Research On Customer Behavior Analysis And Interaction Strategy Considering Customer Response

With the rapid development of national economy and the increasing demand of electric power,Traditional power grid has developed to the smart grid of informatization and interaction,as a secondary energy sources,Electricity began to show more attributes of commodity,this makes marketing management and customer service more and more important in the electric power company.The analysis of user's electricity behavior has become an important work to grasp the direction of marketing management and improve the quality of customer service.Demand response is a variety of short-term behavior of electricity users to adjust the power consumption according to the change of electricity price and incentive policy.The research on the users' demand response behavior,Interactive operation mode between user and power grid and incentive mechanism in smart grid is an important requirement for the safe and economic operation of power grid.It has the feasibility and the necessity,It also provides decision-making basis for energy saving transformation and optimization of power grid.To solve these problems,this paper made research on the user's behavior and the interactive strategy.First of all,the basic theory of the analysis method of user's behavior is analyzed.The advantages and disadvantages of different clustering algorithms,the choice of distance measure and the effect of index evaluation function are studied.Then a model based on the user's response is established,this paper use the elastic price coefficient to simulate the user's behaviour,the response characteristics are extracted,and the user's response behavior and interaction potential are analyzed by the improved clustering algorithm.Then,the load optimization goal is set up to solve the model and the user's optimal power distribution strategy is solved,through the multi attribute decision making method,the user interactive response behavior is evaluated,and the simulation experiment is carried out.The experimental results show that the user load curve is optimized,and the effectiveness of the algorithm is verified.