Literature Review of Methods for Detecting Non-Technical Electricity Losses in Distribution Grids

The widespread deployment of advanced metering infrastructure (AMI) and smart grid technologies has significantly expanded the availability of high-resolution electricity consumption data in distribution networks. This creates new opportunities for addressing non-technical losses (NTL). This paper presents a systematic literature review of data-driven methods for NTL detection based on an analysis of 79 relevant scientific publications from leading databases (IEEE Xplore, Springer, MDPI, and ScienceDirect) indexed in Scopus/Scimago over the period 2020–2025. The reviewed methods are categorized according to a novel triad of criteria: the analytical paradigm, the input data structure, and the required level of grid digitalization. A comparative analysis reveals distinct performance profiles: supervised classification methods, particularly hybrid and ensemble models, enable high-accuracy detection, with many studies reporting F1-scores exceeding 0.95 on benchmark datasets, yet they require extensive labeled data from AMI systems. Unsupervised clustering techniques offer a practical alternative for grids with partial or no AMI by analyzing aggregated consumption patterns. Forecasting-based (regression) methods facilitate continuous consumption monitoring and anomaly detection via deviation analysis, while scenario-modeling techniques provide strategic tools for evaluating the potential impact of NTL reduction measures. This review critically examines the causes of NTL, feature engineering strategies, software tools, and key challenges such as data imbalance, model interpretability, and real-world deployment constraints. The synthesis provides a utility-centric framework for method selection, offering practical recommendations tailored to different infrastructure profiles. The proposed systematization bridges methodological gaps and supports the development of context-aware, data-driven solutions for modern and evolving power systems. Received: 3 July 2025 | Revised: 8 January 2026 | Accepted: 14 January 2026 Conflicts of Interest The authors declare that they have no conflicts of interest to this work. Data Availability Statement Data sharing is not applicable to this article as no new data were created or analyzed in this study. Author Contribution Statement Irbek D. Morgoev: Conceptualization, Methodology, Formal analysis, Investigation, Data curation, Writing – original draft, Writing – review & editing, Visualization. Roman V. Klyuev: Conceptualization, Formal analysis, Writing – review & editing, Supervision. Angelika D. Morgoeva: Validation, Formal analysis, Investigation, Writing – original draft, Writing – review & editing, Project administration.