Research on automatic user identification system of leaked electricity based on Data Mining Technology

Abstract

Leakage of electricity will cause damage to power facilities and lead to large-scale power outage. This not only caused huge economic losses to the country, but also seriously damaged the order of the power supply market, affected the normal life of the majority of users, and even threatened the normal operation of the State Grid. This paper expounds the important research significance of applying data mining to theft and leakage, and introduces the specific functions of each layer. The automatic recognition model is constructed based on LM Neural Network and cart decision tree respectively. The mining analysis is carried out on the same data set, and the accuracy of both can reach 94%. However, after using ROC curve evaluation, the offline area of LM Neural Network is larger, indicating that LM Neural Network has higher accuracy in user classification.