Performance analysis and evaluation of a commercial absorption–refrigeration water–ammonia (ARWA) systemAnalyse de la performance et évaluation d'un système frigorifique commercial à absorption à ammoniac/eau

https://doi.org/10.1016/j.ijrefrig.2008.02.005Get rights and content

Abstract

The Robur absorption–refrigeration water–ammonia (ARWA) system is analyzed using Aspen Plus flowsheet simulator. The results are compared with experimental and some manufacturer data reported in the open literature. Among performance parameters analyzed are coefficient of performance (COP), heat duties of the evaporator, absorber, and the condenser, refrigerant concentration in the weak and strong solution, and flow rates of the weak solution and the flow rate of refrigerant passing through the evaporator. In general, a very good agreement between the simulator's results and the experimental measurements was found. Also, results obtained for the effect of separator (input) heat duty on the COP agree well with the reported experimental data with a maximum percentage deviation of 1.8%. Efficiency of the separator in splitting off the refrigerant at the column top is shown to be of crucial importance; COP increased by 15% in going from 1 to 5 theoretical equivalent mass transfer stages in the separator. Some innovative modifications to Robur cycle aimed at enhancing the separator operation have shown a promising improvement in the COP. In particular, introducing a throttling process directly before the separator can alleviate the separator heat load and enhance the COP by up to 20%. Use of stripping gas injected at the bottom of the boiler is another strategy that has been investigated in this work.

Introduction

Thermal production of steam or heat by burning fossil fuel in many processes results in heat that is rejected to the surrounding as waste. This waste heat can be utilized in useful refrigeration by using vapor-absorption refrigeration systems. This helps reduce problems related to global environment, such as greenhouse effect from CO2 emission resulting from the combustion of fossil fuels in utility power plants and the use of chlorofluorocarbon refrigerants, which is currently thought to affect depletion of the ozone layer. The ban on fluorocarbon fluids has been an impetus towards research into environmental friendly refrigerants such as water. One way to reduce CO2 emissions is to utilize low-grade heat sources in heat powered refrigeration systems such as vapor-absorption. In order to promote the use of absorption systems, further development is required to improve their performance and reduce cost. In the 1950s, a system using lithium bromide/water as the working fluid was introduced for industrial applications. A few years later, a double-effect absorption system was introduced and has been used as an industrial standard for a high performance heat-operated refrigeration system.

Although the vapor-compression cycle, which is around 100 years old, is still most dominating in air-conditioning and refrigerating applications, the well-known refrigerant–absorber systems (e.g., H2O/LiBr and NH3–H2O) are still viable options for certain applications, particularly in the field of gas industry and large-scale water chiller systems (Srikhirin et al., 2001, Yokozeki, 2005). Recently, energy conservation via waste heat recovery has resulted in more attention focused on the use of the NH3–H2O system (Erickson et al., 2004)

In order to analyze and evaluate the performance of vapor-absorption systems, reliable thermodynamic property models to predict temperature–pressure–concentration and enthalpy–temperature are required. These relationships are best furnished through the use of cubic equations of state (EOS). In order for these EOS to predict well the fluid mixture PVT and phase behavior properties at different unit operations in the cycle, the proper interaction parameters between the involved species are required. It is well-known that all cubic EOS are very sensitive to these parameters. One of the difficult problems in using EOS for such mixtures would be how to set up EOS parameters for non-volatile compounds without much information about the critical parameters and vapor-pressure data. Attempts to alleviate these difficulties are reported to bring some success (Yokozeki, 2005). One of the most significant features of using EOS is that, even at supercritical temperatures of refrigerants, thermodynamic properties can be consistently predicted (Yokozeki, 2002)

Finding ways to improve absorption system efficiency has recently become of high research priority. Research activities were mainly focused on finding new potential fluids, development of new or hybrid cycles, and improving the heat and mass transfers of the absorption refrigerator (Wu et al., 2000). One of the good literature reviews about vapor-absorption refrigeration is that of Srikhirin et al. (2001) where a number of research options such as various types of absorption refrigeration systems, research on working fluids, and improvement of absorption processes are discussed.

Aspen Plus (2004) is an excellent modeling tool which is versatile and relatively easy to use in modeling of advanced power cycles. It provides a platform for describing different physico/chemical processes. Use of Aspen Plus leads to an easier way of model creation, troubleshooting and sensitivity analyses since simple sub-sections of complex systems can be created and tested as separate modules before they are integrated. It has an extensive physical–chemical property database with flexibility for incorporating many user-defined model blocks in addition to the many built-in unit models such as heaters, pumps, stream mixers, stream splitters, compressors, etc.

Bram and De Ruyck (1997) discussed and presented a general two-step approach for cycle development and layout optimization using Aspen Plus. The technique is applied to evaporative gas-turbine cycles with one intercooler stage, no reheat and no steam-turbine. Several evaporative cycle layouts are optimized and the feasibility of each cycle is quantified by the exergy destruction and exergy efficiency. In another study (Ong'iro et al., 1995) a computer simulation model in Aspen Plus shell has been developed to simulate the performance of integrated gasification combined cycle (IGCC) and integrated gasification humid air turbine (IGHAT) cycle power plants. The model was used to study the effects of design and performance parameters on the efficiency and emissions from both cycles. Srinophakun et al. (2001) used the energy utilization diagram (EUD) approach to perform sensitivity and exergy analysis study for Rankin power cycles. The authors used Aspen Plus to generate the EUD.

Rosen and Dincer (2003) developed a methodology for the analysis of systems and processes that is based on the quantities exergy, cost, energy and mass. The authors developed a code for these analyses utilizing Aspen Plus simulator and discussed applications of their methodology and another code for the analysis of some engineering processes such as hydrogen and hydrogen derived fuels. Sencan, 2007, Sencan, 2006 presented a linear regression model within data mining process and artificial neural network model for the analysis of ammonia–water absorption refrigeration systems. Fan et al. (2007) presented a state-of-the-art review on the solar sorption research (absorption and adsorption) refrigeration technologies. The basic principles, development history and recent progress in solar sorption refrigeration technologies are reported. The application areas of these technologies are categorized by cooling temperature demand. The authors show that solar powered sorption refrigeration technologies are attractive alternatives that not only can serve the needs for air-conditioning, refrigeration, ice making and congelation purposes, but also can meet demand for energy conservation and environment protection. However, an extensive research work still needs to be done for large-scale applications in industry and for the replacement of conventional refrigeration machines.

An extensive research effort is being invested in the development and improvement of combined power-refrigeration cycles. Vidal et al. (2006) applied the exergy analysis method to evaluate newly developed combined ammonia–water power-cooling cycle (Zheng et al., 2002, Goswami, 1998). The Redlich–Kwong–Soave equation of state was used to calculate the thermodynamic properties. The cycle was simulated as a reversible as well as an irreversible process to clearly show the effect of the irreversibility in each component of the cycle.

Fernández-Seara and Sieres, 2006a, Fernández-Seara and Sieres, 2006b analyzed the effects of ammonia purification and liquid entrainment and blow-down from the evaporator in ammonia–water absorption systems. A mathematical model based on a single-stage system with complete condensation has been developed and ammonia purification is evaluated by means of the Murphree efficiencies of the stripping and rectifying sections of the regenerator column. The entrainment and blow-down are taken into account considering the corresponding flow rates as a fraction of the dry vapor at the evaporator outlet. The influence of the distillation column component's efficiency on the attainable distillate concentration and the effects of the distillate concentration and the liquid entrainment and blow-down on the system operating conditions and performance are analyzed and quantified.

Yokozeki (2005) demonstrated the use of equations of state for consistently describing thermophysical properties of refrigerant–absorbent mixtures where various binary-pairs are used as examples. Adewusi and Zubair (2004) used the second law of thermodynamics to study the performance of single-stage and two-stage ammonia–water absorption refrigeration systems when some input design parameters are varied. Their results show that the two-stage system has higher lost entropy (Stot) and COP, while the single-stage system has a lower Stot and COP.

Performance of absorption refrigeration systems is heavily dependent on the physical–chemical properties of the working fluid (Srikhirin et al., 2001, Perez-Blanco, 1984, Holmberg and Berntsson, 1990) A fundamental requirement of absorbent–refrigerant combination is that, in liquid phase within the operating temperature range of the cycle, they must have a margin of miscibility. The mixture should also be non-corrosive, environmental friendly, of low-cost, chemically stable, non-toxic, non-corrosive, and non-explosive. In addition to these requirements, the elevation of boiling point (the difference in boiling point between the pure refrigerant and the mixture at the same pressure) should be as large as possible. Refrigerant also should have high heat of vaporization and high concentration within the absorbent in order to maintain low circulation rate between the separator and the absorber per unit of cooling capacity. Transport properties (e.g., viscosity, thermal conductivity, and diffusion coefficient) that influence heat and mass transfer should be favorable (Srikhirin et al., 2001, Holmberg and Berntsson, 1990).

Ammonia (refrigerant)–water (absorbent) pair has been widely used for both cooling and heating purposes. Both are highly stable for a wide range of operating temperature and pressure. Ammonia (NH3) also has a high latent heat of vaporization, which is a requirement for efficient performance of the system. It can be used for low temperature applications, as the freezing point of NH3 is −77 °C. Since both NH3 and water are volatile, the cycle requires a condenser to knock down water that normally evaporates with NH3. Without such an auxiliary condenser, the water would accumulate in the evaporator and deteriorate the system performance. The system is also attended by some other disadvantages such as its high pressure, toxicity, and corrosive action to copper and copper alloy. However, water–NH3 is environmental friendly and low-cost. Thermodynamic properties of water–NH3 can be obtained from Srikhirin et al., 2001, Park and Sonntag, 1990, El-Sayed and Tribus, 1985, Ziegler and Trepp, 1984, Herold et al., 1988, and Patek and Klomfae (1995).

In this work the Robur absorption–refrigeration water–ammonia (ARWA) system is analyzed using Aspen Plus flowsheet simulator, and the results are compared with experimental data reported in the open literature. Performance parameters analyzed are coefficient of performance (COP), heat duties of the evaporator, absorber, and the condenser, refrigerant concentration in the weak and strong solution, and flow rates of the weak solution and the flow rate of refrigerant passing through the evaporator. Furthermore, some innovative modifications to Robur cycle aimed at enhancing the separator operation will be explored.

Fig. 1 shows the Robur absorption–refrigeration water–ammonia (ARWA) system simulated in this work. This system has been chosen for study in this work for more than a reason. (1) The cycle represents a commercial unit that is in operation where some performance data are available. (2) Although a one-stage cycle, it has complex flowsheet connectivity (Fig. 2). This is so because of the intermediate extra throttling process after the condenser, the extra refrigerant–refrigerant heat exchanger, the extra air-cooled heat exchanger (absorber) and the use of the strong solution as a heat sink stream in the auxiliary condenser. (3) A cycle, with this complex flowsheet connectivity and recycle loops, is a good testing exercise for Aspen Plus simulator in terms of convergence features and accuracy, and finally (4) the cycle has not been simulated before using a flowsheet simulator like Aspen Plus.

A short description, as presented by Lazzarin et al. (1996) follows (a detailed description of the unit is presented in Horuz and Callander (2004). The machine is basically the well-known ammonia–water chiller Arkla, which is almost a 1 m × 1 m × 1 m size. The Arkla firm was finally taken over by the Italian firm Robur SpA, which actually sells the machines under the brand Robur Co. In the separator (generator), which is located in the center of the unit, the ammonia–water mixture is heated by a typical vertical natural gas or LPG burner: ammonia vapor is separated, but it is mixed up with fractions of water vapor. The ammonia vapor is then purified from water in the auxiliary condenser (rectifier). In this condenser water is knocked down as liquid over a serpentine tube carrying the cold rich solution, which is delivered by the pump. The “almost pure” ammonia vapor proceeds to the air-cooled condenser, after which it undergoes two throttling processes and one cooling process in the refrigerant–refrigerant heat exchanger. In the evaporator, ammonia vapor is again produced by absorbing the required heat from brine water. The vapor is absorbed by the weak solution from the separator (generator) in a first absorber cooled by the solution that has been just used as a cooling medium in the auxiliary condenser (rectifier). The absorption is completed in a second air-cooled absorber. This strong ammonia–water solution is now at the lower cycle pressure, and it must be sent to the higher-pressure separator. This is achieved by a diaphragm pump, driven by an oil rotary pump. The machine is basically a 5 ton absorption chiller with a mixture charge of 13.41 of water and 5.65 kg of ammonia.

The absorber is one of the most critical components of any absorption refrigeration system (Srikhirin et al., 2001, White and O'Neill, 1995). The prevailing state in the absorber is a non-equilibrium one, i.e., for given temperature and pressure in the absorber, the solution absorbs less refrigerant than that of the theoretical value. Therefore, research studies towards understanding and improving the absorption process between the refrigerant vapor and the solution are of primary priority. For water–NH3, literature on absorber designs is also provided (Kang et al., 1997, Jeong and Lee, 1998, Perez-Blanco, 1988). The separator (generator) will also be shown to have a significant effect on the whole cycle performance.

Section snippets

Aspen Plus simulation

Aspen Plus is an advanced flowsheet simulator that has extensive databases available for the physical–chemical properties of chemical substances (Aspen Plus, 2004). It is used in this work to simulate the actual Robur system shown in Fig. 1. The Peng–Robinson (PR) cubic equation of state (EOS) is selected to model and calculate the thermophysical properties of ammonia–water mixture. This equation is known to have a good credibility in predicting phase and PVT behaviors of the system under study

Conclusions

Aspen Plus flowsheet simulator is employed in a comprehensive performance analysis and evaluation for one of the commercial demonstration absorption–refrigeration water–ammonia (ARWA) chillers made by Robur. It has been shown that Aspen Plus provides a very flexible platform for the analysis of power cycles with varieties of options for fluid packages, databases, and simulation tools. The predicted results compared favorably with the experimental and manufacturer data reported in the open

References (38)

Cited by (73)

View all citing articles on Scopus
View full text