Mapping the Electrodialysis Architecture Design Space by Determining Optimal System Configurations for Different Production Outputs

Abstract

Water scarcity is increasing around the world, and it especially affects remote, resource-constrained communities. Many communities with the highest water stress also live in close proximity to slightly saline water sources, while having abundant solar irradiance. Photovoltaic-powered electrodialysis reversal (PV-EDR) systems have been shown to produce water more cost-effectively and energetically efficiently than other desalination technologies. The goal of this work is to establish a framework for designing and optimizing PV-EDR systems for designers to develop low-cost systems that desalinate brackish water in remote, resource-constrained communities of various sizes around the world. By using this framework, the most cost-effective architecture that produces water across a large range of production volumes at the lowest cost can be identified. To potentially produce water more effectively at larger production volumes using variable power, a new architecture was proposed and explored called hybrid operation that utilizes the benefits from both continuous and batch operation. Additionally, this framework can also be used to identify the most cost-effective strategy for employing batteries and managing the energy stored versus used for desalination. Optimizing EDR systems that minimize the capital cost while maximizing their production volume across the design space including different architectures (batch, continuous, hybrid), energy management strategies (predictive, non-predictive, no batteries), feed salinities (100-500 mg/L), target salinities (1000-4000 mg/L), and recovery ratios (50%-90%) allows us to identify the most cost-effective EDR systems designs across a range of production volumes. By comparing the EDR systems designs across the design space, we can identify when each architecture and energy management strategy could be employed. Below 15 m^3 of water production per day, batch systems should be employed over hybrid systems. If users are not sensitive to salinity changes throughout the day, continuous systems should be used when producing more than 65 m^3 of water per day. Conversely, if users are sensitive to salinity changes, or a large buffer volume like a reservoir or pond is not available, hybrid systems should be used when producing more than 80 m^3 of water production per day. Between these production volume thresholds, the specific target salinity, feed salinity and recovery ratio can be used to inform which architecture to use. Incorporating a battery into a PV-EDR system can lower the capital cost of the system by approximately 12.3% for systems that produce between 10 and 100 m^3 of water per day, while producing the same amount of water as a similar EDR system without a battery.