An Ensemble Deep Convolutional Neural Network Model for Electricity Theft Detection in Smart Grids

TL;DR

This paper proposes an ensemble deep convolutional neural network model for detecting electricity theft in smart grids, addressing data imbalance and improving detection accuracy using a voting system.

Contribution

It introduces a novel ensemble deep learning approach with data balancing and voting for enhanced electricity theft detection in smart grids.

Findings

The proposed model outperforms existing methods in accuracy and AUC.

Data imbalance is effectively handled by the random under bagging technique.

The ensemble approach improves detection metrics significantly.

Abstract

Smart grids extremely rely on Information and Communications Technology (ICT) and smart meters to control and manage numerous parameters of the network. However, using these infrastructures make smart grids more vulnerable to cyber threats especially electricity theft. Electricity Theft Detection (EDT) algorithms are typically used for such purpose since this Non-Technical Loss (NTL) may lead to significant challenges in the power system. In this paper, an Ensemble Deep Convolutional Neural Network (EDCNN) algorithm for ETD in smart grids has been proposed. As the first layer of the model, a random under bagging technique is applied to deal with the imbalance data, and then Deep Convolutional Neural Networks (DCNN) are utilized on each subset. Finally, a voting system is embedded, in the last part. The evaluation results based on the Area Under Curve (AUC), precision, recall, f1-score, and accuracy verify the efficiency of the proposed method compared to the existing method in the literature.