Daily predictions of solar radiation utilizing genetic programming techniques

Abstract

The solar radiation prediction in Kuala Terengganu located in Terengganu, Malaysia was investigated in this study to improve the solar system design. Solar radiation data and number of parameters such as solar radiation, temperature, humidity, wind speed and sunshine hours were obtained from Malaysian Meteorological Malaysia MMD. In order to predict the solar radiation, Genetic Programming Techniques (GP) models were develop and tested. Two scenarios were considered in this study in order to validate the efficiency of the proposed model. Coefficients of determination (R2) for the solar radiation during training and testing phases were ranged between 0.99402 to 0.98934 for all months of the year. This study confirms the ability of GP to predict solar radiation values precisely and accurately. The predictions from the GP models could enable scientists to locate and design solar energy systems in Malaysia.