solar radiation, modeling, predictors, South Africa
In this study a generalized model for predicting global solar radiation was developed for South Africa. This was achieved by employing easily measured weather elements like sunshine hour, ambient temperature, relative humidity and wind speed as predictors. The model accuracy was verified by comparing estimated values with measured values of global solar radiation in each province in terms of the following statistical error tests: mean bias error, mean absolute bias error, mean absolute percentage error, root mean square error, and regression coefficient. The values of regression coefficient obtained for each location shows that the developed model is suffice for use to predict global solar radiation and hence, could be used for any location within the country.