Empirical Models for the Estimation of Solar Sky-Diffuse Radiation. A Review and Experimental Analysis

Abstract

Accurate solar radiation data are essential for the development of solar energy application systems. The limited availability of solar radiation data, and especially diffuse irradiance values, makes it vital to develop models to estimate these data. The development of estimation models has been the objective of many studies. This paper presents an extended review of the diffuse ratio (k) vs. clearness index (kt) annual, monthly, daily, and hourly frequency regression models. It is however interesting to note that there is a dearth of such knowledge for diffuse ratio–clearness index regressions that are based on averaged data. Monthly-averaged daily global irradiation data are now easily available from the NASA website for any global location. Using existing models, it is possible to decompose the daily to averaged-hourly global irradiation values. The missing link so far has been hourly averaged diffuse irradiation. This article presents regression equations which could be used to estimate that information. For this purpose, hourly global and diffuse irradiation data was pooled from 19 different locations to obtain three latitude-dependent regression models relating the monthly-averaged hourly diffuse ratio (k) to the clearness index (kt). The results show a high relationship between both variables. These regression equations could be used to estimate the averaged diffuse irradiation values from averaged global irradiation values, which are more easily available.