Spatio-Temporal Grid Intelligence: A Hybrid Graph Neural Network and LSTM Framework for Robust Electricity Theft Detection

TL;DR

This paper presents a hybrid AI framework combining GNNs, LSTM, and machine learning to detect electricity theft accurately by modeling spatio-temporal grid data, significantly improving detection performance over traditional methods.

Contribution

The study introduces a novel hybrid model integrating GNN, LSTM, and supervised learning for robust electricity theft detection, addressing limitations of existing reactive and meter-centric approaches.

Findings

Hybrid model achieves 93.7% accuracy in theft detection.

Balanced precision and recall of 0.55 and 0.50 respectively.

Standalone anomaly detection yields low F1-score of 0.20.

Abstract

Electricity theft, or non-technical loss (NTL), presents a persistent threat to global power systems, driving significant financial deficits and compromising grid stability. Conventional detection methodologies, predominantly reactive and meter-centric, often fail to capture the complex spatio-temporal dynamics and behavioral patterns associated with fraudulent consumption. This study introduces a novel AI-driven Grid Intelligence Framework that fuses Time-Series Anomaly Detection, Supervised Machine Learning, and Graph Neural Networks (GNN) to identify theft with high precision in imbalanced datasets. Leveraging an enriched feature set, including rolling averages, voltage drop estimates, and a critical Grid Imbalance Index, the methodology employs a Long Short-Term Memory (LSTM) autoencoder for temporal anomaly scoring, a Random Forest classifier for tabular feature discrimination, and a GNN to model spatial dependencies across the distribution network. Experimental validation demonstrates that while standalone anomaly detection yields a low theft F1-score of 0.20, the proposed hybrid fusion achieves an overall accuracy of 93.7%. By calibrating decision thresholds via precision-recall analysis, the system attains a balanced theft precision of 0.55 and recall of 0.50, effectively mitigating the false positives inherent in single-model approaches. These results confirm that integrating topological grid awareness with temporal and supervised analytics provides a scalable, risk-based solution for proactive electricity theft detection and enhanced smart grid reliability.