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.

Power Consumption Predicting and Anomaly Detection Based on Long Short-Term Memory Neural Network

An Anomaly Detection Algorithm for the Power Industrial Terminal Security Monitoring

Classification of Electricity Consumption Behavior Based on Improved K-Means and LSTM

Anomaly Detection Algorithm Based on Electric Equipment

Power Load Prediction Model Based on Long Short Term Memory and Sparrow Search Algorithm

Power Consumption-based Anomaly Detection for Relay Protection

A Power Load Forecasting Method Based on Intelligent Data Analysis

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

An Asymmetric Loss with Anomaly Detection LSTM Framework for Power Consumption Prediction

Theft of Electricity Detection Based on Supervised Learning

Short-term Time Series Data Prediction of Power Consumption Based on Deep Neural Network

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

Anomaly Detection in Electricity Consumption Data using Deep Learning

Short-term Prediction of Household Electricity Consumption Using Customized LSTM and GRU Models

Predicting Energy Consumption Using LSTM, Multi-Layer GRU and Drop-GRU Neural Networks

Abnormal Detection of Electricity Consumption of User Based on Particle Swarm Optimization and Long Short Term Memory With the Attention Mechanism

Mat-Transformer-Based State Prediction Method for Information Equipment

A Method Based on LSTM Network for Detection of Abnormal Daily Electricity

Data-Based Line Trip Fault Prediction in Power Systems Using LSTM Networks and SVM.

Demand Forecasting in Smart Grid Using Long Short-Term Memory

Comparative study of long short-term memory (LSTM), bidirectional LSTM, and traditional machine learning approaches for energy consumption prediction