Prediction of Short-Term Solar Radiation Using Machine Learning Methods

Abstract

AbstractThe renewable energy such as wind energy and solar energy has been established to offer benefits of sustainable energy sources and lowered environmental pollution levels. In addition to the benefits they offer, factors such as global warming and the growing energy crisis have led to a progressive increase in demand for renewable energy. To meet such a growing demand, there is a need for an efficient energy management system that promotes more accurate forecasting techniques. Forecasting to solar system output is mainly focused on the prediction of solar radiation. Numerous forecasting methods have been applied for achieving short-term solar radiation forecasting. This paper presents a comparative analysis of forecasting performance of three machine learning methods such as support vector machine, multiple linear regression, and artificial neural network. It is investigated that the prediction accuracy based on the multiple linear regression yields better results than that of the artificial neural network and support vector regression. Subsequent paragraphs, however, are indented.KeywordsSupport vector machineMultiple linear regressionArtificial neural networkSolar radiationSolar power