Experimental studies on combined cooling and power system driven by low-grade heat sources

Energy 128 (2017) 801e812

Contents lists available at ScienceDirect

Energy journal homepage: www.elsevier.com/locate/energy

Experimental studies on combined cooling and power system driven by low-grade heat sources G. Praveen Kumar a, R. Saravanan a, Alberto Coronas b, * a b

Department of Mechanical Engineering, Anna University, Chennai 600025, India CREVER. Mechanical Eng. Dep., Universitat Rovira i Virgili, Tarragona 43007, Spain

a r t i c l e i n f o

a b s t r a c t

Article history: Received 8 December 2016 Received in revised form 5 April 2017 Accepted 12 April 2017 Available online 21 April 2017

An experimental investigation was undertaken to study the actual useful output and performance of a combined power and cooling system that uses low-grade energy. The cycle used was a combination of NH3-H2O absorption refrigeration cycle and Kalina extraction turbine cycle. The expected performance characteristics of the dual output system were first evaluated using an energetic and exergetic approach based on the quality of useful outputs; in the experimental confirmation. It was evaluated in Cooling Alone mode (CA mode) and Combined Cooling-Power mode (CCP mode), for the same operating conditions. The weak solution flow rate and generator temperature were maintained constant at 0.237 kg/s and 133
C respectively throughout the experimental run. The maximum cooling load of 34.26 kW was achieved with a COP's of 0.57 in CA mode. In CCP mode, the system was operated at a split ratio of 0.5 with the useful cooling load of 15.26 kW and estimated expander load of 2.21 kW respectively, with power to cooling ratio of 0.14. The corresponding effective first-law and exergetic efficiencies were 13% and 48%. This study provides a feasible and flexible way to meet the desired combination of power/ cooling ratio to generate varying demand profiles using available low-grade heat sources. © 2017 Elsevier Ltd. All rights reserved.

Keywords: Combined cooling-power Kalina cycle Absorption cycle Ammonia-water Effective first-law efficiency

1. Introduction Low and mid-grade grade heat sources (<300
C) are abundantly available from various heat sources such as solar thermal, geothermal, biomass and waste heat from various thermal and chemical processes. The efficient use of these variable temperature heat sources is a crucial technology, and it can be effectively used for conversion to highly valuable products such as electricity, airconditioning, and refrigeration. Non-azeotropic nature of ammonia-water binary mixture has a relatively low bubble point temperature and it can evaporate in a wide range of temperatures, making it suitable for low and mid-grade heat source utilization. In addition, the broad evaporation temperature of working fluid can provide cooling as well as refrigeration to end users [1]. Kalina proposed a novel power cycle, using this mixture in bottoming cycle and has claimed higher thermal efficiency of 30e60% than comparable with conventional steam Rankine bottoming cycle [2e4]. In order to validate the feasibility of Kalina cycle, pilot power plants have been built and their performance

* Corresponding author. E-mail address: [email protected] (A. Coronas). http://dx.doi.org/10.1016/j.energy.2017.04.066 0360-5442/© 2017 Elsevier Ltd. All rights reserved.

have been investigated [5]. Using solar thermal energy with a maximum temperature of 130
C, Kalina cycle was activated, and it was concluded that performance factor was sensitive to the temperature difference between maximum and minimum temperature of the cycle [6]. By utilizing the geothermal resources in Kalina cycle, the energy and exergy analyses were conducted and it was found that the cycle could be optimized for the turbine outlet pressure and mass fraction of the working fluid [7]. Absorption refrigeration cycle using ammonia-water mixture also plays a major role in low-grade heat source recovery. In recent years, integrated combined power and cooling cycles have been proposed for effective utilization of primary energy. Moreover, the integrated cycles are broadly classified into two groups, with both outputs in the same loop and through different loop. In 1995, Goswami [8] proposed an integrated cycle based on conventional Rankine power and absorption refrigeration cycles for simultaneous power and cooling in the same loop. Further researchers studied the thermodynamically performance of Goswami cycle [9,10]. In Goswami cycle, very high concentration of ammonia vapor is expanded in the turbine to a very low temperature without condensation, which is further used for sensible cooling in the evaporator. Although the cycle can provide power and cooling

802

G.P. Kumar et al. / Energy 128 (2017) 801e812

outputs in the same loop, but power output as a primary goal. Hence, to obtain larger cooling output in the cycle, the cooling effect produced in the evaporator is achieved through the latent heat of evaporation. Zheng et al. [11] proposed a novel absorption power and cooling cycle based on the Kalina cycle with improved performance due to its increased capacity for both power and cooling outputs. The cycle was modified by including the condenser and an evaporator was introduced between the rectifier and second absorber. To enhance the purity of the refrigerant, the flash tank in the Kalina cycle was replaced by a rectifier for more refrigeration output. With these modifications the cycle was able to simultaneously provide dual output. Ziegler [12] proposed the double effect absorption cycle for dual output, which was valid when temperature glide in the heat source was large. In the cycle, separate cooling and power production of 1.2 MW and 100 kW respectively was obtained using thermal energy of 1 MW along with freedom to adjust ratio between cooling to power production. Liu and Zhang [13] developed a series connected dual output system, where a splitting/absorption unit was integrated with Rankine cycle and ammonia refrigeration cycle. The cycle could meet different concentration requirements in heat addition and the condensation processes. Zhang and Lior [14] developed three different cycle configurations namely series, parallel and compound for combined power and cooling. The author also summarized some guidelines for integration of refrigeration and power cycles based on energy and exergy efficiencies. Although the energy and exergy efficiencies were appreciably better, the cycles required high driving source temperatures (450
C). Moreover, the cycles were relatively complicated and necessitated higher capital investment. Wang et al. [15] simplified Zhang's parallel combined refrigeration and power system driven by low-grade heat source. Sun et al. [16] proposed an ammonia-water based power/cooling generation system capable of utilizing low and midgrade heat sources. Power and refrigeration sub-cycles were driven by high and low temperature portions of cascading waste heat respectively. Compared with the primary energy of separate power and refrigeration systems, the combined system consumed 17.1% less heat with the same output. Yu et al. [17] proposed a novel system with adjustable power to cooling ratios. In the proposed system, a modified Kalina and absorption refrigeration sub-cycles were interconnected by splitters, mixers, absorbers and heat exchangers. Lopez-Villada et al. [18] developed a single-stage power and cooling system and compared it with Goswami cycle. Suitability of different solar thermal collectors as a heat source (partly assisted with a load of the conventional heating system) for CCP system was reviewed with variable power to cooling ratio. A simplified CCP cycle based on Kalina and absorption refrigeration was proposed by Wang et al. [19]. To determine the irreversibilities, the exergy analysis was carried out for the major components in the cycle, in which the condenser and exhausted gas from vapor generator had major exergy destruction of 18.7% and 68.34% respectively. Experimental studies are of utmost importance to validate the feasibility of combined power and cooling systems. However, very few experimental studies on ammonia water CCP cycles have been found in literature and a majority of the studies is conducted on prototype setup. The first prototype study of CCP cycle based on Goswami cycle was carried out by Tamm et al. [20]; the authors investigated the cycle both experimentally and theoretically. Through the parametric study, the system could be optimized for first and second law efficiencies besides cooling and power outputs. When the real losses were considered in the analysis, the power and cooling outputs, and the thermal efficiency decreased by 11.8%, 37.7% and 20.6% respectively. There was a good agreement between theoretical and experimental results. Martin [21] conducted

experimental studies on a combined system driven by low temperature thermal sources (<200
C). In the Tamm experimental setup, the rectifier and turbine were included to condition the vapour and to extract power from the working fluid respectively, demonstrating that the turbine exhaust was at sub-ambient temperatures. Demirkaya et al. [22] continued the investigation by using the same experimental arrangement, except for the expansion device, which was replaced by a scroll type expander. The experimental results showed that expander isentropic efficiency was between 30 and 50% and it performed well when the inlet vapour was superheated. Han et al. [23] developed an experimental rig for a CCP system in which the power generation was simulated using a vapor heat exchanger since low capacity efficient expanders are not available in the market. The system was able to provide the output based on user energy demand. When the system was operated as a refrigeration model, the cooling output was 11.67 kW with corresponding COP of 0.465, while on power generation model the simulated net output was 1.02 kW at the pressure ratio of 4. Even though, the combined power and cooling systems are experimentally studied previously by other researchers, the power generation output is maximum of 300e400 W only with cooling output less than 5 kW. In this paper, an attempt is made to design and develop a system which enable the simultaneous production of 15 kW cooling and up to 2 kW net power output and 35 kW only on cooling alone mode. The combined system has integrated generator with solution-cooled rectifier, generator heat exchanger designed to reduce the heat input. By using low-grade heat sources, the CCP system output can be varied from cooling alone to power alone mode along with intermediate operating conditions to meet the demand throughout the year In the CCP system, the experimental study focusses mainly on typical conditions driven by lowtemperature heat sources for cooling alone and dual output condition. Additionally, a sensitivity study is also conducted in CCP system for varied heat source temperatures. 2. Experimental setup and procedure The experimental setup for combined power and cooling cycle driven by a low-grade heat source is shown in Figs. 1 and 2. The designed cooling capacity of the system, at cooling temperature of 10
C, is 35 kW under the cooling alone and 2 kw of power capacity at power alone condition. The P-T-X diagram, P-h diagram and T-s diagram for the base condition operation of combined cooling and power are shown in Fig. 3. The weak and strong solution concentration values for the base condition are 0.3 and 0.22 respectively. The weak and strong concentration values tally with the equilibrium pressure and temperature conditions of the absorber and the generator. 2.1. Experimental setup The experimental setup comprises of an integral absorption refrigeration/power generation system, pressurized hot water heat source simulator, cooling water unit and brine solution secondary refrigeration unit. The absorption refrigeration/power generation system includes both cooling and power generation subsystems. The plant operates in a single stage ammonia-water absorption cycle with parallel flow refrigerant supply arrangement for dual output. The condenser, Condensate Pre-Cooler (CPC) and an evaporator are the major components of the cooling subsystem. For power generation subsystem, an efficient expander with low power generation is not available in the market, hence an orifice plate suitable for a typical pressure ratio of 9.0 (pressure ratio ¼ P16/P17) is used to simulate the expansion process in the experimental

G.P. Kumar et al. / Energy 128 (2017) 801e812

803

Fig. 1. Schematic arrangement of the experiment setup.

setup. An absorber, solution pump, Solution Heat exchanger (SHX) and a generator integrated with Generator Heat exchanger (GHX) and Solution-Cooled Rectifier (REC) are the major components shared by both cooling and power generation subsystem. All the high and low temperature components are insulated to minimize heat loss. For variable power and cooling output ratio, valves V1 and V2 are used to adjust the respective flow rate of refrigerant to corresponding subsystems. If the valve V1 is opened and V2 is closed, the rectified refrigerant is used only for cooling generation. Based on the V1 and V2 valves adjustment, the system can be independently operated under three different modes - as cooling alone, power alone or simultaneous cooling and power. The hot water simulator provides pressurized hot water in closed loop at constant temperature, which acts as a low temperature heat source to the generator, while the cooling water unit consists of a cooling tower that supplies cooling water to the condenser and absorber in a series connection (cw1-cw2-cw3). The brine solution is a mixture of ethylene glycol at a volumetric ratio of 20% in water, which conveys, in a closed loop, the cooling load from the evaporator to the fan coil unit installed inside a cold room. Low pressure weak solution is pumped by a 2 kW multistage centrifugal pump from the absorber (state point 1) to REC at state point 2. The REC is an indirect counter-current gas-liquid heat exchanger, which preheats the weak solution through high

Fig. 2. Photograph of the combined power and cooling system.

temperature refrigerant vapour produced from the integrated generator. Thus, the primary heat of rectification is transferred to be recovered as preheat of weak solution from the absorber. Then the weak solution enters the SHX. The SHX is a shell and coil heat exchanger, where it exchanges sensible heat from the strong solution to weak solution. The shell side weak solution (state point 3) is heated by the strong solution in the tube side of SHX (state point 6). The weak solution enters the GHX at state point 4. The secondary heat of rectification of the refrigerant vapour and the sensible heat of strong solution are thus utilized to the maximum to heat the weak solution in GHX before it reaches the generator. The heat recovery equipment REC, SHX and GHX function effectively with predicted temperature approach, as can be inferred from Table 3. In the generator, the external heat is supplied by the hot water simulator to generate the refrigerant vapour. The rectified ammonia-rich refrigerant leaves the REC at state point 9. By using valves V1 and V2, the rectified ammonia-rich vapor is bifurcated into two parts (state points 10 and 16) for dual outputs. Table 1 shows the type, heat transfer area and the designed heat duty of the various components of the combined power and cooling system. During cooling alone mode, valve V2 is fully closed whereas valve V1 is open. The refrigerant vapour is condensed and leaves the condenser at state point 11. The liquid refrigerant is subcooled in the CPC by the refrigerant vapour coming from the evaporator before being throttled (state point 13). The low pressure liquid refrigerant exchanges its latent heat of vaporization with the brine solution and produces cooling effect. The chilled brine solution is circulated between the evaporator and fan coil unit (state points B1B2-B3-B4). The low pressure ammonia vapour is returned to the absorber at state point 18. During combined cooling and power generation mode, valves V1 and V2 are opened partially according to the turndown requirements of dual outputs. The refrigerant vapour through valve V2 is used for power generation. The high-pressure refrigerant vapour can enter into the expander for extraction power generation process. After expansion, the low pressure vapour leaves at state point 17. The refrigerant vapour flows through valve V1 and produces a cooling effect, as explained in cooling mode. However, in the combined mode the cooling capacity is comparatively reduced

804

G.P. Kumar et al. / Energy 128 (2017) 801e812

(a) Combined cooling and power cycle on a P-T-X diagram.

(b) Combined cooling and power cycle on a P-h diagram.

(c) Combined cooling and power cycle on a T-s diagram. Fig. 3. Combined cooling and power system on a P-T-X diagram.

G.P. Kumar et al. / Energy 128 (2017) 801e812

respect the cooling mode alone due to the reduced refrigerant flow rate. The refrigerant vapour from the expander and CPC are mixed together to enter into the absorber at state point 18, to complete the cycle. A diesel-fired burner is used to heat the hot water simulator and can be operated between the ranges of 80
Ce200
C. To operate at a particular set temperature, a PID controller is provided to control diesel burner output. The hot water and cooling water volumetric flow rates can be adjusted to maintain the different source and sink temperature requirements.

2.2. Measurement and procedure The temperature and pressure in the main components are measured by four wired PT 100 and pressure transducers respectively. The mass flow rates of the weak solution, strong solution and refrigerant flow are measured by Coriolis flow meters. Table 2 shows the measured variables and uncertainties. All the measurements are stored in a data acquisition system, at regular intervals of 1 min. The system is started with cooling water circulation in the absorber and condenser in closed cycle followed by the internal hot water circulation before switching ON the Diesel fired burner with limiting preset hot source temperature. Once the hot water reaches the set temperature, it is circulated between the hot water simulator and the generator followed by weak solution circulation between the absorber and generator. The high and low pressure of the system are decided according to the temperature reached by the condenser and the evaporator, depending on cooling tower and hot water generator performances. When evaporator pressure is reached, the brine circuit starts to exchange the cooling energy to the cold room. The flow rates of the external circuit cooling water, hot water and brine, temperature and pressure of all the major components, and fuel consumption are noted down periodically. The level of the solution tank and pure ammonia tank are maintained for proper liquid flow distributions. Various operating parameters such as pressure, temperature, flow rates are monitored and once the system is under steady state operation with no significant change in the observed parameters, these values are taken for the analysis to evaluate the system performance.

m10 ¼ m11 ¼ m12 ¼ m13 ¼ m14 ¼ m15 ¼ mrc

(6)

m4 X4 ¼ m5 X5 þ m9 X9

(7)

The circulation ratio (CR) value is the ratio between measured mass flow rates of weak solution and the generated refrigerant vapour.

CR ¼

mws mr

(8)

By using Engineering Equation Solver (EES), the enthalpy and entropy are computed from the measured values of temperature, pressure and estimated concentration. In EES, the external routines of Tillner-Roth and Friend are used for estimation of thermophysical properties [25]. The heat loads of the absorber, condenser, generator and evaporator are calculated by

QA ¼ mcw cp;cw ðTcw3  Tcw2 Þ

(9)

QC ¼ mcw cp;cw ðTcw2  Tcw1 Þ

(10)

QG ¼ mhw cp;hw ðThw1  Thw2 Þ

(11)

QE ¼ mb1 cp;b ðTb2  Tb1 Þ

(12)

where cp;cw , cp;hw and cp;b are specific heat at constant pressure of the cooling water, hot water and brine solution respectively, the values of which are taken from literature according to the pressure and temperature [26,27]. The external fluid temperatures of the system such as cooling water, hot water and brine temperature range from 25
C to 40
C, 130
Ce150
C and 5
C to 25
C respectively. The heat recovery of the solution heat exchanger is,

QSHX ¼ mws ðh3  h4 Þ

(13)

The heat recovery of the generator heat exchanger is,

QGHX ¼ mss ðh5  h6 Þ

(14)

The heat recovery of the solution-cooled rectifier is,

QREC ¼ mws ðh3  h2 Þ

3. Data acquisition and processing

805

(15)

The heat recovery of the condensate pre-cooler is, The concentration of the weak solution, strong solution and refrigerant at saturated conditions are estimated by using the correlation equation (1). The coefficient values of the equation are taken from ASHRAE Handbook Fundamentals [24].

vðT; XÞ ¼

3 X 3 X

ai;j X j T i

(1)

i¼0 j¼0

Global mass balance across the components are as follows:

mss ¼ mws  mr

(2)

mr ¼ mrp þ mrc

(3)

m1 ¼ m2 ¼ m3 ¼ m4 ¼ mws

(4)

m5 ¼ m6 ¼ m7 ¼ m8 ¼ mss

(5)

QCPC ¼ mrc ðh11  h12 Þ

(16)

The expander isentropic efficiency is,

ðh16  h17 Þ  h16  h17;s

hi;EXP ¼ 

(17)

The isentropic efficiency is assumed to be a practical achievable value of 85%. The expander work is,

WEXP ¼ mrp ðh16  h17 Þ

(18)

The net mechanical power is calculated by the relation,

Wnet ¼ WEXP  WP

(19)

In this system, two different thermodynamic outputs, cooling and power are simultaneously produced. Hence, at least two independent parameters are required to evaluate and correlate the

806

G.P. Kumar et al. / Energy 128 (2017) 801e812

Table 1 Main system component details. S. No

Component

Heat exchanger type

Heat transfer area (m2)

Heat duty (kW)

1 2 3 4 5 6 7 8

Absorber Condenser Evaporator Generator Generator heat exchanger Rectifier Solution heat exchanger Condensate pre-cooler

Vertical falling film Shell and Tube Vertical shell and Tube Shell and helical Coil Shell and helical Coil Shell and helical Coil Shell and helical Coil Shell and helical Coil

18.7 13.3 4.78 3.25 1.66 1.33 8.7 1.42

66.7 35.0 35.0 63.6 12.3 10.55 57.7 3.4

overall system performance. Accordingly, the effective first law efficiency, effective exergy efficiency and power/cold ratio are used to evaluate the performance parameters analysis. Overall thermal efficiency is based on the first law of thermodynamics and it estimates the cost-benefit analysis. In this efficiency study, equal weights are assigned to the cooling and power which results in overestimation value of produced energy which is inconsistent with thermodynamic limits [28,29]. Consequently, to ascertain the energetic performance of the system, energy quality is also included in the performance calculations. Thus, for a valid expression, the cooling output would have to be weighted differently by dividing the exergy of cooling to reasonable second law efficiency of vapor compression refrigeration. In this work, a typical value of hII;ref equal to 40% is considered. The exergy efficiency is calculated based on a constant reference temperature.

hI;eff ¼

  : . Wnet þ EX E hII;ref QG

(20)

where, EXE is the exergy of the evaporator and hII;ref is the second law of efficiency for the vapor compression refrigeration cycle. Exergy is calculated based on the reference environment of T0 ¼ 303 K, P0 ¼ 101.325 kPa.

hex;eff ¼

  : . Wnet þ EX E hII;ref EXG

(21)

where, EXG is the exergy of the generator. The exergy of the evaporator, EXE is calculated as the working fluid exergy change across the evaporator.

EXE ¼ mb ½ðhb1  hb2 Þ  T0 ðsb1  sb2 Þ

WEXP QE

(24)

The external COP is calculated as the ratio of evaporator load in brine side and generator load in hot water side.

COPext ¼

QE;ext QG;ext

(25)

The internal COP is calculated as the ratio of evaporator load and generator load on working fluid side.

COPint ¼

QE;int QG;int

(26)

Exergy efficiency is calculated for CA mode, based on the ratio between exergy-change across brine in the evaporator to the hot water in the generator.

hex ¼

EXE EXG

(27)

The uncertainty of directly measured parameters such as the temperature, the pressure and the flow rates are provided by the manufacturers, as shown in Table 1. Uncertainties associated with each of the derived quantities such as solution concentration, circulation ratio, internal COP, external COP and exergy efficiency are estimated by using the law of propagation of uncertainty. The uncertainty of derived quantity Y can be calculated by

" Uy ¼

vy Ux vx1 1

2



vy þ Ux vx2 2

2



vy þ…þ Uxn vxn

2 #1=2 (28)

(22)

The exergy of the generator, EXG is calculated as the hot water exergy difference across the generator.

EXG ¼ mhw1 ½ðhhw1  hhw2 Þ  T0 ðshw1  shw2 Þ

R¼

(23)

Power/cooling ratio R, is defined as the ratio of the net mechanical power output of the system to the cooling effect produced in the evaporator:

Where Ux1, Ux2, …, Uxn are the uncertainties of the directly measured parameters of x1, x2, …, xn respectively. Each variable sensitivity associated with the calculation of Y is represented by partial differential parameters such as

vy vy vx1 , vx2 ,

…,

vy vxn [30].

The value of uncertainty calculated for solution concentration is ±1.0%, circulation ratio is ±1.4%, internal COP is ±2%, external COP is ±3%, effective first-law efficiency is ±4% and effective exergy efficiency is ±4% (see Table 2).

Table 2 Measured parameters and details of uncertainties. Measured parameter

Operation range

Accuracy

1 2 3

Temperature Pressure Solution flow rate & density

4

Refrigerant flow rate & density

50 to 200
C 0 bar 25 bar 0.1 kg/h to 8160 kg/h 0.01 kg/m3 to 950 kg/m3 0.1 kg/h to 450 kg/h 0.01 kg/m3 to 800 kg/m3

±0.5 K ± (0.2% x P) bar Mass flow (liquid): (±0.1%) kg/h Density (liquid): (±0.1%)kg/cm3 Mass flow (liquid): (±0.15%) kg/h Density (liquid): (±0.1%) kg/cm3

G.P. Kumar et al. / Energy 128 (2017) 801e812

807

Table 3 Experimental reading for a typical operating condition in CCP mode. TG ¼ 133.1
C, TA ¼ 41
C, TC ¼ 36
C, TE ¼ 8.0
C, SF ¼ 0.5 State points

P (bar)

1 1.5 2 13.1 3 13.1 4 13.1 5 13.1 6 13.1 7 13.1 8 1.5 9 13.1 10 13.1 11 13.1 12 13.1 13 1.5 14 1.5 15 1.5 16 13.1 17 1.5 18 1.5 Performance parameters CR 9.51 QA 59.00 kW QG 56.67 kW QC 15.46 kW QE 15.6 kW 52.05 kW QSHX QCPC

T (
C)

r (kg/m)

X

m_ (kg/s)

40.5 41.5 50.0 98.0 133.5 126.0 63.0 62.0 72.5 72.0 36.0 26.0 8.0 5.5 32.0 45.5 37.0 35.0

836.0 836.0 836.0 836.0 825.0 825.0 825.0 825.0 613.2 613.2 613.2 613.2 613.2 613.2 613.2 613.2 613.2 613.2

0.30 0.30 0.30 0.30 0.22 0.22 0.22 0.22 0.99 0.99 0.99 0.99 0.99 0.99 0.99 0.99 0.99 0.99

0.237 0.237 0.237 0.237 0.213 0.213 0.213 0.213 0.025 0.013 0.013 0.013 0.013 0.013 0.013 0.012 0.012 0.025

QGHX QREC Wsp Wnet

hI;law hex;eff

8.56 kW 10.24 kW 1.33 kW 0.88 kW 13.05% 48.23%

R

0.14

0.63 kW

4. Results and discussions The system characteristics and performance were compared between Cooling Alone mode (CA mode) and Combined Cooling e Power (CCP mode) generation mode under the same operating conditions on two consecutive days at nearly the same atmospheric temperature. In both the cases, weak solution flow rate and hot source inlet temperature were maintained at 0.237 kg/s and 150
C respectively. The system was operated in CA mode on the first day and CCP mode on the next day. Under these operating conditions, the average weak and strong solution concentrations were 0.30 and 0.22 respectively. The base case stream characteristics and performance summary of the system operated in CCP mode are shown in Table 3.

4.1. Cooling alone output Fig. 4 shows the variation of temperature and flow rate for the circuits of cooling water, hot water, brine and absorption system major components temperature profile with respect to time. At a cooling water flow rate of 26 m3/h, the average temperature difference across the condenser and absorber was 0.5
C and 1.5
C respectively. The hot water temperature difference across the generator was 12
C for the flow rate of 4 m3/h. The generator bottom temperature increased steadily and stabilized at around 133
C after nearly 30 min. The variation of condenser and absorber temperature profiles over a period of time followed a similar pattern as that of generator temperature. When the weak solution passed through the SHX, it gained a temperature differential between 47
C and 52
C for the flow rate of 0.237 kg/s. When the level of condensing ammonia in the ammonia tank reached above half of the capacity, the throttling valve was opened to circulate draw-regulated refrigerant to the evaporator. The temperature of the brine coming out from the evaporator steadily reduced from 30
C to 8
C; the same condition prevailed for the entire day of operation except at mid-day, when the cooling load increased

Fig. 4. Variation of temperature on (a) Cooling water circuit (b) Hot water circuit (c) AVAR major components and (d) Brine circuit temperature in CA mode with time.

thereby increasing the chilled brine temperature up to 12
C. The average brine temperature difference across the evaporator was 7
C for the constant flow rate of 4 m3/h. Fig. 5 shows the heat loads of the major components of the system and flow rate of the solution and refrigerant vapour over a period of time. The absorber and condenser loads followed the same trend of the achieved generator load. The absorber load was slightly higher than the generator load with a mean value of around 61 kW. Both the condenser and evaporator loads were around 35 kW respectively. The heat loss accounted for 2% of the total heat supplied. The heat loss of the integrated generator was estimated based on the measured temperature at the generator surface, insulation surface and the thermal conductivity of the insulation material. The total internal heat recovery load remained stable and the value was around 73 kW. The mass flow rate of the weak solution was kept constant at 0.237 kg/s and the corresponding mean generated refrigerant vapour was 0.025 kg/s. Fig. 6 shows the degassing width and pressure ratio of the system over the period of the experimental run. The performance of the system was greatly affected by circulation ratio and pressure ratio, the former was determined by degassing width and the latter was determined by the temperature of the condenser and evaporator. The pressure ratio was also a key parameter for power generation. The average degassing width and pressure ratio value were 0.073 and 9 respectively. The external COP, internal COP and exergy efficiency of the system with respect to time are plotted in Fig. 7. The external COP was determined through hot water load across generator to the brine load across evaporator while internal COP was the ratio of generator load to the evaporator load. The external COP value of cooling alone mode ranged from 0.51 to 0.57 while internal COP ranged from 0.53 to 0.60. The deviation between the external and internal COP was 10e15% due to the effectiveness of the heat exchanger, heat losses and irreversibility in components of the system. The exergy efficiency value of the system ranged from 16% to 24%.

4.2. Combined cooling and power output Fig. 8 shows the temperature variations of cooling water, hot water, brine and major performance components of the system over a period of time. During the experimental run, the hot water temperature difference across the generator was 12
C at the flow rate of 4 m3/h. The cooling water temperature across the condenser and the absorber were 0.3
C and 1.7
C respectively, at the flow rate of 26 m3/h. The temperature at the bottom of the generator was

808

G.P. Kumar et al. / Energy 128 (2017) 801e812

90

0.30

mws =0.237 kg/s Tg=133°C

80

0.25 60

0.20

50 0.15 40 30

Generator Evaporator Internal heat recovery Strong solution

20 10

0.10

Absorber Condenser Weak solution Refrigerant vapour

Flow rate (kgs -1)

Major components heat duty (kW)

70

0.05

0 17:00

17:30

16:00

16:30

15:30

15:00

14:30

14:00

13:30

13:00

12:30

12:00

11:30

11:00

10:30

10:00

09:30

09:00

0.00

Time (h) Fig. 5. Variation of main components heat duty, solution and refrigerant flow rate in CA mode with time.

14 0.14

Degassing width Pressure ratio

12 10

0.1

8

0.08 0.06

6

0.04

4

0.02

Pressure ratio

Degassing width

0.12

2

mws =0.237 kg/s Tg=133°C

0 17:30

17:00

16:30

16:00

15:30

15:00

14:30

14:00

13:30

13:00

12:30

12:00

11:30

11:00

10:30

10:00

09:30

09:00

0

Time (h) Fig. 6. Variation of degassing width and pressure ratio in CA mode with time.

around 133
C, at the weak solution flow rate at 0.237 kg/s. Temperature gain of around 50
C was achieved by the strong solution when it passed through the helical coil type SHX. The temperature measured at the outlet of the condenser and absorber were 34
C and 39
C respectively. The concentrations of the weak and strong solution were estimated to be around 0.30 and 0.22 respectively. The refrigerant liquid entered the evaporator at 7
C after letdown at the flow rate of 0.025 kg/s. The brine left the evaporator at a temperature of 8
C and flow rate of 4 m3/h. The heat duty variation of the main components in the AVAR system such as the condenser, absorber, generator and evaporator along with mass flow rate of strong, weak solution and refrigerant flow are shown in Fig. 9. The generator is a shell and tube heat exchanger and its heat load was about 54 kW in the experimental run, at the weak solution flow rate of 0.237 kg/s. The refrigerant vapour stream produced in the generator after rectification was split into two streams and used for simultaneous power and cooling generation. The condenser and absorber loads were increased

gradually with the increased refrigerant circulation in the system. The absorber heat load was fourfold higher than condenser load due to half of the generated refrigerant being used for cooling entering the condenser, whereas the total refrigerants after cooling and power production reached the absorber. The average heat loads of heat sink components of the condenser and absorber were about 16 kW and 58 kW respectively. The variation in the heat sink components was due to the variation in ambient conditions. The internal heat recovery was estimated by adding the heat loads of all the internal heat recovery components such as SHX, GHX, rectifier and condensate pre-cooler. The heat recovery gradually increased over a period of time, due to the increase in weak solution temperature at the outlet of the heat exchange components. The heat recovery components provided nearly 71 kW of recovered heat load to the generator. The variation of the cycle of useful outputs, i.e. cooling capacity and expander work output variation with respect to time are shown in Fig. 10. At the split ratio of 0.5, the simultaneous cooling

G.P. Kumar et al. / Energy 128 (2017) 801e812

809

0.7

50

mws =0.237 kg/s Tg=133°C

0.6

COP

30

0.4 0.3

20

0.2 10

COP (External) COP (Internal) Exergy efficiency

0.1

Exergy efficiency (%)

40 0.5

17:30

16:30

17:00

16:00

15:30

15:00

14:30

14:00

13:30

13:00

12:00

12:30

11:30

11:00

10:30

10:00

09:30

0 09:00

0

Time (h) Fig. 7. Variation of COP, efficiency in CA mode with respect to time.

and expander work outputs were about 15.6 kW and 2.21 kW respectively. The corresponding expander inlet pressure and temperature were 13.1 bar and 68
C respectively and expander outlet pressure was 1.5 bar. The power output of the expander depended upon many factors such as the high and low pressure of the system, the inlet temperature of the expander, refrigerant purity and isentropic efficiency of the expander. The decrease in high pressure did not affect the expander output much. Although the low pressure of the system increased at CCP mode, it leads to a sacrifice of the expander work output by an increase of the weak solution concentration. This condition held good for generating more ammonia vapour in the generator. Also, the circulation ratio of the system was lowered by increasing degassing width. Fig. 11 shows the variation of degassing width and pressure ratio of the system, with respect to time for a typical operating condition. The degassing width of the system was initially high because of less amount of refrigerant circulation, and it became steady after 1 h from the starting time of the system operation. The concentration of the strong solution was initially high, mainly

90

0.3

mws =0.237 kg/s SR =0.5 Tg=133°C

80

0.25

70 60

0.2

50 Generator Evaporator IHR Strong solution

40 30 20

0.15

Absorber Condenser Weak solution Refrigerant vapour

0.1

Flow rate (kgs-1)

Major components heat duty (kW)

Fig. 8. Variation of temperature on (a) Brine circuit (b) Hot water circuit (c) Cooling water circuit and (d) AVAR major components temperature in CCP mode with time.

0.05

10 17:30

17:00

16:30

16:00

15:30

15:00

14:30

14:00

13:30

13:00

12:30

11:30

12:00

11:00

10:00

10:30

9:30

0 9:00

0

Time (h) Fig. 9. Variation of main components heat duty, solution and refrigerant flow rate in CCP mode with time.

G.P. Kumar et al. / Energy 128 (2017) 801e812

Cooling load (kW)

20

2

mws =0.237 kg/s SR =0.5 Tg=133°C

16

1.6

12

1.2

8

0.8 Cooling load Expander work

4

Expander work (kW)

810

0.4

17:30

17:00

16:30

16:00

15:30

15:00

14:30

14:00

13:30

13:00

12:00

12:30

11:30

10:30

11:00

10:00

9:30

0 9:00

0

Time (h) Fig. 10. Variation of system dual outputs in CCP mode with time.

13 0.14

Degassing width Pressure ratio

11 9

0.1

7

0.08 0.06

5

0.04

3

mws =0.237 kg/s SR =0.5 Tg=133°C

0.02

Pressure ratio

Degassing width

0.12

1

0 17:30

17:00

16:30

16:00

15:30

15:00

14:30

14:00

13:30

12:30

13:00

12:00

11:30

11:00

10:30

10:00

9:30

9:00

-1 Time (h) Fig. 11. Variation of degassing width and pressure ratio in CCP mode with time.

90

1

Effective first law efficiency Effective Exergy Efficiency Power to cooling ratio

70

0.8 0.6

60 50 40

0.4

30 0.2

20 10

0 17:30

17:00

16:30

16:00

15:30

15:00

14:30

14:00

13:30

13:00

12:30

12:00

11:30

11:00

10:30

10:00

9:30

0 9:00

Efficiency (%)

80

mws =0.237 kg/s SR =0.5 Tg=133°C

Time (h) Fig. 12. Variation of efficiency and power to cooling ratio in CCP mode with time.

Power to Cooling ratio

100

G.P. Kumar et al. / Energy 128 (2017) 801e812

811

effective first law and effective exergy efficiency were around 13% and 48% respectively. In the CCP mode, the cooling capacity and net power output were about 15.6 kW and 2.21 kW respectively with the power to cooling ratio of 0.14. The system was able to meet the power/cooling demand profile by easily switching over between the different ratios of power to cooling. The sensitivity study for hot water source temperature was analysed for system performance. This study proves the viability of utilizing low-grade/mid-grade heat sources for producing simultaneous cooling and electric power generation. Acknowledgements

Fig. 13. Effect of hot water temperature in dual output and system performance.

due to the refrigerant that was emptied out of the refrigerant circuit after the end of the previous run and stored in the solution. The high and low pressure of the system was determined by the condenser and evaporator temperatures respectively. Pressure ratio was estimated by the ratio of high to low pressure of the system. For a typical operating condition, the pressure ratio ranged from 8.5 to 9.5. The dual output system performance was estimated by effective first law efficiency, effective exergy efficiency and power/ cooling ratio with respect to time, as shown in Fig. 12. The effective first law and effective exergy efficiency were calculated by considering the cooling output as exergy associated with cooling output; it was divided by reasonable second law efficiency for a refrigeration cycle (assumed 40%) to avoid the over-estimation of cooling output when equal weights were assigned for the thermodynamic quality of the dual outputs (cooling and mechanical power). Heat input of 53.83 kW was required for the system to produce 18.31 kW cooling and 2.43 kW power production yielding effective first law and effective exergetic efficiencies of about 13% and 48% respectively. Moreover, the power to cooling ratio was in the range of 0.13e0.16. The influence of hot water source temperature in the range of 135
Ce150
C in dual output and system performance is shown in Fig. 13. From the hot water temperature of 135
C, ammonia generation began in the generator, corresponding circulation ratio was 18 as the value of strong solution concentration 0.27 left from the generator. This indicates the less ammonia generation. Up to 143
C of hot water, there was steep rise in both cooling output and power output; the slope then gradually reduced. The maximum effective first law and effective exergy efficiencies were 11% and 45% respectively at the hot water temperature of 150
C. Moreover, the effect of hot water temperature was not influenced in the power to cold ratio of the system. 5. Conclusions An experimental investigation of a combined power and cooling system driven by low-grade heat sources has been carried out for both cooling alone and combined cooling-power mode. The system characteristics, as well as the individual and compared performance between CA and CCP mode were discussed at similar heat sink atmospheric conditions. At maintaining the generator bottom temperature of 133
C, and the weak solution flow rate of 0.237 kg/s with same atmospheric conditions, the cooling alone condition gave external, internal COP and exergy efficiency of 0.57, 0.60 and 24% respectively. Moreover, for the same operating conditions for the dual output (cooling and mechanical power) of the system,

This research was supported by the Solar Energy Research Initiative (SERI) Program, Department of Science and Technology, Government of India with grant number DST/TM/SERI/2k12/74(G). Nomenclature Cp EX h m P Q R s SR T X SP

specific heat capacity (kJ/kg K) exergy (kW) specific enthalpy (kJ/kg) mass flow rate (kg/s) pressure (bar) heat load (kW) power/cooling ratio specific entropy (kJ/kg K) split ratio temperature (
C) concentration state point

Acronyms AVAR ammonia vapour absorption refrigeration COP coefficient of performance CR circulation ratio CPC condensate pre-cooler EXP expander GHX generator heat exchanger REC rectifier SHX solution heat exchanger SEV solution expansion valve REV refrigerant expansion valve Subscripts A absorber b brine C condenser E evaporator G generator i isentropic cw cooling water ex exergy hw hot water rc refrigerant for cooling output rp refrigerant for power output ss strong solution (strong in absorbent, weak in refrigerant) sp solution pump ws weak solution (weak in absorbent, strong in refrigerant) eff effective ext external int internal ref refrigeration 1e18 state points in the system corresponding to Fig. 1 1 reference condition

812

I II

G.P. Kumar et al. / Energy 128 (2017) 801e812

first second

Greeks symbols r density (kg/m3) h efficiency (%) References [1] Maloney JD, Robertson RC. Thermodynamic study of ammoniaewater heat power cycles. Oak Ridge National Laboratory Report. 1953. CF-53-8-43. [2] Kalina AI. Combined-cycle system with novel bottoming cycle. J Eng Gas Turbines Power 1984;106(4):737e42. [3] El-Sayed YM, Tribus M. A theoretical comparison of the Rankine and Kalina cycles. ASME Special Publication. 1985. AES-1:97e102. [4] Ibrahim OM, Klein SA. Absorption power cycles. Energy 1996;21:21e7. [5] Leibowitz H. Operating experience on the 3MW Kalina cycle demonstration plant. In: Proceedings of the American power conference, Chicago; 1993. [6] Lolos PA, Rogdakis EDA. Kalina power cycle driven by renewable energy sources. Energy 2009;34:457e64. [7] Nasruddin Usvika R, Rifaldi M, Noor A. Energy and exergy analysis of Kalina cycle system ( KCS ) 34 with mass fraction ammonia-water mixture variation. J Mech Sci Technol 2009;23:1871e6. [8] Yogi Goswami D. Solar thermal power technology: present status and ideas. Energy Sources 1998;20:137e45. [9] Xu F, Goswami DY, Bhagwat SS. A combined power/cooling cycle. Energy 2000;25:233e46. [10] Lu S, Goswami DY. Optimization of a novel combined power/refrigeration thermodynamic cycle. J Sol Energy Eng 2003;125:212e7. [11] Zheng D, Chen B, Qi Y, Jin H. Thermodynamic analysis of a novel absorption power/cooling combined-cycle. Energy 2006;83:311e23. [12] Ziegler F. Novel cycles for power and refrigeration. In: Proceedings 1st European conference on polygeneration 2007 Tarragona (Spain); 2007. p. 169e82. [13] Liu M, Zhang N. Proposal and analysis of a novel ammonia e water cycle for power and refrigeration cogeneration. Energy 2007;32:961e70. [14] Zhang N, Lior N. Methodology for thermal design of novel combined refrigeration/power binary fluid systems. Int J Refrig 2007;30:1072e85. [15] Wang J, Dai Y, Gao L. Parametric analysis and optimization for a combined

power and refrigeration cycle. Appl Energy 2008;85:1071e85. [16] Sun L, Han W, Jing X, Zheng D, Jin H. A power and cooling cogeneration system using mid/low-temperature heat source. Appl Energy 2013;112: 886e97. [17] Yu Z, Han J, Liu H, Zhao H. Theoretical study on a novel ammonia-water cogeneration system with adjustable cooling to power ratios. Appl Energy 2014;122:53e61.  pez-Villada J, Ayou DS, Bruno JC, Coronas A. Modelling, simulation and [18] Lo analysis of solar absorption power-cooling systems. Int J Refrig 2014;39: 125e36. [19] Wang J, Wang J, Zhao P, Dai Y. Thermodynamic analysis of a new combined cooling and power system using ammonia e water mixture. Energy Convers Manag 2016;117:335e42. [20] Tamm G, Goswami D, Lu S, Hasan AA. Theoretical and experimental investigation of an ammonia e water power and refrigeration thermodynamic cycle. Sol Energy 2004;76:217e28. [21] Martin C, Goswami DY. Effectiveness of cooling production with a combined power and cooling thermodynamic cycle. Appl Therm Eng 2006;26:576e82. [22] Demirkaya G, Padilla RV, Goswami DY, Stefanakos E, Rahman MM. Analysis of a combined power and cooling cycle for low-grade heat sources. Int J Energy Res 2011;35:1145e57. [23] Han W, Chen Q, Sun L, Ma S, Zhao T, Zheng D. Experimental studies on a combined refrigeration/power generation system activated by low-grade heat. Energy 2014;74:59e66. [24] Ashrae. ASHRAE Handbook, Fundamentals. 1791. Tullie Circle, Atlanta: ASHRAE; 1993. p. 17e81. [25] Tillner-Roth R, Friend D. A helmholtz free energy formulation of the thermodynamic properties of the mixture {ammonia þ water}. J Phys Chem Ref Data 1998;27:63e6. [26] Melinder A. Thermophysical properties of liquid secondary refrigerants, table and diagrams for the refrigeration industry. Paris: IIF/IIR; 1997. [27] Wagner W, Pruß A. The IAPWS formulation 1995 for the thermodynamic properties of ordinary water substance for general and scientific use. J Phys Chem Ref Data 2002;31:387e538. [28] Ayou DS, Bruno JC, Coronas A. New power and cooling absorption cycles. Int J Therm Environ Eng 2013;5:135e43. [29] Vijayaraghavan S, Goswami DY. On evaluating efficiency of a combined power and cooling cycle. J Energy Resour Technol 2003;125:221e7. [30] Moffat RJ. Describing the uncertainties in experimental results. Exp Therm fluid Sci 1988;1:3e17.