The global capacity of renewable energy is on the rise in order to reach to the carbon neutrality. Among the renewable energy, solar energy is broadly well deployed. Accordingly, solar power plant operation and maintenance (O&M) becomes critical and cost competitiveness in O&M gets more attention in the overall solar power market.

  In response to this demand, Dr. Ko, Suk-whan, a researcher of the Renewable Energy System Research Dept. at the Korea Institute of Energy Research (Yi, Chang-keun, President of KIER), who has performed solar module defect diagnose for the past 10 years, succeeded in AI-powered defect detection technology based on the 10-year accumulated I-V data on solar panels’ degradation, faults, and other problems.

  The technology enables to accurately detect solar panels’ performance status such as defects and degradation once solar panels’ detailed information including serial-parallel solar array and I-V curve data including irradiation and temperature is recorded onto AI learning model. The KIER has carried out evaluation and detection of solar panels’ performance. In the coming years it plans to expand this mission to public service while using this detection technology.

  Dr. Ko, Suk-whan says that many of solar farms, these days, are located in such inaccessible places as waterways, farming lands, on sound-absorbing walls, on the sea, or on the water, preventing from detecting defects and degradation of the solar panels. He emphasizes the technology he has developed helps us to easily detect even minor degradation like contamination to the level of 95% accuracy by using AI learning model analysis developed from normal I-V curve data and labelled defect I-V curve data.

  The technology was accredited as the New Technology by the 2023 New Technology Accreditation Council according to the ‘Industrial Technology Innovation Promotion Law. It is planned to commercialize with research spin-offs and also for public commercial service.