Modeling the Solar Irradiance of Democratic Republing of Congo by Sunshine Duration

Abstract

The electrification rate of the Democratic Republic of Congo (DRC) is very low despite its impressive hydroelectric potential. But, it is hard to imagine that the Congolese national company of electricity could improve this rate in the next five or ten years only by building electrical infrastructure. Building electrical infrastructure requires a big capital that neither this company alone nor the Congolese Government cannot mobilize. So, solar energy seems an alternative to improve this rate. However, sizing and optimization of photovoltaic solar energy systems require knowledge of solar irradiance data related sites. The purpose of this study is to propose general models to predict solar irradiance (in kWh/m2/d) in function of sunshine duration (in h/d) for any province of Democratic Republic of Congo. Then, sunshine duration data and irradiance data related to all the provinces of the Democratic Republic of Congo have been used to establish five statistical models (linear, quadratic, cubic, logarithmic and exponential) to predict solar irradiance in any province of the Democratic Republic of Congo. The models established have been validated by the acceptable values of normalized statistical errors and the satisfaction of the Nash-Sutcliffe criteria (NS) and sufficient value of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathrm{R}^{2}$</tex>, regression coefficient. The results show that by excluding provinces of Nord and Sud Kivu, the cubic model is the best, its value of the regression coefficient is 0.93.