Estimation of Global Solar Radiation Using Empirical Models

Abstract

The dearth of solar radiation data availability has necessitated the development of several mathematical models for estimating global solar radiation (GSR) of regions using the readily available meteorological data of the region. This study was centered on estimating the GSR of the Ihiala region in Sub-Saharan Africa using empirical models. For the last ten years, meteorological data from the Nigerian Meteorological Agency (NIMET) were used. The sunshine-based equation, temperature-based equation, and multivariate polynomial equations were the empirical models employed to estimate the GSR of the region. The performance of the seven models was determined using statistical measures. From the results obtained, the seven models had their respective P-values all less than 5 % significant level for a confidence interval of 95 %. Thereby attesting their suitability for GSR estimation of the region is needed. Also, from the other statistical tools employed, the considered multivariate model had better estimation performance than the other models. Therefore, the considered multivariate model is suitable for estimating the GSR of the Ihiala region in Sub-Saharan Africa.

Keywords

• renewable energy
• global solar radiation
• artificial neural network
• statistical tests