Paper Title
Performance Comparison of Machine Learning Models for Fault Diagnosis in Photovoltaic Systems

Fault detection in photovoltaic systems (PVS) is crucial to mitigate panel degradation. This concern has increased along with panel production even more, knowing this degradation is irreversible and has been found to degrade them at a rate of 1% per year. Early detection of faults can increase the overall efficiency of PVS during its life cycle and helps extend its life span. This study focuses on the evaluation of various machine learning models for detecting and classifying seven types of faults in PVS, including partial shading, a primary cause of hotspots. To achieve this, a database derived from a photovoltaic model in a laboratory has been used. The database, consisting of 2.2*10^6 measurements, includes seven fault types: open circuit, voltage sags, partial shading, inverter, current feedback sensor, and MPPT/IPPT controller in boost converter faults. Both Maximum Power Point Tracking (MPPT) and Limited Power Point Tracking (LPPT) modes are considered. The experimental results reveal distinct performance variations among machine learning models based on the PVS working mode and the specific fault type in the system. Keywords - PV System, Solar Panel, Solar Cell, MATLAB, Predictive Maintenance, Fault Diagnosis, Prediction Models, Machine Learning, Supervised Training, Efficiency, Accuracy.