Application In Bidding For Power Market By Bionic Algorithm With State-space Model

The power industry has been from monopoly to competition world-wide under the promoting of market economy, and many countries established suitable power market according to their national conditions. Power market has been extended and established in our country gradually with the influence of the electricity industry reform, whose goal is to establish generation-side electricity market. Introducing competition in generation side is the characteristic of power market, and profits of generation companies depend on bidding strategies employed. Hence, the power market bidding is the key to power market reform, which not only relates to various members'benefits, but also affects power purchase cost of users, and affects the disposition of the social resources further, it plays an important role in power industry.This thesis introduces some basic situation of power market, and emphatically introduces various bidding method presented by scholars at home and abroad; secondly, the bidding models and principle are described in detail; and then, bionic algorithm is introduced, including basic theory, characteristic and application in several common algorithms derived from bionic algorithm. The thesis also proposes the bionic algorithm based on state-space model of discrete-time system, and describes content of this algorithm. At last, this algorithm is used to solve the bidding in power market.At present, there are some main bidding methods commonly used in power market, such as the dynamic programming method, the linear programming method, the genetic algorithm and so on, but any method couldn't solve the problem effectively and achieve the optimum solution, because they can be used to solve different bidding curves respectively and deal with various sets of constraints. According to characteristics of power market, this thesis use the new bionic algorithm based on state-space model to solve the bidding in power market, giving the encoding style, fitness function, state transition matrix and selection pool, and constructs four bionic calculation models. State transition matrix is the core of the algorithm, and is the decisive factor of the convergence and convergent rate in this algorithm. This algorithm not only retains characteristics of traditional bionic algorithm, but also increases the calculation speed, meanwhile, improves the global convergent ability.The simulation results show that the bidding algorithm can achieves the desired effect effectively. The thesis is a useful try to research bidding method. I hope it would be helpful to researches in further theory and practice.