Advanced exergy analysis of a Joule-Brayton pumped thermal electricity storage system with liquid-phase storage

Highlights

• Advanced exergy analysis of pumped thermal electricity storage systems is proposed.

• The expander has the maximum exergy destruction rate in the recuperated system.

• The largest avoidable exergy destruction arises in the recuperator and expander.

• A quantitative analysis of system performance improvement potential is performed.

• The modified overall exergetic efficiency increases substantially from 37% to 57%.

Abstract

Pumped thermal electricity storage is a thermo-mechanical energy storage technology that has emerged as a promising option for large-scale (grid) storage because of its lack of geographical restrictions and relatively low capital costs. This paper focuses on a 10 MW Joule-Brayton pumped thermal electricity storage system with liquid thermal stores and performs detailed conventional and advanced exergy analyses of this system. Results of the conventional exergy analysis on the recuperated system indicate that the expander during discharge is associated with the maximum exergy destruction rate (13%). The advanced exergy analysis further reveals that, amongst the system components studied, the cold heat exchanger during discharge is associated with the highest share (95%) of the avoidable exergy destruction rate, while during charge the same component is associated with the highest share (64%) of the endogenous exergy destruction rate. Thus, the cold heat exchanger offers the largest potential for improvement in the overall system exergetic efficiency. A quantitative analysis of the overall system performance improvement potential of the recuperated system demonstrates that increasing the isentropic efficiency of the compressor and turbine from 85% to 95% significantly increases the modified overall exergetic efficiency from 37% to 57%. Similarly, by increasing the effectiveness and decreasing the pressure loss factor of all heat exchangers, from 0.90 to 0.98 and from 2.5% to 0.5% respectively, the modified overall exergetic efficiency increases from 34% to 54%. The results of exergy analyses provide novel insight into the innovation, research and development of pumped thermal electricity storage technology.

Introduction

The world is in the transition towards enlarging the exploitation of renewable energy sources (RESs) to tackle the severe environmental problems caused by the utilisation of traditional fossil fuels [1]. However, the high penetration of RESs poses a threat to electric power systems [2], which is attributed to their nature of intermittence, inherent instability and low predictability [3]. To cope with these challenges, operational flexibility of power systems, which is defined as the technical capability of a power unit to modulate electrical power inputs to the grid and/or power outputs from the grid over time [4], has attracted great interest from engineers and researchers worldwide, such as those in France [5], Croatia [6], Portugal [7] and China [8].

Operational flexibility can be improved via supply-side, demand-side and network-side approaches [9]. Conventional thermal power plants are considered as a flexible option in the supple-side approach because they can quickly and dynamically regulate the output power according to the load demand, especially in gas-fired power plants [10], lignite poly-generation systems driven by solar energy [11] and coal-fired power plants by regulating thermal system configurations [12] and revising control strategies for water-fuel ratio [13]. Programmes for demand response and demand flexibility are well-adopted ancillary services in the demand-side approach to reduce peak demands in grids by altering the power consumption strategy [14]. Electrical energy storage (EES) technologies are regarded as an important method in the network-side approach to solve the mismatch of electricity supply and demand and enhance the flexibility and stability of the grid [15]. EES technologies can be categorised on the basis of the form of energy they store as mechanical, electro-chemical, electrical, thermo-chemical, chemical and thermal energy storage [16]. Pumped hydro storage [17] and compressed air energy storage (CAES) [18] are two commercially available energy storage technologies for large-scale electricity storage. Some co-generation systems, such those based on solid oxide fuel cells and CAES [19] and a combination of green CAES with various low- and medium-temperature waste heat recovery cycles, such as the organic Rankine cycle (ORC) and the Kalina cycle [20], have been proposed. Although these options are widely deployed worldwide and have high durability, they also feature a number of drawbacks, such as geographical/geological constraints, environmental issues and long construction times [21]. Therefore, the development of alternative energy storage technologies is strongly encouraged. From these efforts, two recently proposed medium-to-large-scale thermo-mechanical energy storage technologies, namely liquid air energy storage (LAES) [22] and pumped thermal electricity storage (PTES) [23], have emerged. A thermo-economic analysis and comparison of these two technologies [24] showed that PTES can achieve higher roundtrip efficiencies and appears to be economically more competitive at higher electricity-buying prices whereas LAES has lower power/energy capital costs and a lower levelised cost of storage (LCOS). In addition, the efficiency of LAES could be enhanced by utilising waste heat/cold streams, such as a novel LAES combined with high-temperature thermal energy storage, thermoelectric generators and the ORC cycle [25]. Thus, LAES and PTES systems, which are characterised with no geographical constraints, long lifetimes and flexible power ratings, may potentially serve as key options in future electric power systems [26].

A PTES system transforms off-peak electricity into thermal energy and stores it inside thermal reservoirs using a heat pump (HP) cycle; a heat engine (HE) cycle that transforms stored thermal energy back into electricity is then followed. Because the main components of PTES systems have no technical limitation, this type of system has drawn increased attention in recent years. The Joule-Brayton cycle is widely employed in PTES systems and sparked great research interest. Solid thermal reservoirs can be used to store thermal energy in a PTES system. Desrues et al. [27] developed a numerical model to illustrate the feasibility of the process and illustrated that the development of reciprocating compressors and expanders with higher polytropic efficiencies could lead to lower-temperature and smaller-scale applications with decent storage efficiencies of up to 67%. The specific heat transfer characteristics of reciprocating devices have been investigated in other studies [28]. White et al. first presented an analysis of wave propagation [29] and thermodynamic losses [30] in solid packed-bed thermal reservoirs due to irreversible heat transfer and frictional effects and then focused on the thermodynamic aspects of PTES systems, including their energy and power densities, the various sources of irreversibility and their impact on the roundtrip efficiency [31]. The sensitivity of roundtrip efficiency to various loss parameters revealed particular susceptibility to compression and expansion irreversibility [32]. The results also demonstrated that the cycle performance for specific compression and expansion efficiencies was controlled chiefly by the ratio between the highest and lowest temperatures in each reservoir rather than by the cycle pressure ratio [33]. Benato et al. proposed a new PTES configuration that adopted an electric heater to convert electricity into thermal energy and investigated the effects of two heat transfer fluids, nine storage materials and different control strategies on the PTES system performance [34]. The results revealed that the roundtrip efficiency peaked (27%) at a temperature of 950 °C with air as the working fluid [35]. If the grid required high power for a short period of time (e.g. lower than 3 h), storages made of concrete spheres could be adopted; if the requirement was high power for a long period of time (e.g. 4–5 h), hematite (Fe₂O₃) or magnetite (Fe₃O₄) could be used [36]. Smallbone et al. [37] presented an economic analysis of a PTES system using data obtained during the development of the world’s first grid-scale demonstrator project by Newcastle University. The LCOS for a PTES system with a demonstrator size of 2 MW power and a capacity of 16 MWh ranged between 0.07 and 0.11 €/kWh. Chen et al. [38] proposed a high-temperature PTES system based on an additional electric heater and integrated with a parallel ORC to recover waste heat due to the irreversibility of the heating and compression/expansion processes. Results revealed that the combination of a high-temperature PTES with ORC represented a large-scale energy storage technology with the roundtrip efficiency of 48% and the energy storage density of 218.7 MJ/m³. Markides et al. presented a thermo-economic analysis and comparison of PTES and LAES systems and demonstrated that, at the system size intended for commercial applications, LAES (12 MW, 50 MWh) had a lower capital cost and LCOS than PTES (2 MW, 11.5 MWh). However, when considering the required sell-to-buy price ratio, PTES appeared to be economically more competitive beyond an electricity buy price of ~0.15 $/kWh, primarily because of its higher roundtrip efficiency [24]. The authors also integrated the characteristics of these technologies into a whole-electricity system assessment model and assessed the corresponding system-level values in various scenarios for system decarbonisation [39]. Wang et al. developed expressions of thermal front propagation and unbalanced mass flow rates between the inflow and outflow of packed beds in a PTES system [40] and proposed a transient analysis method for assessing system coupling dynamics, heat transfer and thermodynamics [41]. The relevant influencing factors, such as the pressure ratio, polytropic efficiency, particle diameters and structures of thermal energy storage reservoirs, were also analysed. The results indicated that helium (He) with a roundtrip efficiency of 57% presented an advantage over argon (Ar) with an efficiency of 39%.

Compared with solid thermal reservoirs, Joule-Brayton PTES systems with liquid thermal tanks present a number of benefits: (1) the storage liquid may not need to be pressurised which can remarkably reduce the cost of the storage tanks; (2) PTES systems can benefit from this operational experience and cost reductions obtained from the wide deployment of several relevant fluids for concentrating solar power (CSP) plants [42]; (3) finally, hybrid systems that can meet different demands, such as heating, cooling and electricity, may be designed by taking advantage of the indirect heat transfer between a liquid storage material and a working fluid. Many scholars have performed valuable investigations on this topic. Vinnemeier et al. [43] assessed the thermodynamic potential of integrating PTES systems into different types of thermal plants to create large-scale electricity storage units. The results revealed that the roundtrip efficiencies of different heat integration options into different types of thermal plants were roughly in the range of 50%–60%. Laughlin [44] proposed a PTES system with liquid thermal storage tanks that could transfer heat from a cryogenic storage fluid to molten solar salt. In this investigation, the roundtrip efficiency, which was computed as a function of the turbomachinery polytropic efficiency and total heat exchanger steel mass, was found to be similar to that of pumped hydroelectric storage. Farres-Antunez et al. [45] reported a novel combined system in which PTES was operated as a topping cycle and LAES was used as a bottoming cycle. Results indicated that the cycle had a roundtrip efficiency similar to that of the separate systems but provided a significantly larger energy density. The best-combined cycle achieved an increase in thermodynamic efficiency of approximately 10% (from 60% to 70%). Farres-Antunez et al. [46] also proposed two different configurations of solar-PTES systems in which an existing CSP plant was retrofitted with a Brayton heat pump or a new hybrid plant used the Brayton cycle for charge/discharge. Results indicated that heat-to-work efficiencies of approximately 40% (during CSP operation) and roundtrip efficiencies of 55%–60% (during PTES operation) could be achieved with state-of-the-art components in the two schemes. McTigue et al. [47] introduced a PTES variant using supercritical carbon dioxide (sCO₂) as the working fluid. Results revealed that the sCO₂-PTES cycle achieved high roundtrip efficiencies for a given hot-storage temperature (up to 78% at 560 °C). A comparative study of adiabatic-CAES and PTES developed by Xue [48] illustrated that the former had a higher roundtrip efficiency (70%-80%) than the latter (50%-60%), whilst the latter enjoyed higher energy density than the former.

The improvement in roundtrip efficiency are amongst the primary targets for a PTES system; this improvement can be achieved by reducing various sources of irreversibility and exergy dissipation to the surroundings. Therefore, performing an exergy analysis based on the second law of thermodynamics is essential. Conventional exergy analyses can only obtain the overall exergetic efficiency and exergy destruction of an investigated system to determine the component with the largest irreversibility generation. By comparison, advanced exergy analyses, where the exergy destruction is split into avoidable/unavoidable and endogenous/exogenous components [49], can not only identify the component with the largest irreversibility generation, but also determine the components on which design improvement efforts should focus. Such an analysis could also quantify the efficiency improvement potential of the system [50].

Advanced exergy analyses have been extended to many types of energy conservation systems. Wang et al. [51] investigated a 2 MW underwater CAES system using conventional and advanced exergy analyses. The author pointed out that the theoretical maximum efficiency under the unavoidable condition was 84%. Liu et al. [52] presented a comprehensive investigation on a novel two-stage transcritical compressed CO₂ energy storage system using conventional and advanced exergy analyses. Results indicated that the first compressor possessed the highest potential for improvement because its largest avoidable exergy destruction rate (159 kW) accounted for 23% of the total avoidable exergy destruction and 12% of the total exergy destruction. Ebrahimi et al. [53] performed conventional and advanced exergy analyses of a grid-connected underwater CAES facility. The advanced exergy analysis illustrated that 24% of the exergy destruction was unavoidable. He et al. [54] performed conventional and advanced exergy analyses to evaluate the performance of CAES and supercritical compressed CO₂ energy storage (SC-CCES) systems. Results showed that the exergy efficiency of the SC-CCES system, at 57%, was better than that of CAES (51%); moreover, the interactions amongst different components were not extensive because the endogenous exergy destruction rates of the components were larger than the exogenous exergy destruction rates in CAES and SC-CCES. Idrissa and Boulama [55] investigated a combined Brayton/Brayton power cycle using the advanced exergy analysis. Results illustrated that most of the irreversibility generated at the combustion chamber was endogenous (i.e. approximately two-thirds) and unavoidable (86%); by contrast, the irreversibility generated at both turbines and both compressors was largely endogenous and avoidable.

Galindo et al. [56] performed an analysis of a bottoming ORC cycle coupled to an internal combustion engine by using conventional and advanced exergy analyses. The advanced exergy analysis suggested that the priority of improvement should be the expander, followed by the pump, the condenser and the boiler. Razmi et al. [57] proposed an exergo-economic analysis for a co-generation system composed of a CAES, an ORC cycle and an absorption-compression refrigeration cycle; their results illustrated that the respective costs of electricity and chilled water were 0.08 $/kWh and 0.18 $/kWh during the peak period. Boyaghchi and Molaie performed an advanced exergy analysis of a real combined cycle power plant with supplementary firing and presented a parametric study discussing the sensitivity of various performance indicators [58]. Results showed that the combustion chamber concentrated most of the exergy destruction (i.e. over 62%), dominantly in an unavoidable endogenous form; With increasing the turbine inlet temperature and compressor pressure ratio, the avoidable endogenous exergy destruction increased and multiplied by the factors of 1.3 and 8.6, respectively [59]. Wang et al. [60] proposed a cascade absorption heat transformer to utilise industrial low-grade waste heat and used conventional and advanced exergy analyses to determine the cause and avoidable degree of exergy destruction. Results revealed that only 21% of the exergy destruction rates was avoidable by improvement. Fallah et al. performed conventional and advanced exergy analyses to provide detailed information about the improvement potential of the system components for a Kalina cycle [61], a recompression sCO₂ cycle [62] and a steam injection gas turbine system [63].

To the authors’ best knowledge, various irreversibility and exergy losses in the Joule-Brayton PTES system based on solid thermal reservoirs have been presented via conventional exergy analyses. Compared with a Joule-Brayton PTES system using solid thermal reservoirs, PTES systems using liquid thermal stores present some attractive advantages, such as lower storage tank costs and the potential design of hybrid systems. However, current studies on PTES systems based on liquid thermal stores mainly focus on parametric analyses of different hybrid systems and efficiency comparisons with other energy storage technologies; no conventional and advanced exergy analyses have yet been reported for the PTES systems.

The main novelty and contributions of this study are three-fold. (1) The thermodynamic parameter design of the Joule-Brayton PTES system using liquid thermal stores (i.e. solar salt for high-temperature thermal storage and Therminol 66 for intermediate-temperature thermal storage) is performed. (2) Conventional and advanced exergy analyses for non-recuperated and recuperated PTES systems are proposed and compared, and the components on which design improvement efforts should focus are obtained. (3) A quantitative analysis of the overall system performance improvement potential of the recuperated PTES system obtained by improving key parameters related to the turbomachines and heat exchangers is also conduced, and the maximum overall system performance improvement potential is provided. The results of the exergy analyses provide valuable insights into the design, analysis and assessment of large-scale PTES systems.

The rest of this paper is organised as follows. The system description, mathematical models of the main components, exergy analysis models, assumptions and modelling validation are described in Section 2. Detailed conventional and advanced exergy analyses and a quantitative analysis of performance improvement potential are carried out in Section 3. Finally, conclusions are drawn in Section 4.