Watts and Drops: Co-Scheduling Power and Water in Desalination Plants
Ahmed S. Alahmed, Audun Botterud, Saurabh Amin, Ali T. Al-Awami · 2025
Read Watts and Drops: Co-Scheduling Power and Water in Desalination Plants by Ahmed S. Alahmed, Audun Botterud, Saurabh Amin, Ali T. Al-Awami (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-19243.txt.
Introduction
Ahmed S. Alahmed and three co-authors schedule water and power together inside a desalination plant that also generates renewable electricity. The plant mixes thermal and membrane desalination, sells fresh water to a utility at a set price, and trades electricity with the grid in both directions. The question is how to run both technologies hour by hour to maximize profit as sunshine and prices vary.
The optimal policy, they show, has a simple threshold shape set by the plant's own renewable output. As renewable generation rises, thermal water output should shrink steadily while membrane output grows. The authors explain the intuition behind that split and examine special properties of the policy. Co-scheduling the two resources, rather than planning water and power apart, is the paper's central message for plant operators.
Catalog record
| CREATOR | Ahmed S. Alahmed, Audun Botterud, Saurabh Amin, Ali T. Al-Awami |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 22 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2509-19243
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Watts and Drops: Co-Scheduling Power and Water in Desalination Plants Ahmed S. Alahmed, Audun Botterud, Saurabh Amin, Ali T. Al-Awami Abstract We develop a mathematical framework to jointly schedule water and electricity in a profit-maximizing, renewable-colocated water desalination plant that integrates both thermal- and membrane-based technologies. The price-taking desalination plant sells desalinated water to a water utility at a given price and engages in bi-directional electricity transactions with the grid, purchasing or selling power based on its net electricity demand. We show that the optimal scheduling policy depends on the plant’s internal renewable generation and follows a simple threshold structure. Under the optimal policy, thermal-based water output decreases monotonically with renewable ouput, while membrane-based water output increases monotonically. We characterize the structure and intuition behind the threshold policy and examine key special properties. I Introduction The water-energy nexus has garnered growing research attention. The interdependence is bidirectional: electricity powers water extraction, treatment, and desalination, while water is essential for power generation, especially in thermal plants. This mutual reliance has prompted the development of integrated models to optimize resource use and support cross-sectoral planning. Thermal desalination plants (TDPs), commonly used in energy-rich regions, produce freshwater as a primary product while producing electric power as a byproduct. These cogeneration plants benefit from shared heat and power flows and result in tightly coupled operational dynamics. Notably, [ 1 , 2 ] developed dynamic physics-based models for integrated thermal power and thermal water desalination systems, highlighting the importance of coordinated management of thermal power plants and thermal desalination plants. In contrast, reverse osmosis (RO) has recently become the leading desalination technology due to its modularity and improved energy efficiency. In fact, according to some reports, RO now accounts for approximately 69% of the installed desalination capacity globally, with most new projects deploying RO desalination units [ 3 ] . RO desalination plants (RODPs) primarily draw electricity from the grid, positioning them as large, flexible loads within the power system. Recent studies in [ 4 , 5 , 6 ] have explored coupling RODPs with renewable energy sources using hybrid optimization frameworks to minimize costs and emissions. However, these studies treat the power system as an external input and overlook the bidirectional interactions between water and energy. This gap is addressed in [ 7 , 8 , 9 ] by modeling co-optimized water-energy microgrids and coordinated RO-power scheduling platforms, considering dynamic demand, grid constraints, and renewable variability. Their research also explores the use of RODPs for demand-side grid support, including demand response and frequency regulation. However, the existing literature has not addressed the interaction between thermal and RO desalination units when they co-exist within the same territory in the water-energy nexus context. This paper develops a mathematical framework for the joint scheduling of water and power in a profit-maximizing water desalination plant (WDP) equipped with thermal and membrane-based technologies, and local renewable generation. The WDP is modeled as a price-taking agent that engages in water and electricity transactions with a water utility and an electric utility. A key challenge lies in utilizing the complementarities between the two different desalination technologies, especially when coupled with the variability of the renewable energy supply. Our main contribution is the characterization of WDP’s optimal production plan when colocated with renewables. We show that the optimal policy adopts a simple two-threshold structure: when renewable generation is low, the WDP prioritizes thermal-based desalination, and as renewable availability increases, it shifts progressively toward membrane-based production. This structural result provides not only a computationally tractable policy but also valuable operational insights. Specifically, we prove that the thermal-based water output decreases monotonically with renewable availability, while membrane-based output increases accordingly. In addition to the general solution structure, we analyze several special cases to illustrate the practical implications of the proposed policy and to highlight conditions under which either technology is preferred. Our findings have broader relevance for the optimal dispatch of hybrid energy-water systems, particularly in smart grid environments where flexibility and responsiveness are increasingly rewarded. II WDP Framework and Decision Problem We consider a hybrid WDP that jointly optimizes water and electricity schedules to maximize profit, defined as the revenue from selling water and electricity minus operational costs, including electricity procurement from the grid (Fig. 1 ). The WDP utilizes thermal and membrane-based technologies, in addition to renewable energy sources. 1 1 1 Although membrane-based technologies are improving rapidly, thermal desalination remains more robust under high-salinity and harsh climatic conditions, while also offering better grid support capabilities. On one hand, the WDP unidirectionally interacts with the water utility by selling water at a price that is known apriori and may be temporally varying. On the other hand, the WDP bidirectionally transacts with the electricity utility by buying and selling electricity at possibly asymmetric prices of buying and selling based on its net consumption. In both markets, water and electricity, the WDP operates as a price-taker . The WDP electric grid net metering arrangement here is similar to that of energy communities and distributed energy resources aggregation schemes under regulated utilities [ 10 , 11 ] . Fig. 1: Hybrid WDP framework with TDP, RODP, and renewables. The water outputs of the thermal and membrane units are denoted by w_{h},w_{r}\in\mathbb{R}_{+} , respectively. The electricity output of the TDP, RODP, and renewables are denoted by q_{h},q_{r},g\in\mathbb{R}_{+} , respectively. The net electricity consumption of the WDP is denoted by z\in\mathbb{R} . The amount of fuel feeding the TDP is denoted by p_{h}\in\mathbb{R}_{+} . For the remainder of this section, we present the modeling of the TDP, RODP, payment functions of the water and electricity utilities, and the WDP co-optimization problem. II-A TDP and RODP Models II-A 1 TDP The TDP uses a boiler to heat seawater using a fuel source, producing steam that is then condensed to obtain fresh water. The waste heat is used to power a turbine to produce electricity. The TDP, therefore, produces both water and electricity, which we denote by w_{h}\in\mathbb{R}_{+} and q_{h}\in\mathbb{R}_{+} , respectively. The water generated by the TDP is a function of burned fuel w_{h}=f_{h}(p_{h})=\alpha_{h}p_{h} , where \alpha_{h}\geq 0 is a thermal conversion factor. Let f^{\prime}_{h}(p_{h}) be the derivative of f_{h}(p_{h}) with respect to p_{h} . The TDP produces electricity as a by-product of water desalination, governed by the relation w_{h}=\eta_{h}q_{h} , where \eta_{h}>0 denotes the water-to-electricity production ratio. We further define the thermal conversion factor as \beta_{h}:=\alpha_{h}/\eta_{h}\geq 0 , representing the ratio of fuel energy to electricity output. The TDP water flowrate is bounded by the maximum \overline{w}_{h}\in\mathbb{R}_{+} and minimum \underline{w}_{h}\in\mathbb{R}_{+} operational limits that ensure stable operation, i.e., w_{h}\in[\underline{w}_{h},\overline{w}_{h}] . The TDP’s operating cost function is denoted by C_{h}(p_{h})\in\mathbb{R} . We assume the cost function C_{h}(\cdot) to be strictly convex, continuously differentiable, and non-decreasing. II-A 2 RODP The RODP consumes electricity to force pressure seawater through a semi-permeable membrane under high pressure, effectively filtering out salt and impurities. The water production and electricity consumption of the RODP are denoted by w_{r},q_{r}\in\mathbb{R}_{+} , respectively. The amount of water produced by the RODP w_{r} is a linear function of the electricity consumed q_{r} by that module, i.e., w_{r}=f_{r}(q_{r})=\alpha_{r}q_{r} , where \alpha_{r}>0 is the RODP’s conversion factor. We use f_{r}^{\prime} to denote the derivative of f_{r} with respect to q_{r} .