Assessing the Performance of Global Solar Radiation Empirical Models at a Sahelian Site, Sokoto, Nigeria

Abstract

In the region where solar radiation data are scarce, the next alternative method is to use solar radiation models to estimate the data needed for some applications such as simulation of crop performance and the design of solar energy conversion devices. In this paper, the validations of fifteen models for estimating monthly mean daily global solar radiation on the horizontal surfaces were conducted at a location in salielian region where there is great potential for solar energy utilization in electricity generation and irrigation. Evaluations of these models were carried out by using the Mean Bias Error (MBE), Root Mean Square Error (RMSE), Mean Percentage Error (MPE) and t-test. Three of these statistical performance indicators were combined to generate rank score for each model. Temperature based models made the rank of the best three during the wet season. On the annual scale and in the dry season, temperature variations and humidity were combined with sunshine duration to get the best rank of I to 3.