Machine learning models to predict diffuse solar radiation based on diffuse fraction and diffusion coefficient models for humid-subtropical climatic zone of India

• DOI: 10.1016/j.clet.2021.100262
• Publication Date: 2021-08-25T06:51:28Z
• Authors: Shahid Husain; Uzair Ali Khan
• Keywords: Diffuse fraction; Clearness index; Machine learning; Solar radiation; 0211 other engineering and technologies; 0202 electrical engineering, electronic engineering, information engineering; India; TJ807-830; Environmental engineering; 02 engineering and technology; TA170-171; Diffusion coefficient; Renewable energy sources

In the present work, twelve machine learning (ML) models are developed for assessment of monthly average diffuse solar radiation (DSR) with solitary input forecaster as clearness index. Two categories ML were demarcated (i.e. diffusion coefficient and fraction) six each group. The correctness was examined a function some frequently used statistical pointers. A comparision also done between well-recognised available from previous works. results show that perform very well in to literature. top-performing category 1 k-nearest neighbours (KNN) model both training testing data. 2, data random forest (RF) while support vector regression (SVR) well. performance can be slightly improved by using two or more parameters such temperature difference, relative humidity sunshine along index input. Thus, estimate DSR humid-subtropical climate India.