Research Paper AN ENHANCING DEFECT CLASSIFICATION IN SOLAR PANELS WITH ELECTROLUMINESCENCE

Abstract

Solar energy has become one of the most important renewable energy sources worldwide. However, the efficiency and reliability of solar panels are greatly affected by manufacturing defects and aging-related faults such as microcracks, cell breakage, hotspots, and inactive regions. Early and accurate detection of these defects is essential to reduce power loss and maintenance costs. Traditional inspection methods are time-consuming, expensive, and highly dependent on human expertise. This project presents an automated solar panel defect detection and classification system using Electroluminescence (EL) imaging combined with advanced deep learning techniques. EL imaging is a non-destructive technique that captures high-resolution images of solar cells, making invisible defects clearly visible. To efficiently analyze these images, state-of-the-art object detection models YOLOv8 and YOLOv9 are employed. The proposed system involves preprocessing EL images, training deep learning models on labeled defect datasets, and detecting multiple defect types with high accuracy and realtime performance. YOLOv8 and YOLOv9 are chosen due to their fast inference speed, strong feature extraction capability, and improved accuracy compared to traditional machine learning approaches. The system outputs the defect type, location, and confidence score, enabling precise quality assessment. Experimental results demonstrate that the proposed approach achieves high detection accuracy and robustness across different defect categories. This automated solution significantly improves inspection efficiency, reduces manual effort, and supports large-scale solar panel monitoring. The developed system can be effectively applied in solar panel manufacturing, maintenance, and quality control to enhance overall energy production and system reliability.