Prediction of daily global solar radiation by day of the year in four cities located in the sunny regions of Iran

•We test 6 day of year based models for prediction of daily insolation in four cities of Iran.•Via statistical indicators the best model for each city is recognized and introduced.•The models do not have dependency to any meteorological and geographical data.•The hybrid sine and cosine wave model performs best in Bandarabass, Kerman and Tabass.•The 4th order polynomial degree model shows the best performance in Isfahan.

In this study, by using long-term global solar radiation data, 6 day of the year based empirical models were tested for prediction of daily global solar radiation in four Iranian cities of Bandarabass, Isfahan, Kerman and Tabass. The statistical non-linear regression technique was utilized to establish the models then their accuracy evaluated using the statistical indicators of mean absolute percentage error, mean absolute bias error, root mean square error and coefficient of determination. In Bandarabass, Kerman and Tabass, the hybrid sine and cosine wave model and in Isfahan the 4th order polynomial model provided the highest accuracy. To assess the performance of the models in different days of the year a further statistical analysis was performed using the relative percentage error. The number of days falling in its acceptable interval of −10% to 10% for the best models in Bandarabass, Isfahan, Kerman, and Tabass were 346, 322, 334 and 328, respectively. Making comparison with the predictions of some existing monthly averaged models stated the superiority of the newly established day of the year based models. Hence, they can be utilized swimmingly to estimate daily insolation in the corresponding cities and a vast similar climate region around them.