Study On Power Generation Enterprises' Bidding Strategies Based On Price Forecasting

The bidding strategy based on next-day market clearing price forecasting of the power generation enterprise under the electricity market environment are systematically studied in this paper. Firstly, by analyzing the merits and shortcomings of BP neural network and discussing the problem that the criterion-Mean Square Error cost function can't take into account the higher-order statistical behavior of the systems,a neural network price forecasting model based on Minimum Entropy Error to improve the price prediction accuracy is proposed. Secondly,according to the stochastic characteristics of the electricity price time series and considering the impact of load factors to next-day electricity price,a new model of electricity price forecasting based on the Hidden Markov Model is presented.On the basis of the model, a conception of Value at Risk is introduced and employed to obtain a bidding strategy that satisfy risk requirement.Finally,the historical data from a real electricity market is used in the case study,then the result shows the viability and effectiveness of the proposed scheme.