Transmission planning in liberalised electricity markets in the context of market power

Transmission planning is complex, involving consideration of the impact of a transmission augmentation under a large number of future demand and supply scenarios. In principle, the transmission planning problem is well understood in the context of a vertically-integrated electricity industry. In this context, a transmission augmentation has the following primary benefits: It allows for more efficient dispatch (allowing for lower cost remote generation to be used in place of higher cost local generation); it allows inefficient investment in generation to be deferred; and it reduces the need for operating reserves by allowing those reserves to be shared over a wider area. In principle, if the liberalised electricity market is sufficiently competitive, the same tools and techniques that have been developed for transmission planning in the context of an integrated electricity industry can be applied. However, two new issues arise: (a) The first is coordination between generation and transmission investment. How should transmission and generation investment be effectively coordinated? (b) The second issue is the problem of generator market power. Many commentators point out that electricity markets are prone to the exercise of market power. The additional benefits of reducing market power have been referred to as the 'competition benefit.' Although it is widely acknowledged that transmission investment may affect generator market power, there is as yet no widely accepted methodology for computing the competition benefits of a transmission augmentation and, in practice, competition benefits are only estimated on an ad hoc basis, if at all. This thesis sets out a methodology for modelling market power in the context of transmission planning. This methodology is based around a multi-level optimisation problem. The lowest level of this optimisation problem models the dispatch process in a liberalised electricity market, allowing for generator market power. The upper level of this optimisation problem models the behaviour of the transmission network service provider. In the next step, a numerical solution approach, termed the Hybrid Bi-Level Genetic Algorithm/Island Parallel Genetic Algorithm, HB GA/IPGA, was developed to find a good solution of the proposed structures. To further improve the performance of the Hybrid Bi-Level GA/IPGA, high performance computing techniques are employed. The main contributions of this research work are as follows; (1) A systematic modelling of generator market power in a liberalised electricity market through the concepts of simultaneous-move game and worst Nash equilibrium (2) Modelling of the interaction of a transmission network service provider and rival generating companies using a simultaneous-move game nested within a sequential-move game and tackling the multiple Nash equilibria problem through the concept of 'Stackelberg-Worst Nash equilibrium' (3) A game-theoretic framework for modelling the coordination of generation investment and transmission investment (4) A decomposition methodology for decomposing the total benefits of the transmission augmentation policies into the 'Efficiency Benefit', the 'Competition Benefit', and the 'Saving in generation investment cost' (5) The use of high performance computing technologies to improve the performance of the algorithm for solving the proposed constrained-optimisation problem ? in particular, using the 'Threads' model and 'Message Passing' model of parallel programming