String-Level Ground Fault Localization for TN-Earthed Three-Phase Photovoltaic Systems

2026-01-28

AI Summary

This study addresses the challenge of ground fault detection in TN-grounded three-phase photovoltaic (PV) systems, where DC-side grounding faults can induce large currents that damage equipment, and conventional manual inspection methods are inefficient. The authors propose a novel approach that integrates edge AI with the variational information bottleneck (VIB) framework. They develop a PLECS-based simulation model incorporating PV hysteresis effects to generate multi-location fault data and extract correlation features from the four-stage inverter shutdown process. A lightweight fault localization model is then trained using these features. Evaluated under typical sampling rates, the method achieves over 93% accuracy in identifying faulty PV strings while maintaining low computational overhead, demonstrating strong potential for deployment on resource-constrained inverters.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Intelligent Robots: State Estimation
• Computer Vision: Adversarial Attacks & Robustness

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI

Abstract

The DC-side ground fault (GF) poses significant risks to three-phase TN-earthed photovoltaic (PV) systems, as the resulting high fault current can directly damage both PV inverters and PV modules. Once a fault occurs, locating the faulty string through manual string-by-string inspection is highly time-consuming and inefficient. This work presents a comprehensive analysis of GF characteristics through fault-current analysis and a simulation-based case study covering multiple fault locations. Building on these insights, we propose an edge-AI-based GF localization approach tailored for three-phase TN-earthed PV systems. A PLECS-based simulation model that incorporates PV hysteresis effects is developed to generate diverse GF scenarios, from which correlation-based features are extracted throughout the inverter's four-stage shutdown sequence. Using the simulated dataset, a lightweight Variational Information Bottleneck (VIB)-based localization model is designed and trained, achieving over 93% localization accuracy at typical sampling rates with low computational cost, demonstrating strong potential for deployment on resource-constrained PV inverters.

Problem

• ground fault
• photovoltaic systems
• fault localization
• TN-earthed
• string-level

Innovation

• edge AI
• ground fault localization
• Variational Information Bottleneck
• three-phase PV systems
• fault-current analysis