PCA based electricity theft detection in advanced metering infrastructure

Abstract

Advanced metering infrastructure is one of the important components of smart grid and offers an essential link between consumers and their loads, grid, and generation and storage resources. Electricity theft, one of the key concern in AMI, causes million dollar revenue loss every year in developing and developed countries. In this paper, Principal Component Analysis (PCA) based electricity theft detection scheme is proposed. PCA is used to transform a high dimensional dataset into a low dimensional dataset. Using principal components, anomaly score is calculated and compared with a predefined threshold value. The proposed scheme is tested under different attack scenario using real dataset. The results show that the proposed scheme detects electricity theft attacks with high detection rate.