Simulation and experiments on a solid sorptioncombined cooling and power system driven by the exhaust waste heat

Abstract

A solid sorption combined cooling and power system driven byexhaust waste heat is proposed, which consists of a MnCl₂ sorption bed, a CaCl₂ sorptionbed, an evaporator, a condenser, an expansion valve, and a scrollexpander, and ammonia is chosen as the working fluid. First, the theoreticalmodel of the system is established, and the partitioning calculationmethod is proposed for sorption beds. Next, the experimental systemis established, and experimental results show that the refrigeratingcapacity at the refrigerating temperature of –10°C and theresorption time of 30 min is 1.95 kW, and the shaft power is 109.2W. The system can provide approximately 60% of the power for the evaporatorfan and the condenser fan. Finally, the performance of the systemis compared with that of the solid sorption refrigeration system.The refrigerating capacity of two systems is almost the same at thesame operational condition. Therefore, the power generation processdoes not influence the refrigeration process. The exergy efficiencyof the two systems is 0.076 and 0.047, respectively. The feasibilityof the system is determined, which proves that this system is especiallysuitable for the exhaust waste heat recovery.

Keywords

solid sorption / exhaust wasteheat / combined cooling and power system / exergy efficiency

Introduction

According to the China AutomobileIndustry Association, the production and sale of automobiles are 25.07million and 24.95 million from January to November in 2016. For internalengines, approximately 30%–40% of the fuel combustion heat isexpelled directly to the environment through the exhaust gas. Theexhaust emission not only wastes a lot of energy but also impactsthe environment. Consequently, the effective recovery of exhaust heatis becoming increasingly urgent. Considering the fact that the engineexhaust has a relatively high temperature and a large amount of heat,the solid sorption refrigeration cycle is proposed to be used to recoverthe exhaust heat and provide some refrigerating capacity [1–3].

A lot of research has been conductedon adsorption refrigeration systems for vehicles. Zhong et al. [4] proposed a zeolite-water adsorptionair conditioning system to integrate with the recent development ofemission control for heavy-duty vehicles. This system was designedto provide cooling to reduce engine idle for long-haul vehicles. Wuet al. [5] establisheda zeolite 13X-water adsorption refrigeration module driven by exhaustwaste heat. Based on the requirements of the refrigerating capacityin practical applications, the right number of modules could be assembled.Sharafian et al. [6–9] examined the type of adsorbent,the adsorbent mass, the number of adsorption beds, and the structureof the adsorption beds on the performance of a waste heat-driven adsorptionrefrigeration system under different operating conditions, which wasdesigned for vehicle air conditioning applications. Verde et al. [10,11] conducted theoretical and experimental investigationson a silica gel-water adsorption chiller for automotive applications,and developed a non-equilibrium lumped parameter model to predictthe transient performance of the system.

Except for the research on the adsorptionair conditioning systems for vehicles, some research has also beenconducted on the adsorption freezing system for automobiles. Gao etal. [12–14] proposed and established a MnCl₂/CaCl₂-NH₃ two-stage solid sorption freezing system for a refrigerated truck,which was driven directly by exhaust gas. Compound sorbents of MnCl₂ and CaCl₂ developed by the expandednatural graphite treated with sulphuric acid (ENG-TSA) were utilized.Experimental results showed that the refrigerating capacity obtainedwas high enough to meet the requirement of the refrigerated truck.But for the solid sorption refrigeration systems driven by the engineexhaust waste heat, the evaporator and the condenser are generallyequipped with fans, and consequently, some electricity is necessarilyconsumed. If the power generation technology driven by low-grade heatcan be combined with the solid sorption refrigeration system, thezero power consumption of the sorption refrigeration system can beachieved. The power generation technology mainly contains the thermoelectricgeneration technology and the organic Rankine cycle. For the thermoelectricgeneration, the relatively high initial investment greatly restrictsthe practical application. For the organic Rankine cycle, Boretti[15] conducted a researchon the recovery of exhaust and coolant heat with R245fa organic Rankinecycles in a hybrid passenger car with a 1.8 L naturally aspiratedgasoline engine. Yang et al. [16] analyzed the performance of the dual loop ORC system for the enginewaste heat recovery under various operating conditions of the engineand found that the maximum waste heat recovery efficiency of the dualloop ORC system could reach 5.4%. On this basis, some cogenerationcycles with ORC and adsorption technology for the electricity andrefrigeration were investigated [17–19], but the complexityof those cycles greatly restricted their practical application invehicles.

Based on the solid sorption refrigerationcycle, Wang et al. [20] proposed a novel resorption cycle driven by low grade heat forthe cogeneration of electricity and refrigeration, in which the expanderwas installed between the high-temperature salt sorption bed and thelow-temperature salt sorption bed to generate electricity. Additionally,Bao et al. [21,22] established an integration of thechemisorption refrigeration cycle and a scroll expander for the electricityand refrigeration, in which the expander was installed between thesorption bed and the condenser. Due to the simple structure, the resorptioncogeneration cycle is a good candidate for practical application invehicles.

In this paper, a novel solid sorptioncombined cooling and power cycle driven by exhaust waste heat is proposed,and the working pairs of MnCl₂/CaCl₂-NH₃ are chosen. Both the simulationand the experiments are performed, and corresponding results are analyzedin detail. The main goal of the system is to output the refrigeratingcapacity, and the electricity is used by the system itself.

System description

Working principles

The solid sorption combined coolingand power cycle shown in Fig. 1 consists of a high-temperature saltsolid sorption bed (HTS bed), a middle-temperature salt solid sorptionbed (MTS bed), a condenser, an evaporator, an expander, some valves,and so on. The working principles are as follows.

Power generation process betweenthe HTS bed and the MTS bed (Fig. 1(a)): This process can also becalled the resorption process. In this process, the HTS bed is heatedby the heat source with the heat input of Q_des2, and the MTS bed is cooledby the heat sink. The refrigerant desorbed from the HTS bed flowsinto the expander and expands. The electricity generator is drivenby the expander. The exhaust of the expander is sorbed by the MTSbed, and the reaction heat of the Q_ads1 is released to the heat sink.

Sorption/ desorption process (Fig.1(b)): In this process the MTS bed is heated by the heat source withthe heat input of Q_des1, and the refrigerant is desorbed to the condenser at a high pressure P_c. Meanwhile, theHTS bed is cooled by the heat sink and begins to sorb the refrigerantfrom the evaporator at a low pressure P_e, and the vaporization of the refrigerantproduces the refrigerating capacity Q_e at T_e.

The previous research [12] shows that compound sorbents ofMnCl₂ and CaCl₂ developedby the matrix of the ENG-TSA are suitable for the HTS bed and MTSbed, respectively.

System design

The schematic diagram of the systemis presented in Fig. 2, which consists of a MnCl₂ sorption bed (HTS bed), a CaCl₂ sorptionbed (MTS bed), an evaporator, a condenser, an expansion valve (EAV),an expander, four ammonia valves (AV), two exhaust valves (EV), twoair valves (CV) and a liquid storage tank. Ammonia is chosen as therefrigerant.

Two sorption beds consist of manyunit tubes, and the arrangement mode of unit tubes is the staggeredarrangement. The exhaust gas and ambient air exchange heat with theexternal surface of the unit tubes. For the HTS bed, the number ofunit tubes is 66, and for the MTS bed, the number of unit tubes is55. The spacing between two unit tubes is 57 mm. For the evaporator,a plate-fin type heat exchanger is utilized. For the condenser, aparallel-flow heat exchanger is used.

The oil-free scroll expander witha displacement of 12 mL/r is chosen for the expansion process, whichis produced by Air Squared Manufacturing, Inc., and the model numberof the expander is E15H22N4.25.

The detailed working processes ofthe system are as follows.

Power generation process (resorptionprocess): For this process, air valve CV2, exhaust valve EV1, andammonia valve AV4 are open. The HTS bed is heated by the exhaust gasof the engine, and the MTS bed is cooled by ambient air. The ammoniais desorbed from the HTS bed and flows to the scroll expander. Theelectricity generator is driven by the expander. The exhaust of theexpander is sorbed by the MTS bed, and the reaction heat is releasedto ambient air.

Sorption process of the HTS bed anddesorption process of the MTS bed: For this process, air valve CV1,exhaust valve EV2, expansion valve EAV, and ammonia valve AV1 andAV2 are opened. The HTS bed is cooled by ambient air, and the MTSbed is heated by the exhaust gas of the engine. The ammonia is desorbedfrom the MTS bed and condensed into liquid in the condenser. The liquidammonia flows into the liquid storage tank, and then passes throughthe expansion valve (EAV). Consequently, the liquid ammonia changesfrom the high-temperature and high-pressure state to the low-temperatureand low-pressure state. After that, the ammonia flows into the evaporatorand evaporates. The evaporation process of the ammonia inside theevaporator produces the refrigeration effect, and the chilled airtransfers the refrigerating capacity into the cabin. Finally, theammonia is sorbed by the HTS bed.

In order to determine the systemperformance, the theoretical model is established first.

Theoretical model of the system

Model of sorption beds

Considering the fact that the sorptionbed consists of many unit tubes, the sorption bed is divided intoseveral zones along the air/exhaust gas flow direction, as illustratedin Fig. 3(a), and the partitioning calculation method is proposed.It is assumed that the heat transfer of unit tubes in each zone isthe same and the ammonia desorbed from one zone is not sorbed by otherzones. Consequently, the temperature of the air/exhaust gas outletof the last zone is the inlet temperature of the next zone, as expressedin Eq.(1), and the instantaneous refrigerating capacity of the systemis the sum of the instantaneous refrigerating capacity of each zone.

(1)T_{(P-\mathrm{1})-P,\mathrm{out}}^{\left(i\right)}=T_{P-(P+\mathrm{1}),\mathrm{in}}^{\left(i\right)}.

The mathematical model of a unittube displayed in Fig. 3(b) is established. Compound sorbent is filledin the unit tube. The mass ratio between the metal chloride and theENG-TSA is 5:1, and the packing density of the compound sorbent isapproximately 500 kg·m^–3.The length and the diameter of the unit tube are 460 mm and 42 mm,and the external and the internal diameter of the compound sorbentare 38 mm and 12 mm.

The heat-conduction differentialequation of the unit tube in circular cylindrical coordinates is

(2){\rho }_{\mathrm{sor}}{C}_{\mathrm{sor}}{\displaystyle \frac{\partial T}{\partial t}}={\displaystyle \frac{\partial }{\partial z}}\left({\lambda }_{\mathrm{z}}{\displaystyle \frac{\partial T}{\partial z}}\right)+{\displaystyle \frac{\mathrm{1}}{r}}{\displaystyle \frac{\partial }{\partial r}}\left(r{\lambda }_{\mathrm{r}}{\displaystyle \frac{\partial T}{\partial r}}\right)+S,

where r_sor and C_sor are the density and specific heat of thecompound sorbent, l_z and l_r are the thermal conductivityof the compound sorbent in the axial and radial direction, and S is the internal heat source item, whichis

(3)S={\displaystyle \frac{n_{{\mathrm{NH}}_3}}{V}}{\displaystyle \frac{\mathrm{d}X}{\mathrm{d}t}}\mathrm{\Delta }H,

where n_NH3 is the cyclicmolar number of ammonia, V is thevolume of the compound sorbent, dX/dt is the reaction conversion,and ΔH is the reaction heatof the sorbent.

The kinetic model of sorbent MnCl₂ for the synthesis and decomposition are expressed inEqs. (4) and (5) [23].

(4){\displaystyle \frac{\mathrm{d}X}{\mathrm{d}t}}=Ar{\left(\mathrm{1}-X\right)}^{\mathrm{Mr}}{\displaystyle \frac{P_{\mathrm{c}}-P_{\mathrm{eq}}\left(T\right)}{P_{\mathrm{c}}}},

(5){\displaystyle \frac{\mathrm{d}X}{\mathrm{d}t}}=\mathrm{Ar}{X}^{\mathrm{Mr}}{\displaystyle \frac{P_{\mathrm{c}}-P_{\mathrm{eq}}\left(T\right)}{P_{\mathrm{c}}}},

where Ar and Mr are constants, which are 1.0187×10⁻³ s⁻¹ and 1.185,respectively; P_eq(T) and P_c are the equilibrium pressure of the reaction and the control pressure. P_eq(T) is

(6)\ln \left(P_{\mathrm{eq}}\left(T\right)\right)=-{\displaystyle \frac{\mathrm{\Delta }H}{RT}}+{\displaystyle \frac{\mathrm{\Delta }S}{R}},

where ΔS is theentropy change of the reaction.

For the external surface of the unittube

(7)-{\lambda }_{\mathrm{r}}{\displaystyle \frac{\partial T}{\partial r}}={\alpha }_{\mathrm{exh}}\left(T_{\mathrm{w}}-T_{\mathrm{f}}\right)r=\pm {{\displaystyle r}}_{\mathrm{3}},

where α_exh is theconvective heat transfer coefficient, which is

(8){\alpha }_{\mathrm{exh}}=\epsilon \times \mathrm{0.35}\times {\displaystyle \frac{{\lambda }_{\mathrm{exh}}}{\left(\mathrm{2}\times r_{\mathrm{tube}}\right)}}\times {\left({\displaystyle \frac{B_{\mathrm{1}}}{B_{\mathrm{2}}}}\right)}^{\mathrm{0.2}}{\mathrm{Re}}_{\mathrm{exh}}^{\mathrm{0}\mathrm{.6}}{\mathrm{Pr}}_{\mathrm{exh}}^{\mathrm{0.36}}{\left({\mathrm{Pr}}_{\mathrm{exh}}/{\mathrm{Pr}}_{\mathrm{w}}\right)}^{\mathrm{0.25}},{\displaystyle \frac{B_{\mathrm{1}}}{B_{\mathrm{2}}}}\le \mathrm{2},

where ε is thecorrection factor, l_exh is the thermal conductivity of the exhaustgas, and B₁ and B₂ are the longitudinal space and horizontal space of unit tubes.

For the internal surface of the unittube

(9)-{\lambda }_{\mathrm{r}}{\displaystyle \frac{\partial T}{\partial r}}=\mathrm{0}r=\pm r_1\mathrm{.}

At the beginning, the temperatureof the unit tube is

(10)T={{\displaystyle T}}_{\mathrm{c}},t=\mathrm{0.}

The instantaneous refrigerating capacityof the unit tube

(11)Q_{\mathrm{ins,tube,ref}}=\mathrm{\Delta }{M}_{\mathrm{ins,tube,ref}}\times L={\displaystyle \frac{\mathrm{d}X}{\mathrm{d}T}}\times n_{{\mathrm{NH}}_3}\times M_{{\mathrm{NH}}_3}\times L,

where ΔM_ins,tube,ref is the instantaneous molar number of theammonia sorbed by the compound sorbent in the unit tube, and L is the latent heat of the vaporization.

When exhaust gas flows through theunit tube, the temperature difference of the exhaust gas is

(12)\mathrm{\Delta }{T}_{\mathrm{exh,tube}}=T_{\mathrm{exh,tube,in}}-T_{\mathrm{exh,tube,in}}={\displaystyle \frac{Q_{\mathrm{tube}}}{C_{\mathrm{exh}}{m}_{\mathrm{exh,tube}}}}={\displaystyle \frac{{\alpha }_{\mathrm{exh}}\times A_{\mathrm{tube}}\times \left(T_{\mathrm{exh,tube,in}}-T_{\mathrm{w}}\right)}{C_{\mathrm{exh}}{m}_{\mathrm{exh,tube}}}},

where A_tube is the external surface area of the unit tune, and T_w is the wall temperatureof the unit tube.

The average heat input Q_{ave, heat} of thesorption bed in a cycle

(13)Q_{\mathrm{ave,heat}}={\displaystyle \frac{{\displaystyle {\int }_{\mathrm{0}}^{t_{\mathrm{cycle}}}{C}_{\mathrm{exh}}.m_{\mathrm{exh}}.\left(T_{\mathrm{exh, in}}-T_{\mathrm{exh, out}}\right)\mathrm{d}t}}{t_{\mathrm{cycle}}}},

where T_exh,in and T_{exh, out} are the exhaust inlet and outlet temperaturesof the sorption bed.

Model of the evaporator

The instantaneous refrigerating capacityof the evaporator is

(14)Q_{\mathrm{ins,ref}}=C_{\mathrm{chilled air}}.m_{\mathrm{chilled air}}.\left(T_{\mathrm{chilled air,in}}-T_{\mathrm{chilled air,out}}\right),

where T_{chilledair, in} and T_{chilled air, out} are the chilled air inlet and outlettemperatures of the evaporator.

The average refrigerating capacity Q_{ave, ref} in a cycle

(15){\displaystyle \frac{{\displaystyle {\int }_{\mathrm{0}}^{t_{\mathrm{cycle}}}{C}_{\mathrm{chilled air}}.m_{\mathrm{chilled air}}.\left(T_{\mathrm{chilled air,in}}-T_{\mathrm{chilled air,out}}\right)\mathrm{d}t}}{t_{\mathrm{cycle}}}}.Q_{\mathrm{ave, ref}}=

Model of the expander

The average shaft power of the expanderin a cycle

(16){\displaystyle =\frac{\mathrm{1}}{t_{\mathrm{cycle}}}}{\int }_{\mathrm{0}}^{t_{\mathrm{cycle}}}{W}_{\mathrm{ins,exp}}\mathrm{d}t=\frac{\mathrm{1}}{t_{\mathrm{cycle}}}{\displaystyle {\int }_{\mathrm{0}}^{t_{\mathrm{cycle}}}{m}_{\mathrm{ref,exp}}\times {\eta }_{\mathrm{s,exp}}\times \left(h_{\mathrm{HTS,out}}-h_{\mathrm{exp,out,s}}\right)\mathrm{d}t}=\frac{\mathrm{1}}{t_{\mathrm{cycle}}}{\displaystyle {\int }_{\mathrm{0}}^{t_{\mathrm{cycle}}}{m}_{\mathrm{ref,exp}}\times \left(h_{\mathrm{HTS,out}}-h_{\mathrm{MTS,in}}\right)\mathrm{d}t},W_{\mathrm{ave,exp}}

where W_ins,exp is the instantaneous shaft power ofthe scroll expander, m_ref,exp is the mass flow rate desorbed from the HTS bed, h_s,exp is the isentropic efficiency of the expander, h_HTS,out and h_MTS,in are the inlet and outletenthalpy of the expander, and h_exp,out,s is the isentropic outlet enthalpyof the expander.

Performance evaluation of the system

The coefficient of performance (COP)of the system

(17)\mathrm{COP}={\displaystyle \frac{Q_{\mathrm{ave, ref}}+W_{\mathrm{ave,exp}}}{Q_{\mathrm{ave, heat}}+W_{\mathrm{ave, fan}}}},

where W_ave,fan is the averagepower consumed by the evaporator fan and the condenser fan in a cycle.

The exergy efficiency of the system

(18){{\displaystyle \eta }}_{\mathrm{exergy}}={\displaystyle \frac{{{\displaystyle E}}_{\mathrm{ave},\mathrm{ref}}+{{\displaystyle W}}_{\mathrm{ave},\exp }}{{{\displaystyle {{\displaystyle E}}_Q}}_{\mathrm{ave},\mathrm{heat}}+{{\displaystyle W}}_{\mathrm{ave},\mathrm{fan}}}}={\displaystyle \frac{{{\displaystyle Q}}_{\mathrm{ave},\mathrm{ref}}\cdot \left({{\displaystyle T}}_{\mathrm{0}}/{{\displaystyle T}}_{\mathrm{ave},\mathrm{chilled}\mathrm{air}}-\mathrm{1}\right)+{{\displaystyle W}}_{\mathrm{ave},\exp }}{{{\displaystyle Q}}_{\mathrm{ave},\mathrm{heat}}\cdot \left({{\displaystyle \mathrm{1}-T}}_{\mathrm{0}}/{{\displaystyle T}}_{\mathrm{ave},\mathrm{exh}}\right)+{{\displaystyle W}}_{\mathrm{ave},\mathrm{fan}}}},

where E_ave,ref and E_Qave,heat are thecold exergy and the heat exergy, respectively; T_{ave,chilled air} is the averagetemperature of the chilled air; and T_ave,exh is the average temperature of theexhaust gas.

The specific cooling power per kilogramof the sorbent

(19)\mathrm{SCP}={\displaystyle \frac{Q_{\mathrm{ave, ref}}}{M_{\mathrm{tot, sor}}}}.

Results and discussion

Simulation results

Before the simulation is conducted,the following assumptions are made.

(1) The heat conduction of the compoundsorbent in the axial direction is neglected. Therefore, the temperatureof the compound sorbent at the same radius is almost the same.

(2) The operating pressure is constantin the desorption/sorption process.

(3) The heat exchanger capabilityof the evaporator and the condenser is high enough.

Based on the aforementioned modeland assumptions, the performance of the unit tube and the sorptionbed can be simulated by Matlab, and the property parameter is obtainedfrom the Refprop (NIST Standard Reference Database 23, Version 8.0).

Figure 4(a) exhibits the shaft powerof the expander at different expansion ratios of 1.8, 1.6, 1.4 and1.2, and the isentropic efficiency of the expander is assumed to be0.7. The shaft power increases at first and then decreases, and therelatively high expansion ratio corresponds to the large shaft power.When the expansion ratio is 1.6 and 1.2, the maximum shaft power is450 W and 190 W, respectively. Relatively high expansion ratios willlead to the small outlet pressure of the expander at the constantinlet pressure. As a result, the working pressure of the MTS sorptionbed will become lower. Based on the metal chloride and ammonia reactionequilibrium lines [24],the relatively low working pressure corresponds to a relatively lowsorption/desorption equilibrium temperature, and the low sorptionequilibrium temperature requires the low cooling temperature, whichwill hinder the system from adapting to the harsh heat transfer conditions.Consequently, an appropriate expansion ratio depends on the actualheat transfer conditions.

At the beginning, the total instantaneousrefrigerating capacity shown in Fig. 4(b) increases sharply from 0kW to 5.8 kW in a short time, due to the fast sorption reaction rate.After that the refrigerating capacity decreases to 2 kW along withthe gradual decrease of the reaction rate.

Experimental results

The experimental system of the solidsorption combined cooling and power cycle is established, as demonstratedin Fig. 5. For the heating process of sorption beds, the hot air heatedby the electric heater, whose maximum power is 21 kW, is utilizedto simulate the exhaust gas. The hot air temperature of the electricheater can be adjusted from the ambient temperature to 250°C withan accuracy of±3°C. To consume the refrigerating capacity,the electric heater controlled by the temperature controller is installedin the cabin. Temperature sensors (PT100) and pressure transmitters(YSZK-31) are mounted in the experimental system, whose accuraciesare±0.1°C and 0.3%R, respectively. The flow rate of theair is measured by the Pitot tube anemograph (ZC1000-1F). The inletvolumetric flowrate of the expander is measured by a vortex flowmeter(XRLUGB) with an accuracy of±1.5%R. A data acquisition unitof the Agilent 34970A is utilized for the collection of the data.

The working pressure of the sorptionbeds at the resorption time of 30 min is depicted in Fig. 6(a).Thehot air temperature and ambient temperature are 210°C and 15°C,respectively. From 0 min to 30 min, the power generation process betweenthe HTS bed and the MTS bed occurs. In the power generation process,the working pressure of the MnCl₂ bed and theCaCl₂ bed increases with the time at the beginningand then maintains almost constant. Additionally, the working pressureof the MnCl₂ bed is always higher than thatof the CaCl₂ bed, due to the fact that an expanderis installed between two sorption beds. For example, the working pressureof the MnCl₂ bed and the CaCl₂ bed is 0.40 MPa and 0.28 MPa at the time of 15 min, and the expansionration is approximately 1.4.

The ammonia inlet and outlet temperaturesof the scroll expander are shown in Fig. 6(b). After the MnCl₂ bed is heated for about 5 min, the ammonia valve AV4is opened. The high-temperature and high-pressure ammonia flows intothe expander and expands. The ammonia inlet temperature of the expanderincreases rapidly from 41°C to 72°C, and then continues torise to about 104°C. Meanwhile, the outlet temperature of theexpander increases gradually from 20°C to 52°C. After 30 min,the inlet and outlet temperatures of the expander decrease gradually,due to the fact that the ammonia valve AV4 is closed.

The system performance at the resorptiontime of 30 min is shown in Fig. 7, and the refrigerating temperatureis controlled at approximately –10°C. The shaft power ofthe expander increases immediately from 0 W to 241 W at the momentwhen the AV4 is opened. After that, the shaft power gradually increases,and a maximum instantaneous shaft power of 323 W is obtained. Then,the shaft power slowly decreases to 289 W. When the power generationprocess is completed, the sorption process of the HTS bed occurs andthe corresponding refrigerating capacity is output. When the chilledair outlet temperature drops to –10°C, the electric heaterinstalled in the cabin begins to heat the chilled air. After that,the chilled air inlet temperature increases gradually. When the chilledair outlet temperature rises to –10°C, the electric heaterstops heating the chilled air. Based on Eq. (14), the instantaneousrefrigerating capacity is calculated, and it fluctuates continuously,which is shown in Fig. 7(b). The refrigerating capacity is outputonly at the sorption process of the HTS bed.

It can be found that there are somedifferences between the simulation results shown in Fig. 4 and theexperimental results shown in Fig. 7. For the shaft power of the expanderand the refrigerating capacity, the variation tendency in the experimentis similar to that in the simulation, and they increase at first andthen decrease. For the shaft power, the expansion ratio in the experimentfluctuates continually and ranges from 1.3 to 2.0. When the ammoniavalve AV4 is opened, the inlet pressure of the expander increasesgradually from 0.3 MPa to 0.45 MPa and then maintains almost constant.The unstable inlet pressure of the expander leads to the unstableexpansion ratio. In the simulation, the inlet pressure of the expanderis assumed to be constant and is 0.4 MPa.

For the refrigerating capacity, thetime of the maximum value in the experiment comes a little later thanthat in the simulation. This is mainly due to the fact that the temperatureof the evaporator increases gradually during the power generationprocess, and some refrigerating capacity is consumed by the evaporatorin the experiment. In the simulation, the refrigerating capacity consumedby the evaporator is neglected. Simultaneously, the instantaneousrefrigerating capacity presents jagged fluctuations in the experiment,mainly due to the temperature controller in the cabin. When the chilledair outlet temperature of the evaporator drops to the temperatureset, the electric heater installed in the cabin begins to heat thechilled air. When the chilled air outlet temperature rises to thetemperature set, the electric heater stops heating the chilled air.

In addition, different resorptiontimes of 20 min to 40 min are investigated experimentally. The differentresorption times of 20 min, 30 min and 40 min correspond to the differentcycle times of 44.5 min, 64.5 min and 81.5 min, and also correspondsto the different effective shaft power times of 17 min, 25 min and34 min, as shown in Fig. 8(a). The effective shaft power time refersin particular to the time that the shaft power is being output bythe expander. A longer resorption time can effectively extend theshaft power time, due to the fact that much more ammonia is desorbedfrom the HTS bed at a relatively long resorption time. The averagerefrigerating capacity in a cycle at the resorption times of 20 min,30 min and 40 min is 1.89 kW, 1.95 kW, and 1.88 kW, and the averageshaft power in a cycle is 92.4 W, 109.2 W, and 114.7 W.

For solid sorption refrigerationsystems driven by the exhaust waste heat of the engine, since theevaporator and the condenser are generally equipped with fans, someelectricity is unavoidably consumed. Taking the resorption time of30 min as an example, the average power in a cycle consumed by thefans is about 162 W, and under this operational condition, the averageshaft power in a cycle that the expander outputs is about 109.2 W.If the conversion efficiency between the shaft power and the electricityis 0.9, the system can provide 60% of the power for the fans.

Finally, the performance of the systemis compared with that of the earlier-established solid sorption refrigerationsystem. The detailed performance comparison of the two systems isshown as follows.

For the combined cooling and powersystem, the average refrigerating capacity in a cycle at the resorptiontimes of 20 min, 30 min, and 40 min is 1.89 kW, 1.95 kW, and 1.88kW. For the refrigeration system, the average refrigerating capacityin a cycle at the resorption times of 20 min, 30 min, and 40 min is1.9 kW, 2.06 kW, and 1.87 kW. Consequently, the average refrigeratingcapacity of the two systems is almost the same, and the power generationprocess does not influence the refrigeration process.

Because the combined cooling andpower system can supply a part of the power for the fans, the COPand exergy efficiency of the combined cooling and power system ishigher than that of the refrigeration system, which is shown in Fig.9. The COP of the two systems is 0.22 and 0.208 at the resorptiontime of 30 min and the exergy efficiency is 0.076 and 0.047.

Conclusions

Based on the earlier-establishedsolid sorption refrigeration system, a novel solid sorption combinedcooling and power cycle driven by exhaust waste heat is proposed inthis paper. The main goal of the cycle is to output the refrigeratingcapacity, and enable the system itself to use the electricity. Boththe simulation and experiments are conducted and conclusions are madeas follows:

The theoretical model of the solidsorption combined cooling and power system is established. For thesorption beds consisting of many unit tubes, the partitioning calculationmethod is proposed. The simulation results show that the maximum instantaneousshaft power at the expansion ratio of 1.6 is approximately 450 W,and the instantaneous refrigerating capacity increases from 0 kW to5.8 kW and then decreases to 2 kW.

The experimental results show thatwhen the refrigerating temperature is controlled at –10°C,the average refrigerating capacity in a cycle at the resorption timesof 20 min, 30 min, and 40 min is 1.89 kW, 1.95 kW, and 1.88 kW, andthe average shaft power in a cycle is 92.4 W, 109.2 W, and 114.7 W.

The experimental results show thatthis system can provide approximately 60% of the power for the evaporatorfan and the condenser fan. Based on the simulation results, if theexpansion ratio of 1.6 can be maintained in the power generation process,the electricity of the system output is high enough to meet the electricitydemand of the system.

The performance of the solid sorptioncombined cooling and power system is compared with that of the solidsorption refrigeration system. The average refrigerating capacityof the two systems is almost the same under the same operational condition,but the COP and exergy efficiency of the combined cooling and powersystem is higher than that of the refrigeration system. The COP ofthe two systems is 0.22 and 0.208 at the resorption time of 30 minand the exergy efficiency is 0.076 and 0.047.

The feasibility of the solid sorptioncombined cooling and power system is determined, and the MnCl₂/CaCl₂-NH₃ working pairs are employed. The system is especially suitable forexhaust waste heat recovery. In order to make the system provide thecooling capacity output ceaselessly, the heat storage device willbe integrated with the evaporator in future research.

References

[1] Wang R Z, Oliveira  R G. Adsorption refrigeration—an efficient way to make good use of waste heatand solar energy. Progress in Energy &Combustion Science, 2006, 32(4): 424–458

[2] Fernandes M S, Brites G J V N, Costa J J, Gaspar  A R, Costa V A F. Review and future trends of solar adsorption refrigeration systems. Renewable & SustainableEnergy Reviews, 2014, 39(6): 102–123

[3] Hamdy M, Askalany  A A, Harby  K, Kora N. An overview on adsorption cooling systems powered bywaste heat from internal combustion engine. Renewable & Sustainable Energy Reviews, 2015, 51: 1223–1234

[4] Zhong Y, Fang  T, Wert K L. An adsorption air conditioning systemto integrate with the recent development of emission control for heavy-dutyvehicles. Energy, 2011, 36(7): 4125–4135

[5] Wu W D, Zhang  H, Men C L. Performance of a modified zeolite 13X-wateradsorptive cooling module powered by exhaust waste heat. International Journal of Thermal Sciences, 2011, 50(10): 2042–2049

[6] Sharafian A, Bahrami  M. Assessment of adsorber bed designs in waste-heat driven adsorptioncooling systems for vehicle air conditioning and refrigeration. Renewable & Sustainable Energy Reviews, 2014, 30(1): 440–451

[7] Sharafian A, Mc Cague  C, Bahrami M. Impact of fin spacing ontemperature distribution in adsorption cooling system for vehicleA/C applications. International Journal of Refrigeration, 2015, 51: 135–143

[8] Sharafian A, Nemati Mehr  S M, Huttema  W, Bahrami M. Effects of different adsorber bed designs on in-situ water uptake rate measurements of AQSOA FAM-Z02 for vehicle air conditioningapplications. Applied Thermal Engineering, 2016, 98: 568–574

[9] Sharafian A, Nemati Mehr  S M, Thimmaiah  P C, Huttema  W, Bahrami M. Effects of adsorbent mass and number of adsorber beds on the performanceof a waste heat-driven adsorption cooling system for vehicle air conditioningapplications. Energy, 2016, 112: 481–493

[10] Verde M, Harby  K, de Boer R,  Corberán J M. Performance evaluation of a waste-heat driven adsorption system forautomotive air-conditioning: Part I—Modeling and experimentalvalidation. Energy, 2016, 116: 526–538

[11] Verde M, Harby  K, de Boer R,  Corberán J M. Performance evaluation of a waste-heat driven adsorption system forautomotive air-conditioning: Part II— Performance optimization underdifferent real driving conditions. Energy, 2016, 115(360–5442): 996–1009

[12] Gao P, Zhang  X F, Wang  L W, Wang  R Z, Li  D P, Liang  Z W. Study on MnCl₂/CaCl₂–NH₃ two-stage solid sorptionfreezing cycle for refrigerated trucks at low engine load in summer. Energy Conversion & Management, 2016, 109: 1–9

[13] Gao P, Wang  L W, Wang  R Z, Zhang  X F, Li  D P, Liang  Z W. Experimental investigation of a MnCl₂/CaCl₂-NH₃ two-stage solid sorption freezing system for a refrigerated truck. Energy, 2016, 103: 16–26

[14] Gao P, Wang  L W, Wang  R Z, Li  D P, Liang  Z W. Optimization and performance experiments of a MnCl₂/CaCl₂–NH₃ two-stage solid sorption freezing system for a refrigerated truck. International Journal of Refrigeration, 2016, 71: 94–107

[15] Boretti A. Recovery of exhaust and coolant heat with R245fa organicRankine cycles in a hybrid passenger car with a naturally aspiratedgasoline engine. Applied Thermal Engineering, 2012, 36(1): 73–77

[16] Yang F, Dong  X, Zhang H,  Wang Z, Yang  K, Zhang J,  Wang E H,  Liu H, Zhang  G Y. Performance analysis of waste heat recovery with a dual loop organicRankine cycle (ORC) system for diesel engine under various operatingconditions. Energy Conversion & Management, 2014, 80(2): 243–255

[17] Wang L W, Roskilly  A P, Wang  R Z. Solar powered cascading cogeneration cycle with ORC and adsorption technology forelectricity and refrigeration. Heat Transfer Engineering, 2014, 35(11–12): 1028–1034

[18] Jiang L, Wang  L, Wang R,  Gao P, Song  F. Investigation on cascading cogeneration system of ORC (Organic Rankine Cycle) and CaCl₂/BaCl₂ two-stage adsorption freezer. Energy, 2014, 71(12): 377–387

[19] Jiang L, Wang  R Z, Wang  L W, Gao  P, Zhu F Q. Investigation on gradientthermal cycle for power and refrigeration cogeneration. International Journal of Refrigeration, 2017, 76: 42–51

[20] Wang L, Ziegler  F, Roskilly A P,  Wang R Z,  Wang Y. A resorption cycle for the cogeneration of electricity and refrigeration. Applied Energy, 2013, 106(106): 56–64

[21] Bao H, Wang  Y, Charalambous C,  Lu Z, Wang  L, Wang R Z. Chemisorption cooling and electric powercogeneration system driven by low grade heat. Energy, 2014, 72(7): 590–598

[22] Bao H, Wang  Y, Roskilly A P. Modelling of a chemisorptionrefrigeration and power cogeneration system. Applied Energy, 2014, 119(12): 351–362

[23] Lu H B, Mazet  N, Spinner B. Modelling of gas-solid reaction—couplingof heat and mass transfer with chemical reaction. Chemical Engineering Science, 1996, 51(15): 3829–3845

[24] Neveu P, Castaing  J. Solid-gas chemical heat pumps: field of application and performanceof the internal heat of reaction recovery process. Heat Recovery Systems & Chp, 1993, 13(3): 233–251

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag GmbHGermany