Bi-objective optimal design of desalination plants considering the uncertainty of renewable energy sources

Abstract

Desalination systems are commonly used in isolated areas to supply drinking water. However, they usually rely on fossil fuels, which emit greenhouse gases. Renewable energy sources can be used to power desalination systems, but they are intermittent. To address this, the power generation or the capacity of energy and/or water storage can be increased, but this increases the cost. In this paper, we present a bi-objective optimization methodology for designing a desalination system that is powered by renewable energy sources. The methodology aims to maximize reliability while minimizing the levelized cost of water (LCOW) by including the inherent uncertainty in renewable energy generation. We applied this methodology in the Guajira Peninsula, a region of Colombia with serious problems of drinking water scarcity, to evaluate photovoltaic (PV), wind, and PV/wind hybrid systems, determining the optimal combination of wind turbines, photovoltaic (PV) modules, desalination unit characteristics, water storage capacity, and batteries. We also introduced a new indicator, the Reliability - LCOW ratio (RLR), to assist in selecting the optimal configuration. Our results indicate that the renewable systems are feasible, achieving reliability levels between 92% and 95% and LCOW values between 0.82-0.84 USD/m3, when compared to fossil fuel configurations. Despite this, the PV-only system exhibited a significantly higher installed capacity when compared to wind and hybrid systems, while the latter two showed relatively similar characteristics. An analysis about the use of the water tank as a natural battery compared with electricity storage proved that the battery bank was more effective for the PV-only system, while the water tank was more suitable in systems where the use of wind power is predominant. Furthermore, an analysis of the direct CO2 emissions of the fossil fuel configurations shows that the renewable solutions reduce emissions by over 150 tons-eq/y.