Solar radiation models. Part 2: Comparison and developing new models

Ulgen K. , HEPBASLI A.

ENERGY SOURCES, vol.26, no.5, pp.521-530, 2004 (Journal Indexed in SCI) identifier identifier

  • Publication Type: Article / Article
  • Volume: 26 Issue: 5
  • Publication Date: 2004
  • Doi Number: 10.1080/00908310490429704
  • Title of Journal : ENERGY SOURCES
  • Page Numbers: pp.521-530
  • Keywords: clearness index, correlation models, empirical models, global and diffuse radiation, solar radiation, solar radiation models, solar energy system, model comparison, Turkey, SUNSHINE DURATION, CLEARNESS INDEX, TURKEY, COEFFICIENTS


Solar engineers, architects, agriculturists, and hydrologists in many applications of solar energy require a reasonably accurate knowledge about the availability of solar resources at any place. For this purpose, in the past, several empirical models have been developed in order to estimate the solar radiation around the world. The main objective of the present study is twofold, namely: (1) to compare some existing models used for estimating the monthly average daily global solar radiation on a horizontal surface (H) for Turkey's big cities such as Istanbul, Ankara and Izmir which cover approximately 30% of the population of Turkey, and (2) to develop some empirical models for these cities. Experimental data for Izmir were measured in the Solar-Meteorological Station of the Solar Energy Institute at Ege University. The present models were then compared with the 25 models, of which 12 were given in Part I of this study (defined as national models), while the remainder (called international models), presented here, and used for calculating H. The main percentage error (MPE), the root mean square error (RMSE), the main bias error (MBE) and the t-statistic (t-stat) methods were used for evaluating the performance of the models studied. It may be concluded that the present models predict the values of H for the three big cities better than other available models.