Elsevier

Applied Thermal Engineering

Volume 185, 25 February 2021, 116284
Applied Thermal Engineering

Thermodynamic analysis of the optimal operating conditions for a two-stage CO2 refrigeration unit in warm climates with and without ejector

https://doi.org/10.1016/j.applthermaleng.2020.116284Get rights and content

Highlights

  • The parallel compressor combined with ejectors improve the efficiency in warm climates.

  • The gas cooler and flash-tank pressures are used as control variables.

  • The ejector requires the entrainment ratio as an extra control variable.

  • The optimal operating curve with ejector achieves an extra 13% of COP.

  • The available range of the operating conditions is reduced due to the ejector.

Abstract

The natural refrigerant R744 seems to be the long term solution to be imposed in the food retail industry, where low and medium refrigeration temperatures are usually required, despite the technical difficulties related to high the pressure, especially in warm climates, leading to trans-critical operation with the consequent COP reduction. To overcome such difficulties, different approaches have been proposed in literature, being very popular the use of parallel compressor or its combination with ejector expansion devices. The efficient operation of trans-critical R744 systems requires the control of the gas cooler pressure, as well as the pressure of the flash-tank. In the case of ejector expansion devices, the ratio of entrained mass flow rate must be also included in the control variables and, at the same time, its use implies additional restrictions. The current paper presents the optimization methodology for the operating conditions of a two-stage CO2 refrigeration unit for both cases, with and without ejector. The curves of optimal combinations of gas cooler and flash-tank pressures, understood as operational control laws, are provided together with the COP achieved. The operating conditions and performance of the ejector case are compared showing COP improvements of up to 13%. The results also show that the operating region of the control variables is limited due to the use of the ejector.

Introduction

The food retail industry, particularly for supermarket applications, requires low and medium temperature refrigeration for frozen chambers and display cabinets or cold rooms, respectively. Up to 50% of the energy consumed by a supermarket is related to refrigeration systems [4]. Thus, improvements in the systems efficiency are appreciated by the owners of the facilities, who might save costs in the electricity bill.

Former refrigeration systems of supermarkets used HFC refrigerants, like the widely adopted R404A, which are environmentally harmful due to the high Global Warming Potential (GWP) and Ozone Depletion Potential (ODP). In order to achieve a sustainable development, new regulations are being imposed to reduce the environmental impact of refrigeration technologies, such as the Directive of efficiency in Energy Related Products [8], [9] and the traditional refrigerants replacement by natural alternatives according to the F-gas Directive [27]. However, if the replacement of the fluorinated refrigerants is made by less efficient natural refrigerants, the indirect emissions would also increase due to the higher energy consumption of the compressors, i.e., for a given cooling demand, the power consumption of the compressors would increase unless the refrigerant is more efficient. Thus, natural refrigerant solutions must be more efficient than fluorinated refrigerant solutions. This fact can be accounted by implementing solutions with a better Total Equivalent Warming Impact (TEWI), the joint measure that includes the direct emissions resulting from the refrigerant leakage and the indirect resulting from the production and transport of the electricity consumed by the system [29].

In the aforementioned applications, especially commercial refrigeration, the use of refrigeration systems with the natural R744 is becoming the mainstream solution, with more than 20,000 stores worldwide using this technology [31]. The Fig. 1 shows a simplified diagram of a typical trans-critical CO2 supermarket installation with two working temperatures, for fresh and frozen display cabinets. But R744 is not only limited to these applications, small stores [7], hotels, automotive [3] or even ice rinks [23] are beginning to implement R744 systems. However, the main drawbacks of this option are associated to the high pressure and the low efficiency achieved, especially for high outdoor temperatures and trans-critical operating conditions. The high pressure has been identified as a problem many times in literature because of safety reasons. This requires very strength materials which considerably increase the cost of the systems. In addition, the operation of single stage trans-critical systems has been shown to be also very inefficient. In fact, Matthiesen et al. [21] defined the so-called “CO2 equator”, which is an imaginary line splitting Europe in two parts at the Northern Mediterranean shore. In the Northern side, the basic trans-critical cycle is more efficient and cost-effective than HFC-based solutions, and the Southern side, where the basic CO2 cycles cannot improve the efficiency of traditional refrigerants.

Several proposals can be found in the literature to overcome such difficulties and to extend the efficient operating region by pushing the “CO2 equator”. The first approach is to keep the CO2 working on sub-critical conditions by means of cascade systems with R717, R1234ze or R134a as the high pressure stage [5], [26]. It is worth also mentioning the possibility of its combination with a brine chiller to condense in the sub-critical zone. These solutions, although efficient, are usually less cost-effective than single refrigerant circuits and the difficulties associated to the control of both cycles together are remarkable. A second approach is the operation in trans-critical zone by modifying the cycle to increase the efficiency at warmer temperatures. To do so, the most common proposals found in literature are the flash-gas-bypass, the parallel compressor in the high stage, over-fed evaporators, mechanical sub-cooler and ejector expansion devices. A comprehensive review of them can be found in [12], [14], [19], [22], [30]. In addition, Gullo et al. [13] showed with a theoretical analysis how these type of cycles can outperform the R134a/CO2 cascade cycle in two cities of the previously mentioned “CO2 equator”, reducing the TEWI by at least 9.6%. However, the TEWI reduction was attributed to the direct emissions reduction, since the power consumption of all the systems was similar. Anyway, this result foresees a positive impact in the development of a sustainable refrigeration using the CO2.

The ejector is an expansion device that gained considerable ground for the last ten years. The main reasons are its low cost and simplicity, because of the lack of moving parts, and the potential improvement that can be achieved in the COP thanks to the throttling work recovery and the increment in the specific cooling achieved by the ejector expansion, more similar to an isentropic one, [10]. Gullo et al. [15], [14] showed how the use of multi-ejectors in CO2 systems can push the “CO2 equator” to regions below the north of Africa, achieving energy savings of 26,9% and reductions in the environmental impact of 90.9% compared to current solutions with R404A. Indeed, as reported by Haida et al. [17], the multi-ejector increased the COP of a vapour compression rack including parallel compressor by 7%.

Some of the major difficulties to efficiently operate a trans-critical CO2 cycle are the control of the main operating variables to find its optimum combination. Cabello et al. [1] assessed the optimal gas-cooler pressure of a single stage refrigerating CO2 plant. Additionally, the authors analysed the efficiency reduction when the optimal pressure is not well defined. They concluded that it does worth overestimating the optimal pressure since the reduction in COP is smaller than underestimating it. Nebot-Andrés et al. [24] experimentally determined, in a trans-critical plant with an integrated mechanical sub-cooler, two correlations for the optimal gas cooler pressure and sub-cooling degree. Both correlations were functions of the gas cooler outlet temperature and the evaporating temperature. Peñarrocha et al. [25] proposed a model-free real-time optimization and control strategy for CO2 trans-critical refrigeration plants in which the control variables are the compressor speed and the opening of the high pressure valve in order to reduce compressor power consumption. The authors mathematically demonstrated that this strategy is equivalent to maximise the COP. He et al. [18] proposed an on-line optimal quasi cascade controller for an ejector with variable nozzle throat area. The results demonstrated that the optimal working conditions in terms of COP did not mean the maximal ejector efficiency and cooling capacity. Mitsopoulos et al. [22] optimised the gas cooler outlet pressure as a function of the external temperature to find the best performance of the CO2 trans-critical systems with overfed evaporators. However, these authors did not consider the flash tank pressure as an optimisation variable while keeping a constant value of 40 bar. Gullo et al. [15], [14] considered as optimisation variables both, the flash-tank pressure and the gas cooler outlet pressure to assess the performance of the CO2 trans-critical systems, with and without overfed evaporators.

As highlighted by the literature surveyed, there is a strong interest in the application of CO2 to commercial refrigeration and great efforts are being made by researchers to improve the efficiency achieved specially in warm climates. Furthermore, the control of the operating conditions is crucial for the optimal operation of the system. In fact, once the evaporating temperatures of the low and medium temperature evaporators are fixed by the application, the efficiency of the system highly depends on the gas cooler and flash-tank pressures. The first controls the discharge of the medium temperature and parallel compressors and the second controls the suction pressure of the parallel one. Thus, both variables must be optimized to achieve the best performance of the CO2 cycles. Then, a gap can be still identified in the field regarding the search of the optimal operating conditions for CO2 trans-critical systems. As mentioned, most of the optimisation studies are focused on single stage units where only the gas cooler outlet pressure is optimised. Furthermore, the studies which included the flash-tank pressure in the optimisation procedure did not highlight the operational limitations associated to the use of the ejector in an existing system and the comparison of the optimal operating conditions between the systems with and without ejector.

The present manuscript will be focused on two typologies of systems. The first one is a two-stage trans-critical CO2 unit with parallel compressor and the second one is the combination of the former system with a high-pressure ejector. If the optimisation of the operating conditions is not conducted, the performance of the cycle might be poor even using techniques like the parallel compressor or the ejector.

In the units with two evaporating temperatures, the variables to control the optimal operation must be the pressure of the flash-tank and the gas cooler outlet. If an ejector is included, the ratio of entrained mass flow rate to the motive mass flow rate must be also considered as an extra control variable. Thus, the novelty of the present paper, and its main objective, is to find the optimal combination of the operating conditions in the proposed systems, with and without ejector. This optimal combination will be provided through a couple equations which can be implemented as control laws to adjust the parallel compressor speed and the high-pressure valve opening. The comparison of the different optimal operating conditions between the two systems will be provided. In addition, the efficiency improvements achieved by the adoption of the ejector will be addressed, while the implications in the range of the operating conditions and the previously mentioned control laws will be also shown by means of an applicability map. To this end, a thermodynamic model of the unit and the ejector have been developed and coupled to perform the optimisation approach.

The rest of the paper is organized as follows. Section 2 describes the methodology adopted to perform the optimisation analysis. Section 3 describes the results obtained with the analysis for the reference case and for the ejector case providing a comparison between both solutions. Finally, the Section 4 summarises the main conclusions arisen from the results.

Section snippets

Methodology

The present section contains the description of the methodology followed to find the optimal operating conditions for the two cycles proposed. First, the general hypotheses imposed in the models will be detailed with the corresponding limitations. Then, the two systems will be described by mean of its simplified Piping and Instrumentation Diagram (P&ID) and the equations defining the behaviour of the main components. Finally, the procedure followed to find the optimal operating curve of the

Results and discussion

The results obtained as described in the Section 2.4 will be presented in the following subsections. Firstly, the case without ejector is analysed and after that the results including the ejector are also highlighted. Finally, a discussion about the previous results is accomplished where both cases are compared to show the potential improvements that can be achieved due to the ejector while some restrictions appear in the operating conditions.

Conclusions

The literature surveyed reveals that the double-stage trans-critical CO2 systems are becoming the most common solution for the refrigeration in super-market applications. Particularly, the use of parallel compressors in combination with ejectors seems to be a promising solution to achieve the desired efficiency. However, the optimal selection of the operating conditions is crucial for the efficient operation of the systems and several working parameters must be controlled at the same time.

To

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.

Acknowledgments

This study has been partially funded by the ERDF program ITC-20181143 (EJERCER) in collaboration with the company Intarcon S.L.

References (31)

Cited by (21)

  • Life cycle techno-enviro-economic assessment of dual-temperature evaporation transcritical CO<inf>2</inf> high-temperature heat pump systems for industrial waste heat recovery

    2023, Applied Thermal Engineering
    Citation Excerpt :

    Li and Groll [30] established a constant pressure mixing model for the ejector and concluded that for typical air conditioning operation conditions, the COP of transcritical CO2 cycle with ejector is increased by 16% over the basic transcritical CO2 cycle. Expósito-Carrillo et al. [31] presented the optimization methodology for the operating conditions in a CO2 refrigeration cycle and found the ejector system is up to 13% higher COP than the system without ejector. Bai et al. [32] proposed the advanced exergy analysis on transcritical CO2 refrigeration system with an ejector, and the results show the exergy destruction reduction of the system is 43.44%.

View all citing articles on Scopus
View full text