For the degree of Master of Science in Electrical Power Engineering

UMIST 2002 PDF Version 2 Acknowledgments I would like to thank my supervisor, Dr Daniel Kirschen, for his valuable guidance and constant support. Abstract Conventional Economic Modelling (CEM) can be successfully employed to describe different financial equilibria. Its ability, though, to provide insights into the strategic behaviour of generating companies competing in the new, restructured electricity markets is limited. This lead to the development of techniques based on computational modelling and simulation. In this project, a computational model has been developed in order to investigate the bidding behaviour of generating companies in a short-term bilateral market. The generating companies are represented by software agents. Each agent is modelled with specific strategic and operational objectives and bounded reasoning abilities based on principles of reinforcement learning. A number of simulation studies have been carried out based on the software platform developed in this project. These studies have shown that market prices are likely to increase under Hourly Bidding with Pay Bid settlement. They have also shown that when the market share objective of the generating companies takes over their short-term profit maximisation objective, market prices are likely to decrease. Tacit collusion among generators to increase the market prices has also been observed under peak demand. In a market with a small number of generators, agents with large and flexible portfolios were able to exercise market power. This resulted in extremely high market prices. Conversely, in a market with a large number of symmetric generators, prices have experienced a significant reduction.