Photovoltaic Cell Defect Detection Model based-on Extracted Electroluminescence Images using SVM Classifier
Ronnie O. Serfa Juan, Jeha Kim
Abstract
Electroluminescence (EL) imaging is used to analyze the characteristics of solar cells. This technique provides various details about solar panel modules such as solar cell characteristics, materials used, health status, defects, etc. The derived features from solar panel images provide a significant source of information for photovoltaic applications such as fault detection assessment. In this work, a method for classifying between the normal and a defective solar cell was implemented using EL imaging with selected digital image processing techniques through the Support Vector Machine (SVM) classifier. The EL images are processed using feature extraction procedures. The system was observed to provide an accuracy of 95%. The algorithm presented was implemented in MATLAB R2019b programming environment.