Decision support for the strategic behaviour of electricity market players
Ficheiros
Data
2016
Título da revista
ISSN da revista
Título do Volume
Editora
Resumo
O contributo desta tese é dado nas áreas complementares da inteligência artificial e dos sistemas de energia, através da proposta de soluções inovadoras para o apoio à decisão, de
forma inteligente e adaptativa, dos agentes participantes nos mercados de energia elétrica,
considerando a sua participação em múltiplas oportunidades de mercado.
O estado atual dos mercados de energia elétrica deve-se em grande parte ao uso crescente
de fontes de energia renovável. Este aumento tem sido estimulado por novas políticas
energéticas que resultam das preocupações geradas pela escassez dos combustíveis fósseis e
pelo seu impacto no ambiente. Estes incentivos têm dado origem a grandes investimentos no
sector energético. No entanto, a penetração em larga escala de fontes de energia renovável de
natureza intermitente, como o vento e o sol, tem a si associado um grande número de limitações
que restringem, não só a fiabilidade da produção, como também o seu uso
O crescente aumento da integração de recursos energéticos provenientes de fontes
renováveis conduziu a uma inevitável reestruturação do sector energético. Este processo de
reestruturação resultou em mudanças significativas na operação dos mercados de energia
elétrica por todo o mundo. Estes mercados têm como objetivo assegurar uma competição mais
acentuada e justa, desta forma potenciando o decréscimo dos preços da energia elétrica. No
entanto, a reestruturação dos mercados de energia elétrica criou também vários desafios, pois
exigiu a transformação dos modelos conceptuais que dominavam o sector. A reestruturação
trouxe, portanto, um mercado mais competitivo, mas também mais complexo e com maiores
desafios para os seus participantes. O aumento da complexidade e da imprevisibilidade destes
mercados amplifica a dificuldade na tomada de decisão, que é ainda agravada pela crescente
introdução de novos mecanismos de mercado. Por estes motivos as entidades participantes têm
constantemente de repensar o seu comportamento e estratégias de mercado.
A utilização de ferramentas de apoio à decisão torna-se fundamental para que estas
entidades consigam lidar com os novos desafios. A necessidade de perceber o funcionamento
dos mercados e de que forma as interações entre as entidades envolvidas afetam os resultados
de mercado motivou o desenvolvimento de várias ferramentas de simulação. Os sistemas multi-agente têm sido a solução mais adotada, uma vez que são especialmente adequados à análise de
sistemas dinâmicos com interações complexas entre os seus constituintes. Apesar das soluções
existentes permitirem o estudo de diferentes mecanismos de mercado e a análise das relações
entre os intervenientes, estas ferramentas não estão, no entanto, preparadas para fornecer um
apoio à decisão adequado a estes agentes num contexto de negociação em mercado.
O trabalho de investigação deste doutoramento é motivado pela lacuna identificada, e
surge com o objetivo de propor soluções que permitam aos agentes participantes nos mercados
de energia elétrica obterem os melhores resultados possíveis em cada contexto de mercado. A
agregação das várias contribuições deste trabalho resulta num sistema multi-agente de apoio à
decisão: AiD-EM (Adaptive Decision Support for Electricity Markets Negotiations – apoio à
decisão adaptativo para negociações em mercados de energia elétrica). Este sistema inclui uma
metodologia que permite aos agentes de mercado identificar as oportunidades de mercado em
devem apostar em cada momento. O apoio à decisão para a negociação efetiva em cada mercado
é dado por sistemas de apoio à decisão específicos, dirigidos aos diferentes tipos de negociação.
O apoio à participação em mercados com funcionamento por leilão é dado pelo sistema ALBidS
(Adaptive Learning strategic Bidding System – sistema de aprendizagem adaptativa para
licitações estratégicas). Este sistema inclui diversas estratégias de participação em mercado, e
tem a capacidade de aprender quais as abordagens que dão maiores garantias de sucesso em
cada contexto. Para as negociações de contratos bilaterais foi desenvolvido o sistema DECON
(Decision Support for Energy Contracts Negotiation – apoio à decisão para a negociação de
contratos de energia), que inclui metodologias para analisar os perfis de negociação dos
competidores, permitindo assim a adaptação das estratégias de negociação. Por fim, é proposto
um método de análise contextual, que analisa e identifica diferentes contextos de negociação,
permitindo assim uma adaptação contextualizada dos diversos processos de aprendizagem.
As metodologias desenvolvidas foram testadas e validadas através da simulação de
cenários realistas, utilizando dados reais provenientes de diferentes operadores de mercado. Os
resultados promissores que foram obtidos em condições realistas de negociação sustêm a tese
de que uma análise e aprendizagem adequadas utilizando dados históricos pode, de facto,
melhorar os resultados da participação dos agentes em mercados de energia elétrica.
This thesis provides its contribution to the complementary fields of artificial intelligence and power systems by proposing innovative solutions for the intelligent and adaptive decision support of electricity market players, considering their participation in multiple alternative and complementary market opportunities. The current state of worldwide electricity markets is strongly affected by the increasing use of renewable energy sources during the last years. This increase has been stimulated by new energy policies that result from the growing concerns regarding the scarcity of fossil fuels and their impact in the environment. As consequence of the policies and incentives that have been put in place, huge investments have been made in the power and energy sector. However, the large scale integration of fluctuating renewable sources in the power system, such as wind and sun, poses several constraints that limit not only the production reliability but also its use. The large scale integration of renewable based energy resources has led to an unavoidable restructuring of the power and energy sector, which was forced to adapt to the new paradigm. This restructuring process resulted in a deep change in the operation of competitive electricity markets all around the world. These markets aim at ensuring increased and fair competition giving electricity buyers more options and pushing power players to increase their efficiency, thus enabling electricity prices decrease. The electricity markets’ restructuring process brought out, however, several challenges itself, demanding the transformation of the conceptual models that have previously dominated this sector. The restructuring made the market more competitive, but also more complex, placing new challenges to the participants. The growing complexity and unpredictability of the markets’ evolution consequently increases the difficulty of decision making, which is exacerbated by the increasing number of new market types that are continuously being implemented to deal with the new challenges that keep on emerging. Therefore, the intervenient entities are relentlessly forced to rethink their behaviour and market strategies in order to cope with such a constantly changing environment. So that these entities can deal with the new challenges, the use of decision support tools becomes crucial. The need for understanding the market mechanisms and how the involved players’ interaction affects the outcomes of markets has contributed to the emergence of a large number of simulation tools. Multi-agent based software is the most widely adopted solution as this paradigm is particularly suitable to analyse dynamic and adaptive systems with complex interactions among its elements, such as electricity markets. Current software tools allow studying different electricity market mechanisms and analysing the relationships between market entities; however, they are not prepared to provide suitable decision support to the negotiation process of electricity market players. This gap motivates the development of this PhD research work, which arises with the purpose of providing solutions that enable electricity market players to take the best possible outcomes out of each market context. The aggregation of the many contributions of this work ultimately results in an enhanced multi-agent based decision support system: Adaptive Decision Support for Electricity Markets Negotiations (AiD-EM). AiD-EM includes a Portfolio Optimization methodology, which decides in which market opportunities should market players negotiate at each moment. The actual negotiation process in each market is supported by specific decision support systems, directed to different types of negotiation. The participation in auction based markets is supported by the Adaptive Learning strategic Bidding System (ALBidS). This decision support system includes a large number of distinct market participation strategies, and learns which should be used in each context in order to provide the best expected response. Negotiations by means of bilateral contracts are assisted by the Decision Support for Energy Contracts Negotiation (DECON) system, which includes methodologies to analyse competitor players’ negotiation profiles enabling the adaptation of the adopted negotiation strategies and tactics. All methodologies are supported by a context analysis methodology, which allows analysing and identifying different contexts of negotiation, thus enabling a contextual adaptation of the diverse learning processes. The developed decision support methodologies have been tested and validated under realistic simulation scenarios using real data from several electricity market operators. The promising results achieved under realistic negotiation conditions support the thesis that the appropriate analysis and learning from historic data can, in fact, improve players’ outcomes from their participation in electricity markets.
This thesis provides its contribution to the complementary fields of artificial intelligence and power systems by proposing innovative solutions for the intelligent and adaptive decision support of electricity market players, considering their participation in multiple alternative and complementary market opportunities. The current state of worldwide electricity markets is strongly affected by the increasing use of renewable energy sources during the last years. This increase has been stimulated by new energy policies that result from the growing concerns regarding the scarcity of fossil fuels and their impact in the environment. As consequence of the policies and incentives that have been put in place, huge investments have been made in the power and energy sector. However, the large scale integration of fluctuating renewable sources in the power system, such as wind and sun, poses several constraints that limit not only the production reliability but also its use. The large scale integration of renewable based energy resources has led to an unavoidable restructuring of the power and energy sector, which was forced to adapt to the new paradigm. This restructuring process resulted in a deep change in the operation of competitive electricity markets all around the world. These markets aim at ensuring increased and fair competition giving electricity buyers more options and pushing power players to increase their efficiency, thus enabling electricity prices decrease. The electricity markets’ restructuring process brought out, however, several challenges itself, demanding the transformation of the conceptual models that have previously dominated this sector. The restructuring made the market more competitive, but also more complex, placing new challenges to the participants. The growing complexity and unpredictability of the markets’ evolution consequently increases the difficulty of decision making, which is exacerbated by the increasing number of new market types that are continuously being implemented to deal with the new challenges that keep on emerging. Therefore, the intervenient entities are relentlessly forced to rethink their behaviour and market strategies in order to cope with such a constantly changing environment. So that these entities can deal with the new challenges, the use of decision support tools becomes crucial. The need for understanding the market mechanisms and how the involved players’ interaction affects the outcomes of markets has contributed to the emergence of a large number of simulation tools. Multi-agent based software is the most widely adopted solution as this paradigm is particularly suitable to analyse dynamic and adaptive systems with complex interactions among its elements, such as electricity markets. Current software tools allow studying different electricity market mechanisms and analysing the relationships between market entities; however, they are not prepared to provide suitable decision support to the negotiation process of electricity market players. This gap motivates the development of this PhD research work, which arises with the purpose of providing solutions that enable electricity market players to take the best possible outcomes out of each market context. The aggregation of the many contributions of this work ultimately results in an enhanced multi-agent based decision support system: Adaptive Decision Support for Electricity Markets Negotiations (AiD-EM). AiD-EM includes a Portfolio Optimization methodology, which decides in which market opportunities should market players negotiate at each moment. The actual negotiation process in each market is supported by specific decision support systems, directed to different types of negotiation. The participation in auction based markets is supported by the Adaptive Learning strategic Bidding System (ALBidS). This decision support system includes a large number of distinct market participation strategies, and learns which should be used in each context in order to provide the best expected response. Negotiations by means of bilateral contracts are assisted by the Decision Support for Energy Contracts Negotiation (DECON) system, which includes methodologies to analyse competitor players’ negotiation profiles enabling the adaptation of the adopted negotiation strategies and tactics. All methodologies are supported by a context analysis methodology, which allows analysing and identifying different contexts of negotiation, thus enabling a contextual adaptation of the diverse learning processes. The developed decision support methodologies have been tested and validated under realistic simulation scenarios using real data from several electricity market operators. The promising results achieved under realistic negotiation conditions support the thesis that the appropriate analysis and learning from historic data can, in fact, improve players’ outcomes from their participation in electricity markets.
Descrição
Tese de Doutoramento em Informática
Palavras-chave
Inteligência artificial , Energia elétrica , Mercados , Sistemas de apoio à decisão , Aprendizagem adaptativa , Negociação automática