Rerearch On Bidding Behavior Of Power Generation Enterprises Based On Evolutionary Game

Currently,China's electric power system has been constantly reforming.As one of the main players in the market,power generation enterprises,which are engaged in gaining their advantages in the fierce competition in the power market,have been facing with the problem:how to choose the right and reasonable bidding strategy in the market transaction,in order to obtain the maximum profit On the basis of the characteristics and operation rules of China's electricity market,this paper discusses the equilibrium stability scenario analysis of multi-group bidding strategy based on evolutionary game on the power generation side under the background of open electricity market.Firstly,this paper forecasts the pre-market clearing price.Future electricity price plays an important role in the selection of bidding strategies for power generation enterprises.Accurate price prediction can help them adopt more reasonable and scientific bidding strategies.Because of the non-linear and fluctuating characteristics of the electricity price,the neural network algorithm can to the maximum extent,approximate the nonlinear function.Therefore,this paper proposes BAS-BPNN(Longicorn Search algorithm to optimize BP neural network algorithm)electricity price prediction model.Before the model is applied,empirical mode decomposition(EMD)and wavelet denoising are used to process the original data sequence to make the sequence smooth and remove some interference factors.Moreover,the model is verified by the real data of typical power market,and the results,show that the prediction model proposed in this paper can make comparatively accurate price prediction,except for some special cases that cause great fluctuation of the electricity price.Then,based on evolutionary game features and decision-making model of power generation enterprises,three power generation enterprises were chosen and put in the same typical bidding scenario,to build three groups 2x2x2 evolutionary game replication dynamic system.The strategy combination and benefits of combination between them were simulated,the possible each balance system was clearly analyzed,and the requirements to meet the condition of asymptotically sTab.were solved.Finally,the model is verified and analyzed with the background of East China power market.The results show that the bidding model based on evolutionary game can well reflect the dynamic bidding process of the starting electric enterprises,and each participant can forecast the strategy trend of competitors according to the model,so as to adjust their bidding strategies and get more profits.When the supply and demand in the electricity market is tight,the power generation enterprises will choose a high price strategy.With the continuous reduction of market load demand,power generation enterprises will make strategic quotation according to the initial state of the market.Moreover,market load demand,market clearing price,market share of enterprises and their own production costs have a great impact on the bidding income of power generation enterprises.Enterprises with large market share can almost dominate the trend of the final strategic combination of the market,and the generator with low production cost has a large bidding space in the market.These conclusions reflect the problem of centralized bidding at the generation side and shed a light on formulating new market rules for the government.