Detection and Analysis of Deteriorated Areas in Solar PV Modules Using Unsupervised Sensing Algorithms and 3D Augmented Reality

Abstract

Solar Photovoltaic (PV) is increasingly being used to address the global concern of energy security. However, hotspots and snail trails in PV modules, caused mostly by cracks, reduce their efficiency and power capacity. This article presents a new methodology for automatically segmenting and analyzing anomalies like hotspots and snail trails in PV modules, leveraging unsupervised sensing algorithms and 3D Augmented Reality (AR) for visualization. The computer simulations demonstrate enhanced segmentation efficiency compared to current state-of-the-art segmentation techniques, namely Weka and the Meta Segment Anything Model (SAM). These simulations utilized the Cali-Thermal Solar Panels and Solar Panel Infrared Image Datasets for evaluation. The evaluation metrics included the Jaccard Index, Dice coefficient, Precision, and Recall. The calculated values for these metrics were 0.76, 0.82, 0.90, 0.99, and 0.76, respectively. Our objective is to leverage drone technology for real-time, automatic solar panel detection, which would significantly boost the efficacy of PV maintenance. The proposed methodology could improve solar PV maintenance by enabling swift, precise anomaly detection without human intervention. This could lead to significant cost savings, increased energy production, and improved overall performance of solar PV systems. Moreover, the novel combination of unsupervised sensing algorithms with 3D AR visualization heralds new opportunities for future research and development in solar PV maintenance.

Keywords: Solar photovoltaic (PV), Fault and abnormality detection, Unsupervised Segmentation, Image Enhancement, Augmented reality visualization.