A comprehensive review of infrared thermography and deep learning applications for solar photovoltaic systems

Solar photovoltaic (PV) systems are highly promising source of renewable energy for clean energy production and are likely to replace fossil fuels in the upcoming years. However, the output from PV systems is subjected to losses from various defects, including cracks, hot spots, defective modules, etc. Therefore, in order to keep the energy production from PV systems at its maximum level, regular inspection and monitoring techniques are necessary. Infrared Thermography (IRT) has emerged as a non-destructive diagnostic tool for detecting different types of defects associated with PV systems, while deep learning techniques have demonstrated exceptional capabilities in automating and refining defect identification. This review explores the integration of IRT and deep learning for PV system monitoring, highlighting recent advancements, methodologies, and applications. Initially, the review presents an overview of IRT and deep learning in the context of solar PV systems. Key contributions include a synthesis of state-of-the-art developments, offering a succinct summary of the main findings. Furthermore, the review discusses the challenges faced in combining IRT and deep learning for solar PV systems and explores potential future improvements for better evaluation and monitoring, which will ultimately help establish solar PV systems as a leading renewable energy source in the near future.