Preliminary experimental characterization of a three-phase absorption heat pump

i n t e r n a t i o n a l j o u r n a l o f r e f r i g e r a t i o n 3 6 ( 2 0 1 3 ) 7 1 7 e7 2 9

Available online at www.sciencedirect.com

w w w . i i fi i r . o r g

journal homepage: www.elsevier.com/locate/ijrefrig

Preliminary experimental characterization of a three-phase absorption heat pump A. Rosato*, S. Sibilio Seconda Universita` degli Studi di Napoli, Dipartimento di Architettura, via San Lorenzo, 81031 Aversa, Italy

article info

abstract

Article history:

In this paper a recently commercialized three-phase absorption heat pump that is capable

Received 5 August 2012

of storing energy internally in the form of crystallized salt (LiCl) with water as refrigerant

Received in revised form

has been experimentally investigated during summer period. The tests have been per-

23 October 2012

formed with the aim to investigate the operation logic of the machine and to highlight both

Accepted 14 November 2012

the reliability and the efficiency of the system over an operating conditions range of great

Available online 23 November 2012

practical interest.

Keywords:

trically driven vapor compression refrigerating system from an energy, environmental and

Absorption cycle

economic point of view in order to assess the suitability of the absorption heat pump: this

Thermally driven chiller

comparison showed that the absorption system is potentially able to guarantee an energy

Chemical heat pump

saving, a reduction of carbon dioxide emissions and a lower operating cost only in case of

Lithium chloride

the most part (at least 70%) of required thermal energy is supplied by solar collectors.

The measured performance have been compared with those of a conventional elec-

ª 2012 Elsevier Ltd and IIR. All rights reserved.

Solar cooling Trigeneration

Caracte´risation expe´rimentale pre´liminaire d’une pompe a` chaleur a` trois phases Mots cle´s : cycle a` absorption ; refroidisseur a` entraıˆnement thermique ; pompe a` chaleur chimique ; chlorure de lithium ; refroidissement solaire ; trige´ne´ration

1.

Introduction

The worldwide cooling demand has drastically increased over the last few years. This has led to the installation of a large number of electrically driven air conditioning systems (Balaras et al., 2007; Henning, 2007) with a dramatic rise in electricity consumption, which is nowadays mostly generated from fossil fuels. This trend has caused important

environmental problems such as ozone layer depletion and global warming. In this context, there is a clear need to develop more sustainable technologies in order to minimize the environmental impact of cooling applications. Absorption heat pumps have emerged as a promising alternative to conventional vapor compression cycles (Fiskum et al., 1996; Florides et al., 2002; McMullan, 2002; Wang et al., 2011), since they

* Corresponding author. Tel./fax: þ39 081 8122530. E-mail address: [email protected] (A. Rosato). 0140-7007/$ e see front matter ª 2012 Elsevier Ltd and IIR. All rights reserved. http://dx.doi.org/10.1016/j.ijrefrig.2012.11.015

718

i n t e r n a t i o n a l j o u r n a l o f r e f r i g e r a t i o n 3 6 ( 2 0 1 3 ) 7 1 7 e7 2 9

Nomenclature Latin letters B natural gas-fired boiler c specific heat (kJ kg1 K1) C operating cost (V) carbon dioxide equivalent emission (kg CO2) CO2 COP coefficient of performance CW10 ClimateWell10 CWIC2 CW10 internal software natural gas Unit Cost (V Nm3) CUng electric energy Unit Cost (V kWh1) CUel E energy (kJ) EDC electrically driven chiller system fraction of Eth,TDC produced by natural gas-fired EFB boiler FC fan-coil IHE internal heat exchanger HD heat dissipator HWS hot water storage LHV lower heating value (kWh Nm3) M water mass flow meter MCHP micro combined heat and power generation MG natural gas volumetric flow meter P power (kW)/pump PER primary energy ratio (%) PES primary energy saving (%) PHE plate heat exchanger R electric resistance

can use low grade energy sources that are environmentally friendlier instead of electricity. Several scientific papers studied the integration of different types of commercially available absorption systems with cogeneration units by using the surplus of heat coming from the cogeneration device during the warm season for activating the absorption cycle and providing a combination of electric, heat and cooling energy (Angrisani et al., 2012; Chicco and Mancarella, 2009; Hernandez-Santoyo and Sanchez-Cifuentes, 2003; Serra et al., 2009). In comparison to the traditional units based on separate energy production, these plants (called trigeneration systems) showed a significant potential in terms of energy savings and reduction of CO2 emissions (Huicochea et al., 2011; Kavvadias et al., 2010; Li et al., 2006; Lin et al., 2007). There are several technologies of thermally activated chillers commercially available today, e.g. standard absorption system using LiBr/water or NH3/water and salt-water absorption chiller (Srikhirin et al., 2001) and/or chemical heat pump (Wongsuan et al., 2001). Chemical heat pump is a new and promising technology which is capable of operating with low temperature heat sources: salt-water solutions such as lithium chloride (LiCl)/water, sodium sulphite (Na2S)/water, and calcium chlorides (CaCl2)/water, etc. have been used (Boer et al., 2002; Conde, 2004; Ogura et al., 2003). Absorption chillers are more common at medium or larger scale, while small scale units are in process of becoming commercial. In this paper a recently commercialized chemical heat pump using LiCl/water as a working fluid pair has been

SUN T TC0 TDC V_ Greeks a b D h r

Second University of Naples temperature/resistance thermometer temperature of water going towards heat dissipator before by-pass valve ( C) thermally driven chiller volumetric flow rate (m3 s1)

CO2 emission factor for electric energy (kgCO2 kWh1) CO2 emission factor for primary energy (kgCO2 kWh1) difference (%) efficiency density (kg m3)

Subscripts B boiler cool cooling el electric FC fan-coil HD heat dissipator in inlet IHE internal heat exchanger MCHP micro combined heat and power generation ng natural gas out outlet th thermal TDC thermally driven chiller w water

experimentally investigated. It is a three-phase absorption system that is capable of storing energy internally in the form of crystallized salt (LiCl) with water as refrigerant; the triplestate process, so called because it uses solid, liquid and vapor at the same time, makes this thermally driven chiller (TDC) particularly different from other chemical heat pumps or standard absorption processes (which use liquid and vapor phases). The unit was patented in 2000 (Olsson et al., 2000) and it has been developed by the Swedish company ClimateWell via five generations of prototypes. The 4th generation of machines, that was the first to be sold commercially as from 2007 under the name CW10, is installed at the laboratory of Second University of Naples (Fig. 1). It consists of two identical units, so called barrels, that work together. Each barrel consists of four different vessels: the reactor (absorber/ generator), the condenser/evaporator, the solution vessel and the refrigerant vessel. The reactor and condenser/evaporator are the active parts of the unit with a vapor channel between them, while the two other vessels are stores for salt solution and the refrigerant; the unit is operated as a closed system under vacuum conditions and there are heat exchangers in the reactor and condenser/evaporator; solution and refrigerant are pumped from the storage vessels over these heat exchangers and then flow under gravity back to the storage vessels (Bales and Ayadi, 2009). The machine is connected to three external circuits: the thermal supply, the heat sink and the cooling supply. The

i n t e r n a t i o n a l j o u r n a l o f r e f r i g e r a t i o n 3 6 ( 2 0 1 3 ) 7 1 7 e7 2 9

719

Fig. 1 e Schematic of complete CW10 machine on the left (Udomsri and Bales, 2011) and single barrel on the right (Bales and Ayadi, 2009).

process occurring in each barrel works in batch mode, with a separate desorption (charge) phase followed by absorption (discharge) phase: U during the charging phase the reactor is connected to the thermal supply, while the condenser/evaporator is connected to the heat sink; the solution is heated by the thermal source via the heat exchanger in the reactor becoming steadily more concentrated, and when it reaches saturation point further desorption can result in the formation of solid crystals that fall under gravity into the vessel. These then get transferred to the storage vessel. Here they are prevented from following the solution into the pump by a sieve, thus forming a form of slurry in the bottom of the vessel; at the same time water is evaporated and steam is released to the condenser/evaporator; U during the discharging phase the reactor is connected to the heat sink, while the condenser/evaporator is connected to the cooling supply circuit; the saturated solution is pumped over the heat exchanger in the reactor where it absorbs the refrigerant evaporated in the condenser/evaporator. The solution becomes unsaturated in the reactor, but when it goes to the solution store it has to pass through the slurry of crystals, where some of the crystals are dissolved to make the solution fully saturated again. In this way the solution is kept saturated as long as there are crystals available and the net result is a dissolving of the crystals into saturated solution. Since the energy is stored in a chemical form, no energy should be lost to the surroundings; when a barrel is charged, the energy stays stored in the barrel until there is a cooling demand. A by-pass valve is installed in the machine for regulating the water temperature going towards the cooling supply to the set value: by-pass valve position can vary between 100% (bypass valve fully open) and 0% (by-pass valve fully closed). A plumbing unit switches the flows between the external circuits and the relevant heat exchangers in the two barrels. The machine has its own control system that makes all the

“swaps” of the machine which changes the state from charging to discharging and vice versa. The control system also sends signals to the plumbing unit to control all the valves in order to change the circuit connections and it guarantees that the machine works automatically and independently. The unit can be operated so that one barrel is charged while the other one is discharged: this gives quasi-continuous operation, but when the units are swapped at the end of charge/discharge, there is a period without cooling supply. More generally, the CW10 unit can be operated in seven different modes: “manual”, “normal”, “full cycles”, “double”, “timer”, “turbo” and “test”. In this paper the performance of the system have been experimentally investigated during both “normal mode” and “double mode” operation. “Normal mode” is the default and the fully automatic mode, where the barrels alternate in charging and discharging: during this operation mode the machine is always able to both provide cooling energy and use the supplied thermal energy. In “double mode”, both barrels are charged and discharged at the same time. This should result in higher cooling/heating power when discharging and higher charging power when charging; however, running in this mode the discharging delivery and the charging power is not continuous. The machine control system recognizes when a swap should take place. It then sends signals to the plumbing unit which automatically makes all the necessary connections. A swap is performed when one of the following conditions is verified: 1) charging barrel: level reaches 100% will trigger a swap independent of discharging barrel status; 2) charging barrel: level reaches above 80% in combination with condition 3 or condition 4; 3) discharging barrel: level has reached below 40% and TC0/ Tw,TDC,in is higher than 0.67 and TC0 is higher than 15  C in combination with condition 2; 4) discharging barrel: level has been 3% or less for 15 min in combination with condition 2;

720

i n t e r n a t i o n a l j o u r n a l o f r e f r i g e r a t i o n 3 6 ( 2 0 1 3 ) 7 1 7 e7 2 9

where TC0 is the temperature of internal water going towards the heat dissipator before the by-pass valve and Tw,TDC,in is the temperature of water coming from the heat source before entering the machine. The level of each barrel is determined by measuring the weight of the water in the barrels. The same TDC model installed at the laboratory of Second University of Naples (SUN) has been already investigated by Udomsri et al. (2011, 2012), Bales and Nordlander (2005), Bales and Ayadi (2009). Udomsri et al. (2011) presented the monitoring results of CW10 driven by district heat from a network supplied by a centralised combined heat and power fired with municipal waste; they investigated the system during “normal mode” operation and found a maximum thermal coefficient of performance during the hottest period of around 0.50; however, the figure was only 0.41 for the complete monitoring period during the summer of 2008. According to the monitored results obtained from the demonstration, a system simulation model for the TRNSYS environment has been calibrated by Udomsri et al. (2012) and used to find improved system design and control. Bales and Nordlander (2005) carried out just few of the planned experiments on CW10 model during “full cycle” operation due to lack of time before the machines were shipped. Of these, most had missing data due to an error in the logger program that limited the duration of saved data, resulting in an even smaller amount of recorded results; due to these problems, no direct calculations of the thermal coefficient of performance was possible. They tested also the TDC model DB220 produced by ClimateWell, a TDC model less recent than CW10. According to the available measurements obtained for CW10 model, Bales and Ayadi (2009) developed a grey box simulation model for the TRNSYS environment; the TDC unit model was verified against the measured data and showed reasonable agreement, but the authors stated that more data would be needed be needed to make sure the parameters are correct and to verify them properly. The model was also used for parametric studies in order to determine the effect of boundary conditions on the thermal coefficient of performance. Even if some data have been already available in literature, the CW10 unit has not been yet investigated during “double

mode” operation, and the experimental results regarding “normal mode” operation are still quite limited. For these reasons in this paper the performances of CW10 model have been experimentally investigated during both “normal mode” and “double mode” operation in order to better highlight the system operation and performance. In the following the experimental set-up and the results gathered during the experiments (thermal power supplied, cooling power delivered, coefficient of performance, temperature levels, etc.) will be presented and analyzed in detail. In addition the measured data have been used to compare the performance of the experimentally investigated thermally driven chiller with those of a conventional vapor compression refrigerating system from an energy, environmental and economic point of view in order to verify the suitability of CW10 model. The measurements reported in the following can be also used to verify the accuracy of the recently developed TRNSYS simulation model (Udomsri et al., 2011, 2012; Bales and Nordlander, 2005; Bales and Ayadi, 2009) in order to carry out a techno-economic analysis for studying and evaluating the viability of trigeneration plants using the TDC model investigated in this paper.

2.

Experimental set-up

A schematic view of the test apparatus of the Built Environment Control Laboratory of Second University of Naples (SUN), detailing instrumentation components, is shown in Fig. 2. The experimental set-up is located in Frignano, a municipality in the Province of Caserta (around 20 km far from Naples). As stated above, the TDC unit experimentally investigated in this paper is the 4th generation of a chemical heat pump (model CW10), patented in 2000 (Olsson et al., 2000) and sold by the Swedish company ClimateWell. The machine has been described in detail in the previous section. As can be derived from Fig. 2, the unit is supplied by the thermal power recovered from a micro-cogenerator based on a natural gas fuelled reciprocating internal combustion engine (commercialized by AISIN-SEIKI company) and stored in 1000 L

Fig. 2 e Experimental set-up.

721

i n t e r n a t i o n a l j o u r n a l o f r e f r i g e r a t i o n 3 6 ( 2 0 1 3 ) 7 1 7 e7 2 9

tank; taking into consideration that the TDC needs a charging temperature at least 50  C larger than the heat sink temperature to start the absorption process and that the temperature of hot water flowing from the MCHP system into the storage cannot be higher than around 70  C, a natural gas-fired boiler (B) with a rated thermal output of 32 kW has been installed before the inlet of the TDC system. The hot water storage (HWS) is insulated with 50 mm flexible polyurethane layer and is furthermore equipped with an auxiliary 4.0 kW electric resistance (R) fed by the micro-cogenerator. A water to air heat exchanger with a rated power equal to 30 kW is installed as heat sink. The water cooled by the TDC is pumped towards a fan-coil with a rated total cooling capacity equal to 10.95 kW devoted to satisfy the cooling load of a part of the entire laboratory. Variable speed wet rotor pumps (P1, P2, P3 and P4) have been installed in order to circulate the water within the experimental plant; three different pump revolution steps can be manually set for each pump with a maximum mass flow rate equal to 22.8 l min1 for pump P3 and to 14.4 l min1 for the other pumps. The experimental plant is well instrumented (Fig. 2) in order to measure the following parameters:  water temperature in the key-points of the plant (at the inlet and outlet of TDC, FC, HD, B, HWS, MCHP);  ambient temperature;  water volumetric flow rate in the key-points of the plant (flow rate entering TDC, FC, HD, B, HWS, MCHP);  natural gas volumetric flow rate entering both microcogenerator and natural gas-fired boiler;  electric power supplied by micro-cogenerator to the end-user. Water and ambient temperatures are measured by using resistance thermometers Pt100; water mass flow rate is obtained by using an ultrasonic mass flow sensor, while a thermal mass flow meter is installed to evaluate the natural gas volumetric flow rate; three wattmeters measure the electric flows entering and exiting the unit. Two resistance thermometers are used for measuring the hot water temperature within the tank. Table 1 summarizes the main characteristics of the plant instrumentation.

Table 1 e Main characteristics of the plant instrumentation. Parameter

Instrument

Operating range

T

Resistance thermometer Pt100 Ultrasonic volumetric flow meter Thermal volumetric flow meter

50 O 100  C

0.2  C

0 O 50 l min1

2.5% of full scale

0 O 5.0 Nm3 h1

Wattmeter

0 O 6 kW

0.8% of reading 0.2% of full scale 0.2% of full scale

V_ w

V_ ng

Pel,MCHP

0 O 10 kW

Accuracy

The TDC installed at SUN lab is equipped with an internal software (named CWIC2) by means of which several operation system parameters can be monitored and recorded: in particular, the machine internal software provides the values of some parameters that cannot be directly derived by using our instrumentation, i.e. the water temperature TC0, the level of each barrel during system operation, the by-pass valve position, etc. Based on the direct measurements, the parameters listed below are calculated in order to evaluate the plant performances: Pth;TDC ¼ V_ w;TDC $rw $cw $ Tw;TDC;in  Tw;TDC;out Pth;HD ¼ V_ w;HD $rw $cw $ Tw;HD;in  Tw;HD;out Pcool;FC ¼ V_ w;FC $rw $cw $ Tw;FC;out  Tw;FC;in










(1) (2) (3)

where the water specific heat and the water density, respectively, have been assumed equal to cw ¼ 4.187 kJ (kg K)1 and rw ¼ 990 kg m3. The signals coming from the resistance thermometers Pt100 are acquired by three cFP-RTD-124 analog input modules (produced by National Instruments), while the signals coming from the other sensors are managed by two cFP-AI-110 analog input modules (produced by National Instruments). Each acquisition device is a 16-bit resolution system with eight current outputs (4 O 20 mA). The digital data coming from the modules are sent to a personal computer. The software LabView 8.2 is used to define the acquisition frequency and to monitor and/or record all the directly measured and calculated parameters. The experimental data presented in the following sections have been recorded every 10 s. Additional details regarding the above-presented experimental plant can be found in Rosato and Sibilio (2012) and Angrisani et al. (2012).

3.

Experimental results

In the following the operating conditions and the main results gathered during the experiments are highlighted and deeply analyzed. The data are presented separately for “normal” and “double” mode operation. Given the constraints of the experimental set-up, the experiments have not been conducted over the entire range of machine operation; however the achieved results allows to get useful information on the system performance in relation to a range of operating conditions of great interest in the practice not yet fully exploited experimentally. During both the tests in “normal mode” and “double mode” the set value of water temperature going towards the cooling supply was 13  C. In the last section the measured data are used to compare the performance of CW10 unit with those of a conventional electrically driven vapor compression refrigerating system from an energy, environmental and economic point of view.

3.1.

Normal mode operation

The test in “normal mode” has been performed the 19th October 2011 from 11:01 until 17:44. In Figs. 3 and 4 the

722

i n t e r n a t i o n a l j o u r n a l o f r e f r i g e r a t i o n 3 6 ( 2 0 1 3 ) 7 1 7 e7 2 9

Text

Tw,TDC,in

Tw,TDC,out

Tw,FC,in

Tw,FC,out

Tw,HD,in

Tw,HD,out

16:18

17:16

85 80 75

70 65

Temperature (°C)

60 55 50 45 40 35 30 25 20 15

10 11:01

11:30

11:59

12:28

12:57

13:25

13:54

14:23

14:52

15:21

15:49

16:47

Time (hh:mm)

Fig. 3 e Water temperature values measured during “normal mode” operation.

operating conditions related to the experiment performed in “normal mode” are reported as a function of the time: in Fig. 3 the water temperature values measured in the key-points of the plant are depicted, while Fig. 4 shows the level of both barrels and the water volumetric flow rates flowing through the thermally driven chiller, the heat dissipator and the fan-coil. Fig. 5 depicts the thermal power supplied to the TDC system (Pth,TDC), the cooling power produced by the TDC system (Pcool,FC) and the thermal Coefficient Of Performance (COPth,TDC) as a function of the time during “normal mode” operation. Pth,TDC is calculated by using Eq. (1), while Eq. (3) provides Pcool,FC; the instantaneous values of COPth,TDC are defined as follows: COPth;TDC ¼ Pcool;FC =Pth;TDC

(4)

As can be derived from Figs. 3e5, TDC operation in “normal mode” starts around 11:01 with Barrel B charging and Barrel A discharging; at around 13:00 the barrels are “swapped”, so that Barrel A charging starts and cooling is provided thanks to the Barrel B discharging; two additional “swaps” are performed around 14:30 and 16:00, respectively: as a consequence, both Barrel A and Barrel B have been charged and discharged two times through the experiment. Each swap between barrels is due to the fact that charging barrel level reaches 80% and discharging barrel level has reached 40% with both the ratio TC0/Tw,TDC,in higher than 0.67 and TC0 values larger than 15  C. Fig. 3 shows that the water temperature coming from the boiler towards the TDC (Tw,B,out) is around 81.5  C (quite lower than that one suggested by the manufacturer for the TDC, i.e. 85e120  C) and the temperature drop across the machine is about 5e10  C; the temperature of the water coming from the HD towards the machine during charging/discharging periods (Tw,HD,out) oscillates between around 26 and 30  C. Water temperature coming from the TDC towards the fan-coil (Tw,FC,in) is around 15  C, with a minimum value of 12.6  C achieved around 11:10.

Except during the “swap” between the barrels, the volumetric flow rate through both the thermally driven cooling system (V_ w;TDC ) and fan-coil (V_ w;FC ) is 14.4 l min1 (15.0 l min1 is suggested as minimum water flow rate by ClimateWell), while 22.8 l min1 is the water flow (V_ w;HD ) pumped towards the HD (Fig. 4). As can be derived from Fig. 5, during the “swap” between the two barrels the TDC cannot deliver cooling; during the charging/discharging periods, cooling capacity increases till reaching a maximum and then slightly reduces: maximum value of cooling power gathered during the test is about 3.5 kW. The measured data agree well with those reported by the manufacturer that suggests about 3.0 kW as cooling capacity in case of Tw,TDC,in ¼ 80  C, Tw,HD,out ¼ 30  C and Tw,FC,out ¼ 20  C. During Barrel A discharging the values of Pcool,FC are slightly higher than the those achieved during Barrel B discharge. So the plot shows that the two units worked differently, with Barrel B performing poorer than the other: also Bales and Nordlander (2005) found a different performance between two barrels by experimenting the model DB220. Fig. 5 shows that the COPth,TDC (defined by Eq. (4)) is not constant: it increases during discharging phase till reaching a maximum value and then becomes zero when “swap” period starts; the maximum value of COPth,TDC measured during Barrel A discharging is around 0.6, quite higher than the greater value of COPth,TDC achieved during Barrel B discharging. The cumulative cooling energy supplied by the TDC system (Ecool,FC) and the cumulative thermal energy supplied to the TDC system (Eth,TDC) throughout the experiment are equal to 57818.1 kJ and 1836501.1 kJ, respectively; as a consequence, a value of 0.31 can be calculated for the thermal Coefficient of Performance by considering the energies associated to the charge/discharge cycles as follows:

723

i n t e r n a t i o n a l j o u r n a l o f r e f r i g e r a t i o n 3 6 ( 2 0 1 3 ) 7 1 7 e7 2 9

100 95

Barrel A level

90

Barrel B level

Vw,TDC = Vw,FC=14.4 l min-1 Vw,HD = 22.8 l min-1

85 80

Barrel level (%)

75 70

65 60 55 50 45 40 35

30 25 11:01

11:30

11:59

12:28

12:57

13:25

13:54

14:23

14:52

15:21

15:49

16:18

16:47

17:16

Time (hh:mm)

Fig. 4 e Volumetric water flow rate and barrel level during “normal mode” operation.

COPth;TDC ¼ Ecool;FC =Eth;TDC

the results reported above and those achieved during the repeated test has been found. The presented data agrees well also with the values recorded by the CW10 internal software (named CWIC2).

(5)

The values of COPth,TDC found in this work agree quite well with those measured by Udomsri et al. (2011). In Table 2 the duration of both charging/discharging phases and “swap” periods are reported: as can be derived from this table, the three “swaps” between barrels have a duration of around 5 min; regarding the charging/discharging phases, the first one shows a duration quite higher than that the other ones. The experiment described in Figs. 3e5 has been repeated in order to verify its repeatability and a good agreement between

The data related to the experiment carried out in “double mode” have been gathered the 20th October 2011 from 11:11 until 15:41. The water temperatures and volumetric flow rates measured during the test carried out in “double mode” are

Pcool,FC

COPth,TDC SWAP

SWAP

SWAP

BARREL B is discharging BARREL A is charging

BARREL A is discharging BARREL B is charging

Double mode operation

BARREL A is discharging BARREL B is charging

0.70

BARREL B is discharging BARREL A is charging

0.65 0.60 0.55 0.50 0.45 0.40 0.35 0.30 0.25 0.20 0.15 0.10 0.05

0.00 11:30

11:59

12:28

12:57

13:25

13:54

14:23

14:52

15:21

15:49

16:18

16:47

17:16

Time (hh:mm)

Fig. 5 e Pth,TDC, Pcool,FC and COPth values measured during “normal mode” operation.

COPth (-)

Power (kW)

Pth,TDC 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 11:01

3.2.

724

i n t e r n a t i o n a l j o u r n a l o f r e f r i g e r a t i o n 3 6 ( 2 0 1 3 ) 7 1 7 e7 2 9

Table 2 e Duration of both charging/discharging phases and swap periods. 1st charging phase

2nd charging phase

3rd charging phase

4th charging phase

1st swap

2nd swap

3rd swap

116.5

87.3

93.3

89.1

5.3

5.3

5.0

Duration (min)

reported in Figs. 6 and 7; Fig. 8 depicts the thermal power supplied to the TDC system (Pth,TDC) and the cooling power produced by the TDC system (Pcool,FC) as a function of the time during “double mode” operation. Pth,TDC is calculated by using Eq. (1), while Eq. (3) provides Pcool,FC. As can be derived from Figs. 6e8, TDC operation in “double mode” starts around 11:11 with both barrels discharging (cooling power is provided); at around 11:20 the discharging phase stops and both barrels start charging. A “swap” is performed at the end of each charging period due to the fact that Barrel A becomes completely charged (level ¼ 100%). Barrel A and Barrel B have been charged and discharged six times through the experiment. Fig. 6 shows that the water temperature coming from the boiler towards the TDC (Tw,B,out) is around 80.5  C during charging phase and the minimum temperature drop across the machine is around 10  C; the temperature from the HD to the machine (Tw,HD,out) oscillates between around 21 and 27  C during charging phase and between around 28 and 39  C during discharging periods. Minimum water temperature coming from the TDC towards fan-coil (Tw,FC,in) is around 15  C. Except during the “swap” between the barrels, the volumetric flow rate through both the thermally driven cooling machine (V_ w;TDC ) and fan-coil (V_ w;FC ) is 14.4 l min1, while 22.8 l min1 is the water flow pumped towards the HD (Fig. 7). As can be derived from Fig. 8, during both charging and “swap” phases the TDC system cannot provide cooling power; during discharging periods, cooling capacity increases till reaching a maximum and then becomes zero: maximum Text

Tw,B,out

Tw,TDC,out

value of cooling power gathered during the test is about 3.0 kW. The measured values of Pcool,FC are significantly (around 50%) lower than the expected ones: in fact, thanks to the concurrent discharge of both barrels, “double mode” operation should result in higher cooling power in comparison to the “normal mode” operation. This could be due to the low water flow rate entering the absorption system. Compared to the test carried out in “normal mode”, a higher charging power has been measured during “double mode” operation (as expected). However the manufacturer does not provide any information regarding the operation in “double mode” and, therefore, it is not possible a comparison with the measured values. The cumulative cooling energy provided by the TDC system (Ecool,FC) and the cumulative thermal energy supplied to the TDC system (Eth,TDC) throughout the experiment are equal to 8383.3 kJ and 167266.6 kJ, respectively; as a consequence, a very low value (0.05) is obtained for the thermal Coefficient of Performance by using Eq. (5). In Table 3 the duration of both charging/discharging phases and “swap” periods is reported: as can be derived from this table, the five “swaps” have a duration around 6.5 min; the discharging phase has a duration of about 13.5 min; regarding the charging phase, the duration oscillates between 23.7 and 29.8 min. The experiment described in Figs. 6e8 has been repeated in order to verify its repeatability and a good agreement between the results mentioned above and those achieved during the repeated test has been found. The data reported above agrees Tw,FC,in

Tw,FC,out

Tw,HD,in

Tw,HD,out

87 82 77

72 67

Temperature (°C)

62 57 52 47 42 37 32 27 22 17

12 11:11

11:27

11:43

11:59

12:15

12:30

12:46

13:02

13:18

13:34

13:50

14:06

14:21

14:37

14:53

15:09

15:25

Time (hh:mm)

Fig. 6 e Water temperature values measured during “double mode” operation.

15:41

725

i n t e r n a t i o n a l j o u r n a l o f r e f r i g e r a t i o n 3 6 ( 2 0 1 3 ) 7 1 7 e7 2 9

100 96 92

Barrel level (%)

88 84 80 76 72 68

Barrel A level Barrel B level

64

Vw,TDC = Vw,FC = 14.4 l min-1 Vw,HD = 22.8 l min-1

60 11:11 11:27 11:43 11:59 12:15 12:30 12:46 13:02 13:18 13:34 13:50 14:06 14:21 14:37 14:53 15:09 15:25 15:41

Time (hh:mm) Fig. 7 e Volumetric water flow rates and barrel level during “double mode” operation.

well also with the values recorded by the CW10 internal software (named CWIC2).

electrically driven vapor compression chiller (EDC) from an energy, economic and environmental point of view. The comparison is performed by assuming that:

4. Energy, economic and environmental analysis

U TDC operates with the same water temperature and mass flow rates measured during the experiments; U thermal energy required by TDC is supplied by solar collectors with the auxiliary thermal energy, required in case of scarce solar irradiation, provided by a natural gasfired boiler.

In order to assess the suitability of the thermally driven chiller experimentally investigated in this paper, in the following its measured performances are compared with those of an

Pth,TDC

Pcool,FC

36

3.2

34

3.0

32

2.8

30

2.6

28 2.4 26

Pth,TDC (kW)

22

2.0

20

1.8

18

1.6

16

1.4

14

1.2

12

1.0

10 0.8 8 0.6

6 4

0.4

2

0.2

0 11:11

11:27

11:43

11:59

12:15

12:30

12:46

13:02

13:18

13:34

13:50

14:06

14:21

14:37

14:53

15:09

15:25

Time (hh:mm)

Fig. 8 e Pth,TDC and Pcool,FC values measured during “double mode” operation.

0.0 15:41

Pcool,FC (kW)

2.2

24

726

i n t e r n a t i o n a l j o u r n a l o f r e f r i g e r a t i o n 3 6 ( 2 0 1 3 ) 7 1 7 e7 2 9

Table 3 e Duration of both charging/discharging phases and swap periods. Charging phases

Swap periods

1st

2nd

3rd

4th

5th

6th

1st

2nd

3rd

4th

5th

6th

1st

2nd

3rd

4th

5th

29.8

25.7

23.7

27.2

25.5

24.3

13.2

13.5

13.7

13.7

13.5

13.3

6.5

6.3

6.7

6.5

6.7

Due to the poor performance of CW10 during “double mode” operation, the following analysis will be limited to the experimental data gathered during “normal mode” operation. In order to compare TDC with EDC from an energy point of view, the Primary Energy Ratio (PER) has been evaluated. This parameter is defined as the ratio between the useful energy output supplied to the end-user (Ecool,FC) and the primary energy consumption; as a consequence, the values of Primary Energy Ratio (PER) for both TDC and EDC can be calculated as follows: PERTDC ¼ Ecool;FC =ðEFB $Eth;TDC =hB Þ$100

(6)

PEREDC ¼ Ecool;FC =ðEel;EDC =hPP Þ$100 ¼ COPel;EDC $hPP $100

(7)

where Ecool,FC is the cumulative cooling energy provided by the TDC during “normal mode” operation (equal to 57818.1 kJ), Eth,TDC is the cumulative thermal energy supplied to the TDC during “normal mode” operation (equal to186501.1 kJ), hB is the efficiency of the natural gas-fired boiler, hPP is the efficiency of Power Plant (PP) producing electric energy, Eel,EDC is the electric energy required by EDC for providing the same cooling energy Ecool,FC of TDC, COPel,EDC is the electric Coefficient of Performance of EDC (defined as the ratio between the cooling power supplied by EDC and the electric power consumed by EDC), EFB is the fraction of Eth,TDC provided by the natural gas-fired boiler (so that the difference (1  EFB) is the fraction of Eth,TDC recovered from solar collectors).

PER_TDC

Fig. 9 shows the values of both PERTDC and PEREDC at varying EFB from 0.1 (natural gas-fired boiler produces 10% of Eth,TDC) to 0.9 (solar collectors field provides 10% of Eth,TDC). The data depicted in this figure have been obtained by assuming the following values: U hB ¼ 0.9; U hPP ¼ 0.46 (Rosato and Sibilio, 2012); U COPel,EDC ¼ 2. The value of hPP includes transmission and distribution losses. In the same figure the values of Primary Energy Saving (PES ) are also reported. The parameter PES allows to evaluate the potential of primary energy saving; so it is defined as reported below: PES ¼ ½1  ðPEREDC =PERTDC Þ$100

(8)

Positive values of PES mean that TDC allows for an energy saving in comparison to EDC. Fig. 9 denotes that PES increases at decreasing the value of EFB till reaching its maximum value (around 70%) when EFB ¼ 0.1. From this figure it can be derived that the thermally drive chiller investigated in this work is suitable from an energy point of view (PES > 0) if compared to a conventional electrically driven refrigerating system with COPel,EDC ¼ 2 only in case of EFB < 0.3, i.e. only when the most part (at least 70%)

PER_EDC

PES

285

70

265

50

30

245

10

225

-10

PER (%)

205

-30

185

-50

165

-70

145

-90

125

-110 -130

105

-150 85

-170

65

-190

45

-210

25 -230 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 0.60 0.65 0.70 0.75 0.80 0.85 0.90 0.95 1.00

EFB (-)

Fig. 9 e Primary energy ratio and primary energy saving as a function of EFB.

PES (%)

Duration (min)

Discharging phases

727

i n t e r n a t i o n a l j o u r n a l o f r e f r i g e r a t i o n 3 6 ( 2 0 1 3 ) 7 1 7 e7 2 9

CO2_EDC

DeltaCO2

CO2 (%) C

Carbon dioxide equivalent emissions (kgCO2)

CO2_TDC

80 12.0 11.5 60 11.0 10.5 40 10.0 9.5 20 9.0 0 8.5 8.0 -20 7.5 7.0 -40 6.5 -60 6.0 5.5 -80 5.0 4.5 -100 4.0 -120 3.5 3.0 -140 2.5 2.0 -160 1.5 -180 1.0 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 0.60 0.65 0.70 0.75 0.80 0.85 0.90 0.95 1.00

EFB (-)

Fig. 10 e Carbon dioxide equivalent emissions as a function of EFB.

 CO2;TDC ¼ ½b$ðEFB $Eth;TDC =hB Þ 3600

(9)

CO2;EDC ¼ a$Eel;EDC =3600

(10)

where a represents the equivalent CO2 emissions in the power plant for 1 kWh of electric energy consumed and

Operating cost (€)

C_TDC

b represents the equivalent CO2 emissions for 1 kWh of primary energy consumed. The following values have been assumed:  a ¼ 0.523 kgCO2 kWh1 (Rosato and Sibilio, 2012)  b ¼ 0.2 kgCO2 kWh1 (Rosato and Sibilio, 2012). The equivalent CO2 emissions due to electricity production are typical of the mix of technologies adopted in the Italian geographic area. Fig. 10 shows the values of CO2,TDC and CO2,EDC as function of EFB. The percentage difference DCO2 between CO2,TDC and CO2,EDC is also reported in Fig. 10: DCO2 ¼ ½1  ðCO2;TDC =CO2;EDC Þ$100 C_EDC

(11)

DeltaC

4.9 80 4.7 60 4.5 40 4.3 4.1 20 3.9 0 3.7 3.5 -20 3.3 -40 3.1 2.9 -60 2.7 -80 2.5 -100 2.3 2.1 -120 1.9 -140 1.7 1.5 -160 1.3 -180 1.1 0.9 -200 0.7 -220 0.5 0.3 -240 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 0.60 0.65 0.70 0.75 0.80 0.85 0.90 0.95 1.00

EFB (-)

Fig. 11 e Operating cost as a function of EFB during “normal mode” operation.

C (%)

of thermal energy required by the TDC is recovered from solar collectors. However the choice of the energy conversion technology cannot be based only on the energy performances, but it should be also affected by the assessment of the environmental impact. In the following the carbon dioxide equivalent emissions of both TDC and EDC have been assessed by using the following formulas:

728

PES (%), CO C 2 (%), C (%)

i n t e r n a t i o n a l j o u r n a l o f r e f r i g e r a t i o n 3 6 ( 2 0 1 3 ) 7 1 7 e7 2 9

80 60 40 20 0 -20 -40 -60 -80 -100 -120 -140 -160 -180 PES for COP_EDC=1.5 -200 -220 PES for COP_EDC=3 -240 DeltaCO2 for COP_EDC=1.5 -260 -280 DeltaCO2 for COP_EDC=3 -300 -320 DeltaC for COP_EDC=1.5 -340 DeltaC for COP_EDC=3 -360 -380 -400 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 0.60 0.65 0.70 0.75 0.80 0.85 0.90 0.95 1.00

EFB (-)

Fig. 12 e Comparison between absorption heat pump and electric driven chiller.

Data reported in Fig. 10 show that, in comparison to the EDC, thermally driven chiller investigated in this paper allows for a reduction of CO2 emissions only in case the fraction of Eth,TDC provided by the natural gas-fired boiler is lower than around 36%; as a consequence, the thermal energy supplied to the TDC coming from solar collectors has to be larger than 64% in order to guarantee the suitability of the TDC in comparison to the EDC from an environmental point of view. However the evaluation of economic performance indices is also necessary to complete the analysis of the TDC suitability. As known, the assessment of investment profitability depends on country conditions, as feed-in tariffs, bonus payment, market mechanism and even tax rebates. As a consequence, estimating economic benefits is made difficult by the large number of parameters involved and by the fact that incentives are often assigned according to complex schemes. In the following only the operating cost of the TDC has been evaluated and compared to that one of EDC in order to give a general indication. Natural gas and electricity prices in the domestic sector vary largely across Europe: TDC system financial viability in the Italian market is investigatedbyassuminganelectricenergypriceCUel equalto0.18 V kWh1 (Rosato and Sibilio, 2012) and a natural gas price CUng equal to 0.80 V Nm3 (Rosato and Sibilio, 2012). The operating cost of both TDC and EDC has been estimated by using the following equations:
CTDC ¼ EFB $Eth;TDC = 3600$hB $LHVng $CUng

(12)

CEDC ¼ Ecool;TDC =ð3600$COPEDC Þ$CUel

(13)

where LHVng is the Lower Heating Value of natural gas (assumed equal to 9.593 kWh Nm3). The percentage difference between CTDC and CEDC is calculated as follows and reported in Fig. 11: DC ¼ ½1  ðCEDC =CTDC Þ$100

(14)

Fig. 11 shows that, if compared with the EDC, the TDC allows for an operating cost reduction when the parameter EFB becomes lower than around 0.28: this means that the TDC allows to reduce the operating cost only in case the percentage of Eth,TDC recovered from solar collectors is higher than 68%. Taking into consideration that the performance of electric driven chiller is affected by the external weather conditions and loads, the comparison between the absorption chiller and the electric driven chiller has been performed by considering two additional values (1.5 and 3.0) of COPel,EDC. The comparison has been performed from an energy, environmental and economic point of view. Fig. 12 shows the results of this comparison. Fromthis figure it can be derived that, compared to the electric driven chiller with COPel,EDC ¼ 1.5, the thermally drive chiller investigated in this work is suitable from both an energy point of view (PES > 0) and an economic point of view (DC > 0) only in case the thermal energy provided by solar collectors is not lower than 60% of thermal energy required by the TDC; the absorption chiller allows for reducing the carbon dioxide emissions if percentage of Eth,TDC recovered from solar collectors is higher than 52%. In comparison to the electric driven chiller with COPel,EDC ¼ 3.0, the thermally drive chiller investigated in this work allows for saving both energy and money only in case the thermal energy provided by solar collectors is not lower than 80% of thermal energy required by the TDC; the absorption chiller is suitable from an environmental point of view if percentage of Eth,TDC recovered from solar collectors is higher than 75%.

5.

Conclusions

The 4th generation of a three-phase absorption chiller/heat pump that is capable of storing energy internally in the form of crystallized salt (LiCl) with water as refrigerant, patented in 2000 and sold by the Swedish company ClimateWell, has been experimentally investigated. Data have been gathered

i n t e r n a t i o n a l j o u r n a l o f r e f r i g e r a t i o n 3 6 ( 2 0 1 3 ) 7 1 7 e7 2 9

during two different system modes operation: “normal mode” and “double mode”. The performed tests showed a maximum coefficient of performance COPth,TDC equal to about 0.6 while the machine was operating in “normal mode”; the measured system performance during “double mode” was significantly worse than that measured during “normal mode” operation. The measured data have been used to compare the performance of the thermally driven cooling system with that one of a conventional electrically driven refrigerating machine. The comparison has been performed from an energy, economic and environmental point by assuming that the thermal energy required by the TDC is supplied by both a solar collectors field and a natural gas-fired boiler. The comparison pointed out that, in comparison with the EDC, the TDC allows for a reduction of both primary energy consumption, carbon dioxide emissions and operating cost in case of at least 70% of thermal energy required by the TDC is recovered from solar collectors (instead of provided by a conventional natural gas-fired boiler). Comparison between electric driven chiller and absorption heat pump has been also performed by considering two different values of COPel,EDC. However additional tests should be carried out in order to highlight the system performance over a wider range of operating conditions; in addition a comparison of the experimental data against the simulation model developed by Udomsri et al. (2011, 2012) has to be performed in order to verify the accuracy of the model, and the suitability of the model itself for both determining the effect of boundary conditions on the machine efficiency and for evaluating the viability of the thermally driven chiller CW10 in comparison to traditional systems via a techno-economic analysis.

references

AISIN-SEIKI. http://www.aisin.com/ (accessed 26.07.12.). Angrisani, G., Rosato, A., Roselli, C., Sasso, M., Sibilio, S., 2012. Experimental results of a micro-trigeneration installation. Appl. Therm. Eng. 38, 78e90. Balaras, C.A., Gaglia, A.G., Georgopoulou, E., Mirasgedis, S., Sarafidis, Y., Lalas, D.P., 2007. European residential buildings and empirical assessment of the Hellenic building stock, energy consumption, emissions and potential energy savings. Build. Environ. 42, 1298e1314. Bales, C., Ayadi, O., 2009. Modelling of commercial absorption heat pump with integral storage. In: Proceedings Effstock 2009. Swedvac, Stockholm, Sweden. Bales, C., Nordlander, S., 2005. TCA Evaluation, Lab Measurements, Modelling and System Simulations. No. 39809656-4-3. Hogskolan Dalarna, Borlange, Sweden. Boer, R., Haije, W.G., Veldhuis, J.B.J., 2002. Determination of structural, thermodynamic and phase properties in the Na2SeH2O system for application in a chemical heat pump. Thermochim. Acta 395, 3e19. Chicco, G., Mancarella, P., 2009. Distributed multi-generation: a comprehensive view. Renew. Sust. Energ. Rev. 13, 535e551. ClimateWell. http://www.climatewell.com/index.html#/ innovation/Documentation/TechSpecs (accessed 26.07.12.).

729

Conde, M.R., 2004. Properties of aqueous solution of lithium and calcium chlorides: formulations for use in air conditioning equipment design. Int. J. Therm. Sci. 43, 367e382. Fiskum, R.J., Adcock, P.W., DeVault, R.C., 1996. United States Department of Energy thermally activated heat pump program. In: Energy Conversion Engineering Conference. Proceedings of the 31st Intersociety, vol. 732, pp. 736e741. Florides, G.A., Kalogirou, S.A., Tassou, S.A., Wrobel, L.C., 2002. Modelling, simulation and warming impact assessment of a domestic-size absorption solar cooling system. Appl. Therm. Eng. 22, 1313e1325. Henning, H.M., 2007. Solar assisted air conditioning of buildings e an overview. Appl. Therm. Eng. 27, 1734e1749. Hernandez-Santoyo, J., Sanchez-Cifuentes, A., 2003. Trigeneration: an alternative for energy savings. Appl. Energy 76, 219e227. Huicochea, A., Rivera, W., Gutie´rrez-Urueta, G., Bruno, J.C., Coronas, A., 2011. Thermodynamic analysis of a trigeneration system consisting of a micro gas turbine and a double effect absorption chiller. Appl. Therm. Eng. 31, 3347e3353. Kavvadias, K.C., Tosios, A.P., Maroulis, Z.B., 2010. Design of a combined heating, cooling and power system: sizing, operation strategy selection and parametric analysis. Energ. Convers. Manag. 51, 833e845. Li, H., Fu, L., Geng, K., Jiang, Y., 2006. Energy utilization evaluation of CCHP systems. Energy Buildings 38, 253e257. Lin, L., Wang, Y., Al-Shemmeri, T., Ruxton, T., Turner, S., Zeng, S., Huang, J., He, Y., Huang, X., 2007. An experimental investigation of a household size trigeneration. Appl. Therm. Eng. 27, 576e585. McMullan, J.T., 2002. Refrigeration and the environment e issues and strategies for the future. Int. J. Refrigeration 25, 89e99. National instruments. http://www.ni.com/ (accessed 26.07.12.). Ogura, H., Yamamoto, T., Kage, H., 2003. Efficiencies of CaO/H2O/ Ca(OH)2 chemical heat pump for heat storing and heating/ cooling. Energy 28, 1479e1493. Olsson, R., Kaarebring-Olsson, M., Jonsson, S., 2000. A Chemical Heat Pump. Patent No: WO0037864, Sweden. Rosato, A., Sibilio, S., 2012. Calibration and validation of a model for simulating thermal and electric performance of an internal combustion engine-based micro-cogeneration device. Appl. Therm. Eng. 45, 79e98. Serra, L.M., Lozano, M.A., Ramos, J., Ensinas, A.V., Nebra, S.A., 2009. Polygeneration and efficient use of natural resources. Energy 34, 575e586. Srikhirin, P., Aphornratana, S., Chungpaibulpatana, S., 2001. A review of absorption refrigeration technologies. Renew. Sust. Energ. Rev. 5, 343e372. TRNSYS. TRNSYS Coordinator, Solar Energy Laboratory, University of Wisconsin-Madison, 1500 Engineering Drive, 1303 Engineering Research Building Madison, WI 53706, U.S.A. Udomsri, S., Bales, C., Martin, A.R., Martin, V., 2011. Decentralised cooling in district heating network: monitoring results and calibration of simulation model. Energy Buildings 43, 3311e3321. Udomsri, S., Bales, C., Martin, A.R., Martin, V., 2012. Decentralized cooling in district heating network: system simulation and parametric study. Appl. Energy 92, 175e184. Wang, K., Abdelaziz, O., Kisari, P., Vineyard, E.A., 2011. State-ofthe-art review on crystallization control technologies for water/LiBr absorption heat pumps. Int. J. Refrigeration 34, 1325e1337. Wongsuan, W., Kumar, S., Neveu, P., Meunier, F., 2001. A review of chemical heat pump technology and applications. Appl. Therm. Eng. 21, 1489e1519.