Renewable energy-powered water desalination and treatment network under wind power and water demand uncertainty: A possibilistic chance-constrained programming

Abstract

Given the scarcity of freshwater resources, the growing significance of desalination is undeniable. It holds immense potential, particularly in regions grappling with severe water shortages. However, desalination's Achilles heel lies in its voracious energy appetite, requiring roughly ten times more energy than wastewater treatment. Moreover, the prevalent use of fossil fuels in desalination plants poses concerning issues like environmental pollution, fossil fuel depletion, and rising costs. The present study has designed an integrated Water desalination and treatment Network that includes a number of desalination facilities, storage centers, wind farms, and wastewater treatment facilities. The water desalination and treatment network has been structured using a Mixed-Integer Linear Programming (MILP) model, considering uncertainties in wind power and water demand. Employing a chance constraint probabilistic programming approach, this model ensures robustness and balances conservatism with investment attractiveness. It aims to enhance resilience against fluctuations in wind energy and water demand within the water and energy supply chain network. The study applied this model to optimize the locations of desalination plants, treatment centers, and storage facilities. This integrated model ensures autonomy, eliminating the need for external water and energy sources while reliably meeting regional demands. In the context of the Makran coasts case study, our comprehensive mathematical model demonstrates an optimal allocation with 96.67 % attributed to fixed costs and only 3.33 % to variable costs. Moreover, this model precisely optimizes the locations of two desalination centers, two storage facilities, and ten water treatment centers, effectively managing the need for external water resources. Ultimately, through a rigorous sensitivity analysis, we unveiled that the chance constraint parameters have a significant impact on the variable costs.