Fault detection in photovoltaic system using SLIC and thermal images

Solar energy has been gaining a strong momentum as the future clean and renewable source of energy. Optimum utilization of this energy propelled research efforts into many areas of the solar energy system such as photovoltaic (PV) where significant improvement will lead to better systems' efficiency. PV systems operate without any supervisory mechanism but they still can have many faults internally and/or externally hindering its efficiency. In this work, we are focusing on creating a framework for automating defect detection in a solar energy system using thermal imaging to create an accurate and a timely alert system of hazardous conditions. We are proposing to use Simple Linear Iterative Clustering (SLIC) Super-pixel technique as technique for hot spot detection. Experimental results show that the hot spots in the solar panels can be accurately detected using infrared images using SLIC, in a real time implementation. Detection results will give alerts as to where the solar panels may not be working under normal conditions.