A Comprehensive Review of Deep Learning and Machine Learning Techniques for Illegal Electricity Trapping Fraud Detection
Journal Of Recent Trends of Electrical Engineering · 26.03.2026 · 2 (1) · 30-36
DOI: 10.65525/jrtee.v2i1.18 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal Of Recent Trends of Electrical Engineering, online erschienen am 26.03.2026. Verfasst von Anwesha Majumdar und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Abstract (Originalsprache)
For modern power distribution systems, electricity theft is a persistent and complicated problem that significantly contributes to non-technical losses, financial deficits, operational inefficiencies, and public safety issues. Conventional detection methods, such as heuristic rule-based audits and manual inspections, are frequently costly, slow, and inappropriate for large-scale or real-time applications. Data-driven electricity theft detection (ETD) is made easier by the extensive use of smart grids, advanced metering infrastructure (AMI), and Internet of Things (IoT) technologies, which have made it possible to gather high-resolution consumption data. The machine learning (ML), deep learning (DL), hybrid, and Internet of Things (IoT)-assisted ETD approaches described in recent literature are all critically and methodically examined in this review. Methods from sophisticated deep neural architectures like CNNs, LSTMs, BiLSTMs, and ensemble frameworks to traditional machine learning classifiers are examined. Important topics covered include class imbalance, evaluation metrics, feature extraction, data preprocessing, and practical application. In order to facilitate the creation of dependable, scalable, and comprehensible ETD systems, unresolved issues, research gaps, and Finally, potential future paths are emphasized.
Zitieren
APA
Majumdar, A., & Ghoroi, S. (2026). A Comprehensive Review of Deep Learning and Machine Learning Techniques for Illegal Electricity Trapping Fraud Detection. Journal Of Recent Trends of Electrical Engineering, 2(1), 30-36. https://doi.org/10.65525/jrtee.v2i1.18
Vancouver
Majumdar A, Ghoroi S. A Comprehensive Review of Deep Learning and Machine Learning Techniques for Illegal Electricity Trapping Fraud Detection. Journal Of Recent Trends of Electrical Engineering. 2026;2(1):30-36. doi:10.65525/jrtee.v2i1.18
BibTeX
@article{Majumdar2026,
author = {Majumdar, Anwesha and Ghoroi, Sayan},
title = {A Comprehensive Review of Deep Learning and Machine Learning Techniques for Illegal Electricity Trapping Fraud Detection},
journal = {Journal Of Recent Trends of Electrical Engineering},
year = {2026},
volume = {2},
number = {1},
pages = {30--36},
doi = {10.65525/jrtee.v2i1.18}
}
EUVIMED