Work

Effectiveness of supervised machine learning models for electrical fault detection in solar PV systems

Year: 2025

Type: article

Abstract: Even though Photovoltaic (PV) systems have emerged as a viable substitute for non-renewable energy sources, their widespread integration into the electrical grid presents several issues today. On the other hand, various faults are a key concern affecting PV plants' production and longevity. The current study uses Machine Learning (ML) algorithms such as Decision Tree (DT), Naïve Bayes (NB), Random Forest (RF), Support Vector Machine (SVM) and XGBoost to detect and classify PV errors corresponding to Short Circuits (SC), Open Circuits (OC), Ground Faults (GF), and Mismatch Faults (MF). Simulations were conducted in MATLAB/Simulink to analyse voltage, current, and power variations during fault conditions and study their impact. The proposed results show that the effectiveness of ML in electrical fault detection, with the following classification accuracies: SVM - 97.40%, DT- 97.20%, RF - 97.20%, NB - 97.60%, and XGBoost - 98.0%. The effectiveness of the classification is confirmed through confusion matrices and correlation heatmaps. This research highlights the need for integrating intelligent monitoring, real-time IoT-based detection, and prediction analytics to improve PV system reliability.

Source: Scientific Reports

DOI: 10.1038/s41598-025-18802-4

Authors: Ved Khandeparkar, Shreshtha, Senthil Kumar Ramu

Institution: Vellore Institute of Technology University

Language: English

Topic: Electrical Fault Detection and Protection

Subfield: Electrical and Electronic Engineering

Field: Engineering

Domain: Physical Sciences

SDGs: Affordable and clean energy, Industry, innovation and infrastructure

Open Access status: gold

Estimated APC paid: $2,690