•  
  •  
 

Keywords

stealing electricity energy, anomaly detection, the abnormal event of electric energy mete, outlier score, unsupervised learning, improved FCM clustering

Abstract

Power stealing detection can find power stealing users efficiently and then reduce the operation cost of the power system. In this paper, a detection model of power stealing behavior based on improved fuzzy cmeans clustering is proposed, which is suitable for the situation without many known power stealing user samples. The model includes factor analysis, local outlier calculation based on improved fuzzy cmeans clustering, model evaluation and parameter adjustment with ROC curve, and the selection of the best detection threshold. Firstly, through factor analysis, the dimension specification of the user's power consumption characteristics (including power load data and abnormal events of electric energy meter) is carried out to improve the efficiency of model detection. Then, the FCM clustering algorithm integrated with a genetic simulated annealing algorithm is applied to detect the user's power consumption characteristics. Finally, compared with the existing mature algorithm, the result indicates that the model has a high detection accuracy for electricity theft. The detection model can output the power consumption behavior outlier and all the tested users' power stealing probability order. The output of the detection model can detect the power stealing users with high precision. The results can be employed into the field inspection, and the efficiency of antistealing work will be improved.

DOI

10.19781/j.issn.1673-9140.2021.06.020

First Page

164

Last Page

170

Share

COinS