Photovoltaic (PV) module monitoringand upkeep are essential for a dependable and effective operation. Due to hotspots in PV modules brought on by a variety of flaws and operational issues, the dependability of the PV system may be put in jeopardy. Hotspots should be found and categorized from a monitoring perspective for later maintenance. In this study, hotspots are identified, assessed, and categorized using thermal pictures of PV modules and a machine learning technique. To do this, categorization is based on the texture and histogramof gradient (HOG) features of thermal pictures of PV modules. The machine learning method like Naive Bayes (nBayes) classifier is used to train the images in order to identify the hotspots and classifies them into defective and non-defective images.

Keywords : Hotspots, Monitoring, Photovoltaic (PV) Modules, Naive Bayes Classifier, Texture and Histogram of Gradients (HOG) Thermal Images.