Optimization and design of a low-cost, village-scale, photovoltaic-powered, electrodialysis reversal desalination system for rural India
Introduction
This paper presents the parametric theory and system modeling used to design a cost-optimized, constant voltage, and constant pumping power photovoltaic-powered electrodialysis reversal (PV-EDR) desalination system for rural India that can be built from off-the-shelf components. The EDR system was based on GE Water's electrodialysis stack model number AQ3-1-2-50/35 [1], which is a readily available product, and was previously studied, modeled, and tested by Wright and Winter [2]. The system was chosen to operate in batch mode at constant voltage and constant pumping power to accommodate varying levels of input salinity. The costs of off-the-shelf components were estimated by linear cost models, and performance estimates were based on data from OEMs and vendors. These data help establish a baseline for the lowest cost PV-EDR system that meets our desired performance requirements and can be built with readily available parts and materials. This study also investigates the cost sensitivity of PV-EDR systems to determine where future research and R&D efforts should be focused to enable further cost reductions.
The theoretical system developed in this study was designed for the Indian village of Chelluru, which lies 70 km northeast of Hyderabad. Chelluru has a population of approximately 2000 people, putting it in the median Indian village population range of 2000–5000 people [3]. The groundwater salinity is 1600 mg/L, which is within the typical Indian groundwater range of 1000–2000 mg/L. Assuming 3 L of daily water consumption per person per day [4], the median village water requirement in India is 6–15 m3/day. While Chelluru's average daily water demand is 6 m3/day, this study aimed to create a 10 m3/day system in the interest of targeting the most common village size. Finally, a product water salinity of 300 mg/L was selected for satisfactory palatability, which is well below the Bureau of Indian Standards for Drinking Water recommendation of 500 mg/L [5].
Brackish groundwater of salinity at or above the acceptable threshold set by the Bureau of Indian Standards for Drinking Water (500 mg/L) underlies approximately 60% of India's land area (Fig. 1) [6]. This fact makes providing desalinated water to the majority of the country imperative. As of 2015, 96.7% of Indian villages had been electrified, meaning they have some form of grid electricity [8]. However, the grid electricity in most villages is not reliable, nor do many households benefit from it. In 2011, only 55.3% of rural households used electricity for lighting [9], which suggests that not all the households in electrified villages have reliable access to even basic electricity. Even those that do have access to electricity experience intermittent power outages and may only have access for a few hours per day. This makes photovoltaic (PV)-powered desalination systems attractive, especially given that India has a high average daily global horizontal irradiance solar resource of 6 kWh/m2 [10] (Fig. 2).
Tata Projects Limited, a sponsor of this research, has been working to mitigate the lack of access people have to safe drinking water sources. They have installed approximately 2200 reverse osmosis (RO) systems, all of which are grid-connected, in villages across India to desalinate the available water sources to safe drinking levels [11]. However, there is a need for more cost-effective solutions in areas where grid connection is nonexistent or unreliable. Current RO desalination solutions have been rendered cost-prohibitive for off-grid rural applications; off-grid RO systems cost more than double that of an equivalent capacity grid-connected system, at $11,250 compared to $4500. As a result, the local non-governmental organizations (NGOs) or village municipalities that purchase desalination systems are currently limited to grid-powered solutions even in semi-reliable grid electricity environments [11]. If the cost of off-grid desalination systems could be reduced, it would open up a substantial and untapped market of villages without reliable grid connection.
Electrodialysis (ED) requires less than 50% of the specific energy compared to RO to desalinate water below 2000 mg/L to a product water concentration of 350 mg/L [2]. The majority of India's brackish groundwater is below 2000 mg/L [6]. To first order, this implies that the cost of the power subsystem for PV-EDR would be half that of PV-RO. These factors suggest that ED could provide a lower-cost, off-grid brackish water desalination solution compared to RO [2]. In addition to energy savings, ED can achieve high recovery ratios of 80–90 %, compared to only 30–60 % typically achieved by the RO systems used in Indian villages [2]. The authors have observed these systems running with recoveries as low as 15%, even when blending feed and desalinated water. Adoption of ED could lead to less water waste, an important factor given that India's groundwater resources are rapidly being depleted [[12], [13], [14]]. Additionally, while RO membranes have an expected lifetime of 3–5 years, ED membranes have an expected lifetime of 10+ years [2], which could improve maintenance and serviceability.
The use of PV power for ED and EDR systems has been studied in the past. Laboratory-scale work has been completed to model and test a PV-ED system [15]. A number of field pilots have also been conducted. In 1987, Adiga et al. [16] completed a pilot PV-ED project in the Thar Desert, though the product water was 1000 mg/L – too brackish for our application. Additionally, PV power systems and batteries at the time were less efficient and much more expensive than they are now.
In the same year, Kuroda et al. [17] designed and constructed a batch mode PV-ED seawater desalination system in Nagasaki which operated continuously from June 1986 to March 1988 to produce 2–5 m3 of drinking water at 400 mg/L per day. The system was meant to be optimized by matching the power consumption of the ED desalination process with the power generation from the PV panels. In 1992 Soma et al. [18] constructed a similar PV-ED system for brackish water desalination, and monitored the seasonal variation of water production.
Similar to the work presented in this paper, both systems were designed with the motivation to minimize the cost of the PV-ED desalination system. However, the tests were conducted approximately thirty years ago, produced water of a higher salinity than our targets for an Indian village, and listed no concrete cost, power, or energy consumption values for present-day comparison. Additionally, advancements in PV and battery technology have enabled different PV-ED configurations and lower costs than what was previously achievable [19]. These developments warrant a fresh, present-day investigation into cost-optimization and development of PV-EDR desalination systems.
Cost optimization of PV-powered RO desalination systems is a related area of research. Bilton et al. [[20], [21], [22], [23]] investigated the impacts of location-specific environmental and demand parameters on the optimal design of modular PV-RO desalination systems using genetic algorithms. They also worked extensively on examining energy generation methods considering not only PV, but also wind turbines and diesel generators, and optimizing them together with RO systems to determine a high-reliability system configuration with the lowest lifecycle cost. Koutroulis and Kolokotsa [24] also investigated community-scale RO systems; they found for their context that a hybrid system using PV and wind power was lower cost than using a single power source.
The work presented here has similar goals in optimizing for minimal system cost while achieving high reliability of off-grid desalination systems in under-served communities. However, the performance and costs of components such as solar panels and batteries are generalized, rather than picking specific components from an inventory. Furthermore, the optimization analysis is focused on a single location for which we have water and solar irradiance data, and where a village-scale RO system is currently running. The novelty of this work lies in the parametric theory to design the lowest cost, constant voltage and pumping power PV-EDR system that can be built from off-the-shelf parts. The major insights from the study are that co-optimization of the PV and ED subsystems leads to lower cost than a serial optimization of each, and that flexible operation of the system allows for reduced battery costs by overproducing water on sunny days and storing it in tanks. The knowledge presented herein is generalizable to many locations and applications beyond Chelluru, India, and will enable engineers to design cost effective PV-EDR systems for other size scales, salinities, and contexts.
Section snippets
Electrodialysis reversal batch system behavior
In this study, EDR operated in a batch mode configuration is examined. EDR operates identically to ED, with the addition of a periodic reversal of the diluate and concentrate streams as well as the stack voltage polarity. In batch mode, the water in the diluate and brine tanks is recirculated through the ED stack continuously until the desired diluate salinity is reached. After a reversal, the diluate streams flow in the channels where the concentrate streams flowed previously, and vice versa.
PV-EDR system optimization
When the model described in the previous section is coupled to a particle swarm optimization (PSO) [42] algorithm, multiple designs are randomly initialized and then varied and eventually converge on a low-cost design with acceptable reliability. A PV-EDR design was characterized as a combination of the design variables listed in Table 2. Due to the coupled nature of the PV and EDR subsystems, it is nontrivial to determine what configuration of the ED stack, pump models, and quantities of PV
Comparison to PV-EDR system designed using conventional methods
For comparison, let the optimized design found in the previous section be referred to as Design A, in which the optimization and design of the PV and EDR subsystems were performed jointly, resulting in the design of a co-optimized system. On the other hand, let Design B be a PV-EDR system designed using the conventional method of designing the load – in this case, the ED desalination system – and then the power system sequentially.
Design A has been detailed in the Section 3. For Design B, the
Conclusions
In this study, the lowest capital cost village-scale, photovoltaic-powered electrodialysis desalination system for rural India was identified based on current component prices and performance. This was achieved through investigation of the parametric relationships that govern the characteristics of the electrodialysis process and the photovoltaic power system, and creation of a model to predict a PV-EDR system's performance. The model developed is generalized to take inputs of local conditions
Nomenclature
- αp
temperature coefficient
- A
active ED membrane area
- APV
area of the photovoltaic panel array
concentration of concentrate stream at the inlet of the ED stack
- Cconc
concentration of the concentrate stream at the outlet of the ED stack
concentration on the surface of the anion exchange membrane at the boundary of the concentrate stream
concentration on the surface of the cation exchange membrane at the boundary of the concentrate stream
concentration of the diluate stream at the
Acknowledgments
This work was sponsored by Tata Projects Limited, the United States Agency for International Development (contract number AID-OAA-C-14-00185), the United States Bureau of Reclamation (agreement number R16AC00122), the MIT Energy Initiative, and the Tata Center for Technology and Design at MIT.
References (47)
- et al.
Justification for community-scale photovoltaic powered electrodialysis desalination systems for inland rural villages in India
Desalination
(2014) - et al.
Photovoltaic electrodialysis system for brackish water desalination: modeling of global process
J. Membr. Sci.
(Apr. 2006) - et al.
Performance analysis of photovoltaic electrodialysis desalination plant at Tanote in Thar Desert, Desalination
Desalination
(1987) - et al.
An electrodialysis sea water desalination system powered by photovoltaic cells
Desalination
(1987) - et al.
Design optimization of desalination systems power-supplied by PV and W/G energy sources
Desalination
(2010) - et al.
A Robust Model of Brackish Water Electrodialysis Desalination with Experimental Comparison at Different Size Scales
Desalination
(2018) - et al.
Brackish water desalination by electrodialysis: batch recirculation operation modeling
J. Membr. Sci.
(2005) Electrodialysis, a mature technology with a multitude of new applications
Desalination
(2010)- et al.
Advantages of Operation Flexibility and Load Sizing for PV-Powered System Design
Solar Energy
(2018) GE Aquamite EDR Systems Fact Sheet
Retrieved from the GE Water & Process Technologies Document Library website
(1998)
Basic water requirements for human activities: meeting basic needs
Water Int.
Drinking Water - Specification (Second Revision IS 10500)
Ground water quality in shallow aquifers of India
India Solar Resource
Census of India 2011 - household amenities and assets: source of lighting
Technical Report, Government of India
India SolarResource - Global Horizontal Irradiance
Personal Conversation with Nataraja C. B.
Impending Water Crisis in India and Comparing Clean Water Standards among Developing and Developed Nations
ResearchReport 39: Water Scarcity and the Role of Storage in Development
Groundwater storage and depletion trends in Tamil Nadu State, India
Water Resour. Manag.
The practical operation of photovoltaic desalination systems
Int. J. Sol. Energy
Cited by (39)
Improving efficiency and discharge power of acid-base flow battery via a bi-objective optimisation
2023, Journal of Energy StorageDesalinating RO retentate employing NF coupled with CDI: A path towards cleaner production
2023, Journal of Cleaner ProductionRenewable energy sources utilized for membrane desalination processes
2023, Green Membrane Technologies towards Environmental Sustainability