TITLE:
Fault Diagnosis in Photovoltaic Systems Using Machine Learning Algorithms
AUTHORS:
Patience Tifuh Taah, Derek Ajesam Asoh, Jerome Ndam Mungwe, Jean-de-Dieu Nguimfack, Daniel Agoons
KEYWORDS:
Machine Learning, Fault Diagnosis, PV Systems, Neural Networks, PV Faults
JOURNAL NAME:
Smart Grid and Renewable Energy,
Vol.17 No.6,
June
15,
2026
ABSTRACT: The global energy landscape is undergoing a significant shift from fossil fuels to renewable energy sources, driven by environmental concerns. Solar energy, particularly through photovoltaic (PV) technology, has emerged as a prominent renewable energy source. However, PV systems face challenges due to the occurrence of faults, which negatively impact their efficiency and power output. Machine learning algorithms have revolutionized various sectors, including PV systems, offering potential solutions to fault diagnosis. This study focuses on utilizing machine learning for fault diagnosis in PV systems. Through a MATLAB simulation of a 7.5 kW PV system, three fault scenarios were implemented to generate the dataset. Machine learning algorithms were trained and tested using the classification application, with cross-validation and hold-out validation techniques employed for model validation. Results revealed that the wide neural network algorithm achieved the highest accuracy, reaching 98.88% and 98.23% with the cross-validation and hold-out respectively. The ensemble algorithms also demonstrated promising results, achieving an accuracy of 95.02%. These findings underscore the effectiveness of machine learning algorithms in accurately diagnosing faults in PV systems, offering valuable insights for enhancing system efficiency, reliability, and availability.