Global Solar Radiation Forecasting Based on Hybrid Model with Combinations of Meteorological Parameters: Morocco Case Study

The adequate modeling and estimation of solar radiation plays a vital role in designing energy applications. In fact, unnecessary environmental changes result several problems with the components photovoltaic affects generation network. Various computational algorithms have been developed over past decades to improve efficiency predicting various input characteristics. This research provides five approaches for forecasting daily global (GSR) two Moroccan cities, Tetouan Tangier. this regard, autoregressive integrated moving average (ARIMA), (ARMA), feed forward back propagation neural networks (FFBP), hybrid ARIMA-FFBP, ARMA-FFBP were selected compare forecast different combinations meteorological parameters. addition, performance three has calculated terms statistical metric correlation coefficient (R2), root means square error (RMSE), stand deviation (σ), slope best fit (SBF), legate’s (LCE), Wilmott’s index agreement (WIA). model is by using computed metric, which present, optimal value. R2 forecasted ARIMA, ARMA, FFBP, models varying between 0.9472% 0.9931%. range value SPE 0.8435 0.9296. LCE 0.8954 0.9696 WIA 0.9491 0.9945. outcomes show that ARIMA–FFBP ARMA–FFBP techniques are more effective than other due improved (R2).