Market Clearing Price Forecasting Based On Regional Power Market

Electricity price is the core content in electricity market environment. All parties taking participates in power market are urgent to need high-accuracy price forecasting methods. For generating businessmen, if they can accurately predict next-day market clearing price, it is helpful for them to formulate an optimal bidding strategy to maximize their profits. From the point of view of buyer of electricity energy, market clearing price constitutes the unit cost of electricity-purchasing, market price forecasting make it possible for buyer to control their own dynamic cost; From the point of view of market regulation, market clearing price forecasting is not only helpful for healthy, stable and orderly competition and development, but also provides scientific basis for policy-making. Therefore, market clearing price forecasting for all participants is of great significance. It has become an important part of Electricity market, this paper, on ground of actual electricity market, researches on the regional market clearing price to provide a basis for power plant bidding.This paper mainly researches on market clearing price forecasting based on regional electricity market. It proposes the following three models:⑴Improvement of hierarchical genetic algorithm and BP neural network model. Improving the inherent weaknesses of BP neural network model, this model can improve the accuracy of the prediction.⑵chaotic hierarchical genetic algorithm and BP neural network combination forecasting model. Examples show that this method, comparing with the nonlinear combination forecasting model based on neural network BP, is able to improve the combination forecasting accuracy.⑶the model which combines fuzzy neural network and immune clustering algorithm is established. By testing, the method is proved effective. This paper describes in detail the various modeling Principle and the improvement of existing models, and then uses the relevant data to test the models, have achieved a better result. It demonstrates that these models are effective and practical. In Chapter VI, this paper suggests several measures to improve prediction accuracy. Finally, pointed out the areas that require further study.