Power Demand Side Control Technology For Electricity Customers In Urban And Rural Areas

This paper proposes and designs a demand-side responsiveness evaluation model that takes into account the customer’s intention label by studying the existing technologies and behavioral research methods of residential electricity customer demand-side management.Firstly,the main Tensor Flow is used to build a deep learning platform.Then,the density-based hybrid attribute clustering algorithm is used to cluster these users.Finally,according to the above clustering results,the demand-side management strategy considering user behavior is formulated,which is the background of power system reform.Firstly,a big data analysis platform was built based on Tensor Flow.Then,the deep neuron network classification algorithm is used to classify different responsive residential users.Finally,according to the above clustering results,the demand-side management strategy considering user behavior is formulated,which is the background of power system reform.The demand side management and control of residential electricity customers provide strong support.The main research results are as follows:1.Introduce the demand side response potential evaluation method of customer intention category label.Residents’ electricity consumption customers are highly sensitive to price,and their electricity consumption habits have personalized characteristics.Therefore,in order to more accurately evaluate the responsiveness of residents,price sensitive and power failure sensitive labels are introduced by analyzing customer demands to more accurately evaluate the possibility of customers participating in demand side response.2.Assisted demandside management and control decisions with the analysis results of massive behavioral data of residential electricity customers.Through big data analysis of massive behavioral data of customers such as electricity consumption,payment and demand,etc.,the behavior rules and characteristics of customers are explored,and personalized demand-side management and control strategies are formulated to effectively improve the response rate of residents’ demand-side.3.Implement customer behavior analysis algorithm optimization and application.Density-based Clustering Algorithm for Mixed Attributes and Multi-dimension Data(MMDBC)is adopted to cluster target customers.Considering the hybrid attribute and density clustering,this algorithm is more suitable for multi-type data clustering of power users.