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

TL;DR

This paper introduces a novel edge-AI-based method for accurately localizing ground faults at the string level in TN-earthed PV systems, significantly improving speed and efficiency over manual inspection.

Contribution

It presents a new GF localization approach using a Variational Information Bottleneck model trained on simulation data, tailored for resource-limited PV inverters.

Findings

Achieves over 93% localization accuracy.

Demonstrates low computational cost suitable for real-time deployment.

Provides comprehensive analysis of GF characteristics and simulation-based validation.

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.