Optimal Dispatch In Generation Company Under The Multi-market Mode And Risk Assessment

As the deregulation of power industry, the GENCOS have been one of the independent market entities. Hydroelectric Company has to face the market price and runoff uncertainty in the deregulated system. How can the supplier balance the expected objective benefit and expected risk is an important issue. The paper discusses the hydroelectric Company's optimal dispatch under the multi-market mode. A short-term and long-term optimal dispatch model is built considering the risk constraints. The chaotic Particle Swarm optimization (PSO) algorithm is applied to solve the problem. The paper uses er'tan hydro station's data to prove the feasibility of the model.This paper comprises six chapters. In chapter I, the significance of the thesis, current research works of domestic and foreign, and the risk of hydro plant are introduced. In chapter II, Generation Company's bidding model and the multi-market modes are discussed. In chapter III, An improved PSO algorithm called chaotic PSO is developed to solve the model. It also compares two different kinds of chaotic PSO. In chapter IV, the short-term optimal dispatch model is built in the day-head market and ancillary market. The chaotic PSO algorithm is presented to solve the problem. In chapter V, the long-term optimal dispatch model is built in the contract market; the generation rights is also discussed in the chapter. The main works of the paper are summarized in the last chapter.The innovations of the paper are as follows.1. Through the Monte Carlo model simulating the several price and runoff scenarios, the paper builds the optimal dispatch model under the multi-market mode. The strength of the model is to consider the punishment factor to expected risk.2. In the deregulation market, the paper discusses the optimal dispatch in contract market, day-head market, and ancillary market.3. In solving the optimal problem such as dynamic economic dispatch, the new particles may not satisfy the constraints. A new repair strategy is proposed in the thesis, which can improve the efficiency of the algorithm.The paper is funded by the National Natural Science fund (No. 50539140).