Research On Pricing Strategy Of Residential Peak And Valley Time-sharing Price Based On Data Mining

The contradiction between power supply and demand is becoming more and more serious due to the increase of residential electricity consumption year by year and the serious delay of power system construction.In view of the above problems,this paper puts forward the pricing strategy research of residential peak-valley time-sharing electricity price based on data mining.Through the leverage of electricity price,it is of great significance to guide the resident user to change the power consumption mode spontaneously,to reduce the maximum peak-valley load difference,to reduce the power waste and to build a resource-saving society.By means of data mining,this paper predicts the average daily load curve of residents in the future by using multivariate linear regression and support vector regression combined with variable weight model according to the related historical data of residents.Then the fuzzy clustering idea is used to determine the optimal threshold,and then the membership function is adjusted,and finally the peak-valley time division scheme is given.Then,using Markov chain algorithm,the load transfer probability of residents in different peak-valley electricity prices is solved.On the premise of ensuring that the consumers' electricity expenses do not increase,taking the maximization of the comprehensive income of the power grid and residents as the objective function,a scientific and reasonable peak-valley electricity price is given.At last,taking the second step residential users in a certain area of Guiyang City as an example,the simulation results prove the validity of this method.