A survey on detecting and diagnosing of electrical faults for photovoltaic panel systems utilizing artificial intelligence techniques

Abstract Global energy demand growth and the challenges of climate change necessitate a shift to renewable energy sources, among which Photovoltaic (PV) solar energy plays a vital and effective role due to its environmental friendliness, easy of deployment, diversity and sustainability. A crucial aspect of ensuring the reliability and performance of PV is the timely detection and identification of electrical faults caused by long operational life, external influences, and internal system factors. Recently, Artificial Intelligence (AI) has been used as advanced monitoring techniques in this field, contributing to the precise enhancement and improvement of efficiency, accuracy, and effectiveness. This paper presents a comprehensive review of relevant AI techniques, including Machine Learning (ML), Deep Learning (DL) and other AI techniques. The paper thoroughly highlights their methodologies, advantages, limitations, associated datasets, and evaluation metrics. Accordingly, a comprehensive study is also conducted on the used datasets, their types, the total number of samples in each dataset and the number of faults that have been considered. Ultimately, the paper provides suggestions and recommendations for developing advanced systems of detecting and identifying electrical faults in PV systems.