A recent study has made a significant breakthrough in solar energy planning by identifying an artificial intelligence (AI) model that can accurately predict the sun’s seasonal strength across India’s diverse climates. The Gaussian Process Regression (GPR) machine learning model was found to be the most reliable tool for forecasting seasonal Global Horizontal Irradiance (GHI), a crucial parameter for assessing the available solar resource for Stand-Alone Photovoltaic (SAPV) systems, such as solar power plants.
The study, conducted by researchers from Vellore Institute of Technology, Chennai, used data from four climatically distinct locations in India: Chennai, Jaisalmer, Leh, and Mawsynram. The GPR model demonstrated exceptional precision, achieving a Root Mean Square Error (RMSE) of 0.0030 and a Coefficient of Determination of 0.9999, representing a reduction in error of up to 189.1% compared to other tested models.
The research aimed to enhance the precision of long-term solar resource evaluation and planning across diverse industries by analyzing historical climatic fluctuations and their influence on solar irradiance. Accurately predicting the GHI seasonal average is essential for estimating the power a solar panel can absorb and optimizing SAPV system planning. The study used three years of hourly meteorological data from 2017 to 2019 from the National Solar Radiation Database (NSRDB) and compared the performance of three machine learning models: Efficient Linear Regression (ELR), Regression Trees (RT), and Gaussian Process Regression (GPR).
The results confirmed the exceptional performance of the GPR model, which consistently delivered the lowest prediction errors across all four locations and three seasons. The study provides a validated tool for solar energy developers and planners, enabling them to accurately forecast seasonal GHI and plan SAPV systems for maximum efficiency and reliability, even in India’s most challenging climates. By leveraging machine learning techniques, the study can uncover complex patterns in GHI and climatic data that conventional statistical or physical models often overlook, making it a significant breakthrough in solar energy planning.