Simplified Independent Model for Predicting Global Solar Radiation

Abstract

In this investigation, five existing independent empirical models were calibrated and evaluated to calculate daily and monthly mean global solar radiation on a horizontal surface in Adrar city located in the south of Algeria, using meteorological data measured from 2013 to 2018. The measured data were divided into two periods; the first period (2015-2018) was used to calculate the empirical coefficients of the models, while the second period (2013-2014) was used to validate the correlations. Additionally, the best model (Al-Salaymeh model) was compared with five dependent empirical extreme air temperature models. In general, the results show that dependent models exhibited privileged performance than independent models. However, Al-Salaymeh regression independent model can contend with regression dependent models. Because of they use only the day number as a key input with smaller relative errors. It is found that daily statistical tests mean absolute bias error, root mean square error and coefficient of determination were equal to 2.0117 MJ/m², 2.4612 MJ/m² and 0.8014 respectively. The best independent empirical model was also compared with the modified Algerian solar atlas model to show their effectiveness. As a conclusion, the simple independent models can satisfactorily describe the horizontal global solar radiation for Adrar.