TFC. The following table summarizes the considerations developed so far. Table 3. Comparison of different renewable technologies in terms of e fficiency of conversion and area occupied. Method

Organic biomass

Approx. efficiency of light of conversion (ratio incident solar eneigy to electric power delivered to users)

-

0.10/0

Solar direct (PV or solar thermoelectric)

5%-100/0

Solar indirect (wind)

n.a.

Order of magnitude deliverable powel pel area occupied [footprint) for an iiradiation of 1200 kWh/rn'/year. (kWh/m'/year)

-1

-

-

100 10000 (on suitable

sites only)

At this point, we may proceed with embedding the data found so far within the concept of agriculture. That means to estimate what fraction of the land is needed to generate energy comparable to the present day use. The relevant numbers here are listed in Table 4 (data from refs. 17 and 29). Table 4. Land areas of the world presently used for different purposes.

Millions square km . Land "potentially suitable for agriculture"

130

Land used for food production

50

Arable land (crops production)

1.5


24 1

There is a certain degree of uncertainty in these data [18], but the values reported here can be considered as prudent estimates. The concept of “land used for food production” is the sum of the arable land plus the grassland used for pasture. It is reasonably safe to assume in both cases the land is accessible by human beings and equipment and that it exists in human-compatible conditions of climate and temperature. Hence this area is also suitable for setting up energy producing plants. From these data, it is possible to estimate the extents of land required for two targets taken as the energy equivalent to the present value of world electric power generation (target 1) and equivalent to the present value of world TFC. The results are listed in Table 5 . Table 5 . Land areas needed for energy production using different renewable technologies

Land area needed as percentage of land used for food production (50~10~ Km2) Technology generation

BIOMASS conversion to electric power SOLAR DIRECT conversion (PV or solar concentration)

0.2%

SOLAR INDIRECT conversion (wind)

0.0030/0

1.5% I

0.03%

The values reported in Table 5 are, obviously,just orders of magnitude and are also to be considered as conservative. The fraction of land needed for the stated targets could become smaller for technological improvements and also taking account that a considerable fraction of the plants could be placed in areas which are not “agricultural land”, for instance offshore or on the roofs of existing buildings. The data, anyway, lead to the conclusion that both solar direct and solar indirect plants embedded in agricultural land would have a very small impact on agricultural production even for the relatively ambitious targets outlined above. Indeed, the human metabolic requirement of energy is only 5x106 GWWyear and with the known efficiency of the best crops it could be managed in just a few percent of the world land area. The impact of energy plants on agriculture would be, therefore, smaller than that of phenomena such as erosion and soil degradation. It is also worth comparing the surface needed for energy production with that of urban areas. Here, the data available are very uncertain. Kitajma [30] estimates the fraction of “urban land” as ca. 2% whereas Goldsmith [31] reports data which indicate that the fraction of land covered with “permanent structures” may be as high as ca. 15%. The large discrepancy among these values is probably due to differences in the definition of “urban land”. However, Kitajima’s estimation, compared with the data of Table 3,

242

U. Bardi

indicates that at the land area needed to provide a town with sufficient energy is of the same order of magnitude as the urban area. In other words, a town should dedicate a land area of extent approximately equal to the area of the town itself to the production of energy. Micro-production of energy at the level of single roofs is a possible solution, but a much easier approach is to disperse this production over the agricultural land surrounding the town. There exist very large urban spreads in the world where this approach may be difficult, but in most OECD countries towns are surrounded by agricultural areas much larger than the towns themselves. It may also be remarked that it may be possible to make even solar direct technologies compatible with organic agriculture occupying the same areas. Right now, the shadow of a solar plant (photovoltaic or thermoelectric) makes it impossible to cultivate the area Underneath. However, photosynthesis uses only a fraction of the solar light spectrum and solar direct plants could be modified to let energy pass through in the right wavelengths only. The mirrors of a solar concentrating plant, for instance, could be coated in such a way to reflect only the part the spectrum not used by photosynthesis, letting instead pass the rest. The efficiency of the solar plant would be lowered, but it would be possible to cultivate the area underneath. This technology is probably unnecessary at present because of the small fraction of area needed by solar plants, but it is mentioned here as a possibility for the future.

3. costs Estimations of the costs of renewable energy are common in the literature, a review can be found, for instance, in [8]. In most of the studies published so far, the calculations arrive to values that favor fossil fueled plants over renewables. In general, electric power generated by wind turbines is only marginally more expensive than that generated by fossil fuels. For photovoltaic panels, the gap is large, of approximately a factor 5 (see, for instance, ref. 10). The “solar power agriculture” paradigm offers a way to look at the financial support for renewables which goes beyond the cost calculations reported in the literature. The misleading factor of these calculations lies in the fact of producing a single, static number. However, the cost of energy is dynamic and it varies over the whole lifetime of a plant which, for the case of renewable plants may be very long, typically of a few decades. We saw in the previous section that the “new renewables” produce a positive, and often highly positive, energy return (EROEI). Since energy is worth money, there follows that the financial return from these technologies will be positive in the long run. It may argued on whether other investments will produce a quicker return, but not on the fact that the money and resources placed into the deployment of renewable energy plants will be, eventually, paid back and will subsequently produce a profit. A case could be made that renewables are not only an effective investment but also the safest possible investment since they produce a commodity, energy, which is the basis of human civilization and hence will never suffer market crashes. However, it doesn’t appear that investors have picked up yet the potential of renewable energy. This may be due to the bad image of renewables which still lingers in


243

the press and in the public opinion. In this respect, it may be worth citing here as an example an article by J. Dvorak [32]: “People still equate photovoltaics with the hippydippy 1970s. A recent article in the New York Times showed some hippies in Ukiah, California, with a solar panel outside their yurt-like home with a sod-covered roof. Presumably, the solar panel was used to power a lava lamp.” Apparently, the promotion of renewable energy done so far has been off target with the emphasis made on the concept that renewable energy is something “soft” as opposed to fossil fuels which, somehow, are “hard”. Switching to “soft” energy, apparently, would change our way of living, maybe carrying all of us to the Aquarian Age, to a higher state of consciousness or something like that. It isn’t clear how the kWh at a domestic outlet can be defined “hard” or “soft” and anyway these concepts don’t go well with investors who are more likely to think in terms of financial returns rather than in terms of holistic happiness. However, the economic return of renewables looks different when placed within a context, agriculture, which is known to be reliable, solid, and effective. Maybe agriculture never provided such a rapid return on investments as startup companies from Silicon Valley, but everybody understands that we can’t eat silicon. So, there is a logic in investing in agriculture and there are people interested in investments which privilege safety over speed and which also have a certain degree of “ethical” value. Here is where the “agriculture paradigm” as opposed to the “industrial paradigm” can be most useful. As long as energy production is managed by power utilities, as it is the case today, energy is produced within the “industrial paradigm” since power utilities are industrial companies which have a tradition of generating energy from the combustion of fossil fuels. In the industrial paradigm, the capital is provided by shareholders and companies will normally choose the road that leads to the fastest return on investments. In “solar power agriculture”, however, the situation is different and the focus moves from capital to land. The question that a stock market investor asks is “how can I increase the value of the capital I own?” while the question that a farmer or agricultural operator asks is “how can I increase the value of the land I own?” The answer to the latter question is to produce something on the land: it may be crops, it may be wine, it maybe wood, it may also be electric power in the paradigm if solar power agriculture. Large financial resources could be unlocked and utilized for the diffusion of renewable energy as soon as the concept of solar power agriculture becomes an accepted paradigm. For instance, the budget of the European Union for “rural development” in the period 2000-2006 is of approximately 50 billion euros. It is worth noting that in the 2003 document describing the EU agricultural policies [33], the world “energy” is not even mentioned, although there is one picture showing 19‘hcentury windmills. If governments realize that agriculture can be supported in an effective manner by means of the concept of solar power agriculture, part of these resources can be utilized for the diffusion of solar energy in farmlands; this benefits not only farmers, but all the sectors of society. It remains to consider how fast renewable energy could develop in order to cover the targets outlined here. This depends on the amount of financial resources that can be reasonably allocated to renewables and that may be estimated from the historical record. During the 1990-2000 decade wind power grew at an average rate of 22.4% per year and

244

U. Bardi

photovoltaic of 28.9% in the countries belonging to the OECD (Organization for Economical Cooperation and Development) [81. According to the BP statistical energy review [34] the growth of wind power worldwide was of 29% from 2001 to 2002, with the wind share of total global electricity supply increasing four-fold since 1996, to 0.4% in 2002. In those countries where wind technology can be considered as “mature”, the data [35] growth rates higher than 20% have been observed over at least one decade as for instance in Denmark (22.0% yearly) and in Germany (62.9% yearly). As a comparison, the highest rate of growth of crude oil production was 7% yearly for the period from 1930 to 1971 [I]. At a yearly growth rate of 20% and starting with 0.5% of the present supply, renewables could grow to produce the equivalent of the electricity produced today by fossil fuels in 30 years. This is, obviously, just an order of magnitude estimation. However it shows that renewables have a considerable potential to replace fossil fuels. 4. Public Acceptance and Environmental Concerns

Opinion polls show that most people are, theoretically, favorable to renewable energy (e.g. 36). However, when it is question of turning ideas into practice and actually build plants, the public opinion may not remain as favorable as opinion polls claim it to be. Basically, people object to renewable energies for two reasons: I. Renewable energies are seen as ineffective, that is they are not taken seriously a source of power. This concept leads to the related opinion that money and resources should be better spent in energy conservation measures. Renewable energies are seen as polluting, not in the same way as fossil fuels or 2. nuclear plants, but still causing visual pollution, damaging wildlife, producing noise, etc. These two arguments are closely linked to each other. Surely, people would be much more willing to accept the noise and the visual pollution of renewable plants if they were convinced that they are a real option to get rid of fossil fuels. Evidently, this is not the case since in recent times a remarkable backlash has been observed against wind power (e.g. ref. 37). The negative public attitude is not limited to wind turbines only. Mini-hydro plants are also accused to kill fish [8] and photovoltaic power is accused of using up too much land (see e.g. ref. 38). The emergence of such negative attitudes towards renewable plants is often related to policy mistakes, in this case in the way the new plants are proposed. Local communities often resent being invaded by power utilities with their gigantic wind towers. People feel (in most cases, correctly) that the value of their property will be reduced by the presence of large and visible energy plants in the vicinity. In these conditions, local communities cannot be blamed for opposing renewable plants. Here, the paradigm of “solar power agriculture” can reduce public opposition to renewable energies. By linking the new renewables with a well known and accepted technology such as agriculture, the gain in image is impressive. Solar power technologies, such as wind turbines and PV panels, cease to be seen as toys for hippies or bird-killing machines but become part of a respected and reliable production system which has fed

Solar Power Agriculture: A Mew Paradigm for Energy Production

245

mankind for some ten thousand years and is still feeding 6 billion and a half people on the earth. If, as it is the case, solar power agriculture is just another form of conventional agriculture, then it shares with it the same sturdiness, reliability, and social value. At the same time, the visual impact of the new renewables is considerably reduced in relative terms when it is looked at in the context of the impact of traditional agriculture. It may be argued that no activity of mankind on earth has had a more important impact than agriculture. Indeed, with about 40% of the land area used for food production, and a comparable fraction used for fiber and biomass production, we can say that traditional agriculture didn’t just have an impact on the environment; it radically changed the face of the planet. In many cases, actually, agriculture has been devastating for the environment, leading to irreversible salinization of desertification of large areas [7]. In comparison, turning less than 1% of the land from conventional agriculture andor forestry to solar power agriculture represents no major ecological impact. 5. Conclusion and Perspectives for the Future

The paradigm of “solar power agriculture” or “agri-energy” outlined here falls within some well known and accepted paradigms of agriculture and it goes one step beyond the idea that renewables are useful for supplying isolated areas with energy. Here, it is the opposite case: the existing power lines can be used for carrying the rurally produced electricity to towns and industrial centers at little or no additional costs. Carrying their products to town is what the rural world has been doing during the past 10,000 years or so. The concept of “solar power agriculture” permits to approach the problem of introducing renewable energy in the world supply in a completely different way than the presently accepted industrial paradigm. The use of renewables for isolated or remote areas is probably reaching saturation whereas the idea of placing large solar plants in deserted areas has not materialized yet for the high investments needed and for the lack of a suitable infrastructure for transporting the energy produced. However, once electricity (or other forms of energy) are placed in the same range of products as agricultural products, we immediately set a path of growth which is gradual, economically sound, requires low investments, and which can be put into practice in OECD countries without revolutionary changes. The present work has calculated that the fraction of land needed for present-day renewable technology to provide the world with an amount of energy comparable to the energy produced by fossil fuels would have only a minimal impact on food and textile agricultural production. The experience of the past decade shows that renewable energy production can grow at a yearly rate higher than 20%, much larger than the present trends of increase in energy production (around 2% year). If these trends could be maintained, even ambitious targets such as the complete replacement of fossil fuels by renewables are not impossible in less than one century.

246

U. Bardi

References [ 11 Association for the Study of Peak Oil (ASPO) 2004, www.oilpeak.net. [ 2 ] Bentley RW. (2002) Global oil & gas depletion: An overview. Energy Policy 30 189-

205. [3] Deffeyes KS. (2001) Hubbert's Peak, the impending world oil shortage, Princeton. 141 Campbell CJ, Laherrere JH. (1998) The end of cheap oil. ScientlJic American March: 60-65. [5] Hubbert MK. (1962) Nat. Res. Council Publ. 1000-D Washington DC, 54. [6] La1 R. (2003) Global potential of soil carbon sequestration to mitigate the greenhouse effect. Environment International 29 (4): 437-450. [7] Ponting C. (1991) A green history of the world, New York. [8] International Energy Agency (IEA) (2003) Report on renewables, www.iea.org. [9] Muneer TA, Kubie M. (2003) Generation and transmission prospects for solar electricity: UK and global markets. J. Energy Conversion and Management 44 (1): 35-52. [lo] Bal L, Chabot BC. (2001) Les knergies renouvelables. Etat de I'art et perspectives de developpement. R. Acad Sci. Paris, Sciences de la terre et des planetes 333: 827-834. [ l l ] Pearce JM.(2002) Photovoltaics-A path to sustainable htures. Futures, 34 (7): 663-674. [ 121 Schlaich J. (1999) Tension structures for solar electricity generation. J. Engineering Structures 21 (8): 658-668. [ 131 Trieb F, Langniss 0, Klaiss H. (1997) Solar electricity generation--A comparative view of technologies, costs and environmental impact. Solar Energy 59 (1-3): 8999. [14] Pimentel D, Patzek TW. (2005) Ethanol production using corn, switchgrass, and wood; biodiesel production using soybean and sunflower. Natural Resources Research 14 (1): 65-76. [15] Turrini, E. (1991) La Via del Sole, Edizioni Cultura della pace, San Domenico di Fiesole (Fi). [16] Hoogwijk M, Faaij A, Van den Brock R, Berndes G, Gielen D, Turkenburg W. (2003) Exploration of the ranges of the global potential of biomass for energy. Biomass and Bioenergy 29 (4): 225-257. [17] Wolf J, Bindraban PS, Luijten JC, Vleeshouwers LM. (2003) Exploratory study on the land area required for global food supply and the potential global production of bioenergy. Agricultural Systems 76 (3) 84 1-861. [ 181 F A 0 (2004) www.fao.org, Statistical Database. [ 191 International Energy Agency (IEA) (2001) www.iea.org. Statistics on World Energy. [20] Tiezzi E. (2001) Tempi Storici, Tempi Biologici, Venthnni Dopo. Donzelli Ed. Roma.

Solar Power Agriculture: A New Paradigmfor Energy Production

247

[21] Pazek TW. (1997) http:/lpatzek.berkeley.edu/Ell/Photosynthesis.pdf, based on Energy, Plants and Man by David Walker, OxyGraphics, England, Second Edition, 1993. [22] Kling GW. University of Michigan, http://www.globalchange.umich.edu/globalchange1/current/lectures/kling/energyfl ow/energyflow.html (accessed in 2004). [23] Haberl H,Wackernagel M, Wrbka T. (2004) Land use and sustainability indicators. An introduction. Land Use Policy 21 (3): 193-198. [24] Swenson FU3. (2000) Solar energy meets the new global challenge. http://www.ecotopia.com/apollo2/ises2OOO.htmISES Millennium Solar Forum, Mexico City. [25] Shapouri H,Duffield, JA, Graboski S. (1995) Estimating the net energy balance of corn ethanol. U.S. Department of Agriculture Economic Research Service, Office of Energy. Agricultural Economic Report No. 72 1. [26] Knapp KE,Jester TL. (2000) Proceedings of the Solar 2000 Conference, Madison, Wisconsin. [27] White SW, Kulcinsky GL. (2000) Birth to death analysis of the energy payback ratio and C 0 2gas emission rates from coal, fission, wind and DT-fusion electrical power plants. Fusion Engineering and Design 48 (3-4): 473-48 1. [28] Pimentel D, Eisenfeld HS, Olander L, Carroquino M, Corson C, McDade J, Chung Y, Cannon W, Roberts J. (1994) Achieving a secure energy future: environmental and economic issues. Ecological Economics 9 (3): 201-219. [29] Kitajima Y. (1997) www.formal.stanford.edu/jmc/nature/nodel 1.html (accessed January 2003) [30] Goldsmith E.(1972) Blueprint for Survival, London. [31] Dvorak JC. (2001) Sunshine in the forecast. Forbes Global Magazine 05.14. [32] Rural development in the European Union, Fact Sheet 2003, http://europa.eu.inticommlagriculture/rurlpublilindex_en. htm. [33] BP energy report. 2003, www.bp.com. [34] 2003 World Wind Energy Association (WWEA), http://www.wwindea.org/pdf/WWEA~InstallationFigures~World2003 .pdf. [35] Farhar BC. (1996) Energy and the environment: The public view. Renewable Energy Policy Project, College Park, Md, October. [36] Heberling M. (2002) Environment news, The Heartland Institute, editor. [37] Palazzetti M, Pallante M. (1997) L’uso razionale dell’energia, Bollati Boringhieri ed.


POSSIBILITIY ASSESSMENT OF WIND ENERGY UTILIZATION IN BOSNIA AND HERZEGOVINA F. BEGIC (l), A. KAZAGIC (l), N.H. AFGAN (*) (I)

JP Elektroprivreda BiH, Sarajevo, Bosnia and Herzegovina, Lisbon, Portugal

(2)

Instituto Superior Tecnico,

The limited reserves of fossil fuels and the need for environmentally friendly electricity generation, requires investigation of possibilities for the utilization of renewable resources in Bosnia and Herzegovina. One of the possible renewable sources is wind energy. The research of potential utilization of wind energy in Bosnia and Herzegovina presently is being carried out. As a part of the research, some locations are investigated using knowledge of atmosphere physics and meteorological conditions for given locations. The research is aimed to create basis for possible utilization of wind energy for electricity generation in Bosnia and Herzegovina, through the assessment of meteorological, technical, technological, economical, social, and environmental aspects of wind energy utilization at selected locations. The research comprises the following: Specification of decentralized wind energy systems - potential share and demand; Data collection of the direction, frequency and velocity of wind at the selected macro locations: in Sarajevo area (Mountain Bjelasnica and settlement Butmir) and in Mostar area (Mountain Velez); Selecting the potential micro-locations for installing the wind turbines field; Selection of the type and construction of the wind turbines; Sustainability assessment of electricity generation from wind turbines on the base of precisely defined criteria and sustainability indicators - Multi-criteria Sustainability Assessment; Optimization of the wind energy systems. The investigation is especially interesting for similar - prevailing mountain regions, and could answer whether utilization of wind energy for electricity generation in such regions is feasible. In the paper, investigation on wind characteristics in Sarajevo area is presented, as well as sustainability assessment of wind energy utilization being used within the research.

1. Introduction In the past several years, a continuous growing of the electricity consumption and generation in Bosnia and Herzegovina can be noticed. Among the current process of reconstruction of the electric power system in Bosnia and Herzegovina, common for countries in transition, refurbishment of the existing power plants, including investment in environment protection, is a process that is expected to continue. In the last five years, thanks to intensive efforts in this direction, electricity supply has been improved. However, assessments of electricity consumption in domain of JP Elektroprivreda BiH have shown that the growing of electricity consumption in the next period (till year 2020) should continue. Consequently, extending of energy power system by additional 100 MW of installed capacity is expected to be realised in close future. Due to unsatisfied the overall environmental issue, preferably high carbon dioxide emission, extending of energy system by renewable resources is optioned as well. In that sense, research on possibilities of utilisation of wind energy in Bosnia and Herzegovina, that has currently being carried out, is of fundamental significance [ 1,2,3,4].

2. Potential for Wind Energy Utilisation in Bosnia and Herzegovina On the base of data on wind characteristics from The Hydro and Weather Forecast Institute of Bosnia and Herzegovina, preliminary data base on wind frequencies and velocities for selected regions is created. A part of the results for Sarajevo area is 249

250

F. Begic, A. Kazagic & N.H. Afgan

presented here. Charts in Figs. 1 and 2 show frequencies and velocities of wind in respective directions for locations of mountain Bjelasnica and settlement Butmir, [2,3]. It can be seen that the average wind velocity for Butmir is about 2 m / s , so this macro location is rejected from further evaluation. However, the wind velocity for area of mountain Bjelasnica is potentially very convenient for installing the wind turbines, especially for directions NE-SW and N-S.

Fig. 1. Wind frequencies in respective directions in regions of Bjelasnica and Butmir sites.

Fig. 2. Wind velocities in respective directions for regions of mountain Bjelasnica and settlement Butmir.

The wind velocity at different altitudes can be estimated by relation (1):

Wind Energy Utilization in Bosnia and Hevzegovina

25 1

where: v wind velocity in Bjelasnica site vo wind velocity in Butmir site H altitude of top of Bjelasnica Ho altitude of settlement Butmir a average value of 4 prevailing directions Results of this calculation are given in Fig. 3.

I

I

500

7W

900

11W

13w

1500

17W

1900

21W

Atilude, m

I

Fig. 3. Estimated wind velocities for different altitudes.

On the base of presented data, two micro locations are selected for possible installing the wind turbines in Sarajevo area; Top of Mountain Bjelasnica and Lisicija Glava, see Table 1. Table 1. Selected locations for installing the wind turbines.

Average velocity Altitude Maximal power coefficient Rotor diameter Power Wind turbo-generator power

1. Top of Bjelasnica 12 m i s 2067 m A S . 0,4 40 m 451 kW 600 kW

2. Lisicija Glava 10 m i s 1762 m A.S. 0,4 40 m 451 kW 600 kW

3. Sustainability Assessment of the Wind Energy Utilisation Multicriteria sustainability assessment based on ASPID methodology is used for sustainability evaluation of wind energy utilisation in Bosnia and Herzegovina. The point of the used method is in synthesis of specific criteria which describe various aspects of the objects (in this case energy systems of various sources) on the base of initial attributes or indicators, and which are being evaluated by General index characterising all options under consideration. This basic method-MGI (Method General Index) is adopted by ASPID methodology (Analysis and Synthesis of Index at Information Dejciency) [ 5 , 61.

F. Begic, A. Kazagic h N.H. Afgan

252

The aspects considered in the analysis include consumption of resources, environmental issue, social aspect and economic aspect for all options under consideration. Option of wind turbines is compared against other seven options of energy systems for electricity generation of various sources. In this consideration, it is supposed that 7875 GWh of electricity is to be provided in projected lifetime of 15 years for all options under consideration. In calculation of sustainability indicators for wind turbines, in order to get realistic and reliable results, it is assuemd that materials for the wind turbines construction are produced by the electricity generated in a 100 MW pulvurised coal based power station an option under consideration that has a low efficiency (heat consumption of 10673 kJ/kWh or efficiency of 33%) and strong environment impacts. It is calculated that 438 units of 600 kW wind turbines are to be installed to provide aimed electricity generation of 7875 GWh over period of 15 years. Basic data related to wind energy utilisation from this analysis are given in Table 2. At the same time, data from this table presents input data for calculation of sustainability indicators of wind based electricity generation. Table 2. Data on wind energy utilization based on required electricity production.

Required electricily production

7 875 GWh

Life time

15 years

Annual Electricity production

525 G W a

Annual operation time

2000 Wa

Total Power installed

262 500 kW

Power installedper unit

600 kW

Number of units

438

Used area

1000 m2IGWh

Zauzetost zemljiita

7,9 x

Efficiency

25 %

Assumed energy costs

8 cent/kWh

Investment costs

1050 EURlkW

0 & M costs Total 0 & M costs

1,4 cent/kWh

lo5mz/kW

7 350 000 E m g o d .

3.1. Sustainability Indicators of Wind Based Electricity Generation A sub-stoichiometric quantity of oxygen, or air, is used in the partial oxidation fuel reaction. This small amount of oxidant produces some heat and water that reacts with the non-oxidized fuel to generate hydrogen and carbon dioxide, considering theoretical approach showed in Eq. (3). However, practical systems, based on partial oxidation technology, generate syngas with less hydrogen than other methods such as autothermal or steam-reforming. A large amount of carbon monoxide is yielded. The process temperature is also higher, around 1000°C.

Wind Energy Utilization in Bosnia and Herzegovina

253

3.1.I . Resources Indicators (H) Materials needed for construction of a 600 kW wind turbine are given in Table 3. Table 3. Materials for wind turbine construction.

456 134 12 3 3 lOppm) to avoid poisoning of the electrocatalyst of the FC. Internal reforming and electrochemical conversion of CO (CO is produced after reforming of hydrocarbon fuel) is only possible with high-temperature SOFC and MCFC and this essentially distinguishes them from low-temperature FC. Currently, the main fuel for all of these types of FC is hydrogen. Chemical storage of hydrogen in liquid fuels is considered to be one of the most advantageous way for supplying hydrogen to fuel cells. A variety of liquid fuels, such as methanol, ethanol, and hydrocarbon are suitable for this purpose. Among liquid fuels, ethanol is a promising source of hydrogen while it is an easily accessible and non-polluting raw material which can be readily produced from renewable sources by fermenting the sugars founding grains, such as corn and wheat, as well as potato wastes, cheese whey, corn fiber, rise straw, sawdust, urban wastes, and yard clippings. Hydrogen production from non-fossil, renewable energy sources such as ethanol to became technically, economically an ecologically important for sustainable development in the 2 1St. century. 335

V. Galvita

336

Thermodynamic aspects of ethanol steam reforming have attention in the literature [ 11. The overall reaction of hydrogen production from ethanol is strongly endotermic and corresponds to the formation of 6 mol of H2 per mol of ethanol: C2H50H + 3H20 = 2C02 + 6H2

(1)

However, other undesirable products such as CO, CH4, CHlCHO, and C2H4 are also usually formed during reaction. In addition, the formation of coke on the surface of the catalyst is also not uncommon. Coke formation may occur by Boudouard reaction: 2co+

coz + c

(4

Another possible rote for the carbon formation is through ethylene. At, present, two processes for ethanol SR (step (i)) are available, namely a onestep process on supported metal catalysts [2, 31 and two-step process in a two-layer reactor [4] (Fig. 1). The layer in such reactor are made of different type of catalysts which working at different operating temperature. Within the first layer ethanol is converted into a mixture of CH4, CO, and H2 [4]. Over the second catalyst layer, the products of the ethanol decomposition are converted by autothermal reforming to a gas mixture enrich with hydrogen Fig. 1.

CZHSOH ~~0 Catalyst Layer 1 H~ Ethanol decomposition ,

..

HTFuelCeU

Catalyst Layer 2 Autothermal Reforming

4 I(02) Air

coz H2

..

b

i HT Fuel Cell

Fig. 1. Scheme of a two-layer catalytic reactor for ethanol conversion into a hydrogen-rich gas mixture.

2. Experimental 2.1. Catalysts Preparation The catalyst Pd (1 wt.%) supported on Sibunit catalyst (PdC) prepared as described in [5] was used in the present study. The catalyst consisted of sphere particles of size 1.2-1.5 mm, pore volume and the BET surface area were 0.74 cm3/g and 400 m2/g, respectively. The CeOz-ZrOz was prepared via urea hydrolysis. The ceria and zirconia oxide samples was prepared from Ce(N03)3 and ZrO(NO&. The starting metal salt was dissolved in distilled water to the desired concentration 0.1 M. Then, the mixed metal salt solution will be added to a 0.4 M urea solution with the salt to urea solution ratio of 2: 1,

Hydrogen Production from Ethanol for Fuel Cell

337

and the mixture will be kept 48 h at 100 OC. The sample will be then allowed to cool to room temperature prior to being centrifuged to separate a gel product from the solution. The gel product will be washed with ethanol and dried for 12 h in an oven at 105 "C. Then, the product was calcinated at 400 "C for 4 h. The catalyst consisted of sphere particles of size 1-1.2 mm, pore volume and the BET surface area were 0.44 cm3/g and 120 m2/g,respectively. Catalysts containing 5wt.% Cu and 1 wt.% Pd will be prepared by incipient wet impregnation of Ce02-Zr02(or Pd/C for Cu-Pd/C catalyst) with an aqueous solution of Cu nitrate and Pd nitrate. The samples obtained were dried for 12 h in an oven at 105OC and then heated at 400 "C in air for Ce02-Zr02and in He flow for Pd/C. The CeOZ-ZrO2-CaO/Al2O3was prepared by the impregnation method. Commercial alumina granules were immersed in the solution containing metal components. The samples obtained were dned for 12 h in an oven at 105OC and then heated 6h.at 1000°C in air. Ni loading was carrying out by the incipient wet impregnation of Ce0z-Zr02-CaO/A1z03with an aqueous solution of Ni nitrate.

2.2. Catalytic Activity Measurements Experiments of ethanol decomposition were performed under atmospheric pressure, temperature of 200-450°C, WHSV = 1600-10000 cm3/h.g and inlet watedethanol molar ratio of 8.1-1.04. In a typical experiment a desired flux of helium and water-ethanol mixture to a plug flow reactor, which contained 25 mg of the catalyst diluted with an equal amount and size of quartz particles. The composition of the reactor effluent was analyzed by means of Agilent Mass Spectrometer (for HZ,CH4,C0,C02 EtOH and Ac). For autothermal reforming, a gas mixture composed of 10 vol. % of CH4, 10 vol.% COz, 20 vol. % H2, 4.25 vol.% O2 and 55.75 vol.% H20. The catalytic performance was measured with increasing reaction temperature 85OOC.

3. Results and Discussion 3.1. Effect of Reaction Temperature Homogeneous (empty reactor) steam reforming of ethanol was primarily investigated. It was observed that for temperature values ranging from 200 to 7OO0C the main reaction product was acetaldehyde, carbon oxides and methane Fig. 2, which appear at temperature above 40OoC. It is see that as temperature increases up to about 7OO0C the concentration of acetaldehyde passes through maximum, while a carbon oxides concentration increased with increasing temperature. For the same temperature interval the methane concentration was almost constant (Fig. 2). The ethanol conversion increases with increasing temperature and attains 16% at 70OoC. Furthermore, a considerable carbon imbalance was observed in the temperature range 400-700°C and the visual inspection showed, after a 2 h operation the wall of reactor was covered with carbon.

V. Galvita

338

In ref. [6] we have already discussed the effect of temperature on ethanol conversion and products distribution upon ethanol decomposition over P d C catalyst in steam presence. It was shown that when water-ethanol mixture (H20/CzH50Hmolar ratio

EtOH = 22.5 vol.% H,O = 67.5 vol.% He = 10 vol.% EtOHI H,O = 113

F 1.6 cm31s

o l Temperature, 'C

Fig. 2. Homogeneous (empty reactor) reforming of in the presence of steam and ethanol.

8.1-1.04) was fed into the reactor at WHSV 1600-2200 cm3/g.h, the reaction products were H2, CO, C02 and CH4, and 100% ethanol conversion was attained at 330-360°C. Analysis of the temperature dependencies of the outlet H2, CO, C02 and CHq concentrations proves that within the whole temperature interval the concentrations obeyed the following equations:

This means that CH4,CO and H2 were the products of ethanol decomposition by reaction:

CzH-jOH= C&

+ CO + H2

(31

while COz was the product of the water gas-shift reaction:

It is known [7-lo], however, that ethanol decomposition to C&, CO and H2 over metallic palladium containing catalysts proceeds by reactions:

CzH-jOH-+ CH3CHO + H2

(5)

CH3CHO + CH4 + CO

(6)

Hydrogen Production from Ethanolfor Fuel Cell

339

with the limiting stage being the ethanol dehydration, whereas acetaldehyde decomposition proceeding at high rate (Fig. 3).

0 O

Slow \(,

step

2 H 3I- 4 H + H2

Pd

Fast

step

___ __ __ Pd

Fig. 3. Mechanism of ethanol decomposition at Pd-catalysts

The knowledge of the ethanol decomposition mechanism allows to offer the usage of the bimetallic, bifunctional catalyst for an intensification of process in a two layer fixed bed reactor and for price reduction of the catalyst in the first layer. Ethanol-to-acetaldehyde dehydration will proceed on the first component of such catalyst, with acetaldehyde decomposition to carbon oxide and methane proceeding on the second one (Fig. 4). As reaction (3) effectively and with high selectivity proceeds on I-B group metals [4], copper might be used as the first component of the first layer. As the second component of the first layer it is necessary to retain Pd, which as against Co and Ni, does not give rise to the surface carbon deposition [4, 11, 121.

Fig. 4. Mechanism of ethanol decomposition at bimetallic-bifunctional Cu-Pd-catalysts

Fig. S shows comparative temperature dependences of ethanol conversion during its decomposition over bimetallic catalysts 5wt. %Cu-lwt.% Pd/C and 5wt. %Cu-lwt.% Pd/CeOz-ZrOz, and over lwt.% PdK. One can see that total conversion of ethanol is reached at 330' C over the bimetallic catalysts, whereas for the P d C catalyst at this temperature the conversion does not exceed 80 %. The reaction products for those catalysts were H2, CO, C02, CH4 and CH,CHO, except for Cu-Pd- catalysts at 33OoC, were CH3CHO was absent. The data of Fig. Sa prove the bimetallic catalysts being essentially more active and selective then the catalyst containing Pd only. Moreover, bimetallic catalysts increase the water gas shift reaction. Fig. Sb shows that CO

V. Galvita

340

selectivity for bimetallic catalysts decreased with increasing temperature. Selectivity PdIC at experimental condition was stable.

210

240

270

300

330

Temperature, OC Fig. 5a. Effect of temperature on ethanol conversion upon ethanol decomposition over bimetallic catalysts 5wt.% Cu - lwt.% Pd, and over lwt.% P d C . Experimental conditions: WHSV = 5000 cm3/h,g,inlet composition 8 vol.% CzHsOH + 92 vol.% HzO.

60

b

-

Iwt.% Pd 5w.%CulCeO,-ZrO,

I I 225

\\

I

I

I

I

250

275

300

325

b

, 33

Temperature, ' C

Fig. 5b. Effect of temperature on CO selectivity (b) upon ethanol decomposition over bimetallic catalysts 5wt.% Cu-lwt.% Pd, and over lwt.% PdC. Experimental conditions: WHSV = 5000 cm'hg, inlet composition 8 vol.% CzHSOH+ 92 vol.% HzO.

To estimate the feasibility to use Pd-Cu/Ce02-Zr02catalyst in ethanol decomposition in steam presence, it seemed reasonable to determine its stability under the reaction conditions. Under experimental conditions, (temperature 275"C, ethanol conversion was ca. 99% (100% ethanol conversion was attained at 280°C)); only H2, CH4, CO, C 0 2 and

Hydrogen Production from Ethanol,for Fuel Cell

341

C2H50H were detected at the reactor outlet. The outlet concentrations were stable for 50 h. There is no carbon imbalance (within f 5 % accuracy) was observed during this experimental run. Thus, the use of the bifunctional, bimetallic catalyst as the catalyst for the first layer of the two-layer reactor for ethanol decomposition proved to be quite promising for the steam reforming of ethanol to syngas. 3.2. Autothermal Reforming of Methane over Supported Ni Catalysts The effect of temperature and amount of Ni in catalysts In the autothermal reforming of methane is shown in Fig. 6. Methane conversion increases with increasing reaction temperature. Catalytic activity increase with increasing of amount of Ni. The thermodynamic equilibrium of methane reforming were attained only at temperature high 75OoC and for catalysts containing 5 and 10 wt % of Ni. Thus, for autothermal reforming of methane the Ni (10 wt.%) catalyst supported over Ce02-ZrO2-CaO/Al203showed high activity and stability. We can use the Ni- Ce02Zr02-CaO/A1203catalyst as the catalyst for the second layer of the two-layer reactor for methane autothermal reforming. 100

s

95

C

.-0

$ C

CH,"= 10 vol.%

co,,"= 10 vol.%

90

H,," = 20 vol.%

8

H,O,"= 20 vol.%

85

O,,"= 4.25vol.%

m

5 80

2 wt.% Ni

625

650

675

700

725

750

775

800

Temperature, OC Fig. 6. Temperature dependence of methane conversion for autothermal reforming of methane over Ni- Ce02Zr02-CaO/A120, catalysts.

4. Conclusions

The concept of the two-layer fixed-bed reactor seems to be quite promising for the steam or autothermal reforming of ethanol to syngas. According to this concept, ethanol is first converted to the H2, CO, C02 and CH4 gas mixture over a bimetallic catalyst (for example, Cu-Pd/Ce02-Zr02) and then the mixture produced is converted to syngas over another Ni-contain catalyst by autothermal reforming. It has been shown that the use of

V. Galvita

342

the two-layer fixed-bed reactor prevents coke formation and provides the syngas yield closed to equilibrium. It was found that the bimetallic catalyst being essentially more active then the catalyst containing Pd only. For autothermal reforming of methane (produced after ethanol decomposition in the first layer), the Ni (10 wt.%) catalyst supported over Ce02-Zr02-CaO/A1203showed high activity and stability for autothermal reforming of methane. References [I] Ioannides T. (2001) Thermodynamic analysis of ethanol processors for fuel cell

application. J. Power Sources 92 (1-2): 17-25. [2] Aupretre F, Descorme C, Duprez D. (2002) Bio-ethanol catalytic steam reforming

over supported metal catalysts. Catal. Communication 3 (6): 263-267. [3] Fatsikostas A, Kondarides D, Verykios X. (2002) Production of hydrogen for fuel cells by reformation of biomass-derived ethanol. Catal. Today 75 (1-4): 145-155. [4] Galvita V, Semin G, Belyaev V, Semikolenov V, Tsiakaras P, Sobyanin VA. (2001) Synthesis gas production by steam reforming of ethanol. Appl. Catal. A. General 220 (1-2): 123-127. [5] Semikolenov V. (1997) Zhurnal Prikladnoi Khimii 70: 785 (in Russ.). [6] Galvita V, Belyaev V, Semikolenov V, Tsiakaras P, Frumin A, Sobyanin V. (2002) Ethanol decomposition over pd-based catalysts in the presence of steam. React. Kinet. Catal. Lett. 76 (2): 343-35 1. [7] Isawa N, Yamamoto 0 , Tamura R, Nishikubo M, Takezawa N. (1999) Catal. Lett. 62 (2-4): 179-184. [8] Takezawa N, Isawa N. (1997) Steam reforming and dehydrogenation of methanol: Difference in the catalytic functions of copper and group VIII metals. Cutal. Today 36 (1): 45-56. [9] Davis JL, Barteau MA. (1987) Decarbonylation and decomposition pathways of alcohols on Pd (1 11). Surf: Sci. 187 (1-2): 386-387. [ 101 Shekhar R, Barteau MA. (1995) Structure sensivity of alcohol reactions on (1 10) and (1 11) palladium surfaces. Catal. Lett. 31 (2-3): 221-237. [ 111 Freni S, Mondello N, Cavallaro S, Cacciola G, Parmon VN, Sobyanin VA. (2000) Hydrogen production by steam reforming of ethanol: a two step process. React., Kinet. Catal. Lett. 71 (1): 143-152. [I21 Cavallaro S, Mondello N, Freni S. (2001) Hydrogen produced from ethanol for internal reforming molten carbonate fuel cell. J. Power Sources 102 (1-2): 198204.

HYDROGEN FUEL CELL URBAN BUSES OPERATING IN THE CITY OF PORT0 G.A. GONCALVES('), T.L. FANAS('), R. TEIXEIRA(2),A. SILVA(3) (1) IST-Instituto Superior Tecnico, DEM-STE, 1049-001 Lisboa, Portugal (2) STCP-Sociedade de Transportes Colectivos do Porto, Av. Fernao de Magalhaes 1862-13, 4350-158 Porto, Portugal (3) Portugal Comercio de Combustiveis e Lubrificantes S.A., Lagoas Park -Edijcio 3 2740-244 Porto Salvo The aim of the CUTE project was to develop and demonstrate an emission-free and low-noise transport system, including the accompanying energy infrastructure, which has great potential for reducing the global greenhouse effect towards the Kyoto protocol, improving the quality of the atmosphere and life in densely populated areas and conserving fossil resources. For the CUTE project a total of 27 buses were built by EvoBus. These vehicles are based on the Mercedes-Benz Citaro model city bus and run on hydrogen using fuel cells for power generation. One of the participating cities was Porto, where the local bus company (STCP - Sociedade de Transportes Colectivos do Porto) is operating three buses. The hydrogen infrastructure is supplied by BP. The hydrogen is produced centrally by electrolysis from grid electricity. In the project, IST is conducting the study of the environmental impact (based on a well to wheel approach) of the operation of the buses and the comparison with conventional technologies (Diesel, compressed natural gas and electricity). This will include measurements of the actual consumptions of the vehicles for the different routes, identification of the road profiles, driving behavior, load and ambient conditions.

1. Objectives The aim of the CUTE [5] project is to develop and demonstrate an emission-free and lownoise transport system, including the accompanying energy infrastructure, which has great potential for reducing the global greenhouse effect towards the Kyoto protocol, improving the quality of the atmosphere and life in densely populated areas and conserving fossil resources. For this purpose, 27 hydrogen fuelled fuel-cell buses were tested for 2 years in 9 European cities between 2003 and 2005. This project has financial support from the European Community and is currently the largest project worldwide in he1 cell technology demonstration in transports. 2. Project Partners

The participants of the project included the bus operators and local authorities for the 9 cities, BP and Shell as suppliers of hydrogen and the infrastructure and several research organizations. A more detailed list is presented on Table 1.

3. CUTE, ECTOS and STEP: 3 Projects Running in Parallel The ECTOS project aimed to implement and demonstrate the state-of-the-art hydrogen technology by running part of the public transport system with fuel cell busses within Reykjavik, Iceland. The hydrogen was produced from hydro and geothermal power, enabling a C02-free production chain. In the ECTOS project, three fuels cell buses similar to the ones used in CUTE were tested, and the project ran parallel in Reykjavik with the CUTE project. The STEP project is part of the Western Australian Government's commitment to working towards sustainable and 'zero emission' transportation in Western Australia. 343

G.A. Goncalves et al.

344

Table 1. Project participants.

s t a t h a f t SF Storstockholms Lokaltraffik (SL) Stuttgarter StraBenbahn AG (SSB) Sydkraft AB Transports Metropolitans de Barcelona S.A. (TMB) University of Iceland ECTOS University of Stuttgart (USTUTT) ~

NO SE DE SE

ES IS

DE

Energy supplier Public transport authority Public transport authority / company Hydrogen facility manufacturer Public transport authority / company Research Research

Three hydrogen fuel cell buses similar to the ones used in the CUTE project were trialed on normal Perth service routes for two years, from early 2004 to 2006. 4. The Vehicle

The vehicle being used in the project (Fig. 1) is based on the standard Diesel powered Citaro bus commercialized by EvoBus (a subsidiary of the DaimlerChrysler group). The engine and transmission are located in the same position and the vehicle offers the same

Hydrogen Fuel Cell Urban Buses Operating in the City of Porto

345

passenger space. In order to increase the r e ~ ~ a b iofl ithe ~ vehicle, the changes have been reduced to a m ~ n ~keeping ~ u ~as ,much of the conventional bus parts as possible.

Fig. 1 . 1 l e CUTE project bus.

The entire -fuel cell system i s located on the roof of the bus (Fig. 2). This required some changes in the suspension and chassis in order to cope with the extra weight. in a high location, This has been decided based on several considerations: Safety in terns of accidents, as the rooftop ofthe bus i s rarely damaged in the case of a collision; ]Inthe case o f leakage, the hydrogen is ~ ~ e d ~vented a t toe the ~ atmosphere; ~ The location provides easy access for maintenance. X 1. ~~~r~~~~Sforage

The hydrogen is stored on board in compressed form at a pressure of 350 bar. 'The storage ~ n each ~ with a volume of module uses 9 cylinders (carbon fiber wrapped a l ~ m liner), 205 Wes. The total storage capacity (at the nominal pressure of 350bar) is 44kg of hydrogen, suflficient to fulfill the typical daily range requirements for city buses.

Fig. 2. Location of the main elements.

346

G.A. Gomalves et a1

4.2. Fuel Cell The fuel cell is of the PEM type, and the two stacks are capable of providing a total gross power of 250 kW. The DC current produced is converted into AC in an inverter before being fed to the main electric engine (Fig. 3). In order to ensure the simplicity of the system, all other electrical components are supplied from the main engine.

4.3. Engine and Power Train The electrical traction engine is placed in the same position as in a conventional bus (Fig. 4). It is designed to a maximum power of 205kW, comparable to the power of conventional diesel and natural gas city buses. The rear axle and automatic transmission are similar to the ones used in the conventional Citaro model, the changes are essentially in the gearbox gear ratios, which have been adapted for the specific characteristics of the engine. The presence of the gearbox ensures an easier adaptation of the drivers.

I I

Cwnprerror

Cooling

Expander

Pump

Hydrogen

wafer

mmf E

Inverter

Electricat Traction

Tank

Engine

Fig. 3 . Operation of the fuel cell system (Source DaimlerChrysler AG).

Fig. 4. Engine and gearbox location.


347

5. Hydrogen Supply Infrastructure One of the most important factors when developing an alternative fuel transportation system is the capacity to supply the fuel, especially if the nature of the fuel prevents it from being distributed through conventional refueling stations. This is the case for hydrogen, which requires a specific fuelling infrastructure. Some of the bus companies that participate in the project already have recognized experience with gaseous fuels, owning a significant number of natural gas buses. In the case of hydrogen, another problem is the need to manufacture the fuel. This can be done in several ways and from a wide variety of primary energy sources, and was one of the goals of the CUTE project to analyze different combinations of production and distribution of hydrogen.

5.1. Hydrogen Supply Routes One of the main drivers to the introduction of hydrogen is the possibility to have a local as well as global C 0 2 free system. For that purpose, a renewable energy source must be used, and that is the case when using hydro or geothermal power. However, cost and availability issues bring the possibility of using conventional fossil fuels as primary energy sources. At the present stage, steam reforming of natural gas is the most probable route to produce hydrogen on a large scale (needed if it is to be introduced in widespread use as transportation fuel). For the CUTE project, given its demonstration status, it was been decided to cover a broad range of alternatives, not only in what refers to primary energy source but also if the production is centralized or on-site. The routes taken arc summarized in Table 2 . Different filling stations were used in each city, and the hydrogen is either stored at a high pressure (-450bar) prior to refueling or is compressed to a suitable pressure during filling. In any case, the filling time is not expected to exceed 10 minutes. The use of different filling station technologies allowed drawing conclusions to what is the best way to provide hydrogen at such high pressures, regarding not only the energy consumed for compressing the hydrogen, but also the availability of the filling station. Table 2. Hydrogen supply routes used in the CUTE project.

Cities within the project Amsterdam, the Netherlands Barcelona, Spain Hamburg, Germany

I Primary energy source

I Hydrogen Production

IRenewable (solar)

IOn site (electrolysis)

I Renewable (meen electricitvl

I Renewable (wind)

I On site (electrolvsisl

I On site (electrolysis)

London, United Kingdom

Crude oil

Centralized (transport in liquid form)

City of Luxembourg

Natural gas

Centralized (transport in gaseous form)

Madrid, Spain

Crude oil + Natural gas

Porto, Portugal Reykjavik, Iceland - ECTOS

I IVariable (grid mix)


I + On site (Steam ieformei)


Renewable (Geothermal and hydro energy)

On site (electrolysis)

Stockholm, Sweden

Renewable (green electricity)

On site (electrolysis)

Stuttgart, Germany

Natural gas

On site (Steam reformer)

G.A. Goncalves ed QE.

448

5.2. Hydrogen Infrastructure in Porio

The hydrogen ~ f r a s ~ inc Port0 ~ e (Fig. 5 ) was supplied by BP. The hydrogen is produced in the Linde factory located in Alenquer ~ o u g an h electrolysis process using grid elechicity. It is then an sported by road in tankers (in gaseous form at 200 bar) to the falling station. The trailers are then connected to the station, sewing as s t a ~ ~ o n a ~ storage. As the storage presswe in the buses has a m ~ i m u mof 350 bar, in order to transfer the hydrogen to the buses an additional compression stage is required, raising the filling presswe to 400 bar (the system must also cope with tlae reduction in the p r e s s ~ eof the t a k e r as the hydrogen is transferred). The station has parking space and connecting equipment for two trailers.

Fig. 5. Hydrogen tanker, i ~ t e ~ ~ d storage i a t e and filling station with dummy test tank.

6. city $pecific ~ o n ~ ~ ~ o n ~

Adding to the different. hydrogen production routes, each city presents specific operation e s . factors are s ~ ~ conditions in trafGc, topography and average t e ~ p e r a ~ ~These in Table 3. 'The number of passenger transported indudes all the modes of transport operated by the transport company. Table 3. City specific characteristics.

Cities Within the project

Barcelona, Spain

I 1.300.000

I

7. Bus ~~~~~~~o~ in Porto

The buses were operated by STCP. Maintenance infrastrucwes were created that meet the safety regulations for operation with highly volatile and flammable gaseous &el

~


349

(Fig. 6). Equipment installed includes H2 detectors, gas extraction equipment and fire detectors.

Fig. 6 . Hydrogen bus depot location at the Francos station.

The route for the operation of the buses was selected for the high gradients and good visibility to the population. It’s a circular line (Fig. 7) and the hydrogen buses were operated as complement to the Diesel and natural gas vehicles. Therefore all the vehicles were operated under the same conditions. 8. Environmental Impact of the Hydrogen Fuel Cell Bus (Well-To-Wheel Analysis)

When using hydrogen as fuel, the local emissions of a fuel cell bus consist only of water. However, to evaluate the global environmental impact of this fuel, a complete analysis must be made covering all the steps in the production and use of the fuel, from the primary energy source to the end use - a well-to-wheel analysis. This task is being performed by IST together with the University of Stuttgart and consists of two phases.

Fig. 7. Route layout.

Within the first phase, a well-to tank analysis was be performed by the University of Stuttgart. This analysis covers the life of the fuel from the primary energy source (renewable, crude oil or natural gas) to the moment it is filled to the on-board reservoir.

350

G.A. Goncalves et al.

Using a Life Cycle assessment tool (GAB1 Software [ 11, develop at the University of Stuttgart), the environmental impact of the production of hydrogen was compared with the impact of conventional fuels, as can be seen in Fig. 8. WELL-TO-TANK

011

Fig. 8. Well-to-Tank analysis.

In the second phase, the Tank-to-Wheel analysis was performed by IST in order to quantify the impact of the operation of the fuel cell bus and compare it to conventional Diesel and Natural Gas buses (see Fig. 9). TANK-TO-WHEEL \

/

VEHICLE AND FUEL

OPERATION PARAMETERS

\

I

TAILPIPE EMISSIONS

TOPOGRAPHY

I

CLIMATE

1

gas bus

\ I Hydrogen bus

~

Fig. 9. Tank-to-Wheel analysis.


351

For this purpose three tools were used: 0 Mobile6 [2], a tool developed by the EPA (Environment Protection Agency, United States) to calculate the emissions of road vehicles; Copert [3], which fulfils a similar role and represents the European market; 0 EcoGest [4] is a tool developed at IST that uses as input the characteristics of the vehicle, driving behavior and topography and, based on those inputs, calculates on a second by second basis the power needed and the resulting emissions and fuel consumption. These three tools were used in order to ensure reliable results from the simulations: Mobile and Copert are proven tools that calculate moderately accurate results, while EcoGest allows for much more precise calculations. In addition to the simulations, the vehicles will be monitored during their regular operation. This monitorization will be extended to the conventional Diesel and natural gas vehicles. In this case the parameters monitored will be dynamic performance and fuel consumption. The combination of the life-cycle analysis of the production and distribution of the fuels, together with the values from the operation of the different technologies produced a well-to-wheel analysis for the fuel cell bus as well as conventional Diesel and natural gas buses, and thus allowing for the completion of the remaining IST tasks within the CUTE project: Evaluation of the Environmental Impact of the bus operation; 0 Compare Hydrogen with conventional Diesel/Natural Gas technologies; 0 Contribution to the objectives of the Kyoto Protocol; Monitoring of the Energy and Environmental performance of the vehicles. Acknowledgments

The CUTE project was co-financed by the European Union. CUTE is one of the research projects supported by the European Commission, DG Energy and Transport (NNE5/113/2000). One of the partners, STCP, has support from DGTT (Direccao Geral dos Transportes Terrestres). References [ 11 IKP (1992-2002): GaBi Software-System and database for Life Cycle Engineering; www.gabisoftware.com [2] US EPA User's Guide to Mobile6.1 and Mobile6.2: Mobile Source Emission Factor Model, Office of Transportation and Air Quality, United States Environmental Protection Agency, 2002. [3] Ntziachristos L, Samaras Z. (2000) COPERT 111: Computer program to calculate emissions from road transport - Methodology and emission factors (Version 2.1), European Environment Agency. [4] Silva CM, Farias TL, Mendes-Lopes JMC (2002) EcoGest - Numerical modeling of the dynamic, fuel consumption and tailpipe emissions of vehicles equipped with spark ignition engines. Proceedings of Urban Transport and the Environment 2002 - Eighth International Conference on Urban Transport and The Environment in the 2 1st Century, 13-15 March, Seville, Spain. [5] http://www.fuel-cell-bus-club.com.


ENERGY IN SLOVENIA AND CROATIA: CHALLENGES AND POSSIBILITIES FOR SUSTAINABILITY A. ZIDANSEK“32’,R. BLINC‘22”,1. SLAUS (3), D. NAJDOVSKI‘4’ “)Jozef Stefan International Postgraduate School, Ljubljana, Slovenia ‘2)J.Stefan Institute, Ljubljana, Slovenia “R. Boskovic Institute, Zagreb, Croatia

‘‘jX3data, Ljubljana, Slovenia Three components of a strategy to reduce the negative environmental impact of energy use in Slovenia and Croatia are proposed. This strategy is based on energy production and energy consumption data, which are are presented and analyzed for Slovenia and Croatia. The first component is maximising the use of renewable energy sources. The second strategy component comprises scenarios for changing consumption patterns to meet the sustainability requirements. The third one is emphasising fundamental curiosity driven research that throughout the history has proven to be the most effective solution for various problems including those of energy production. These strategies are being analyzed for Slovenia and Croatia. Besides the technology fix we discuss also social implications of a change in consumption patterns. These three components of the strategy aim not only to meet the requirements of the Kyoto protocol and EU standards, but to make significant progress toward achieving genuine sustainable development according to the original Brundtland Commission definition.

1. Introduction Development is interconnected with energy. Sustainable development requires that energy is being produced and consumed in a way that does not produce a negative impact on the environment, and that the quality of life and development of future generations are not jeopardized. Each country has to establish its own best strategy for achieving sustainable development. Slovenia has just become a member of the EU, and Croatia is bordering with two members of EU and is aspiring to the full membership in EU. Data on energy production and energy consumption for Slovenia and Croatia are presented and analysed and then we proceed to outline the strategy for reducing the negative environmental impact of energy use. 2. Analysis of the Current Energy Production and Consumption in Slovenia and Croatia

Primary energy production, energy consumption, energy consumption per capita and energy consumption per unit GDP for Slovenia and Croatia in comparison with Austria and Macedonia are shown in Figs. 1-4. At present renewable energy sources in Slovenia amount to only 7% of the end energy consumption, if small hydroelectric power plants are also counted as renewable energy. This involves mainly biomass, small hydroelectric power plants, geothermal and solar energy, while large hydroelectric power plants are here not considered as renewable energy sources. In Croatia the renewable energy includes mainly biomass-fuel wood [l-31.

353

354

A. Zidansek et al.

A

A

-0-

Croatia

I

I

I

I

I

1992

1994

1996

1998

2000

2002

Year Fig. 1. Primary energy production for Slovenia and Croatia in comparison with Austria and Macedonia in EJ (lOIR J) [l].

1

I

I

I

I

I

1

a

-v-

v--r-v--v-v-v

-.--\v1992

Macedonia

I

I

I

I

1994

1996

1998

2000

2002

Year Fig. 2. Total primary energy consumption for Slovenia and Croatia in comparison with Austria and Macedonia in EJ ( l o l a J) [l].

355

Energy in Slovenia and Croatia

180

$, 170 Q

8

160 150

L

140

c

0 .c

130

;

a 120 110

8

100

h

p 90 (u

5

80

2 m

70

.E

60

?I

1992

1994

1996

1998

2000

2002

Year Fig. 3. Energy consumption per capita for Slovenia and Croatia in comparison with Austria and Macedonia in GJ (lo9 J) 111.

1992

1994

1996

1998

2000

2002

Year Fig. 4. Energy consumption per unit GDP for Slovenia and Croatia in comparison with Austria and Macedonia in H/US $ GDP [l].

356

A . Zidansek et al.

Table 1 summarizes selected energy data for Slovenia and Croatia. Slovenia and Croatia are equal owners of the Nuclear plant Krsko (NEK), which is located in Slovenia. Table 1. Energy data for Slovenia and Croatia in 2001 [1-3].

Total Electricity Production

Slovenia

Croatia

13.7

12.1

13.8

14.3

35.2%

33.6%

36.8% 1.1 53 1040

3.0 60 1080

in TWh Total Electricity Consumption in TWh ~

Electricity Production by Source

Net Electricity Import Net Oil Import Net Natural Gas Import

Nuclear in TWh in thousand bbl/day in million cubic meters

The fraction of renewable energy sources, excluding hydro-energy, in the electrical energy both in Slovenia and Croatia is even lower as the following discussion shows. About 100 000 square meters of solar collectors are currently installed in Slovenia. The current peak power of the installed photovoltaic systems is about 100 kW. A large part of the photovoltaic systems has been installed in the rural area of the Triglav national park. The average daily solar irradiation in Koper at the Adriatic coast is 5800 WWm2 in summer and 1000 WNm2 in winter, whereas in central Slovenia (Ljubljana) the corresponding irradiation is 5300 WWm2 in summer and 600 WNm2 in winter. The cost of solar energy is however relatively high. The estimated geothermal energy potential of Slovenia is about 50-70 MW. The biomass potential is estimated to 40 MW. Croatia is the owner of half of the nuclear power plant in KrSko, so that a fraction of its energy comes from this source. About 150 000 square meters of photovoltaic systems are currently installed in private houses and hotel roofs. The total installed capacity of solar energy in Croatia was 6 MW in the year 2000, while the total potential of the solar energy is 100 PJ. The solar energy panels in Croatia are mainly installed along the coast and in the islands. The installed biomass capacities amount to 7 MW.

3. Components of a Strategy to Reduce the Negative Environmental Impact of Energy Use We propose three components of a strategy to reduce the negative environmental impact of energy use. The first one is maximising the use of renewable energy sources. The second strategy component comprises scenarios for changing consumption patterns to meet the sustainability requirements. The third one is emphasising fundamental curiosity driven research which throughout the history has proven to be the most effective solution for various problems including those of energy production.


351

A comparison of C 0 2 emissions per capita for Croatia and Slovenia (Fig. 5) demonstrates that the emissions in Croatia are increasing in recent years, while emissions in Slovenia are already at a very high level which is comparable to developed countries. The increase in Slovenian emissions between 1994 and 1999 was mainly due to the fact that prices of gas were significantly lower than in the neighbouring countries.

i-

/

/

-m--0-

Slovenia Croatia

-v- Macedonia

1992

1994

0

1996

1998

2000

2002

Year

Fig. 5 . COz emissions in metric tons carbon equivalent per capita for Slovenia and Croatia in comparison with Austria and Macedonia [l].

3.1. Increasing Renewable Energy To reach lower emission of greenhouse gasses in accordance with the Kyoto protocol and European Union standards Slovenia has to increase renewable energy sources (without hydroenergy) to reach 12% till 2010, and Croatia plans to reach 12.2% of the end energy consumption by 2030. So far no commercial wind energy plants are operating neither in Croatia nor in Slovenia, but several pilot projects are planned and some of them are in the testing phase. Slovenia plans that photovoltaic systems with 3 MW peak power should is installed in the next years. A geothermal pilot power plant of 5 MW is planned to be installed till 2010. Until 2010 the installed biomass power should be 8 - 10 MW. The biogas power plant potential at animal farms and waste treatment facilities in cities on the other hand is estimated to additional 10-30 MW. The wind energy sources have a relatively large potential and the installed power till 2010 is estimated to 40-80 MW

[1,21. The total wind energy potential on Croatian islands is 209 MW, while it is 163 MW at the Adriatic coast. The total installed capacity of geothermal energy sources in

358

A . Zidansek et al.

Croatia in 2000 was 36.7 MW, whereas the total geothermal potential is estimated at 839 MW [ l , 31. 3.2. Sustainable Consumption Patterns Presently, an energy consumption pattern is determined mainly by economic considerations, as well as by personal and local convenience. This has resulted in a large fraction of obsolete energy components producing large negative environmental impact. Currently, energy consumption is characterized by large inefficiency. Analyses and practices in several countries have shown that it is possible to maintain and even increase GDPicapita without increasing energy consumption and even decreasing it. This strategy component is based on modem technologies which are energy frugal, notably on information and communication technologies. It is necessary to simultaneously analyse for each country social, economic and environmental impact introduced by these modem technologies. Implementation of sustainable consumption patterns is slow, because production cost for renewable energy is much bigger than for the energy from fossil fuels or nuclear energy. This cost estimate becomes completely different, if external costs are also taken into account in addition to direct costs of energy production. Here also the price of the damage caused to the environment is taken into account. In Slovenia this damage due to climate changes alone exceeds 100 million Euros annually. In Slovenia until 2010 greenhouse gas emissions will rise by about 10% according to the EEA estimates instead of decreasing by 8% versus the 1986 baseline as agreed by the Kyoto protocol. According to most climate models the total world greenhouse gas emissions should be reduced well beyond Kyoto requirements to about one half of the current emissions in order to stabilize the climate and precent the negative effects caused by the threat of climate changes. In Slovenia C 0 2 emissions alone represent 16.2 million tons anually. These emissions cause external damage estimated to 18 euros per tone by EEA. If we assume that one half of these emissions should be reduced in order to stabilize the global climate, this one half of the emissions amounts to the total of 140 million euros external costs. If Kyoto protocal was accepted, Slovenia should already buy the emission rights for more than 3 million tons COz anually, which amounts to about 60 million euros anually. If this amount is used to support renewable energy sources and in this way take into account the external costs of energy use, the renewable energy sources would become competitive. This can of course not be done at a single country level, but a global agreement should be reached. If we want to achieve sustainable consumption patterns, a global green tax reform is therefore needed. The cost of environmental damage produced by production of energy must be included into the consumer price.


359

3.3. Promising R&D Directions Fundamental curiosity driven research throughout the history has shown that major breakthroughs almost always have large social implications. For the energy related field the well known example is Faraday’s discovery of electromagnetic induction. It is this type of research that has to be encouraged and stimulated. However, often it is said that small countries should try to concentrate on applied research, but we strongly believe that this is not enough since: research resulting in breakthroughs is enriched by differences in culture and environment; it is impossible to find the best solution for a country by just applying outside recipes without strength which is assured only by fundamental research; applications come from fundamental research, which then is the prerequisite for any successful application. It is not the aim of this paper to go into details of what kind of fundamental curiosity driven research should be done in these two countries. It suffices to say that high quality fundamental research must be supported at an adequate level. While fundamental curiosity driven research can’t be planned, mission oriented fundamental and applied research should be planned. In the following some promising R&D directions are presented.

3.3.1. Hydrogen Research

A promising energy source is hydrogen. In principle hydrogen is a completely clean energy source and its only emission at the end user is clean water. However, the problem is that the most efficient way to produce hydrogen is electrolysis, which demands electricity usually obtained from the fossil fuels. A possible solution is to obtain the electricity from a renewable source. In this case hydrogen acts basically as substance to transmit energy. Starr suggests using the nuclear energy as a source for the Continental Super Grid, where the energy is transmitted to the end user via hydrogen, which feeds fuel cells [4]. A better solution would be the use of biologically derived hydrogen. While the research in this field is still in the initial phase, there are some promising results with bacteria [ 5 ] and green algae [6]. The research in nanotube with a possible application for hydrogen storage is beeing developed by the laboratories at J. Stefan Institute. Fuel cell cars have recently been introduced as a possibility to use hydrogen as a clean energy source. The advantage of these cars is that they create no polution in the city, although there is still polution created in producing hydrogen. 3.3.2. Zirconium Ceramicsfor Nuclear Waste Storage

Slovenia and Croatia own a joint nuclear power plant which produces significantly less COz emissions per MWh than fossil fuels. It does not require burning fossil fuels, but uses uranium fuel instead. Therefore there is only a small amount of indirect emissions, connected with the uranium fuel cylcle, and even this only in case that fossil fuels are used in preparation, transport and treatment of the nuclear fuel. Nuclear energy is

360

A. Zidansek et al.

however connected with other risks. One of the most important issues is the necessity to solve the problem of nuclear waste. A possibility to secure sustainable treatment of nuclear waste is the use of zircon ceramics to embed spent reactor fuel. Several billion years old natural zircon crystallites clearly show alpha particle tracks due to uranium impurities, which proves the efficiency of zircon based materials to assure long term containment of uranium [7]. Both J. Stefan Institute and R. BoSkoviC Institute are planning R&D projects in this direction.

3.3.3. Piezo and Solar Panels as Noise Barriers along the Highways One of the possibility to increase the use of photovoltaic power sources is to utilize solar cell panels as anti-noise barriers along the highways. Since anti-noise barriers amount to about 20% of the cost of building the highway roads, the cost of solar energy obtained in this way would be much lower than that from other commercial solar sources [8]. Two projects are currently in pilot phase at two different sections of the highway under construction in Slovenia. The first project is based on the use of highway noise barriers equipped with photovoltaics and piezomaterials for energy production. Technical documentation for the test field at the KrSka vas - Obre2je highway in the length of 100 m has been prepared. In this case the noise barrier will be equipped with photovoltaics and piezomaterials. This project is aimed to demonstrate how much energy can be obtained from the noise barrier while the noise protection is improved. The noise barrier device will serve as a self organized energy device where on one hand the energy is obtained from photovoltaics and piezomaterials, and on the other hand the device also needs energy for absorption of high frequency acoustic noise emissions. The goal will be to satisfy these energy needs with only part of the produced energy, so that the energy balance of the device is positive. Another similar project is designed for a windy area of the Razdrto-Vipava highway in the initial length of 500 m with a possible extension to 8 km.Here the noise barrier will also serve as a wind barrier. Also in this case the energy balance will be studied. Due to high wind loads the wind represents an additional source of energy, which will improve the energy balance of the device. 4. Conclusions

Suggested energy strategies for Slovenia and Croatia are summarised as follows: 1. Increase GDPkapita maintaining energy consumption per capita constant or at most keeping GDP/capita growth twice as large as the energy consumptiodcapita growth. 2. Increase the fraction of solar, geothermal, biomass and small hydro energy in 2010 to be 50% larger than it is in 2004. 3. Maintain and develop nuclear science and nuclear technology skills so that strategies to reduce nuclear waste problems can be implemented. 4. Change the pattern of energy consumption by emphasising renewable energy sources and reducing fossil fuel.


361

5.

Increase R&D support and R&D activity to provide opportunities for major breakthroughs and to assure incremental improvements in energy technologies, in particular renewable technologies. The negative environmental impact of energy use in Slovenia and Croatia has been analyzed. The analysis leads to a pragmatic policy to keep all energy options open allowing each one to seek its role in a competitive and economical energy mix. Discussed changes in the consumption patterns represent an important option for reducing negative impacts and are therefore a vital component of a successful strategy for sustainable development. The fastest way to change the consumption patterns is the implementation of green tax reform, where external costs of the environmental damage are also included into the consumer price. This would however significantly increase the price of fossil fuel energy, and is thus only acceptable, if a global agreement is reached. A proper combination of green tax reform and research can thus lead to new energy consumption patterns which are both environmentally sustainable and economically viable. Here environmental sustainability aims to reduce the fossil fuel use well beyond the proposed Kyoto targets in order to stabilize the global climate, and increased support for energy research is expected to decrease the price of clean energy so that it becomes competitive with the fossil fuels. If only a small fraction of external environmental costs were spent on research, clean and renewable energy technologies would become cheaper, new energy sources like fusion and energy producing bacteria closer, and radioactive waste storage safer. References [I] Central Intelligence Agency (2003) The World Factbook. Potomac Books, Washington, D.C. [2] Ministry of the Environment, Spatial Planning and Energy of the Republic of Slovenia (2002) Porocilo o stanju okolja v Sloveniji 2002. Ljubljana. [ 3 ] Vuk B, Marusic D. (2002) Energija u Hntatskoj 2001 - Energy in Croatia 2001, Ministry of Economy of the Republic of Croatia, Zagreb. [4] Starr C. (2001) National Energy Planning for the Century. ANS Winter Meeting, November 13, Reno, Nevada, USA. [5] Van Ooteghem SA, Beer SK, Yue PC. (2002) Hydrogen production by the thermophilic bacterium Thermotoga neapolitana. Applied Biochemistvy and Biotechnology 98: 177-189. [6] Melis A, Happe T. (2001) Hydrogen production. Green algae as a source of energy. Plant Physiology 127: 740-748. [7] Scott JF, private communication, 2003. [8] Nordmann T, Frolich A, Durr M, Goetzberger A. (2000) First experience with a bifacial PV noise barrier. 16th European Photovoltaic Solar Energy Conference and Exhibition, Glasgow, United Kingdom.


SUSTAINABILITY ASSESSMENT AS A BASIS FOR THE DECISION MAKING IN SELECTION OF ENERGY SYSTEM F. BEGIC"), N.H. AFGAN(*) (I)

J.P.Elektroprivreda Bosnia and Herzegovina, Sarajevo, Bosnia and Herzegovina (2) Instituto Superior Tecnico, Lisbon, Portugal

In this paper the multicriteria sustainability assessment of various options of the energy power system of the JP Elektroprivreda of Bosnia and Herzegovina is performed. The rehabilitation of a 110 MW Thermal Power Unit, is considered in comparison with some other comparable options, such as Thermal power unit with coal fuelled boiler with combustion in fluidized bed, Combined cycle gas turbine plants, Hydro Power Plant, Power plants based on solar energy (PV systems), Wind turbines and Bio-mass power plants. Used methodology is based on a system of stochastic models of uncertainty, enabling to realize the assessment from various supporting systems, as well as enabling to obtain various normalization indexes by using Non-numeric (ordinal), Non-exact (interval) and Non-complete information (NNN-information) obtained from sources of various reliability and probability. Trough the analysis of multicriteria assessment of various options the decision-makers are offered to evaluate different options and make selection of the optimal option which meet the muticriteria evaluation.

1. Introduction

Most of decision models used in evaluation of complex systems today are based on single criteria analysis. However, in our modern age, decision-making based of single criteria analysis becomes unacceptable, if we set priorities that include costs, requirements, possibilities and risks. It is often necessary to consider several aspects at the same time. Energy power system is a complex system which evaluation depends on a number of economic, environmental, social and technological parameters. One of the perspective methods for evaluation of energy system quality is multicriteria sustainability assessment based on the analysis and synthesis of indicators expressing different aspects of the system. Application of this method in the cases of information deficiency (ASPID methodology) enables evaluation of various energy systems. In this work, multicriteria sustainability assessment of energy systems of various sources is applied to the Energy Power System of Bosnia and Herzegovine in order to investigate this complex system.

2. Energy Power System Under Consideration In the last five years, as the result of intensive efforts of power plants rehabilitation and refurbishment of entire electric power system, electricity supply in Bosnia Herzegovina has been improved. In year 2000, the electricity generation reached 63% of electricity generation in pre-war 1990 year, Fig. 1. The assessments of electricity consumption in domain of JP Elektroprivreda BiH, estimated by various consulting companies, have shown that the growing of electricity consumption in the next period (till year 2020) should continue. Today, the combined capacity of all power plants in Bosnia and Herzegovina is 3940 MW, of which 1983 MW is in hydropower plants (not including the small hydropower plants) and 1957 MW is in thermal power plants. The capacity of power plants belonging to JP Elektroprivreda BiH - a largest power company in Bosnia, is 1839 MW, or 47% of total capacity in Bosnia and Herzegovina. From this power generation, 482 MW is in hydropower plants, and 1357 MW in thermal power plants. An additional 3 63

F. Begic & N.H. Afgan

364

1990

1991

1992

1993

1Y94

1995

1996

1597

1998

1999

2000

Fig. 1. Trend of power generation and consumption within a realistic and complex energy power system under consideration.

11 MW is available to JP Elektroprivreda BiH from small hydropower plants, bringing total capacity to 1850 MW. Thermal power plants are taking the mean load in the electricity generation in Bosnia and Herzegovina. Nowadays, the electricity generation in thermal power plants is app.73% of overall electricity generation. Present growing of electricity consumption demand requires installation a new power capacities very soon. In this light, the multicriteria sustainability assessment based on ASPID methodology is used for the evaluation of sustainability of different options of energy power systems and various sources.

3. Sustainability Assessment of Energy Systems of Various Sources 3.1. Option Under Consideration Eight options for providing the additional power system options within Energy Power System of Bosnia and Herzegovina are considered. Renewable sources and fossil fuel clean technologies are included into consideration. The following options are considered: Option 1: Reconstruction of pulverized coal fired unit in condensing regime; Option 2: Reconstruction of coal fired unit in cogeneration regime; Option 3: Fluidized bed combustion unit-New power station; Option 4: Combined cycle gas turbine power station-New power station; 0 Option 5: Reconstruction of hydro power plant; 0 Option 6: Power plants on solar energy (PV systems)-New power station; Option 7: Wind turbines power-New power units; Option 8: Biomass power plants-New power station. Requirement for all options of energy systems under consideration is production of 7875 GWh of electricity over lifetime of 15 years. This gives average annually electricity production of 525 GWh. Some of energy systems under consideration can produce simultaneously electricity and heat energy (cogeneration), taking into account that this cogeneration plants also must produced 7875 GWh of electricity in projected lifetime.

Sustainability Assessment for the Decision Making

365

Within option 1-Reconstruction of pulverized coal fired unit in condensing regime, the 110 MW pulverized coal fired power plant with slag type boiler furnace is considered, operating in condensing regime-without steam extraction from the steam turbine. Investment at this option is aimed to improve boiler efficiency and upgrade the power. By reconstruction of the boiler, the boiler efficiency is increased from 85% up to 87%; the slag tap furnace is extended by 2 m and new burning system (low NOx burners) is introduced. Additionally, upgrading of the plant is performed through reconstruction of the steam turbine; output power is increased from 110 MW up to 118,55 MW. All these measures affect environment issues, reducing emissions of CO2, NO,, and SO2 per produced kWh. In the option 2-Reconstruction of coal-based unit in cogeneration regime-the 110 MW pulverized coal fired power plant with slag tap boiler furnace operating in cogeneration regime is considered. Additional investment compared to the option 1 is related to the steam extraction control on the steam turbine. Additional 80 MW of heat power installed is provided by steam turbine heat extraction. Option 3 presents a new power station-1 10 MW fluidized bed combustion unit. The power plant will operate in cogeneration regime, producing additional 80 MWh/h of heat. The modern design of the fluidized bed furnace with internal circulation enables burning of low valuable coals, at heat consumption of 8050 kJ/kWh (efficiency of 45%). Option 4 of the considered energy options is combined cycle gas turbine power plant. Installed power of this plant is 102 MW, and it is assumed that all input energy is used for electricity generation. The fuel is natural gas with heating value of 45500 kJ/kg. Estimated heat consumption of the plant is 6545 kJkWh. In the case of option 5-Reconstruction of hydro power plant of 6 x 25 MW power installed, upgrading of the plant by additional 30 MW of installed power is obtained, as well as increasing of efficiency by 4%. This is provided through increasing of design head by 6 m, increasing of installed flow by 10 YO,and introducing new modernized Francis turbines. Option 6 presents solar power plant (PV systems) with total installed capacity of 210 MW. Covered surface of this solar field is 266 000 m2, or 1.26 m2/kW. Option 7 presents utilization of wind energy. According to the available data of wind characteristics on selected micro locations, it is calculated that 438 single units of 600 kW power each is to be installed to provide average annual production of 525 GWh. This gives average annual operation of 2000 h. Within option 8, installing of new biomass power plant is considered. This new plant has 80 MW of installed power, and overall efficiency of 20%. Emision of C02 for this option is reduced by the amount of carbondioxide which is being absorbed by the plants used as fuel in this power station.

3.2. Sustainability Indicators The sustainability assessment is based on the definition and calculation of sustainability indicators. In this work, sustainability indicators reflecting 4 criteria of sustainability are defined and given in Table 1. Indicators from Table 1 are calculated for all options under consideration; on the base of that single criterion assessment can be performed.

366

F. Begic & N.H. Afgan Table 1. Sustainability indicators.

I Tvue of Indicator

I Name o f Indicator

I

unit

I

knlkWh kglkWh kglkWh kglkWh EUlUkWh EUR/kWh l1kWh h/kWh

Fuel indicator

Resource Indicator (RI)

Cupmm indicator

I Environment Indicator (EI)

Economic Indicator (Ecl)

I I

Social Indicator (SI)

Insulation indicator C02 indicator SO2 indicator NOx indicator Energy costs indicator Investment indicator Efficiency indicator Job Indicator Diversity indicator

I I

I

I

I

I

3.3. Results of Single Criteria Analysis The results of single criteria assessment within the analysis are presented. Comparing the considered options according to the environment indicators, e.g. see Fig. 2-indicator of sulfur dioxide emission, it can be noticed that renewable resources, hydropower plant, and clean fossil fuel technologies as Combined cycle gas turbine station and fluidised-bed combustion, are in advantage in reference to the reconstmcted conventional coal-fueled stations.

20 .

15

2

10

.

~

. .

..

..

1 rn . .~ .

5

.

.

.. . .

...

.. ._. . .

0

1

2

3

4

5

6

7

6

0pt ions

Fig. 2. Comparision of SO2 indicators within single criteria analysis

The opposite to this, the comparison of economic indicators, e.g. indicator of investment-see Fig. 3, options of coal-fired power plant is preferable in comparison to the renewable resources. These are obvious examples of unreliability of single criteria analysis, when selection of the most convenient energy system depends exclusively on selected criteria. Consequently, subjectivity of decision-making process is strongly expressed.

Sustainability Assessment,fov the Decision Making

__

8

367

I______

7 r

,

-

................

0 . -

1

2

3

4 5 ODtions

6

7

8

Fig. 3. Comparison of investment indicators within single criteria analysis.

3.4. Multicriteria Sustainability Assessment

The multicriteria sustainability assessment based on ASPID methodology (Analysis and Synthesis of Index at Information Deficiency) [ 5 , 61 of the energy power system under consideration is performed. Used methodology is based on the system of stochastic model of uncertainty, enabling to obtain the assessment from of various supporting systems by computer, as well as enabling to obtain various normalization indexes by using Nonnumeric (ordinal), Non-exact (interval) and Non-complete information (NNNinformation), obtained from sources of various reliability and probability. The multicriteria analysis is based on the determination of sustainability indicators, described in previous chapter, and specific criteria, adopted by weighting factors, and which are being agglomerated into General sustainability indices. General indices are formed through the following procedure: 1. Formation of vectors x = (x, .....)x, of all input attributes (characteristics) which are necessary for full evaluation of quality of the options under considerations; in the work, the attributes are expressed by 4 group of indicators: Resource Indicator, Environment Indicators, Social Indicators and Economic Indicators. 2. Formation of vectors of specific criteria q = ( q l ,........, q m), by which input attributes (indicators) x,, ...,xm are to be evaluated. 3. Introducing of weighting factors, by which multicriteria sustainability assessment of the options under consideration is expressed by means of additive aggregate function, or sintetized function given by relation (1): 4. Selection of vectors w = (y.....w,), wi2 0, wI + ....w m = 1 , i.e. weighting factors. In praxis, vectors w = (w,, .... w,) often can not be exactly determined due to information deficiency. In such a case, method of randomization is used, which enables the values of weighting factors for each considered case to be determined. 5. As final result of this procedure, a list of priority of options under consideration at defined criteria is obtained. The multicriteria assessment taking into account all criteria at the same time, where the different criteria are adopted by respective weighting factors, gives a realistic and reliable sustainability rating of the options under consideration for a lifetime.

368

F. Begic & N.H. Afgan

Fig. 4. Weighting factors for case EcIz > EI2 = SI = RI within multicriteria assessment

Considering the case when the sustainability assessment criteria is based on the economic indicator with domination of investment indicator is preferable in reference to the other indicators, the Fig. 4, it can be noticed that option with the highest rating of the sustainability index is option 5-Reconstruction of hydro power plant. Option 2Reconstruction of coal-based unit in cogeneration regime-is ranked on third place in this case, although this option according to the single criteria assessment is preferable option.

Fig. 5. General indices for the case EcI2 > EI2 = SI = RI within multicriteria assessment.

Additionally, in the case when economic indicator with the domination of the indicator of investment cost, have an advantage in reference to the other indicators, and when the remained indicators are not equal, see Fig. 6, the option 2 falls down on 4thplace in the list of priority, under influence of weighting factors of the other indicators, Fig. 7.

Fig. 6. Weighting factors for case E c l ~ > RI > EIz > SI within multicriteria assessment.

Fig. 7. General indices for the case EcI2 > RI > EI2 > SI within multicnteria assessment.

Sustainability Assessment for the Decision Making

369

4. Conclusions

Most of the models for decision-making for the adoption of power system option are based on single criterion assessment. Presently the sustainability assessment of power system is needed, the models are being advanced and sustainability multicriteria assessment of power system is adapted. However, lack of information doesn’t enable us to obtain a clear assessment of quality of power system option. One of prospective methods to estimate sustainability of the energy system is multicritaeria assessment of sustainability based on the analysis and synthesis of indexes under deficiency of information or ASPID methodology. This methodology is based on the stochastic model of uncertainty, and the assessment from of various supporting systems, to obtain various normalization indexes by using Non-numeric (ordinal), and Non-complete information (NNN-information) obtained from sources of various reliability and probability. Multicriteria sustainability assessment of various energy systems is based on definition of following indicators: resource indicators, environment indicators, social indicators and economic indicators, including also the weight factors. In this paper, the multicriteria assessment of selected options of the energy power system of the Public Enterprise Elektroprivreda of Bosnia and Herzegovina is performed, based on the real and actual example of increasing the energy power system by new power capacity. Through the analysis of multicriteria assessment of selected options, the decisionmakers could be enabled to form opinion related to the selection of an optimum option based on the sustainability assessment. References [I] Begic F. (2003) Method of sustainability assessment of the energy power systems of various sources in Bosnia and Herzegovina. (in Bosnian), PhD theses, University of Sarajevo, Mechanical Engineering Faculty Sarajevo, October. [2] Afgan NH. (2000) Multicriteria indicators for clean air technologies. UNESCO Chair holder, Instituto Superior Tecnico, Lisbon, Portugal. [3] Afgan NH, Carvalho MG. (2000) Sustainable Assessment Method for Energy Systems, Instituto Superior Tecnico, Lisbon, Portugal, Kluwer Publisher. [4] Begic F, Kazagic A. (2002) Sustainability Development of Energy System - Basis of Society Development Strategy (in Bosnian), 5th International Conference “Energetska i procesna postrojenja - Dubrovnik 2002”, Proceedings, Energetika Marketing, Zagreb, May. [5] Hovanov N, Fedotov Yu, Zakharov V. (1999) The making of index numbers under uncertainty. Environmental Indices: Systems Analysis Approach, EOLSS Publishers Co., Oxford, pp. 83-99. [6] Hovanov N. (1996) Analysis and Synthesis of Parameters under Information DeJiciency. St. Petersburg State University Press. (Monograph, in Russian).

2004 New and Renewable Energy Technologies for Sustainable Development: Evora, Protugal, 28 June-1 July 2004

Renewable Energy Technologies

Energy Statistics Yearbook 2004

China's Energy Outlook 2004

Association for Jewish studies 2004-28(1)

New Technologies for Energy Efficiency

New Technologies for Energy Efficiency

New Technologies for Energy Efficiency

Association for Jewish studies 2004-28(2)

State Magazine (July-August 2004)

Sound on Sound (July 2004)

Megatrends for Energy Efficiency and Renewable Energy

World Energy Outlook 2004 (World Energy Outlook)

Frommer's New Orleans 2004

2004)

Sustainable Biotechnology: Sources of Renewable Energy

2004

2004

Assessment of hydrogen energy for sustainable development

Renewable and Sustainable Polymers

Energy Balances and Electricity Profiles 2004

Fueling Development Energy Technologies for Developing Countries

Renewable Energy

The Solar Economy: Renewable Energy for a Sustainable Global Future

Government Promotion of Renewable Energy Technologies

Renewable Energy

Renewable Energy

The Solar Economy: Renewable Energy for a Sustainable Global Future

Quicken 2004 for Dummies

EXERGY: Energy, Environment and Sustainable Development

Agent Communication: International Workshop on Agent Communication, AC 2004, New York, NY, July 19, 2004

2004 New and Renewable Energy Technologies for Sustainable Development: Evora, Protugal, 28 June-1 July 2004

Renewable Energy Technologies

Energy Statistics Yearbook 2004

China's Energy Outlook 2004

Association for Jewish studies 2004-28(1)

New Technologies for Energy Efficiency

New Technologies for Energy Efficiency

New Technologies for Energy Efficiency

Association for Jewish studies 2004-28(2)

State Magazine (July-August 2004)

Sound on Sound (July 2004)

Megatrends for Energy Efficiency and Renewable Energy

World Energy Outlook 2004 (World Energy Outlook)

Frommer's New Orleans 2004

2004)

Sustainable Biotechnology: Sources of Renewable Energy

2004

2004

Assessment of hydrogen energy for sustainable development

Renewable and Sustainable Polymers

Energy Balances and Electricity Profiles 2004

Fueling Development Energy Technologies for Developing Countries

Renewable Energy

The Solar Economy: Renewable Energy for a Sustainable Global Future

Government Promotion of Renewable Energy Technologies

Renewable Energy

Renewable Energy

The Solar Economy: Renewable Energy for a Sustainable Global Future

Quicken 2004 for Dummies

EXERGY: Energy, Environment and Sustainable Development

Agent Communication: International Workshop on Agent Communication, AC 2004, New York, NY, July 19, 2004

Recommend Documents