Research On Bidding Strategy Of Power Producers In Competitive Power Market

As a special commodity,with the reform of the market,it is particularly important to open up the regulation of the electricity market and provide a fair and free competitive environment for power generation companies involved in power trading.The experience of foreign power market reform shows that the deregulated power market can ensure that each market member can give full play to its own advantages and improve the efficiency of power production while reducing the price of electricity.The reform of power marketization is a process of repeated research and continuous improvement in practice.As one of the main members participating in the electricity market trading,power producers usually adopt certain strategies to compete in order to participate in market transactions.Their behavior strategies are related to the operation of the entire electricity market.In this paper,for the competitive power market,in order to ensure the profit of power producers,the following research is conducted on the bidding strategies of power producers participating in the competitive power market:(1)Aiming at the prediction problem of clearing electricity price in electricity market,an adaptive Kalman filtering algorithm is proposed to predict the market clearing price.According to the historical data of electricity market transactions,an adaptive Kalman filtering model for clearing electricity price forecasting in electricity market is established,and the market parameters are used to correct the model parameters.The predicted electricity market will be cleared and submitted to the power producer for reference to the quotation strategy.The power producer will set a bid price at a bidding moment slightly lower than the predicted market clearing price to ensure that the generating capacity of the unit can fully participate in the market trading.To ensure that their profits are maximized.(2)For the study of the game behavior of the power producers in the competitive power market,the adaptive Kalman filter is combined with the game theory to construct the dynamic game model of the generator.All generators participating in the competition are equivalent to one competitor,regardless of individual competitor behavior.The gamer behavior of the entire electricity market is between the generator and the equivalent competitor.Kalman filtering is used to predict the behavior of competitors.As a reference for the game of power producers,a dynamic game model of power producers that fully considers the behavior of competitors is constructed.The quotation strategy is adjusted according to the estimated competitor response,and the Nash equilibrium solution of the game model is solved.The Nash equilibrium strategy combination of the game model is the best strategy for generators to participate in the game.(3)For the adaptive learning of the bidding strategy of the power producer,the Kalman filter is combined with the improved Q learning algorithm of the dynamic gamer of the power producer to construct an adaptive learning model of the generator's bidding strategy.Considering the power producer as an agent,the power producer can judge the state of the power market based on historical transaction data and adopt a quotation strategy that is beneficial to its own interests.Kalman filtering is used to estimate the behavior of competitors,predict the behavior status of competitors and the strategies that competitors may adopt in the current state,further improve the quotation mechanism of power producers,and improve the quotation efficiency of power producers.