Research On Time-of-use Price Optimization Based On User’s Demand Response

With the rapid development of China’s economy,the energy problem is becoming more and more prominent.How to improve energy efficiency has become an urgent problem to be solved.As the most direct and effective economic means at present,time-of-use(TOU)price,by setting different prices in different periods,guides users to transfer part of the electricity in peak periods to valley periods,which is conducive to improve the load rate of power units,optimize the power consumption mode of power users,alleviate the problem of energy shortage,and realize the scientific and coordinated development of power,economy and natural environment.According to the current practical problems of electric energy in China,this paper puts forward a scientific and effective TOU price time division and price formulation method,and makes an in-depth theoretical and Experimental Research on the related problems.Firstly,this paper studies the time division under TOU price,and proposes a peak valley time division model based on long-term and short-term memory neural network and EK-means clustering algorithm.Through the feature extraction and clustering of users’ load data for a long time,the time division results with long-term applicability are obtained.Secondly,based on the principle of consumer psychology,this paper constructs the user response model,studies the solution steps of the user response model,and puts forward the correction method of the mismatch between the user load data and the time period,so that the obtained user response model is closer to the actual user response law.In addition,this paper also gives the quantitative method of user satisfaction from two aspects of user power consumption mode and user power expenditure,which lays a foundation for the establishment of subsequent objective function.Then,according to the actual needs of power grid companies and residential users,this paper constructs an objective function that takes into account the effect of peak shaving and valley filling and user satisfaction,and sets the corresponding constraints;The particle swarm optimization algorithm for electricity price optimization is studied,and the improved methods of inertia weight and learning factor are proposed to improve the convergence speed and accuracy of particle swarm optimization algorithm.The performance of the improved method is verified by test function.Finally,the mathematical models mentioned in the previous chapters are optimized and simulated through practical examples.The results show that: in terms of time division,the average value and variance of the contour coefficient of the time division results obtained by the proposed method are0.4921 and 0.0103 respectively,which are better than 0.4826 and 0.0568 of the classical time division method,and can better reflect the actual power consumption characteristics and laws of users;In terms of user response model,the Pearson correlation coefficient between the fitting curve and the actual load curve after the correction according to the method in this paper is 0.9928,which is greater than 0.9866 of the traditional uncorrected method,which can better describe the user’s response law to the price;In terms of tou pricing strategy,this paper sets different weights according to the power consumption in different regions,and uses the improved particle swarm optimization algorithm to obtain different optimized price results.The power grid company can flexibly configure according to the power consumption in different regions,so as to obtain the optimal effect of taking into account peak shaving and valley filling and user satisfaction.This paper makes theoretical and Experimental Research on the time division of TOU price and the formulation method of TOU price,and verifies and discusses the relevant problems through specific examples,so as to lay a certain foundation for the theoretical development of TOU price.