Abstract:With the rapid development of infrastructure including power grids, managers in power industry these days are faced with an increasingly severe problem of theft of electricity. Theft of electricity has negative effects on many socioeconomic aspects, including impacting stable growth of economic for power enterprises and social development. The traditional anti-theft means require officers checking the integrity of kilowatt-hour meter and the correctness of wiring house by house, which requires enormous manpower and material resources. With the advancement of information collecting technology, power enterprises now possess relatively complete database of power consumption. As a result, performing data mining on existing database and identifying abnormal users has become a hot topic in the field of information technology. The purpose of this paper is to identify abnormal users with machine learning algorithms, providing power enterprises with means to detect theft of electricity at lower cost. The contributions of this paper are listed as follows: 1) Propose various features, providing means to describe users’ behaviors. First, monistic features (mean, difference, coefficient of variation, range, standard deviation), binary features (cosine similarity, Pearson product-moment correlation coefficient) and multivariate features are calculated based on user power consumption (monthly, seasonally, yearly, per holiday, per workday). Then principal component analysis is used to perform dimensionality reduction on the dataset. The dataset is finally scaled and oversampled to form the feature dataset. 2) Study various classification algorithms, optimize hyperparameters, providing models to detect theft of electricity. In this paper, the mechanism behind support vector machine, back-propagation neural network, random forest and XGBoost is studied and the hyperparameters are optimized. 3) Compare and evaluate different classifiers, and provide suggestions for real-world application. In this paper, different classifiers are evaluated with different metrics and the best classifier is recommended based on real-world dataset and different application scenarios.

Autoencoder and Kernel Density Estimation-Based Anomaly Detection Method for Power Consumption among Distribution Network Customers

Power Consumption-based Anomaly Detection for Relay Protection

Anomaly Detection for Power Consumption Patterns Based on Unsupervised Learning

Anomaly Detection Algorithm Based on Electric Equipment

Anomaly electricity detection method based on entropy weight method and isolated forest algorithm

Feature Extraction of Anomaly Electricity Usage Behavior in Residence Using Autoencoder

A Distributed Anomaly Detection Method of Operation Energy Consumption Using Smart Meter Data

High-Dimensional Energy Consumption Anomaly Detection: A Deep Learning-Based Method for Detecting Anomalies

Anomaly Detection in Electricity Consumption Data using Deep Learning

An Efficient Method for Detecting Abnormal Electricity Behavior

A deep learning approach for anomaly detection and prediction in power consumption data

Power Consumption Predicting and Anomaly Detection Based on Transformer and K-Means

A New Anomaly Detection System for School Electricity Consumption Data

Enhancing power utilization analysis: detecting aberrant patterns of electricity consumption

Detection and Classification of Anomalies in Power Distribution System Using Outlier Filtered Weighted Least Square

Theft of Electricity Detection Based on Supervised Learning

An Unsupervised Abnormal Power Consumption Detection Method Combining Multi-Cluster Feature Selection and the Gaussian Mixture Model

Anomaly Detection of Smart Metering System for Power Management with Battery Storage System/Electric Vehicle

Identification of Abnormal PV Output Power Based on kernel Density Estimation and Consistency Method

Anomaly Detection for Power Consumption Data Based on Isolated Forest

Analytical Investigation of Autoencoder-Based Methods for Unsupervised Anomaly Detection in Building Energy Data