Diagnosis of Photovoltaic (PV) Panel Defects Based on Testing and Evaluation of Thermal Image

Abstract

Abstract Photovoltaic (PV) solar energy can only be economical if the PV module operates reliably for 25–30 years under field conditions. The PV module and it overall reliability can be radically affected by faults during the manufacturing process, in real field conditions, transportation, and installation. So, there is a need for diagnosing defects in PV modules to improve their reliability. Operating temperature plays the key role for improving the efficiency of PV panels. The temperature within the PV cell unevenly increases because of such defects in the cell. As such, it is very important to monitor the temperature and temperature distribution in PV panels in order to locate such defects. Infrared thermography (IRT) plays a major role in predictive and preventive maintenance of PV panels and can determine the severity of the problem. This article investigates the delamination, snail trails, and bubbled faults of PV panels using digital thermal image analysis and their feature extraction. Real time experiments were conducted, and the test results are presented in this article. Thermal images of panels are captured using a (FLIR) T420bx® thermal imager. The thermal images of panels are analyzed by segmenting the image using the k-means clustering algorithm. Histogram statistical features such as mean, standard deviation, variance, entropy, skew, and kurtosis are extracted from the segmented thermal image. Based on these features, the defects in PV panels are identified with reasonable accuracy.