Artificial neural networks for forecasting the 24 hours ahead of global solar irradiance

Abstract

The global situation of climate changes is becoming increasingly serious as a result of the exhaustion of fossil energy. In this study, we focus on solar energy, which has been receiving increased amounts of attention in the last few decades. The integration of solar energy into electricity networks requires reliable forecast information of solar resources enabling it to quantify the available energy and allowing it to optimally manage the transition between intermittent and conventional energies. After we have investigated in our previous study different forecasting techniques in order to find which one is appropriate for forecasting the daily global solar irradiance for the region of Rabat. Throughout this work, we investigated the hourly case. The first-tested approach is linear modeling based on classical SARIMA models. The second approach proposes non-linear modeling based on Artificial Neural Networks (ANNs) models in the univariate case and the multivariate case. Numerous research has demonstrated the ability of ANNs to predict time series of weather data. In this study, we examined a particular structure of ANNs, Multilayer Perceptron (MLP), The results showed that ANN performed better than SARIMA models.