Infrared Thermography based Defect Detection and Classification of solar Photovoltaic panel using machine learning

Abstract

The surge in renewable energy capacity is predominantly fueled by newlinesolar photovoltaics (PV), constituting over 50% of renewable electricity newlineinstallations. Projections indicate further expansion by 2030, according to newlinethe 2022 international energy agency world energy report. PV panels or newlinesolar panels, harness sunlight radiation through the photovoltaic effect to newlinegenerate electricity, Serving as a cornerstone in renewable energy systems. newlineThe 2020s witnessed substantial growth in PV module manufacturing, newlinespearheaded by China, the United States and India. newlineEnsuring a reliable, efficient and secure PV system is crucial to newlinemitigate energy losses, mismatches and safety issues. Detecting, identifying newlineand quantifying defects in PV modules are essential tasks, stemming from newlinemanufacturing, installation and environmental factors. These defects newlinecompound, leading to reduced power generation capacity. Advanced and newlineintelligent Operation and Condition Maintenance (OACM) emerge as a newlinecompelling solution, guaranteeing efficiency in PV plants. Traditional newlineanomaly detection techniques include visual inspections, physical thermal newlineimaging and electrical test like I-V curves. Among this infrared newlinethermography image-based defective module detection stands out for its newlinenon-disruptive nature and high accuracy. Infrared thermography (IRT), newlineproven effective in the electrical and medical sectors has gained popularity newlinefor error checking in PV modules with recent studies leveraging infrared newlinethermography for error detection. newline