Abstract

This paper proposes a method to evaluate solar energy on flat ground for different state of sky (clear sky, moderately covered sky and covered sky). Four semi-empirical models (Link Cloudiness Factor (LCF), Perrin-Brichambaut (PB), Ghouard (GD), Bird Hulstrom (BH)) have been considered to predict solar radiation. An influence analysis of the seasonal variation on the quality of solar energy potential was carried out. In this study, two seasons (dry seasons and rainy season) were identified for the site of Nouakchott. Results of the study showed three cases: The first one is the clear sky, which includes two cases one was during the rainy season, where the model of the LCF was the best with Normalized Mean Square Error NMSE (0.073) and Correlation Coefficient R (0.975), and the other was during the dry season, where the model of BH was most suitable with NMSE (0.069) and R (0.973). The second case is the moderately covered sky, where the GD model was the best model during the dry season whereas BH showed the best results with NMSE (0.242) and R (0.911) during the rainy season. The third case is the covered sky, where GD model with NMSE (0.880) and R (0.831) in the dry season and PB model with NMSE (0.615) and R (0.831) in the rainy season were the most adapted ones.

Keywords: Solar energy, state of sky, prediction, solar radiation, clearness index.