Optimal design and operation of an islanded water-energy network including a combined electrodialysis-reverse osmosis desalination unit

Highlights

• An optimization model for a combined water-energy system is proposed.

• Water system involves a hybrid electrodialysis-reverse osmosis desalination plant.

• Effect of water storage tank on energy reduction is studied.

• Effect of pump’s speed on resource allocation of water-energy system is studied.

• Operating water system with renewable energy and battery storage is optimized.

Abstract

This paper proposes an optimization framework to simultaneously optimize the saline and freshwater water sources as well as decentralized renewable and conventional energy sources, which consequently results in an economic and efficient water-energy system operation. The inextricable links between water and energy systems are taken into considerations, and a mixed integer nonlinear programming formulation is used to solve the proposed optimization model. A hybrid electrodialysis-reverse osmosis is integrated in the water distribution system and its electricity consumption is minimized. Also, the day-ahead economic dispatch problem of the energy system incorporates the random behavior of solar and wind generation units, along with charging and discharging modes of battery storage systems. The outcomes of the proposed models highlight: 1) integrating the hybrid desalination module in the water-energy system will reduce peak demands by more than 30% if the water and energy systems are solely supplied by saline water and renewable energy sources, respectively; 2) presence of an extra water storage tank also decreases peak demands by 13%. The impact of key elements of the combined system such as pumps’ speeds, water tank constraints, and presence of renewable sources on the output cost, generated energy, and hydraulic parameters is investigated through case studies.

Introduction

Scarcity of freshwater is a globally recognized issue. According to the United Nations [1], the water usage has increased twice the population’s growth over the last century, causing an escalation in the number of regions that have already reached their capacity to sustainably deliver potable water. As the rate of urbanization and climate change increases, this shortage will be exacerbated, especially in more deserted areas facing drought. Hence, water distribution systems (WDSs) in the world have no choice but looking for alternative sources of water, such as seawater, groundwater, and rainwater. On the other hand, substituting freshwater sources with saline or brackish water would require additional processes which are often energy intensive. Ironically, these social and environmental parameters that are expediting the water shortage in the world are also contributing to the energy sources depletion [2]. It is estimated that with the current global rate of energy consumption, the demand for energy in the year 2035 will be 40% higher than what it was in 2010 [3]. Additionally, water and energy systems are interlinked, and thus shortage of one source will cause dearth of the other. Water is required in energy production processes, while energy is utilized to pump and transport water in the water distribution system, as well as in collecting and treating wastewater to be released to the environment.

Because of the increasing scarcity of water and energy sources and the fact that water and energy systems are highly interdependent, it is crucial to understand how (1) sustainable sources can be integrated in each system and (2) how the two systems can collectively be operated within one optimized framework developed for the combined system. As such, the main motivation behind this work is to develop a mathematical model that can optimize the performance of water-energy systems with desalination modules, renewable sources, conventional generation, water distribution system, and battery energy storage systems to minimize the energy consumption of the nexus.

Desalination in general refers to a process that removes dissolved salts (i.e., sodium chloride in this paper) from saline or brackish water, with the purpose of producing freshwater used for drinking, irrigation, or industrial usage. There are various methods currently used for desalination such as capacitive deionization [4], reverse osmosis (RO) [5], solar thermal desalination [6], and electrodialysis (ED) [7]. Integrating ED and RO systems can create a more efficient system than each stated individual systems by offering unique advantages being high permeate recovery and the capacity to scaling up [8]. However, both methods are considered to be high energy intensive. Thereby, a hybrid ED-RO system is selected to be the design of the desalination model in this study, with the goal of minimizing its electricity consumption.

Several co-optimization models of water and energy systems exist in the literature, in which the costs, energy consumption, or load demands of the water system were minimized [[9], [10], [11], [12]]. However, desalination processes were out of the scope of all models, and freshwater was considered to be the only source of water. Energy minimization is vital to economic feasibility of a desalination plant. One of the greatest challenges of desalination processes is their intensive electrical power demands. For instance, the electricity usage of a reverse osmosis plant changes within the range of 4–7 kW h/m³. This means that the energy required to run an RO plant supplying water to 48,000 households is equivalent to providing electricity to 10,300 same size households [13]. Such high energy demand of desalination processes is often an economic burden to the WDS, as more than 11% and 44% of brackish and seawater desalination process cost is attributed to electricity, respectively [14].

The optimization of desalination processes has been studied from different aspects in the existing literature. For example, the RO desalination plants were optimized in terms of increasing the removal efficiency [15], minimizing the membrane’s fouling and saturation [16,17], increasing the capacity of the permeate [18], and enhancing the configuration and operation [19]. For electrodialysis systems, prior research was mostly focused on design, testing, and experimental studies. Factors such as voltage, flow rate, membrane type, and feed water source were tested to obtain the optimum operational conditions [[20], [21], [22]]. Furthermore, the optimization models present in the literature are associated with standalone desalination plants [23]. For instance, the cost of a photovoltaic-powered electrodialysis reversal (PV-EDR) system [24,25], a small PV-RO system [26], and a small PV-wind-RO system [27] were minimized to achieve reduced amount of energy and water consumption in the desalination plants. Also, a mixed integer nonlinear programming (MINLP) was developed to minimize the energy of an RO desalination system to be implemented in Atacama desert, using energy recovery devices [28]. The cost function of the RO system coupling with an energy production plant developed based on a direct osmosis was minimized using an optimization problem [29]. A multi-stage RO system was optimized to attain optimal water recovery, net specific energy consumption and osmotic pressure [30]. Nonetheless, none of the above studies considered integration of water and energy systems, and they mainly focused on standalone water desalination units.

There are a few research studies focused on developing optimization models for integrated water-energy systems including desalination modules. For instance, the cost function of a combined water and energy plant was optimized with the objectives of attaining economic benefits and lower CO₂ emissions, in which the total cost of the system was minimized [31]. Since the main scope was on minimizing the total cost of the system, energy consumption of units and resource allocation were not included in the model. As such, the formulation in [31] resulted in oversimplified equations in both water and energy systems, thus detailed equations associated with water tank constraints, pump’s head equations, pump’s status and speed, hydraulic equations, specific parameters of desalination design and its constraints were disregarded. An optimization model was proposed to optimize a water distribution system including an RO desalination unit with a conventional thermal power plant-based energy system in [32]. The formulation, however, only considered conventional generation units and ignored the importance of renewable energy sources and storage units in future of water-energy systems. In addition, the water system model only included an RO system in the distribution system to remove salinity from seawater, whereas RO systems alone are neither suitable nor sustainable to be used as the front-line and sole desalination system for high saline waters like seawater [33].

While the prior research has added insightful knowledge to the design and operation of desalination processes to substitute and supplement freshwater sources, the optimization of a water distribution system incorporated with desalination units, reservoirs, water tanks, operated with an energy generation unit composed of renewable sources, conventional generation, and storage units has been out of the scope of all the studies. To address the existing limitations, a novel framework is developed in this study to synchronously optimize the operation of a combined water distribution-energy generation system such that.

• The energy usage of the water distribution network including a hybrid ED-RO desalination unit, reservoir, and storage tanks with variable-speed pumps is minimized on a daily basis,

• The day-ahead economic dispatch of a water-energy system including conventional and renewable energy sources, as well as battery storage units is formulated and fluctuations and peak demands are eliminated,

• The feasibility of operating the WDS supported by an “off-grid” energy generation system with only renewable energy and battery storage systems is studied,

• A flexible desalination model is optimized to be able to be operated in combination with reservoir or alone as the sole water supplier, and to be compatible with any salinity level in the feed stream,

• The effect of increasing water storage on the energy consumption of the studied WDS is investigated,

• The effect of pump’s speed on resource allocation of water-energy system is studied through case studies.

The rest of the paper is organized in the following order. Section 2 explains the problem formulation, section 3 discuses the case studies and the outcomes of the research, and section 4 concludes the paper.