Detection and classification of faults in photovoltaic modules

Abstract

Solar photovoltaic systems are being widely used in green energy harvesting recently. At the same rate of growth, the modules that come to the end of life are growing fast. Therefore, rapid fault detection and classification of PV modules can help to increase the reliability of the PV systems and reduce operating costs. Inspection and maintenance of solar modules are important to increase the lifetime, reduce energy loss, and environmental protection. A combination of infrared thermography and machine learning methods has been proven effective in the automatic detection of faults in large-scale PV plants. However, so far, few studies have assessed the challenges and efficiency of these methods applied to the classification of different defect classes in PV modules. In this dissertation, an efficient PV fault detection and classification method is proposed to classify different types of PV module anomalies using thermographic images utilizing convolutional neural networks (CNN) and artificial neural networks (ANN). Eleven types of PV module faults such as cracking, diode, hot spot, offline module, and other faults are utilized. Several evaluation metrics were used toassess the performance namely accuracy, recall, precision, and F1 score. The testing accuracy was obtained as 91% for the detection of anomalies in PV modules and 91% to classify defects for four classes and 73% for twelve classes. In conclusion, the integration of advanced imaging techniques and machine learning algorithms presents a promising avenue for enhancing the reliability of PV systems. As the demand for clean and sustainable energy continues to grow, such innovations will be instrumental in meeting global energy needs while minimizing environmental impact. The advancements outlined in this thesis represent a significant step forward in the pursuit of more efficient and resilient renewable energy infrastructures.