Multi-objective optimization of a renewable power supply system with underwater compressed air energy storage for seawater reverse osmosis under two different operation schemes
Introduction
Water is essential for each living being on our planet. Although nearly 71% Earth surface is covered by water, about 97% in it is salty water and only 3% is drinkable fresh water. The fresh water shortage is a significant threat to safety of the human society [1]. Nowadays, due to the rapid growth of global population and accelerated development in industrial and agricultural sectors, the water scarcity issue is more and more serious [2]. Thus, to seek a feasible solution to solve this problem is urgent and imperative globally.
Desalination is considered as an effective water treatment process to produce fresh water from the saline water via removing the dissolved salts and other minerals. From the perspective of work principle, the desalination technology can be classified into two main categories: the phase-based thermal method and the pressure difference based membrane method [3]. The phase-based thermal method uses external heat to drive a phase change process, including the humidification dehumidification (HDH), the multi-stage flash (MSF), the multi-effect distillation (MED) and the thermal/mechanical vapor compression (TVC/MVC). On the other hand, the pressure difference based membrane method usually operates in a process to produce pressure difference through electricity supply, containing the reverse/forward osmosis (RO/FO), the capacitive deionization (CDI) and the electro-dialysis reversal (EDR) [4].
In fact, the desalination technology is a typical energy extensive process whether the membrane method or the thermal method [5]. Initially, the desalination plant depends on the energy from combusting fossil fuels severely. However, this type of power source can result in environmental pollution eventually. Together with the depletion of the traditional fossil fuel resources, extensive research efforts have been carried out in progress to exploit alternative energy resources for desalination in current years, such as the nuclear energy [6], wind energy [7], solar energy [8], geothermal energy [9], ocean energy [10] and their combination [11]. Undoubtedly, the renewable types of energy sources listed previously can support to achieve the sustainable goals. Besides these renewable energy types, the hydropower is another good choice. Although the large-scale hydropower plant is the main utilization form of hydropower, and lots of huge hydropower plants have been built widely around the world, the relatively small-scale run-of-river hydropower system is also attractive [12]. Moreover, in some arid regions, the PV panels can plant over irrigation canals to reduce the water evaporation rate and also to produce extra electricity via PV [13].
Among the previous desalination methods mentioned above, the RO method is considered as the most common technology, which occupies more than 65% of the global desalination installed capacity [6]. Commonly, the RO process is a typical membrane method which adopts a series of semipermeable membranes to separate the multiple mediums with different solute concentration with the help of pressure difference. The main load in a RO plant is the high pressure pump (HPP). In spite of the application of energy recovery device (ERD) is widely existed in modern RO plant for energy saving, the HPP still accounts for the tremendous majority of the total plant power [14].
In the journey for building a sustainable society, thanks to the benefits of carbon footprint reduction and operation cost saving, the renewable driven RO system draws the most attentions in scientific and industrial communities nowadays, especially in the rural and remote islands or the coastal areas. Usually, the renewable power supply system can be designed either in the grid-connected mode or in the off-grid mode which depends on the utility grid coverage scope. In spite of the renewable energies often rely on the unforeseen meteorological condition, some types of renewable energies still have complementary characteristic between them in nature, and thus the combination of multiple energy sources is the main application form [15]. The most favorable candidate is the integration of wind and solar. Moreover, due to the intermittent nature of renewables, the energy storage device or other supplement power source must be collaborated in the renewable driven RO system to enhance the system reliability.
For a certain location, it is a complex job to implement a renewable powered RO plant sizing process, and a good trade-off among reliability, economic and environmental aspects should be considered. In this sense, the multi-objective optimization (MOO) is indispensable in the system design process, which has the ability to optimize multiple objectives simultaneously. Elmaadawy et al. [16] conducted an optimal sizing and techno-enviro-economic feasibility assessment of a large-scale off-grid RO plant powered by hybrid renewable energy sources. They found that the photovoltaic (PV)/wind/diesel with battery storage has an excellent performance. Zhou et al. [17] carried out a multi-objective optimal operation of a coastal hydro-electrical energy system with seawater RO based on the third generation of the constrained nondominated sorting genetic algorithm (NSGA-III) algorithm. This proposed grid-connected system used the pumped storage plant as the energy buffer and the optimization objectives were the minimization of the total period cost and the tie-line power fluctuations. Campana et al. [18] implemented a 100% renewable wastewater treatment plants with a techno-economic assessment. This grid-connected system used the multi-energy storage technology based on battery and hydrogen to absorb the surplus renewable power. The optimization targets were minimizing the net present cost and maximizing the renewable share. In addition, a summary of the optimization of the renewable energy driven RO desalination system can be found in literature [19].
As mentioned above, it can be recognized that the wind/PV combination is a promising power source to driven the RO plant, and the energy storage device in this system is still indispensable. Meanwhile, due to the uncertainty on both the renewable power generation side and the RO power consumption side, the MOO for this type of PV/wind/storage system is also required for the system sizing and operation. From the related literatures survey above, the energy storage devices involved in this type of system include the battery storage, the hydrogen storage and the pumped hydro storage. However, although some researches have selected the compressed air energy storage (CAES) as the energy buffer [14], the corresponding multi-objective optimization issue has not been investigated at all. Especially, because the RO system is established in islands or coastal areas, the underwater CAES (UW-CAES) system is more suitable. The MOO for the related RO power suppled system with PV/wind/UW-CAES configuration is lack of research in this area.
To fill this gap, the main objective of this paper is to find an optimal size/configuration and the related energy management strategy for a renewable power supply system of a seawater RO plant under two different operation schemes, by considering the trade-off among reliability, economic and environmental aspects. Firstly, the proposal of a renewable power supply system for seawater RO under off-grid and grid-connected schemes is presented and the multi-dimension system models are built. Then, based on the RO system load requirement and the meteorological condition, a multi-objective optimization problem is given to seek the optimal system size/configuration and the optimal energy management strategy. This MOO issue is solved by the multi-objective particle swarm optimization (MOPSO) algorithm, and the optimal solution is selected by the Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS) method. Finally, a case study in China is carried out to verify this method.
The contributes of the current work can be summarized as follows.
- (1)
At the energy supply side, the power source of this renewable power supply system is the PV/wind combination owing to their spatio-temporal complementary characteristics, aiming at reducing the PV and wind capacities. Besides, due to the uncertainty of renewable sources, the diesel engine generator in off-grid scheme and the main grid power in grid-connected scheme are served as the supplement power sources.
- (2)
At the system level, the UW-CAES device is adopted as the energy buffer. In which the flexible energy bag has a fixed pressure during whole operation, implying a constant pressure charging and discharging processes. Thereby, the related compressor and turbine have a high efficiency and a simple structure.
- (3)
As for the system operation aspect, the grid-connected scheme and the off-grid scheme are involved and the performances are compared. This can provide a reference for seawater RO system design for different locations.
- (4)
At the algorithmic level, a bi-level method by combination of MOPSO algorithm and TOPSIS method is used to solve the MOO problem, and thus determine the optimal capacity and configuration of system components. In detail, the MOPSO generates a set of alternative optimization results (Pareto front set) and the TOPSIS selects the optimal solution.
Section snippets
System structure
Fig. 1 gives the architectures of the proposed renewable seawater RO system under two different operation schemes, i.e. the grid-connected scheme and the off-grid scheme. In both structures, three main parts are involved, which are the power supply section, the RO section and the energy storage section.
The power supply section is settled to produce electricity to the system load, mainly from the wind and the sunlight in nature. In such system, a wind energy conversion system (WECS) with several
System modelling
Some assumptions are made to simplify the performance investigation as follows:
- (1)
The air is treated as ideal gas, which is composed of 76% nitrogen and 24% oxygen in mass fraction.
- (2)
Heat dissipation and pressure loss in pipes and heat exchangers are not considered.
- (3)
The hot tank is assumed to be well insulated and no heat loss existed.
- (4)
The isentropic efficiencies of compressor and turbine in UW-CAES system are fixed.
The energy aspect
In current paper, two performance criteria are employed to assess the system power/energy supply reliability, which are listed as follows.
The loss of power supply probability (LPSP) is used to evaluate the power supply reliability. In charge mode, the mismatch power never exists. Thus, the LPSP can be determined by:where denotes the discharge time, is the supplementary power source, which is the DEG power in
Natural resources condition and load demand requirement
A long-term renewable power generation series is essential to assess the performance of any renewable driven system. In general, the renewable power is dominated by the meteorological condition. Thus, according to the components’ models mentioned above, the ambient temperature, the wind speed and the solar irradiance are needed.
The Dawanshan island in Guangdong province of China is chosen as the location for case study. This island locates at latitude 21.947 N, longitude 113.733 E, which is in
Discussion
From the simulation and the case study above, the following discussion remarks and method limitation can be concluded.
- -
The employed bi-level multi-objective optimization algorithm is widely recognized as a superior performance method and have been applied in many areas [13,27,33]. The simulation of this paper is credible for decision makers.
- -
For the selected site, the installed capacity of PV and WT is larger than the load requirement due to the intermittent nature. Although the energy storage
Conclusion
On the path to constructing a sustainable society, the renewable driven power supply system is a flourishing development trend in current energy sector. In this paper, a multi-objective optimization of a renewable power supply system with the support of UW-CAES for RO plant in both grid-connected scheme and off-grid scheme is implemented. This optimization problem is solved by the MOPSO and the TOPSIS method. The total energy requirement of this given RO plant in a year is 178.53 MWh according
CRediT authorship contribution statement
Pan Zhao: Conceptualization, Methodology, Software, Formal analysis, Investigation, Writing – original draft, Writing – review & editing. Feifei Gou: Visualization. Wenpan Xu: Visualization. Jiangfeng Wang: Supervision. Yiping Dai: Supervision.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgement
The authors gratefully acknowledge the financial support by National Natural Science Foundation of China (Grant No.51876152).
References (34)
Integration of hybrid power (wind-photovoltaic-diesel-battery) and seawater reverse osmosis systems for small-scale desalination applications
Desalination
(2018)- et al.
Performance analysis of a stand-alone hybrid energy system for desalination unit in Egypt
Energy Convers. Manag.
(2020) - et al.
Energy for desalination: a state-of-the-art review
Desalination
(2020) - et al.
Design and economic evaluation of solar-powered hybrid multi effect and reverse osmosis system for seawater desalination
Desalination
(2019) - et al.
Nuclear desalination: a state-of-the-art review
Desalination
(2019) - et al.
Wind-driven SWRO desalination prototype with and without batteries: a performance simulation using machine learning models
Desalination
(2018) - et al.
Solar powered desalination – technology, energy and future outlook
Desalination
(2019) - et al.
Techno-economic analysis of MED and RO desalination powered by low-enthalpy geothermal energy
Desalination
(2015) Technical feasibility of wavepower for seawater desalination using the hydro-ram (Hydram)
Desalination
(2003)- et al.
Sustainable and reliable design of reverse osmosis desalination with hybrid renewable energy systems through supply chain forecasting using recurrent neural networks
Energy
(2019)