Photovoltaic Module Health Assessment Based on Unmanned Aerial Vehicle Infrared Images

Photovoltaic (PV) power generation systems are a new type of power generation system which directly converts solar radiation into electrical energy using the photovoltaic effect of semiconductor materials. However, various problems inevitably occur in the operation of PV systems due to the aging of solar cells and environmental conditions. The hot spot effect is one of the most common faults in PV modules, which not only causes abnormal temperature on the surface of the module but also damages the photovoltaic panel, affecting the overall power generation efficiency. In this paper, based on the multiscale data features of images in PV module defect diagnosis, the data correlations in different scenarios are explored to investigate the impact mechanism of several hot spot features (position, area, brightness difference) on the health condition of PV modules directly from the perspective of unmanned aerial vehicle (UAV) infrared images. By incorporating a multiscale data fusion algorithm, we established a comprehensive health assessment system for PV modules. Experimental analysis demonstrates the effectiveness of the proposed method.