ANFIS Technique to Estimate Daily Global Solar Radiation by Day in Southern Algeria

Abstract

In this paper, the performance and accuracy of the adaptive neuro-fuzzy inference system (ANFIS) technique is assessed for estimating the daily horizontal global solar radiation H in South Algeria. 21 years of experimental data for three cities in South Algeria Adrar, Timimoun and Bordj Badji Mokhtar (BBM) were used to train and test the ANFIS model. The results verified that the ANFIS model provides accurate predictions based on statistical formulas such as: the root mean square error (RMSE), mean absolute percentage error (MAPE and the coefficient of determination (R). In a nutshell, the results highly encouraged the implementation of ANFIS to estimate the daily horizontal global solar radiation using only the number of the day of years nd.