Machine Learning Approach for Photovoltaic Panels Cleanliness Detection

Solar Energy is one of the primary renewable energy sources. Solar Photovoltaic systems depend on absorbed solar irradiance, which is significantly affected by depositions and shadowing. Machine learning and image processing techniques are widely used in the detection of depositions. However, current work depends on primitive features, fixed camera, fixed background, and manually cropped images. In this paper, a machine learning approach is proposed to determine the amount of depositions. The proposed method overcomes the limitations of earlier work by utilizing terrestrial and aerial images and applies automatic background removal along with the extraction of sophisticated features. The results demonstrate the efficiency and the robustness of the proposed approach under different classification algorithms.