Calibration and Correction of the Beam Solar Radiation Models Using High Temporal Resolution Measurements

Abstract

The conversion of the Direct Normal solar Irradiance (DNI) energy into thermal or electrical energy is performed by concentrating solar power systems. They use the received solar radiation to expose an industrial working fluid to high temperature. A good knowledge of the availability of the DNI in the place of settlements of a solar power plant plays a crucial role in the design of solar collectors and the optimization of operations within the industrial applications. In regions where no solar radiation measurements are available, the modeling of the DNI is commonly adopted to determine the solar radiation potential of the site. However, the DNI models could predict the direct solar radiation with high uncertainty. This paper presents a performance analysis of three DNI clear sky models: Ineichen, Iqbal, and Solis models to predict the Direct Normal Irradiance. The performance analysis conducted in our study consisted of validation and calibration of these models using 10 min resolution measurements performed in a southern Tunisian region (Tataouine). The purpose of our research is to improve the accuracy of the DNI model prediction and to offer more reliable estimations. The results of the sub-hourly statistical validation of these models show that the Iqbal model achieves the lowest uncertainty of prediction. After performing calibration with the in situ measurements, The results show that, the uncertainty of estimation, represented by the relative root mean square error (rmse%), has been generally reduced by 3%. For a high DNI intensity, the lowest errors are recorded by the corrected Solis model (rmse% = 6.2%) while, for a low DNI intensity, it reaches 25.6% which is recorded by the corrected Iqbal model.