Identification of Electricity Consumers'Electricity Theft Behavior Based on K-MEANS Clustering of Electricity Metering Data
Due to the poor application of the current method in the identification of power theft behavior of power users,the AUC value is relatively low and cannot achieve the expected identification effect.Aiming at the deficiencies and defects of the current methods,this paper proposes the identification of electricity theft behavior of power users based on K-MEANS clustering of electric energy metering data.The use of ETL technology for energy metering data load data extraction,using K-MEANS clustering algorithm for data clustering analysis,diagnosis and identification of user power theft,so as to achieve the identification of power theft based on energy metering data K-MEANS clustering of power users.Experimentally proved that the AUC value of this method is above 0.95,which can realize the accurate identification of power theft behavior of power users.
electricity metering dataK-MEANS clusteringelectricity theft behaviorETL technologyload data
万方数据
