Wavelet-Gaussian Process Regression Model for Regression Daily Solar Radiation in Ghardaia, Algeria

Abstract

Received: 18 October 2020 Accepted: 22 February 2021 Several methods have been used to predict daily solar radiation in recent years, such as artificial intelligence and hybrid models. In this paper, a Wavelet coupled Gaussian Process Regression (W-GPR) model was proposed to predict the daily solar radiation received on a horizontal surface in Ghardaia (Algeria). A statistical period of four years (2013 -2016) was used where the first three years (2013-2015) are used to train model and the last year (2016) to test the model for predicting daily total solar radiation. Different types of wave mother and different combinations of input data were evaluated based on the minimum air temperature, relative humidity and extraterrestrial solar radiation on a horizontal surface. The results demonstrated the effectiveness of the new hybrid model WGPR compared to the classical GPR model in terms of Root Mean Square Error (RMSE), relative Root Mean Square Error (rRMSE), Mean Absolute Error (MAE) and determination coefficient (R).