Predictive Maintenance of Solar Photovoltaic Systems: A Comprehensive Review

Abstract

Over recent decades, solar photovoltaic (PV) technologies have transformed the energy market, becoming a cornerstone of renewable energy systems. Ensuring the reliability of critical components within PV systems is essential to maximise their lifespan and minimise unexpected failures and downtime. Predictive maintenance, which leverages equipment condition modelling to anticipate faults and schedule maintenance, has emerged as a promising approach. However, forecasting PV equipment faults remains complex due to the indirect measurability of equipment status and the susceptibility of systems to various adverse conditions that may compromise system performance. In response to these challenges, there has been growing research interest in developing predictive analytics tools to optimise operational management. This study provides a comprehensive review of PV system design, key components, operation, different faults and maintenance strategies. Furthermore, it conducts a comparative analysis of artificial intelligence, including machine learning and deep learning models, to evaluate their performance in predictive maintenance applications for PV systems. This article reviews and analyses both established and emerging techniques used in PV systems, with particular emphasis on their effectiveness in addressing predictive maintenance. Its findings aim to inform the development of advanced fault prediction methods to improve the reliability and efficiency of solar PV systems. In addition, the paper serves as a valuable reference for researchers in this field, offering a clear overview of current approaches. It also identifies the main challenges and outlines key recommendations for future research directions, helping to guide innovation and progress in PV system maintenance and performance.