The Clearness Index Model for Estimation of Global Solar Radiation in Thailand

Abstract

The aim of this paper is to present a novel model to estimate monthly global solar radiation in Thailand. The novel model was investigated based upon the regression technique. Meteorological data sets obtained from the ground-based measurement between 1995-2004 at Khon Kaen, Ubon Ratchathani, Nakhon Sawan, Bangkok and Songkhla were used in modeling. The monthly clearness index and normalized monthly sunshine duration were correlated in order to search the regression coefficients using 5 different models i.e. linear, quadratic, linear-logarithmic, logarithmic, and power equations. The MBE, MPE and RMSE were used as the main criteria for selecting the best model for estimation of global solar radiation. Results showed that, the linear equation and power equation were the best fit for Nakhon Sawan whereas the quadratic equation was the best fit for Ubon Ratchathani and Bangkok, while the linear-logarithmic was the best fit for Khon Kaen and Songkhla. Finally, the quadratic equation performed the best fit for all regions compared to other models, as the RMSE was minimum.