Scholarlink

Solar power is one of the fastest-growing renewable energy sources, and the demand for the same has increased owing to rapid increase in industrialization, population growth, and due to the continuous need to reduce carbon footprint. With such an increase in solar adoption throughout the world, there is a requirement for fast and accurate photovoltaic (PV) fault detection and diagnosis (FDD) methods as well. The most commonly used method for FDD in large-scale PV farms is thermal photography. Due to the high cost associated with both these processes, the frequency of such surveys is often less, and hence, there is a need to combine such surveys with the measurement data for cost-effective modelling. In this paper, we detail the methodology employed to predict power loss in a real-world solar farm in India by utilizing both thermal imaging and measurement data. The methodology involves the development of the theoretical model of the system based on a single diode model for the PV cell. The system performance and production losses are calculated based on the measurement data. Thermal imaging is carried out to identify the number and type of faults in the solar farm, and the performance loss value is mapped to the faults in the system using a multivariate regression model. The developed model is found to predict power loss in the system with an accuracy of about 94.25%.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Power Loss and Performance Analysis of Large-Scale Solar Farms Using a Hybrid Approach Combining Measurement Data and Thermal Imaging

  • 章节

  • 发表日期：

    2026年

查看全文
  • Anjana G. Rajakumar,
  • Ila Thakur,
  • Remish Leonard Minz,
  • Nikhil Kulkarni,
  • Prateek Mital,
  • Nilanjan Chakravortty

摘要

Solar power is one of the fastest-growing renewable energy sources, and the demand for the same has increased owing to rapid increase in industrialization, population growth, and due to the continuous need to reduce carbon footprint. With such an increase in solar adoption throughout the world, there is a requirement for fast and accurate photovoltaic (PV) fault detection and diagnosis (FDD) methods as well. The most commonly used method for FDD in large-scale PV farms is thermal photography. Due to the high cost associated with both these processes, the frequency of such surveys is often less, and hence, there is a need to combine such surveys with the measurement data for cost-effective modelling. In this paper, we detail the methodology employed to predict power loss in a real-world solar farm in India by utilizing both thermal imaging and measurement data. The methodology involves the development of the theoretical model of the system based on a single diode model for the PV cell. The system performance and production losses are calculated based on the measurement data. Thermal imaging is carried out to identify the number and type of faults in the solar farm, and the performance loss value is mapped to the faults in the system using a multivariate regression model. The developed model is found to predict power loss in the system with an accuracy of about 94.25%.