Machine Learning-Based Optimization Technique for Forecasting the Solar Radiation

Abstract

Solar radiation measurement determines how often electricity a given area absorbs from the sun. This light is the key source of energy for conversion into solar thermal and photovoltaic plants. The radiation incident is not stable and relies on the temperature records, contributing to intermittent activity and electricity supply changes. This justifies designing a method to forecast and estimate incident radiation to predict improvements in photovoltaic systems' performance. In this paper, the support vector machine (SVM) based machine learning is proposed to improve solar radiation prediction accuracy. The designed system results are compared with existing models that predicted the radiation and the global solar radiation is predicted accurately with efficient.