Optimization using artificial intelligence techniques of an autonomous photovoltaic solar water pumping system: Applications to drilling for the water supply of the Great Green Wall (GMV) in Ferlo (Senegal)

This study aims to find a solution to the water supply problems encountered by the rural world of the Great Green Wall of Ferlo (Senegal). The water supply of these localities is done by drillings running on diesel. The latter consume fuel that is harmful to the planet and these areas are very far from the electricity grid. These populations do not meet their demand for water. In order to provide a positive answer, we propose an optimization by artificial intelligence techniques of an autonomous solar photovoltaic water pumping system applied to these drillings. A multi-objective optimization technique will be used to find a set of optimal solutions. The probability of loss of power supply (LPSP) is the criterion that will evaluate the reliability of the system while the total life cycle cost (TLCC) will be used for the economic evaluation. The objective is to minimize the TLCC while maintaining good satisfaction of the system for a desired LPSP under certain constraints taking into account the real data of the site. The expressions of LPSP and TLCC are transformed into mathematical expressions that are a function of two decision variables (PV power and battery capacity). We will develop two solar cell temperature models in the optimization method and will simulate the operating system in MATLAB. The first model depends on the ambient temperature and the second on the ambient temperature and wind speed to see which one performs better.