Sunshine-Based Estimation of Global Solar Radiation on Horizontal Surface at Lake Van Region (Turkey)

Abstract

In this study several sunshine-based regression models have been evaluated to estimate monthly average daily global solar radiation on horizontal surface of Lake Van region in the Eastern Anatolia region in Turkey by using data obtained from seven different meteorological stations. These models are derived from Angstrom-Prescott linear regression model and its derivatives such as quadratic, cubic, logarithmic and exponential. The performance of this regression models were evaluated by comparing the calculated clearness index and the measured clearness index. Several statistical tests were used to control the validation and goodness of the regression models in terms of the coefficient of determination, mean percent error, mean absolute percent error, mean biased error, mean absolute biased error, root mean square error and t-statistic. The results of all the regression models are within acceptable limits according to the statistical tests. However, the best performances are obtained by cubic regression model for Bitlis, Gevas, Hakkari, Mus stations and by quadratic regression model for Malazgirt, Tatvan and Van stations to predict global solar radiation. The spatial distributions of the monthly average daily global solar radiation around the lake Van region were obtained with interpolation of calculated solar radiation data that acquired from best fit models of the stations. The annual average solar energy potential for Lake Van region is obtained between 750 kWh/m(2) and 2485 kWh/m(2) with annual average of 1610 kWh/m(2) (C) 2012 Elsevier Ltd. All rights reserved.