- Email: [email protected]
Experimental evaluation of a variable effect LiBr–water absorption chiller designed for high-efficient solar cooling system
Accepted Manuscript Title: Experimental evaluation of a variable effect LiBr-water absorption chiller designed for high-efficient solar cooling system Author: Z.Y. Xu, R.Z. Wang, H.B. Wang PII: DOI: Reference:
S0140-7007(15)00226-1 http://dx.doi.org/doi: 10.1016/j.ijrefrig.2015.07.019 JIJR 3107
To appear in:
International Journal of Refrigeration
Received date: Revised date: Accepted date:
6-10-2014 7-4-2015 15-7-2015
Please cite this article as: Z.Y. Xu, R.Z. Wang, H.B. Wang, Experimental evaluation of a variable effect LiBr-water absorption chiller designed for high-efficient solar cooling system, International Journal of Refrigeration (2015), http://dx.doi.org/doi: 10.1016/j.ijrefrig.2015.07.019. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Experimental evaluation of a variable effect LiBr-water absorption chiller designed for high-efficient solar cooling system
Z.Y. Xu a, R.Z. Wang a,*, H.B. Wang b
a
Institute of Refrigeration and Cryogenics, Key Laboratory for Power Machinery and Engineering of M.O.E., Shanghai Jiao Tong University, Shanghai 200240, China b
Shandong Normal University – Lishan College, Shandong 262500, China
* Corresponding author. Tel.: +86 21 34206548. E-mail address: [email protected] (R.Z. Wang).
Highlights ► A variable effect LiBr-water absorption chiller is studied. ► The chiller is evaluated based upon a real developed 50 kW prototype. ► COP increased from 0.69 to 1.08 under generation temperature from 95 oC to 120oC. ► The effects of several parameters on COP and cooling power were analyzed.
Abstract A variable effect LiBr-water absorption chiller is studied in this paper based upon a real developed 50 kW prototype. The chiller is designed specifically for the high-efficient utilization of the solar power with variable temperature. It can obtain the optimized COP and cooling power under different heat source temperatures. The construction, circulation and testing system of the chiller were introduced. A typical running condition of the chiller from the starting to the steady operating was given to show the dynamic performance. Several groups of the temperatures and COPs were given to show the steady state performance. These data showed that the COP increased from 0.69 to 1.08 under generation temperature from 95oC to 120oC. Besides, the effects of chilled water temperature, cooling water temperature, pump frequency and opening of valve on COP and cooling power were analyzed respectively.
Keywords
-0Page 1 of 14
Absorption chiller, Variable effect, Solar cooling, LiBr-water
Nomenclature HG
high pressure generator
HA
high pressure absorber
CON
condenser
LG1
the first low pressure generator
LG2
the second low pressure generator
EVA
evaporator
ABS
absorber
SHX
solution heat exchanger
COP
coefficient of performance
t
temperature, oC
c
specific heat capacity, kJ/(kg*K)
ρ
density, kg/m3
q
volume flow rate, m3/s
Subscripts ch1
chilling water inlet
ch2
chilling water outlet
c1
cooling water inlet
c2
cooling water outlet
out
outlet
1. Introduction Nowadays, environmental protecting and energy saving issues become more and more urgent. Solar power or waste heat power utilization can reduce these problems. As one of the promising options, absorption refrigeration technology has been researched a lot (Srikhirin, et al., 2001). Thermally driven property of the absorption chiller makes it appropriate for solar cooling system (Kim
-1Page 2 of 14
and Ferreira, 2008; Zhai and Wang, 2009; Wang, et al., 2009; Wang, et al., 2013), CCHP system (Zhang, et al., 2011) or other low grade heat recovery systems (Kalinowski, et al., 2009). Among the different absorption refrigeration technologies, LiBr-water absorption chiller is developed and commercialized well, especially the single effect and double effect LiBr-water absorption chillers. They were studied both experimentally (Asdrubali and Grignaffini, 2005; Izquierdo, et al., 2012; Chaudhari, et al., 1985)and numerically (Florides, et al., 2003; Yin, et al., 2010). To get refrigeration of 5 oC under cooling temperature of 35 oC, single effect and double effect absorption chillers need heat sources about 85-110 oC and higher than 140 oC respectively. Meanwhile, the COPs of these chillers do not rise with the generation temperature, which means the exergy efficiencies decrease when generation temperature rises. Due to the small working range and steady COP, these conventional absorption chillers don't always couple well with the solar power or waste heat. To solve this problem and fulfill the different needs of absorption refrigeration, advanced absorption refrigeration cycles were proposed in the past decades (Kang, et al., 2000). Yattara, et al. (2003) studied the SE/DL (Single effect/Double Lift) absorption cycle. Liu and Wang (2004) studied the system performance of a solar/gas driving double effect LiBr-water absorption system which had a gas driving high pressure generator and a solar driving low pressure generator. Yan, et al. (2013) proposed a novel absorption refrigeration cycle for heat sources with large temperature change which had the similar objective with the SE/DL absorption cycle. Erickson (1991) and Wang and Zheng (2009) proposed several different one-and-a-half effect absorption cycles to fill up the blank area between single effect and double effect absorption cycles. Hong, et al. (2011) proposed the EAX (Evaporator Absorber eXchange) cycle which worked under the similar heat source temperature with the one-and-a-half effect absorption cycles. Biermann (1985) and Kauffman (1984) proposed the variable effect absorption cycle.
Dao (1990) proposed the regenerative absorption cycles which also obtained
variable effect absorption refrigeration. Besides, several cycles are proposed based on the GAX (Generator Absorber eXchange) concept to improve both the working range and system performance (Sabir, et al., 2004; Kang, et al., 1999; Jawahar and Saravanan, 2010). Although these novel cycles extended the working range of absorption refrigeration, they were seldom studied experimentally. In this paper, the performances of a 50 kW variable effect LiBr-water absorption chiller are evaluated. The chiller is built based on a novel variable effect absorption
-2Page 3 of 14
refrigeration cycle (Xu, et al., 2013) whose aim is to provide a choice for solar cooling system activated by medium temperature solar collector.
2. The variable effect LiBr-water absorption chiller The variable effect LiBr-water absorption chiller is expected to get continuously rising COP under rising heat source temperature. It should work between single effect and double effect refrigeration. The construction and circulation schematic of the variable effect LiBr-water absorption chiller invented by the authors (SJTU) is shown in Fig.1. Construction, circulation and heat exchange processes of the chiller are introduced in the following 3 paragraphs. In this chiller, 7 phase-change heat exchangers are distributed in 3 chambers with different pressures. HG (1), HA (2) and inside of LG 1(3) are included in the high pressure chamber. Outside of LG1 (3), LG 2 (4), CON (5) are included in the medium pressure chamber. ABS (6) and EVA (7) are included in the low pressure chamber. Besides, other components including the SHX (8), throttling valve (9), solution sprayer (10) and pressured water tank (11) are also indispensable. HA valve (12),
evaporation pump (13), generation pump (14), absorption pump (15) and hot water pump (16) are controllable components. Heat source supplies heat to the generation process in HG (1). ABS (6) and CON (5) release absorption heat and condensation heat to the cooling water. EVA (7) supplies cooling output to the chilled water. Condensation heat of the high pressure steam supplies heat to the generation process in LG1 (3). Absorption heat from HA (2) is transferred to the LG2 (4) through the closed hot water circuit. According to the theoretical calculation of the variable effect absorption cycle, temperature of the closed hot water circuit may achieve 100oC under some conditions. In this case, a pressured water tank is added in the closed hot water loop. There are two solution circuits in the chiller. In one of the circuits, the solution is pumped out of ABS (6), preheated in SHX (8), boiled in HG (1), cooled down in SHX (8), and flows back to the ABS (6). In the other circuits, the solution is pumped out of ABS (6), preheated in SHX (8), diluted in HA (2), boiled in LG1 (3) and LG 2 (4), cooled down in SHX (8), and flows back to the ABS (6). The
-3Page 4 of 14
refrigerant is generated in HG (1), LG1 (3) and LG2 (4), absorbed in HA (2) and ABS (4), condensed inside the LG1 (3) and evaporated in EVA (7). In the chiller, high pressure vapor is generated from HG (1). It is then partially condensed inside LG1 (3) and partially absorbed in HA (2). The condensation heat and absorption heat are all utilized for low pressure generation, but the difference is that the condensed high pressure vapor becomes refrigerant fluid and flows into EVA (7). The condensed vapor gets double effect refrigeration while the absorbed vapor gets single effect refrigeration.
A 50 kW variable effect absorption chiller was designed and manufactured as shown in Fig.2. Its rated conditions are generation temperature of 125oC, condensation temperature of 40oC, absorption temperature of 35oC and evaporation temperature of 5oC. Shell of the chiller is made of carbon steel. Except that the solution heat exchanger is a flat plate heat exchanger, other heat exchangers are shell and tube heat exchangers. Horizontal bronze finned tubes are used in these heat exchangers. Condenser uses finned tubes with 12mm outer diameter and others use finned tubes with 15.88mm outer diameter. The numbers of finned tubes for the HG (1), HA (2), LG1 (3), LG2 (4), CON (5), ABS (6), and EVA (7) are 108, 76, 64, 92, 84, 196 and 132 respectively. Absorption heat, generation heat, condensation heats and evaporation heats are transferred through falling film heat transfer processes. The chiller was filled with 180kg LiBr-water solution with 50% concentration and 7.5kg pure water after the leakage examination. The pumps in the system are all variable frequency canned motor pump. There are 4 canned motor pumps in this chiller and their rated powers are all 300W, hence the total pump consumption is about 1.2 kW.
3. Experimental process The chiller was tested in the system shown in Fig.3. The system was composed by a steam boiler, the absorption chiller, a cooling tower, a chilled water tank, a heat exchanger between chilled water and cooling water, a data logger and the control system.
As is shown in Fig.3, heat source pressure was controlled by valve (1) between the steam boiler and the absorption chiller. Heat source flow rate was controlled by valve (2) between the absorption
-4Page 5 of 14
chiller and the ambient and valve (1). Valve (3) controlled the cooling water flow rate. Valve (4) controlled the proportion of cooling water flows into the cooling tower. Valve (3), valve (4) and the fan of the cooling tower decided the cooling water temperature. Valve (5) controlled the chilled water flow rate. Three-way valve (6) controlled the proportion of chilled water flows through the heat exchanger which was heated by the cooling water. Valve (5) and three-way valve (6) controlled the chilled water temperature and affected the cooling water temperature. Pictures of the water circuits and the chiller under testing are shown in Fig.4. The chiller control strategy was as follows. As is shown in Fig.1, the absorption pump (14) and the generation pump (15) should have same working frequency to keep a steady solution level in the absorber, so the frequency of the two pumps were controlled together by one frequency converter. It affected the circulating ratio of the chiller. Frequency of hot water pump (16) was controlled by another frequency converter and it affected the heat transfer between HA (2) and LG2 (4). Valve (12) controlled the flow rate ratio between HG (1) and HA (2). This ratio affected the COP of the absorption cycle according to ref (Xu, et al., 2013). Heat source temperature, generation temperature of HG and LG 2, inlet and outlet temperature of chilled water and cooling water were measured by PT100 temperature sensors and collected by Agilent 34970A and 34972A every 5 seconds. Flow rates of chilled water and cooling water were tested by vortex flow meters. PT100 sensor had an accuracy of 0.1oC and was connected to the data logger with four wires. Flow meter had an accuracy of 0.5%.
4. Results and analysis In this testing system, heat source temperature, chilled water temperature, cooling water temperature could be changed. In this chiller, the HA valve opening, frequencies of the generation pump and the hot water pump could be adjusted. So the effects of these parameters on the performance of the chiller are discussed in this part. Both steady state and dynamic performances are included.
4.1. Typical operating condition Fig.5 shows the variations of the heat source temperature, HG outlet temperature, LG2 outlet temperature, cooling water inlet/outlet temperature, and chilled water inlet/outlet temperature from the warming-up to the steady state of the chiller. The temperatures were recorded from the time when heat source started to flow into the generator.
-5Page 6 of 14
Heat source temperature ascended immediately when the system began to work. The steam started to preheat the solution when it was hot enough. During the preheating, the heat source temperature rose slowly while the HG outlet temperature and the LG2 outlet temperature rose rapidly. In this period, steam flow rate was small to keep the solution temperature growing steady. Otherwise, temperature distribution of the chiller could be uneven and some solution might splash into the pure refrigerant. This would cause the refrigerant pollution which should be avoided. This period lasted about 900s.
After the chiller was warmed up, solution in HG started to be boiled. Pressure of HG began to rise. HG outlet temperature and LG2 outlet temperature also started to grow slowly. Inlet and outlet temperature difference of cooling water became large which meant the chiller started to output cooling power. A sudden falling of the heat source temperature occurred for the heat consumption increased a lot after the chiller started to output cooling power. The falling was held back by increasing the heat source flow rate. The system worked steady after the warming-up period as shown in Fig.5. There were still some fluctuations of heat source temperature and generator temperature during the steady operating. This was caused by the pressure decrease and pressure recharge of the steam boiler.
4.2. Performances under different conditions The COP of the chiller is calculated from the Eq. (1). COP
QE
(1)
QG
Here QE is the cooling output, and QG is the heat input. Part of the steam condensed after flowing out of generator. This made the actual amount of steam consumed by the generator hard to be measured accurately. Considering this, the heat input was calculated from Eq. (2). This equation ignored the heat dissipation that was negligible for two reasons. On one hand, only one side and the top of HG/HA were contacted with the ambient. The heat dissipation area was small. On the other hand, HG and HA were all falling film exchanger. The high pressure chamber was filled with vapor, and the temperature of vapor was much lower than the solution.
QG QC Q E
(2)
-6Page 7 of 14
Where QE and QC were obtained from Eq. (3) and Eq. (4) respectively:
Q E c q ch1 ( t ch1 t ch 2 )
(3)
Q C c q c1 ( t c1 t c 2 )
(4)
The density and heat capacity of water were considered constant for the calculation. In this experiment, the flow rate of cooling water and chilled water were fixed at 20m3/h and 10m3/h. With these calculations, several groups of parameters describing the steady state condition were obtained in Table 1.
As is shown in Table 1, cooling water inlet temperature ranged from 27.6 oC to 32.6 oC, chilled water outlet temperature ranged from 7.2 oC to 13.4 oC. As the generation temperature varied from 95 o
C to 120 oC, the thermal COP ascended from 0.69 to 1.08. According to the thermodynamic
calculation of the variable effect absorption cycle, the COP rises from 0.8 to 1.06 under generation temperature from 95 oC to 135 oC (Xu, et al., 2013). The experimental COP rose just like the expectation. The experimental COP reached 1.08 under a lower generation temperature of 120 oC because the cooling temperatures and evaporation temperatures are different. The cooling power ranged from 33.4 kW to 51.9 kW. The cooling power under low generation temperature was small. The LG2 outlet temperature varied from 74.7 oC to 86.1 oC and it stayed stable around 85 oC under high generation temperatures. LG2 outlet temperature and the condensation pressure decided the outlet concentration of LG2. Similar LG2 outlet temperatures illustrated that the generation in LG2 stayed stable when generation temperature was higher than 110oC.
4.3. Effects of chilled water temperature
Fig. 6 shows the COPs and cooling powers under different chilled water outlet temperatures. The triangle points represent the COP values and the circle points represent the cooling power values (the same with the Fig.7, Fig.8 and Fig.10). This group of data was collected under the following condition. Opening of HA valve was 0.5, generation pump frequency was 21 Hz, hot water pump frequency was 30 Hz, generation temperature ranged between 108.8~111.7oC, cooling water temperature ranged between 28.3~28.6 oC. The COP rose from 0.83 to 1.02 and the cooling power varied between 33.7 kW and 42.4 kW when chilled water outlet temperature rose from 10.7 oC to 14.1 oC.
-7Page 8 of 14
As is shown in Fig.6, COP and cooling power rise with the evaporation temperature. This is similar to other absorption chillers. When evaporation temperature becomes higher, the evaporator pressure becomes higher. This reduces the absorber outlet concentration. The concentration change in HG is expanded. More refrigerant vapors are generated in HG and the pressure of the HG rises. As solution in LG1 is boiled by the high pressure condensation heat, LG1 outlet temperature is increased and more vapors are condensed in LG1. According to the analysis in part 2, more double effect refrigeration are yielded, therefore the COP rises.
4.3. Effects of cooling water temperature
Fig.7 shows the COPs and cooling powers under different cooling water temperatures. This group of data was obtained under the following condition. Opening of HA valve was 0.25, generation pump frequency was 18 Hz, hot water pump frequency was 20 Hz, generation temperature ranged between 117.5~118.7oC, chilled water outlet temperature ranged between 10.5~11.3 oC. The COP rose from 0.56 to 0.89 and the cooling power varied between 29.6 kW and 44.6 kW when cooling water inlet temperature rose from 29.5 oC to 34.1 oC. When cooling temperature increases, absorption temperature and condensation temperature are all lifted. The effects of higher absorption temperature are very similar with lower evaporation temperature. It increases the solution concentration out of absorber and yields less double effect refrigeration. On the other hand, higher condensation temperature lifts the pressure in LG1 and LG2. High pressure results in high saturated temperature of solution. This impairs the generation in low pressure generators. Both sides reduce the COP that the COP is affected more deeply by the cooling water temperature than by the chilled water temperature.
4.4. Effects of generation pump frequency
Fig.8 shows the COPs and cooling powers under different generation pump frequencies. Frequency of the generation pump stands for the flow rate of weak solution or circulation ratio of the chiller. This group of data was collected under the following condition. Opening of HA valve was 0.5, hot water pump frequency was 40 Hz, generation temperature ranged between 120~122oC, cooling
-8Page 9 of 14
water temperature ranged between 28.4~29.6 oC, chilled water outlet temperature ranged between 9.1~10.1 oC. The COP varied between 0.92 and 1.03 while cooling power rose from 32.1 kW to 53.3 kW when the generation pump frequency rose from 20 Hz to 23 Hz. More solution pumped into the generator consumed more heat from the heat source, and the chiller obtained more refrigeration. This makes the cooling power rise with the pump frequency. COP rose at first and fell after the pump frequency exceeded 21 Hz. When pump frequency was higher, the solution concentration change in generator was reduced. More heat input was consumed by the pre-heating of the solution which decreased COP.
4.5. Effects of the HA valve opening As mentioned in Ref (Xu, et al., 2013), each generation temperature corresponds with a constant flow rate ratio (between HG inlet and LA outlet) for the optimized COP. This optimized flow rate ratio is small for low generation temperature and big for high generation temperature. In this case, the relation between HA valve opening and COP under low generation temperature and high generation temperature are all recorded and shown in Figure 9.
Fig.9 (a) shows the COP variation under low generation temperature. This group of data was collected under the following condition. Hot water pump frequency was 50 Hz, generation pump frequency was 19.5 Hz, generation temperature ranged between 106.5~108.7oC, cooling water temperature ranged between 30.2~32.9 oC, chilled water outlet temperature ranged between 14.2~15.0 o
C. The COP decreased from 0.99 to 0.60 when the opening of HA valve rose from 0.25 to 1.00. Fig.9 (b) shows the COP variation under high generation temperature. This group of data was
collected under the following condition. Hot water pump frequency was 40 Hz, generation pump frequency was 20 Hz, generation temperature ranged between 116.1~113.1oC, cooling water temperature ranged between 30.5~28.1 oC, chilled water outlet temperature ranged between 13.5~14.5 o
C. The COP rose from 0.70 to 0.94 when the opening of HA valve rose from 0.25 to 1.00. Opening of the HA valve represents the flow rate ratio between HA inlet and HG inlet. Small HA
valve opening means small flow rate at HA inlet, and vice versa. Small flow rate at HA inlet ensures big concentration decrease in HA. This results in a lower HA outlet concentration (which is also the
-9Page 10 of 14
LG1 inlet) and lower saturation temperature at LG1 inlet. This increases the temperature difference in LG1 thus increasing the COP. This is the positive effect. However, when generation temperature increases, the condensation temperature at LG1 becomes higher which amplifies the temperature difference in LG1. The importance of small HA valve opening is weakened. On the other hand, small flow rate in HA also have some negative effect on the COP. The solution flows through HA, LG1 and LG2 in series. Small flow rate in HA will also limit the amount of vapor generated in LG1 and LG2, thus limiting the growth of COP. The negative effect does not change its influence with the generation temperature. Under low generation temperature, the barrier to increase the COP is the low condensation temperature or small temperature difference in LG1. Positive effect of small HA valve opening is fierce. Small HA valve opening get high COP. When generation temperature rises, the positive effect is dwindled and the negative effect remains unchanged. Optimized opening of HA valve rises.
4.6. Effects of hot water pump frequency Fig.10 shows the COPs and cooling powers under different hot water pump frequencies. The hot water pump frequency represents the heat transfer efficiency between HA and LG2.This group of data was collected under the following condition. Generation pump frequency was 21 Hz, HA valve opening was 0.5, generation temperature ranged between 119.0~119.9oC, cooling water temperature ranged between 27.6~28.4 oC, chilled water outlet temperature ranged between 9.0~10.4 oC. The COP rose from 0.84 to 1.08 while the cooling power varied between 34.8kW and 44.6kW when the hot water pump frequency rose from 0 Hz to 30 Hz.
In this chiller, the heat transfer between HA and LG2 was momentous. This part of heat transfer makes it possible to obtain the extra refrigeration in LG1. When the frequency is reduced, heat transfer efficiency is impaired and the amount of extra refrigeration decreases. On the contrary, COP rises with the increasing of hot water pump frequency.
5. Conclusion In this paper, circulation of the variable effect LiBr-water absorption chiller was introduced. Experimental evaluation of a 50 kW variable effect absorption chiller was carried out. Different performances of the chiller were analyzed and the results are as follows:
-10Page 11 of 14
(1).A typical running condition of the chiller from the starting to the steady operating state was given to show the dynamic performance. The chiller parameters kept steady after the warming-up process. (2).A group of data of the steady state parameters was given to show the steady state performance. The chiller can obtain rising COPs from 0.69 to 1.08 under generation temperature from 95oC to 120oC. The corresponding overall COP varied from 0.67 to 1.05. (3).The effects of chilled water temperature, cooling water temperature, generation pump frequency, HA valve opening and hot water pump frequency were analyzed respectively. The chiller performance will be higher under high chilled water temperature, low cooling water temperature, proper frequency of generation pump, optimized HA valve opening and high frequency of hot water pump. These tendencies conform to the qualitative analysis of the chiller.
Acknowledge This work was supported by the key project of the Natural Science Foundation of China for international academic exchanges under the contract No. 51020105010. The support from the Ministry of Education innovation team (IRT 1159) was also appreciated.
References Srikhirin, P., Aphornratana, S., Chungpaibulpatana, S. A review of absorption refrigeration technologies. Renew. Sust. Energ Rev. 2001; 5: 343-72. Kim, D.S., Ferreira, C.A.I. Solar refrigeration options - a state-of-the-art review. Int. J. Refrigeration 2008; 31: 3-15. Zhai, X.Q., Wang, R.Z. A review for absorbtion and adsorbtion solar cooling systems in China. Renew. Sust. Energ Rev. 2009; 13: 1523-31. Wang, R.Z., Ge, T.S., Chen, C.J., Ma, Q., Xiong, Z.Q. Solar sorption cooling systems for residential applications: options and guidelines. Int. J. Refrigeration 2009; 32: 638-60. Wang, R.Z., Yu, X., Ge, T.S., Li, T.X. The present and future of residential refrigeration, power generation and energy storage. Appl. Therm. Eng. 2013; 53: 256-70. Zhang, C.Z., Yang, M., Lu, M., Shan, Y.H., Zhu, J.X. Experimental research on LiBr refrigeration Heat pump system applied in CCHP system. Appl. Therm. Eng. 2011; 31: 3706-12. Kalinowski, P., Hwang, Y., Radermacher, R., Hashimi, S., Rodgers, P. Application of waste heat powered absorption refrigeration system to the LNG recovery process. Int. J. Refrigeration 2009; 32: 687-94. Asdrubali, F., Grignaffini, S. Experimental evaluation of the performances of a H2O-LiBr absorption refrigerator under different service conditions. Int. J. Refrigeration 2005; 28: 489-97.
-11Page 12 of 14
Izquierdo, M., Marcos, J.D., Palacios, M.E., Gonzalez-Gil, A. Experimental evaluation of a low-power direct air-cooled double-effect LiBr-H2O absorption prototype. Energy 2012; 37: 737-48. Chaudhari, S., Paranjape, D., Eisa, M., Holland, F. A study of the operating characteristics of a water-lithium bromide absroption heat pump. Journal of heat recovery systems 1985; 5: 285-97. Florides, G.A., Kalogirou, S.A., Tassou, S.A., Wrobel, L.C. Design and construction of a LiBr-water absorption machine. Energ. Convers. Manage. 2003; 44: 2483-508. Yin, H.X., Qu, M., Archer, D.H. Model based experimental performance analysis of a microscale LiBr-H2O steam-driven double-effect absorption Chiller. Appl. Therm. Eng. 2010; 30: 1741-50. Kang, Y.T., Kunugi, Y., Kashiwagi, T. Review of advanced absorption cycles: performance improvement and temperature lift enhancement. Int. J. Refrigeration 2000; 23: 388-401. Yattara, A., Zhu, Y., Ali, M.M. Comparison between solar single-effect and single-effect double-lift absorption machines (Part I). Appl. Therm. Eng. 2003; 23: 1981-92. Liu, Y.L., Wang, R.Z. Performance prediction of a solar/gas driving double effect LiBr-H2O absorption system. Renew. Energy 2004; 29: 1677-95. Yan, X.N., Chen, G.M., Hong, D.L., Lin, S.R., Tang, L.M. A novel absorption refrigeration cycle for heat sources with large temperature change. Appl. Therm. Eng. 2013; 52: 179-86. Erickson, D.C. One-and-a-half effect absorption cycle.US Patents, Patent; 1991. Wang, J.Z., Zheng, D.X. Performance of one and a half-effect absorption cooling cycle of H2O/LiBr system. Energ. Convers. Manage. 2009; 50: 3087-95. Hong, D.L., Chen, G.M., Tang, L.M., He, Y.J. Simulation research on an EAX (Evaporator-Absorber-Exchange) absorption refrigeration cycle. Energy 2011; 36: 94-8. Biermann, W.J. Variable effect desorber-resorber absorption cycle.US Patents, Patent; 1985. Kauffman, K.W. Variable effect absorption machine and process.US Patents, Patent; 1984. Dao, K. Advanced regenerative absorption refrigeration cycles.US Patents, Patent; 1990. Sabir, H.M., Chretienneau, R., ElHag, Y.B.M. Analytical study of a novel GAX-R heat driven refrigeration cycle. Appl. Therm. Eng. 2004; 24: 2083-99. Kang, Y.T., Akisawa, A., Kashiwagi, T. An advanced GAX cycle for waste heat recovery WGAX cycle. Appl. Therm. Eng. 1999; 19: 933-47. Jawahar, C.P., Saravanan, R. Generator absorber heat exchange based absorption cycle-A review. Renew. Sust. Energ Rev. 2010; 14: 2372-82. Xu, Z.Y., Wang, R.Z., Xia, Z.Z. A novel variable effect LiBr-water absorption refrigeration cycle. Energy 2013; 60: 457-63.
Fig.1. Schematic of the SJTU variable effect absorption chiller Fig.2. Picture of the SJTU variable effect absorption chiller Fig.3. Schematic of the testing system Fig.4. Pictures of the testing system
Fig.5. Typical operating condition of the chiller Fig.6. Effects of chilled water temperature on COP & cooling power Fig.7. Effects of cooling water temperature on COP & cooling power Fig.8. Effects of generation pump frequency Fig.9. Effects of HA valve opening
-12Page 13 of 14
(a) Under low generation temperature (b) Under high generation temperature Fig.10. Effects of hot water pump frequency
Table 1. Steady state parameters of the chiller Cooling
No.
tHG,out
tLG2,out
tch1 (oC)
tch2 (oC)
tc1 (oC)
tc2 (oC)
1
95.0
78.3
12.5
9.4
31.2
34.9
35.9
0.69
2
100.2
80.4
10.2
7.5
32.5
35.9
33.4
0.71
3
101.9
81.1
10.1
7.2
32.6
36.0
33.4
0.72
4
104.4
74.7
16.4
13.2
29.8
34.2
45.5
0.78
5
107.7
75.1
16.6
13.4
28.6
32.2
37.4
0.82
6
110.9
84.3
15.0
10.7
32.4
37.2
43.5
0.85
7
115.8
86.1
14.7
10.5
30.9
35.1
49.0
1.00
8
118.8
84.6
11.0
7.2
28.2
32.0
43.7
1.02
9
120.0
85.4
13.5
9.0
27.6
31.9
51.9
1.08
power(kW)
COP
-13Page 14 of 14
S0140-7007(15)00226-1 http://dx.doi.org/doi: 10.1016/j.ijrefrig.2015.07.019 JIJR 3107
To appear in:
International Journal of Refrigeration
Received date: Revised date: Accepted date:
6-10-2014 7-4-2015 15-7-2015
Please cite this article as: Z.Y. Xu, R.Z. Wang, H.B. Wang, Experimental evaluation of a variable effect LiBr-water absorption chiller designed for high-efficient solar cooling system, International Journal of Refrigeration (2015), http://dx.doi.org/doi: 10.1016/j.ijrefrig.2015.07.019. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Experimental evaluation of a variable effect LiBr-water absorption chiller designed for high-efficient solar cooling system
Z.Y. Xu a, R.Z. Wang a,*, H.B. Wang b
a
Institute of Refrigeration and Cryogenics, Key Laboratory for Power Machinery and Engineering of M.O.E., Shanghai Jiao Tong University, Shanghai 200240, China b
Shandong Normal University – Lishan College, Shandong 262500, China
* Corresponding author. Tel.: +86 21 34206548. E-mail address: [email protected] (R.Z. Wang).
Highlights ► A variable effect LiBr-water absorption chiller is studied. ► The chiller is evaluated based upon a real developed 50 kW prototype. ► COP increased from 0.69 to 1.08 under generation temperature from 95 oC to 120oC. ► The effects of several parameters on COP and cooling power were analyzed.
Abstract A variable effect LiBr-water absorption chiller is studied in this paper based upon a real developed 50 kW prototype. The chiller is designed specifically for the high-efficient utilization of the solar power with variable temperature. It can obtain the optimized COP and cooling power under different heat source temperatures. The construction, circulation and testing system of the chiller were introduced. A typical running condition of the chiller from the starting to the steady operating was given to show the dynamic performance. Several groups of the temperatures and COPs were given to show the steady state performance. These data showed that the COP increased from 0.69 to 1.08 under generation temperature from 95oC to 120oC. Besides, the effects of chilled water temperature, cooling water temperature, pump frequency and opening of valve on COP and cooling power were analyzed respectively.
Keywords
-0Page 1 of 14
Absorption chiller, Variable effect, Solar cooling, LiBr-water
Nomenclature HG
high pressure generator
HA
high pressure absorber
CON
condenser
LG1
the first low pressure generator
LG2
the second low pressure generator
EVA
evaporator
ABS
absorber
SHX
solution heat exchanger
COP
coefficient of performance
t
temperature, oC
c
specific heat capacity, kJ/(kg*K)
ρ
density, kg/m3
q
volume flow rate, m3/s
Subscripts ch1
chilling water inlet
ch2
chilling water outlet
c1
cooling water inlet
c2
cooling water outlet
out
outlet
1. Introduction Nowadays, environmental protecting and energy saving issues become more and more urgent. Solar power or waste heat power utilization can reduce these problems. As one of the promising options, absorption refrigeration technology has been researched a lot (Srikhirin, et al., 2001). Thermally driven property of the absorption chiller makes it appropriate for solar cooling system (Kim
-1Page 2 of 14
and Ferreira, 2008; Zhai and Wang, 2009; Wang, et al., 2009; Wang, et al., 2013), CCHP system (Zhang, et al., 2011) or other low grade heat recovery systems (Kalinowski, et al., 2009). Among the different absorption refrigeration technologies, LiBr-water absorption chiller is developed and commercialized well, especially the single effect and double effect LiBr-water absorption chillers. They were studied both experimentally (Asdrubali and Grignaffini, 2005; Izquierdo, et al., 2012; Chaudhari, et al., 1985)and numerically (Florides, et al., 2003; Yin, et al., 2010). To get refrigeration of 5 oC under cooling temperature of 35 oC, single effect and double effect absorption chillers need heat sources about 85-110 oC and higher than 140 oC respectively. Meanwhile, the COPs of these chillers do not rise with the generation temperature, which means the exergy efficiencies decrease when generation temperature rises. Due to the small working range and steady COP, these conventional absorption chillers don't always couple well with the solar power or waste heat. To solve this problem and fulfill the different needs of absorption refrigeration, advanced absorption refrigeration cycles were proposed in the past decades (Kang, et al., 2000). Yattara, et al. (2003) studied the SE/DL (Single effect/Double Lift) absorption cycle. Liu and Wang (2004) studied the system performance of a solar/gas driving double effect LiBr-water absorption system which had a gas driving high pressure generator and a solar driving low pressure generator. Yan, et al. (2013) proposed a novel absorption refrigeration cycle for heat sources with large temperature change which had the similar objective with the SE/DL absorption cycle. Erickson (1991) and Wang and Zheng (2009) proposed several different one-and-a-half effect absorption cycles to fill up the blank area between single effect and double effect absorption cycles. Hong, et al. (2011) proposed the EAX (Evaporator Absorber eXchange) cycle which worked under the similar heat source temperature with the one-and-a-half effect absorption cycles. Biermann (1985) and Kauffman (1984) proposed the variable effect absorption cycle.
Dao (1990) proposed the regenerative absorption cycles which also obtained
variable effect absorption refrigeration. Besides, several cycles are proposed based on the GAX (Generator Absorber eXchange) concept to improve both the working range and system performance (Sabir, et al., 2004; Kang, et al., 1999; Jawahar and Saravanan, 2010). Although these novel cycles extended the working range of absorption refrigeration, they were seldom studied experimentally. In this paper, the performances of a 50 kW variable effect LiBr-water absorption chiller are evaluated. The chiller is built based on a novel variable effect absorption
-2Page 3 of 14
refrigeration cycle (Xu, et al., 2013) whose aim is to provide a choice for solar cooling system activated by medium temperature solar collector.
2. The variable effect LiBr-water absorption chiller The variable effect LiBr-water absorption chiller is expected to get continuously rising COP under rising heat source temperature. It should work between single effect and double effect refrigeration. The construction and circulation schematic of the variable effect LiBr-water absorption chiller invented by the authors (SJTU) is shown in Fig.1. Construction, circulation and heat exchange processes of the chiller are introduced in the following 3 paragraphs. In this chiller, 7 phase-change heat exchangers are distributed in 3 chambers with different pressures. HG (1), HA (2) and inside of LG 1(3) are included in the high pressure chamber. Outside of LG1 (3), LG 2 (4), CON (5) are included in the medium pressure chamber. ABS (6) and EVA (7) are included in the low pressure chamber. Besides, other components including the SHX (8), throttling valve (9), solution sprayer (10) and pressured water tank (11) are also indispensable. HA valve (12),
evaporation pump (13), generation pump (14), absorption pump (15) and hot water pump (16) are controllable components. Heat source supplies heat to the generation process in HG (1). ABS (6) and CON (5) release absorption heat and condensation heat to the cooling water. EVA (7) supplies cooling output to the chilled water. Condensation heat of the high pressure steam supplies heat to the generation process in LG1 (3). Absorption heat from HA (2) is transferred to the LG2 (4) through the closed hot water circuit. According to the theoretical calculation of the variable effect absorption cycle, temperature of the closed hot water circuit may achieve 100oC under some conditions. In this case, a pressured water tank is added in the closed hot water loop. There are two solution circuits in the chiller. In one of the circuits, the solution is pumped out of ABS (6), preheated in SHX (8), boiled in HG (1), cooled down in SHX (8), and flows back to the ABS (6). In the other circuits, the solution is pumped out of ABS (6), preheated in SHX (8), diluted in HA (2), boiled in LG1 (3) and LG 2 (4), cooled down in SHX (8), and flows back to the ABS (6). The
-3Page 4 of 14
refrigerant is generated in HG (1), LG1 (3) and LG2 (4), absorbed in HA (2) and ABS (4), condensed inside the LG1 (3) and evaporated in EVA (7). In the chiller, high pressure vapor is generated from HG (1). It is then partially condensed inside LG1 (3) and partially absorbed in HA (2). The condensation heat and absorption heat are all utilized for low pressure generation, but the difference is that the condensed high pressure vapor becomes refrigerant fluid and flows into EVA (7). The condensed vapor gets double effect refrigeration while the absorbed vapor gets single effect refrigeration.
A 50 kW variable effect absorption chiller was designed and manufactured as shown in Fig.2. Its rated conditions are generation temperature of 125oC, condensation temperature of 40oC, absorption temperature of 35oC and evaporation temperature of 5oC. Shell of the chiller is made of carbon steel. Except that the solution heat exchanger is a flat plate heat exchanger, other heat exchangers are shell and tube heat exchangers. Horizontal bronze finned tubes are used in these heat exchangers. Condenser uses finned tubes with 12mm outer diameter and others use finned tubes with 15.88mm outer diameter. The numbers of finned tubes for the HG (1), HA (2), LG1 (3), LG2 (4), CON (5), ABS (6), and EVA (7) are 108, 76, 64, 92, 84, 196 and 132 respectively. Absorption heat, generation heat, condensation heats and evaporation heats are transferred through falling film heat transfer processes. The chiller was filled with 180kg LiBr-water solution with 50% concentration and 7.5kg pure water after the leakage examination. The pumps in the system are all variable frequency canned motor pump. There are 4 canned motor pumps in this chiller and their rated powers are all 300W, hence the total pump consumption is about 1.2 kW.
3. Experimental process The chiller was tested in the system shown in Fig.3. The system was composed by a steam boiler, the absorption chiller, a cooling tower, a chilled water tank, a heat exchanger between chilled water and cooling water, a data logger and the control system.
As is shown in Fig.3, heat source pressure was controlled by valve (1) between the steam boiler and the absorption chiller. Heat source flow rate was controlled by valve (2) between the absorption
-4Page 5 of 14
chiller and the ambient and valve (1). Valve (3) controlled the cooling water flow rate. Valve (4) controlled the proportion of cooling water flows into the cooling tower. Valve (3), valve (4) and the fan of the cooling tower decided the cooling water temperature. Valve (5) controlled the chilled water flow rate. Three-way valve (6) controlled the proportion of chilled water flows through the heat exchanger which was heated by the cooling water. Valve (5) and three-way valve (6) controlled the chilled water temperature and affected the cooling water temperature. Pictures of the water circuits and the chiller under testing are shown in Fig.4. The chiller control strategy was as follows. As is shown in Fig.1, the absorption pump (14) and the generation pump (15) should have same working frequency to keep a steady solution level in the absorber, so the frequency of the two pumps were controlled together by one frequency converter. It affected the circulating ratio of the chiller. Frequency of hot water pump (16) was controlled by another frequency converter and it affected the heat transfer between HA (2) and LG2 (4). Valve (12) controlled the flow rate ratio between HG (1) and HA (2). This ratio affected the COP of the absorption cycle according to ref (Xu, et al., 2013). Heat source temperature, generation temperature of HG and LG 2, inlet and outlet temperature of chilled water and cooling water were measured by PT100 temperature sensors and collected by Agilent 34970A and 34972A every 5 seconds. Flow rates of chilled water and cooling water were tested by vortex flow meters. PT100 sensor had an accuracy of 0.1oC and was connected to the data logger with four wires. Flow meter had an accuracy of 0.5%.
4. Results and analysis In this testing system, heat source temperature, chilled water temperature, cooling water temperature could be changed. In this chiller, the HA valve opening, frequencies of the generation pump and the hot water pump could be adjusted. So the effects of these parameters on the performance of the chiller are discussed in this part. Both steady state and dynamic performances are included.
4.1. Typical operating condition Fig.5 shows the variations of the heat source temperature, HG outlet temperature, LG2 outlet temperature, cooling water inlet/outlet temperature, and chilled water inlet/outlet temperature from the warming-up to the steady state of the chiller. The temperatures were recorded from the time when heat source started to flow into the generator.
-5Page 6 of 14
Heat source temperature ascended immediately when the system began to work. The steam started to preheat the solution when it was hot enough. During the preheating, the heat source temperature rose slowly while the HG outlet temperature and the LG2 outlet temperature rose rapidly. In this period, steam flow rate was small to keep the solution temperature growing steady. Otherwise, temperature distribution of the chiller could be uneven and some solution might splash into the pure refrigerant. This would cause the refrigerant pollution which should be avoided. This period lasted about 900s.
After the chiller was warmed up, solution in HG started to be boiled. Pressure of HG began to rise. HG outlet temperature and LG2 outlet temperature also started to grow slowly. Inlet and outlet temperature difference of cooling water became large which meant the chiller started to output cooling power. A sudden falling of the heat source temperature occurred for the heat consumption increased a lot after the chiller started to output cooling power. The falling was held back by increasing the heat source flow rate. The system worked steady after the warming-up period as shown in Fig.5. There were still some fluctuations of heat source temperature and generator temperature during the steady operating. This was caused by the pressure decrease and pressure recharge of the steam boiler.
4.2. Performances under different conditions The COP of the chiller is calculated from the Eq. (1). COP
QE
(1)
QG
Here QE is the cooling output, and QG is the heat input. Part of the steam condensed after flowing out of generator. This made the actual amount of steam consumed by the generator hard to be measured accurately. Considering this, the heat input was calculated from Eq. (2). This equation ignored the heat dissipation that was negligible for two reasons. On one hand, only one side and the top of HG/HA were contacted with the ambient. The heat dissipation area was small. On the other hand, HG and HA were all falling film exchanger. The high pressure chamber was filled with vapor, and the temperature of vapor was much lower than the solution.
QG QC Q E
(2)
-6Page 7 of 14
Where QE and QC were obtained from Eq. (3) and Eq. (4) respectively:
Q E c q ch1 ( t ch1 t ch 2 )
(3)
Q C c q c1 ( t c1 t c 2 )
(4)
The density and heat capacity of water were considered constant for the calculation. In this experiment, the flow rate of cooling water and chilled water were fixed at 20m3/h and 10m3/h. With these calculations, several groups of parameters describing the steady state condition were obtained in Table 1.
As is shown in Table 1, cooling water inlet temperature ranged from 27.6 oC to 32.6 oC, chilled water outlet temperature ranged from 7.2 oC to 13.4 oC. As the generation temperature varied from 95 o
C to 120 oC, the thermal COP ascended from 0.69 to 1.08. According to the thermodynamic
calculation of the variable effect absorption cycle, the COP rises from 0.8 to 1.06 under generation temperature from 95 oC to 135 oC (Xu, et al., 2013). The experimental COP rose just like the expectation. The experimental COP reached 1.08 under a lower generation temperature of 120 oC because the cooling temperatures and evaporation temperatures are different. The cooling power ranged from 33.4 kW to 51.9 kW. The cooling power under low generation temperature was small. The LG2 outlet temperature varied from 74.7 oC to 86.1 oC and it stayed stable around 85 oC under high generation temperatures. LG2 outlet temperature and the condensation pressure decided the outlet concentration of LG2. Similar LG2 outlet temperatures illustrated that the generation in LG2 stayed stable when generation temperature was higher than 110oC.
4.3. Effects of chilled water temperature
Fig. 6 shows the COPs and cooling powers under different chilled water outlet temperatures. The triangle points represent the COP values and the circle points represent the cooling power values (the same with the Fig.7, Fig.8 and Fig.10). This group of data was collected under the following condition. Opening of HA valve was 0.5, generation pump frequency was 21 Hz, hot water pump frequency was 30 Hz, generation temperature ranged between 108.8~111.7oC, cooling water temperature ranged between 28.3~28.6 oC. The COP rose from 0.83 to 1.02 and the cooling power varied between 33.7 kW and 42.4 kW when chilled water outlet temperature rose from 10.7 oC to 14.1 oC.
-7Page 8 of 14
As is shown in Fig.6, COP and cooling power rise with the evaporation temperature. This is similar to other absorption chillers. When evaporation temperature becomes higher, the evaporator pressure becomes higher. This reduces the absorber outlet concentration. The concentration change in HG is expanded. More refrigerant vapors are generated in HG and the pressure of the HG rises. As solution in LG1 is boiled by the high pressure condensation heat, LG1 outlet temperature is increased and more vapors are condensed in LG1. According to the analysis in part 2, more double effect refrigeration are yielded, therefore the COP rises.
4.3. Effects of cooling water temperature
Fig.7 shows the COPs and cooling powers under different cooling water temperatures. This group of data was obtained under the following condition. Opening of HA valve was 0.25, generation pump frequency was 18 Hz, hot water pump frequency was 20 Hz, generation temperature ranged between 117.5~118.7oC, chilled water outlet temperature ranged between 10.5~11.3 oC. The COP rose from 0.56 to 0.89 and the cooling power varied between 29.6 kW and 44.6 kW when cooling water inlet temperature rose from 29.5 oC to 34.1 oC. When cooling temperature increases, absorption temperature and condensation temperature are all lifted. The effects of higher absorption temperature are very similar with lower evaporation temperature. It increases the solution concentration out of absorber and yields less double effect refrigeration. On the other hand, higher condensation temperature lifts the pressure in LG1 and LG2. High pressure results in high saturated temperature of solution. This impairs the generation in low pressure generators. Both sides reduce the COP that the COP is affected more deeply by the cooling water temperature than by the chilled water temperature.
4.4. Effects of generation pump frequency
Fig.8 shows the COPs and cooling powers under different generation pump frequencies. Frequency of the generation pump stands for the flow rate of weak solution or circulation ratio of the chiller. This group of data was collected under the following condition. Opening of HA valve was 0.5, hot water pump frequency was 40 Hz, generation temperature ranged between 120~122oC, cooling
-8Page 9 of 14
water temperature ranged between 28.4~29.6 oC, chilled water outlet temperature ranged between 9.1~10.1 oC. The COP varied between 0.92 and 1.03 while cooling power rose from 32.1 kW to 53.3 kW when the generation pump frequency rose from 20 Hz to 23 Hz. More solution pumped into the generator consumed more heat from the heat source, and the chiller obtained more refrigeration. This makes the cooling power rise with the pump frequency. COP rose at first and fell after the pump frequency exceeded 21 Hz. When pump frequency was higher, the solution concentration change in generator was reduced. More heat input was consumed by the pre-heating of the solution which decreased COP.
4.5. Effects of the HA valve opening As mentioned in Ref (Xu, et al., 2013), each generation temperature corresponds with a constant flow rate ratio (between HG inlet and LA outlet) for the optimized COP. This optimized flow rate ratio is small for low generation temperature and big for high generation temperature. In this case, the relation between HA valve opening and COP under low generation temperature and high generation temperature are all recorded and shown in Figure 9.
Fig.9 (a) shows the COP variation under low generation temperature. This group of data was collected under the following condition. Hot water pump frequency was 50 Hz, generation pump frequency was 19.5 Hz, generation temperature ranged between 106.5~108.7oC, cooling water temperature ranged between 30.2~32.9 oC, chilled water outlet temperature ranged between 14.2~15.0 o
C. The COP decreased from 0.99 to 0.60 when the opening of HA valve rose from 0.25 to 1.00. Fig.9 (b) shows the COP variation under high generation temperature. This group of data was
collected under the following condition. Hot water pump frequency was 40 Hz, generation pump frequency was 20 Hz, generation temperature ranged between 116.1~113.1oC, cooling water temperature ranged between 30.5~28.1 oC, chilled water outlet temperature ranged between 13.5~14.5 o
C. The COP rose from 0.70 to 0.94 when the opening of HA valve rose from 0.25 to 1.00. Opening of the HA valve represents the flow rate ratio between HA inlet and HG inlet. Small HA
valve opening means small flow rate at HA inlet, and vice versa. Small flow rate at HA inlet ensures big concentration decrease in HA. This results in a lower HA outlet concentration (which is also the
-9Page 10 of 14
LG1 inlet) and lower saturation temperature at LG1 inlet. This increases the temperature difference in LG1 thus increasing the COP. This is the positive effect. However, when generation temperature increases, the condensation temperature at LG1 becomes higher which amplifies the temperature difference in LG1. The importance of small HA valve opening is weakened. On the other hand, small flow rate in HA also have some negative effect on the COP. The solution flows through HA, LG1 and LG2 in series. Small flow rate in HA will also limit the amount of vapor generated in LG1 and LG2, thus limiting the growth of COP. The negative effect does not change its influence with the generation temperature. Under low generation temperature, the barrier to increase the COP is the low condensation temperature or small temperature difference in LG1. Positive effect of small HA valve opening is fierce. Small HA valve opening get high COP. When generation temperature rises, the positive effect is dwindled and the negative effect remains unchanged. Optimized opening of HA valve rises.
4.6. Effects of hot water pump frequency Fig.10 shows the COPs and cooling powers under different hot water pump frequencies. The hot water pump frequency represents the heat transfer efficiency between HA and LG2.This group of data was collected under the following condition. Generation pump frequency was 21 Hz, HA valve opening was 0.5, generation temperature ranged between 119.0~119.9oC, cooling water temperature ranged between 27.6~28.4 oC, chilled water outlet temperature ranged between 9.0~10.4 oC. The COP rose from 0.84 to 1.08 while the cooling power varied between 34.8kW and 44.6kW when the hot water pump frequency rose from 0 Hz to 30 Hz.
In this chiller, the heat transfer between HA and LG2 was momentous. This part of heat transfer makes it possible to obtain the extra refrigeration in LG1. When the frequency is reduced, heat transfer efficiency is impaired and the amount of extra refrigeration decreases. On the contrary, COP rises with the increasing of hot water pump frequency.
5. Conclusion In this paper, circulation of the variable effect LiBr-water absorption chiller was introduced. Experimental evaluation of a 50 kW variable effect absorption chiller was carried out. Different performances of the chiller were analyzed and the results are as follows:
-10Page 11 of 14
(1).A typical running condition of the chiller from the starting to the steady operating state was given to show the dynamic performance. The chiller parameters kept steady after the warming-up process. (2).A group of data of the steady state parameters was given to show the steady state performance. The chiller can obtain rising COPs from 0.69 to 1.08 under generation temperature from 95oC to 120oC. The corresponding overall COP varied from 0.67 to 1.05. (3).The effects of chilled water temperature, cooling water temperature, generation pump frequency, HA valve opening and hot water pump frequency were analyzed respectively. The chiller performance will be higher under high chilled water temperature, low cooling water temperature, proper frequency of generation pump, optimized HA valve opening and high frequency of hot water pump. These tendencies conform to the qualitative analysis of the chiller.
Acknowledge This work was supported by the key project of the Natural Science Foundation of China for international academic exchanges under the contract No. 51020105010. The support from the Ministry of Education innovation team (IRT 1159) was also appreciated.
References Srikhirin, P., Aphornratana, S., Chungpaibulpatana, S. A review of absorption refrigeration technologies. Renew. Sust. Energ Rev. 2001; 5: 343-72. Kim, D.S., Ferreira, C.A.I. Solar refrigeration options - a state-of-the-art review. Int. J. Refrigeration 2008; 31: 3-15. Zhai, X.Q., Wang, R.Z. A review for absorbtion and adsorbtion solar cooling systems in China. Renew. Sust. Energ Rev. 2009; 13: 1523-31. Wang, R.Z., Ge, T.S., Chen, C.J., Ma, Q., Xiong, Z.Q. Solar sorption cooling systems for residential applications: options and guidelines. Int. J. Refrigeration 2009; 32: 638-60. Wang, R.Z., Yu, X., Ge, T.S., Li, T.X. The present and future of residential refrigeration, power generation and energy storage. Appl. Therm. Eng. 2013; 53: 256-70. Zhang, C.Z., Yang, M., Lu, M., Shan, Y.H., Zhu, J.X. Experimental research on LiBr refrigeration Heat pump system applied in CCHP system. Appl. Therm. Eng. 2011; 31: 3706-12. Kalinowski, P., Hwang, Y., Radermacher, R., Hashimi, S., Rodgers, P. Application of waste heat powered absorption refrigeration system to the LNG recovery process. Int. J. Refrigeration 2009; 32: 687-94. Asdrubali, F., Grignaffini, S. Experimental evaluation of the performances of a H2O-LiBr absorption refrigerator under different service conditions. Int. J. Refrigeration 2005; 28: 489-97.
-11Page 12 of 14
Izquierdo, M., Marcos, J.D., Palacios, M.E., Gonzalez-Gil, A. Experimental evaluation of a low-power direct air-cooled double-effect LiBr-H2O absorption prototype. Energy 2012; 37: 737-48. Chaudhari, S., Paranjape, D., Eisa, M., Holland, F. A study of the operating characteristics of a water-lithium bromide absroption heat pump. Journal of heat recovery systems 1985; 5: 285-97. Florides, G.A., Kalogirou, S.A., Tassou, S.A., Wrobel, L.C. Design and construction of a LiBr-water absorption machine. Energ. Convers. Manage. 2003; 44: 2483-508. Yin, H.X., Qu, M., Archer, D.H. Model based experimental performance analysis of a microscale LiBr-H2O steam-driven double-effect absorption Chiller. Appl. Therm. Eng. 2010; 30: 1741-50. Kang, Y.T., Kunugi, Y., Kashiwagi, T. Review of advanced absorption cycles: performance improvement and temperature lift enhancement. Int. J. Refrigeration 2000; 23: 388-401. Yattara, A., Zhu, Y., Ali, M.M. Comparison between solar single-effect and single-effect double-lift absorption machines (Part I). Appl. Therm. Eng. 2003; 23: 1981-92. Liu, Y.L., Wang, R.Z. Performance prediction of a solar/gas driving double effect LiBr-H2O absorption system. Renew. Energy 2004; 29: 1677-95. Yan, X.N., Chen, G.M., Hong, D.L., Lin, S.R., Tang, L.M. A novel absorption refrigeration cycle for heat sources with large temperature change. Appl. Therm. Eng. 2013; 52: 179-86. Erickson, D.C. One-and-a-half effect absorption cycle.US Patents, Patent; 1991. Wang, J.Z., Zheng, D.X. Performance of one and a half-effect absorption cooling cycle of H2O/LiBr system. Energ. Convers. Manage. 2009; 50: 3087-95. Hong, D.L., Chen, G.M., Tang, L.M., He, Y.J. Simulation research on an EAX (Evaporator-Absorber-Exchange) absorption refrigeration cycle. Energy 2011; 36: 94-8. Biermann, W.J. Variable effect desorber-resorber absorption cycle.US Patents, Patent; 1985. Kauffman, K.W. Variable effect absorption machine and process.US Patents, Patent; 1984. Dao, K. Advanced regenerative absorption refrigeration cycles.US Patents, Patent; 1990. Sabir, H.M., Chretienneau, R., ElHag, Y.B.M. Analytical study of a novel GAX-R heat driven refrigeration cycle. Appl. Therm. Eng. 2004; 24: 2083-99. Kang, Y.T., Akisawa, A., Kashiwagi, T. An advanced GAX cycle for waste heat recovery WGAX cycle. Appl. Therm. Eng. 1999; 19: 933-47. Jawahar, C.P., Saravanan, R. Generator absorber heat exchange based absorption cycle-A review. Renew. Sust. Energ Rev. 2010; 14: 2372-82. Xu, Z.Y., Wang, R.Z., Xia, Z.Z. A novel variable effect LiBr-water absorption refrigeration cycle. Energy 2013; 60: 457-63.
Fig.1. Schematic of the SJTU variable effect absorption chiller Fig.2. Picture of the SJTU variable effect absorption chiller Fig.3. Schematic of the testing system Fig.4. Pictures of the testing system
Fig.5. Typical operating condition of the chiller Fig.6. Effects of chilled water temperature on COP & cooling power Fig.7. Effects of cooling water temperature on COP & cooling power Fig.8. Effects of generation pump frequency Fig.9. Effects of HA valve opening
-12Page 13 of 14
(a) Under low generation temperature (b) Under high generation temperature Fig.10. Effects of hot water pump frequency
Table 1. Steady state parameters of the chiller Cooling
No.
tHG,out
tLG2,out
tch1 (oC)
tch2 (oC)
tc1 (oC)
tc2 (oC)
1
95.0
78.3
12.5
9.4
31.2
34.9
35.9
0.69
2
100.2
80.4
10.2
7.5
32.5
35.9
33.4
0.71
3
101.9
81.1
10.1
7.2
32.6
36.0
33.4
0.72
4
104.4
74.7
16.4
13.2
29.8
34.2
45.5
0.78
5
107.7
75.1
16.6
13.4
28.6
32.2
37.4
0.82
6
110.9
84.3
15.0
10.7
32.4
37.2
43.5
0.85
7
115.8
86.1
14.7
10.5
30.9
35.1
49.0
1.00
8
118.8
84.6
11.0
7.2
28.2
32.0
43.7
1.02
9
120.0
85.4
13.5
9.0
27.6
31.9
51.9
1.08
power(kW)
COP
-13Page 14 of 14