Elsevier

Applied Energy

Volume 242, 15 May 2019, Pages 1198-1208
Applied Energy

Conventional and advanced exergy analysis of a grid connected underwater compressed air energy storage facility

https://doi.org/10.1016/j.apenergy.2019.03.135Get rights and content

Highlights

  • The world’s first grid connected UWCAES facility was thermodynamically analyzed.

  • Conventional and advanced exergy analyses were conducted.

  • The primary sources of energy loss were identified.

  • Conventional analysis showed that 47.1% of the exergy destruction was avoidable.

  • Advanced analysis revealed that 23.6% of the exergy destruction was unavoidable.

Abstract

A data driven exergy analysis has been conducted for the first known grid connected Underwater Compressed Air Energy Storage facility, located in Toronto, Canada. Further to examining the plant through conventional exergy analysis, results were enhanced by splitting exergy destruction rates into avoidable and unavoidable, as well as endogenous and exogenous parts via advanced exergy analysis. The conventional exergy analysis showed that under real operational conditions, the exergy destruction ratio was 47.1%, while under the theoretical unavoidable operational conditions it could be reduced to 15.9%. The overall outcome of the conventional exergy analysis was confirmed by the advanced exergy analysis, the details, however, were quite different. The results of advanced exergy analysis assigned the improvement priority to heat exchanger 4, followed by the turbine and the stage 3 compressor. Conversely, the conventional exergy analysis indicated that the total exergy destruction of the turbine was higher than that for heat exchanger 3. The advanced exergy analysis also revealed that 76.4% of the exergy destruction was avoidable, highlighting the significant potential of the system for performance improvement.

Introduction

Energy storage technology has been rapidly developing to support the need for grid flexibility [1], [2], [3]. The energy storage landscape includes solutions for voltage support, frequency regulation, arbitrage, and a number of other ancillary services for an evolving grid. Compressed Air Energy Storage technology has a proven track record for bulk energy storage [4], [5], [6] and holds promise as an increasingly fast acting demand response and load supply asset [7]. Underwater Compressed Air Energy Storage (UWCAES) is one of the newest branches of this technology [8], [9]. In UWCAES, air is compressed to the hydrostatic pressure that corresponds to the submerged depth of the air storage containers (accumulators). In a marine environment the system is quite flexible and can scale with depth. In 2015, the first grid connected UWCAES system was implemented in Lake Ontario by Hydrostor Inc. [10].

CAES systems rely on the compression and expansion of air which typically necessitates careful thermal management. Historically, isothermal and adiabatic approaches to CAES thermal management have been considered. In this study, an adiabatic design is utilized. This entire UWCAES process is managed by a number of interdependent components. The strategic arrangement of these components can lead to many different facility configurations. One of the more insightful studies of the thermodynamic performance of these configurations was realized by exergy analysis [11], [12]. Compared to a first-law energy analysis, exergy approaches consider both the quantity and quality of energy. In the literature, intensive and novel exergy analyses of different configurations of CAES systems have been proposed. Safaei and Keith [13] proposed a distributed compressed air energy storage system for recovering waste heat during the compression phase. Their technique suggested a potential reduction of the fuel consumption of the McIntosh plant by about 25%. Barbour et al. [14] analyzed a 2-stage adiabatic CAES system with an elevated temperature packed bed thermal strategy. They defined efficiency of the system as the total useful work released by the expansion, over the total input work required to run the compression. By storing the generated heat of the compressors in a solid material they could increase the defined efficiency up to 71%. Wolf and Budt [15] applied an adiabatic low-temperature CAES approach to obtain a roundtrip efficiency (defined as the total output power in the discharge phase over the total input power in the charging phase) of 52–60%. Their work concluded that using this approach led to start-up times of less than 5 min with cost-minimized system components. Related thermodynamic and exergy analysis of other CAES-related systems have also been performed [16], [17], [18]. Najjar and Zamout [19] investigated the performance of a CAES plant for arid regions. They showed that CAES had the common advantages of peak load gas turbine power plants and that of the pumped storage scheme. The characteristics of Huntorf CAES have also been studied through exergy analysis [20]. This work introduced an efficient methodology for the estimate of the cavern volume of cavern based CAES plants. Kim and Favrat [21] performed exergy analysis of a micro-CAES system. The results showed that this system especially with quasi-isothermal compression and expansion processes were very effective for distributed power networks. Exergy analyses of CAES combined with thermal energy storage operating on hot water was done by Bagdanavicius and Jenkins [22]. Mohammadi et al. [23] performed an exergy analysis of a Combined Cooling, Heating and Power system integrated with wind turbine and conventional CAES.

In the majority of the aforementioned studies, conventional exergy analyses provided very useful information about the thermodynamic performance of the system and was effective in determining the exergy destruction of the components. The analysis, however, does not differentiate unavoidable exergy destruction from the improvable avoidable one. Furthermore, it treats each component as a separate entity, whereas in reality, the interlinking components of the integrated system interact with each other. The so called, ‘advanced’ exergy analysis, alternatively, provides such information in detail. With advanced exergy analysis, the exergy destruction in each system component is split into endogenous and exogenous; as well as avoidable and unavoidable parts. By applying advanced exergy analysis, components of the system that influence the overall process performance can be identified. This is very useful information when trying to ascertain the avoidable parts of endogenous and exogenous exergy destruction [24]. The idea of subdividing the exergy destruction was first proposed by Tsatsaronis [25]. The application of this technique has become popular in recent years [26]. Açıkkalp et al. [27] conducted a thermodynamic performance evaluation of a cogeneration system and identified the potential for the economic improvement of each component. They also calculated the effects of system components on one another in terms of exergy destruction rate. Soltani et al. [28] performed an advanced exergy analysis on an externally fired combined cycle power plant and concluded that the exergy destruction rate in most of the system components was unavoidable and endogenous. Tan et al. [29] modeled a geothermal district heating system with advanced exergy analysis and showed system’s great operational modification potential. Fallah et al. [30] applied both conventional and advanced exergy analysis for the study of a low-temperature Kalina cycle geothermal application. They reported that according to the conventional exergy analysis, the exergy destruction of the evaporator was highest while the improvement priority was given to the condenser. More research in the application of advanced exergy analysis can be found in [31], [32].

From the review of the literature herein, it is clear that the advanced exergy analysis is a promising and popular tool for the thermodynamic scrutiny of CAES system performance. That said, the conventional and advanced exergy study of UWCAES facilities remains rare. In one of the limited number of studies, Wang et al. [33] studied a UWCAES unit with both conventional and advanced exergy analysis. They defined efficiency as the total output power to/from over the total input power of the system. They reported that under the theoretical real condition the exergy efficiency was 53.6% while under the unavoidable condition it was 84.3%. In another study, they investigated the efficiency of a multi-level UWCAES system and showed that the exergy efficiency of the system tends toward 62% when more energy is stored in the CAES subsystem while even higher efficiencies of 81% were possible when more energy was stored in the battery packs [34].

According to the authors’ knowledge, there are neither conventional nor advanced exergy analyses performed for a UWCAES system based on real operational data. This paper is thus one of the first attempts to fill this gap by revealing sources of thermodynamic irreversibility and identifying components in UWCAES systems with the highest exergy destruction rate. To achieve this, exergy destruction subdivisions and exergy efficiency of the system components are calculated by both conventional and advanced exergy analysis. The interaction between system components will also be studied to identify the weak points of the UWCAES system. It is hoped that results from this study will not only help to improve the Toronto Island UWCAES plant but also other next generation CAES facilities.

Section snippets

The Toronto Island UWCAES plant

The process flow diagram of the Toronto Island UWCAES plant is shown in Fig. 1. The system consists of six main subsystems including the air compression unit, compressed air storage unit, transmission pipeline unit, expansion unit, thermal energy storage unit and heat exchangers. Each component in Fig. 1 is given a 3-letter code. The Toronto Island UWCAES plant is equipped with sensors and flow analyzers throughout to measure the properties of fluids in the inlet and outlet of each component.

Thermodynamic analysis

The mass, energy and exergy balance of each component of the system can be written as:ṁin-ṁout=0Q̇-Ẇ=ṁouthout-ṁinhinQ̇-Ẇ=ṁouteout-ṁinein+ĖDwhere ṁ is the mass flow rate (kg/s), h is the specific enthalpy (kJ/kg), Q̇ is the heat transfer rate (kW) to the control volume, Ẇ is the rate of work leaving the control volume (kW), e is the specific exergy (kJ/kg) and ĖD is the exergy destruction rate.

The overall exergy of the system can be expressed by thermomechanical and chemical

The advanced exergy analysis approach and theoretical assumptions

Various approaches have been proposed in the literature for advanced exergy analyses [38]. The most common approaches are the thermodynamic cycle, the exergy balance, the engineering method, the structural theory and the equivalent components method. When the thermodynamic cycle of the system is well defined, the thermodynamic cycle method is well suited and provides higher prediction accuracy than others [30]. Thus we have chosen this method for our UWCAES analysis. Also real, unavoidable and

Conventional exergy analysis

Thermodynamic models presented by Wang et al. [33] were applied to define the thermodynamic properties of the Toronto Island UWCAES plant presented in Fig. 1. During the ramp-up phase the status of the system changes rapidly. Thus, we apply the averages of the thermodynamic properties and mass flow rates for the unavoidable and ideal parts in the analysis. In Table 3, the average of temperature, pressure and mass flow rate from sensors for real conditions are compared with calculated values for

Conclusion

In the present work, conventional and advanced exergy analyses were carried out to analyze the world’s first grid-connected underwater compressed air energy storage plant, located in Toronto, Canada. The main results are as follows:

  • Both advanced and conventional exergy analyses indicated the notable improvement potential of the plant. However, details of the analyses were quite different. Conventional exergy analysis showed that the highest exergy destruction occurs in the turbine, followed by

References (39)

Cited by (78)

View all citing articles on Scopus
View full text