Scholarlink

The planning and control of solar power face unpredictability mainly due to factors such as local weather dependencies, irradiance fluctuations, humidity, and atmospheric pollutants. This study investigates the impact of atmospheric, environmental, and pollution parameters on solar output. Datasets from 1 year of operation of a 2.64 kW polycrystalline rooftop panel in Jaipur went through statistical and regression analysis. Following this, the energy production was correlated with 20 parameters, including 2022’s solar flares and CMP outbursts. From the parameters under investigation, ten exhibited a significant correlation with solar output. These were six atmospheric variables, namely, air temperature, relative humidity, cloud opacity, GHI, DNI, DHI, and four pollutant parameters (PM10, NOx, SO2, benzene). Using the regression technique in machine learning, a solar output model was developed, surpassing traditional empirical correlation-based models and improving input–output accuracy.

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

Machine Learning Perspectives on Solar Panel Efficiency: The Impact of Pollutants and Environmental Factors

  • 章节

  • 发表日期：

    2025年

查看全文
  • Richa Verma,
  • Bhupender Parashar,
  • Priyanka Kulshrestha,
  • Bishnu Kant Shukla

摘要

The planning and control of solar power face unpredictability mainly due to factors such as local weather dependencies, irradiance fluctuations, humidity, and atmospheric pollutants. This study investigates the impact of atmospheric, environmental, and pollution parameters on solar output. Datasets from 1 year of operation of a 2.64 kW polycrystalline rooftop panel in Jaipur went through statistical and regression analysis. Following this, the energy production was correlated with 20 parameters, including 2022’s solar flares and CMP outbursts. From the parameters under investigation, ten exhibited a significant correlation with solar output. These were six atmospheric variables, namely, air temperature, relative humidity, cloud opacity, GHI, DNI, DHI, and four pollutant parameters (PM10, NOx, SO2, benzene). Using the regression technique in machine learning, a solar output model was developed, surpassing traditional empirical correlation-based models and improving input–output accuracy.