Research on the Method of Electric User Payment Based on SOM Neural Network Clustering Analysis

Abstract

With the rapid development of economy, the demand of customers' electricity service presents a diversified trend. In order to meet the diversified demands of power customers for services, power supply enterprises have continuously introduced new payment methods and actively guided power customers to pay through new payment channels. In this paper, a comprehensive index system is built based on the payment behavior data of electric power users, and strive to fit the diversified characteristics of users. Firstly, SOM neural network clustering algorithm is used to subdivide the payment behavior of electricity customers; Then, in order to optimize the clustering effect, the optimal number of clustering clusters is obtained through cluster optimization analysis, and then the SOM neural network is used to cluster the power customers, and the attributes of various groups and the overall attributes are compared and analyzed to obtain the customer segmentation results; Finally, the sales scheme for response type users is specified according to the clustering results, so as to achieve the evaluation of the target service quality level and ultimately improve the service level.