Modeling of diffuse solar radiation exergy in Turkey: Comparison of diffuse fraction and diffusion coefficient models

In this study, 19 regression models were established for the prediction of exergy value of monthly average diffuse solar radiation with a single or multiple input predictor (Relative sunshine duration, clearness index, relative humidity, and air temperature). Five model categories were defined depending on the diffuse fraction and diffusion coefficient. The most frequently used statistical indicators were used to determine the performance of the empirical models. The global performance indicator (GPI) method was used to determine the accuracy ranking of the models. Consequently, the results show a good agreement with the values estimated from the regression models and the measurement data. The regression models established in this study can be used with great accuracy to evaluate the monthly average diffuse solar radiation exergy in the regions under consideration or in similar climatic conditions.