Multi-Criteria Optimization of Solar Photovoltaic-Based Water Lifting Systems

Abstract

In many fast-growing cities, the increasing cost of electricity has turned water pumping into a serious financial burden, especially for residents of multi-storey apartment buildings. At present, this problem is often addressed through the use of numerous small, grid-connected pumps installed independently by households, a solution that is neither energy-efficient nor economically sustainable. This study explores the use of a solar photovoltaic water pumping system as a practical alternative to such decentralized pumping schemes. A nonlinear mathematical model was developed to represent real operating conditions in Jizzakh, Uzbekistan, taking into account local solar irradiation, average daily water demand per household, and regional electricity tariffs. Since the system design involves several competing objectives — reducing energy consumption, limiting installation costs, and minimizing the required rooftop area — a multi-objective optimization approach based on the Non-dominated Sorting Genetic Algorithm II (NSGA-II) was applied. The model was implemented in Python using the DEAP library and validated using real operational data. The results indicate that the proposed solar photovoltaic pumping system can substantially reduce both energy consumption and annual operating costs when compared with conventional grid-based solutions. Overall, the findings suggest that the proposed approach can be used as a practical decision-support tool for engineers, urban planners, and policymakers aiming to improve the sustainability of residential water supply systems, particularly in regions with high solar potential and limited resources, such as Central Asia.