Académique Documents
Professionnel Documents
Culture Documents
[21.10.201411:41am]
[120]
//blrnas3.glyph.com/cenpro/ApplicationFiles/Journals/SAGE/3B2/BSEJ/Vol00000/140024/APPFile/SG-BSEJ140024.3d (BSE) [PREPRINTER
stage]
Article
Abstract
This paper studies the substitution of air condensation by water condensation in the chiller of a university
building. This building has a surface of about 30,000 m2, is located in a city of the Spanish north coast, and
refrigeration is needed for six months a year. In this paper, the novelty lies in the cooling system of the
condensation water, since the typical cooling tower is not going to be considered; instead, the domestic
cold water consumed in the building would be used to cool the chiller condensing water. As the domestic
water consumption in the building is neither constant nor coincident with the cooling requirements of the
chiller, the use of the fire water tank as heat storage was proposed. Thus, the chiller is cooled by the
water contained in the fire tank and this fire water is then cooled by the domestic water. With this
solution, the reduction in the annual energy consumption of the chiller is around 18.6%. The requirements
to implement this solution in buildings are: cool water for air conditioning produced in a centralized way, a
relatively high building domestic water demand and a building fire water tank.
Practical application:The solution proposed in this paper can contribute to primary energy savings by
improving the energy efficiency of HVAC systems. It can be adopted in all buildings that require refrigeration, have a significant consumption of domestic water, and have a water storage tank for fire or other
uses. The article includes the preliminary design of a waterwater heat exchanger as a possible solution
for cooling the chiller, but alternative solutions could be adopted. Finally, a cost-benefit analysis has been
performed, showing that the proposed solution is feasible.
Keywords
Air conditioning, chiller, water condensation, energy efficiency, thermal storage
Introduction
Corresponding author:
CJ Renedo, Electrical and Energy Engineering Department,
University of Cantabria, E.T.S.I.I. y T, Avda. de los Castros,
39005 Santander, Spain.
Email: renedoc@unican.es
31% Industrial
24% Buildings
5% Agriculture
38% Services
62% Residential
46% Heating
25% Heating
33% Offices
11% Hospitals
7% Lighting
22% Commercial
27% Others
28% Lighting
30% Restaurants
16% Others
4% Educational
Renedo et al.
Compared with
Improvement
No control
Conventional system
Conventional system
Reference
7
11
12
13
14
15
47
14
14
14
14
22
(continued)
Renedo et al.
Table 1. Continued.
Studied system
Compared with
Improvement
Free cooling
20
Onoff control
PID and Fuzzy controllers
26
30
21
Conventional system
22
22
Existing operation
44
Evaporative cooling
Passive cooling dissipation with
earth to air heat exchangers
Passive cooling dissipation with
indirect evaporative cooling
Passive cooling dissipation with
indirect/direct evaporative
cooling
Passive cooling dissipation with
indirect/direct evaporative
cooling
Central chilling system using
optimal control settings
Smart chiller sequencing
System control with artificial
neural networks and genetic
algorithms
Water pump control strategy
Conventional system
Existing operation
Existing operation
Reference
20
27
24
21
22
22
6,47
42
(continued)
Downloaded from bse.sagepub.com at COLUMBIA UNIV on March 1, 2015
Table 1. Continued.
Studied system
Compared with
Improvement
47
R22 chiller
45
Combining variable air volumebased chilled water air conditioning system with thermal
energy storage under demand
controlled combined with the
economizer cycle ventilation
Swimming pools as heat sinks
Aquifer thermal storage
Variable Refrigerant Flow (VRF)
Multi-split VRF
Multi-split VRF
Multi-split VRV system with the
heat recovery ventilation units
time with the cooling requirements of the chiller.52 In order to harmonize this temporal mismatch (batch process),53 the re water tank is
going to be used as an intermediate heat storage.
This article includes an analysis of the system,
showing 18.6% reduction in the annual energy
consumption of the chiller. A sensitivity analysis
has been performed that takes into account four
aspects: (1) variation in water demand (WD) of
the building, (2) variation of the supply water
temperature, (3) change in distribution temperature of domestic water (TW) inside the building,
and (4) increased CD.
Case study
The building studied is located in Santander,
Spanish city located on the coast of
Cantabrian Sea. The prevailing weather in the
city is shown in Table 2,5456 and the building
needs cooling from May to October. The
Reference
28
31
46
30
30
30
30
Renedo et al.
3.82 W
Cooling
Dry bulb Wet bulb Dry bulb Wet bulb Dry bulb Wet bulb
64 m
19.5
1%
24.7
19.4
2%
23.7
19
September
18.3
October
15.4
22.5
19.6
14.1
11.3
85
135
78
77
157
127
8.4
8.3
90 m
N
Faculty
of Law
110 m
Faculty of
Economics
Table 3. Natural gas consumed by the chiller, water demand, and water inlet temperature in the period of building
cooling.
Working days
3
NGC (Nm )
TWD (m3)
TTW.in ( C)
Month
Day
Month
Day
May
June
July
August
September
October
22
21
22
22
21
22
459.8
20.9
2140
97.3
13.0
907.7
43.2
2179
103.8
15.0
1755.3
79.8
2320
105.5
16.0
1429.7
65.0
1940
88.2
16.0
1592.7
75.8
2498
119.0
16.0
770.3
35.0
2113
96.0
14.0
Proposed facility
For the design of the facility, the substitution of
air condensation by water condensation in the
chiller has been studied; water condensation
being energetically more favorable since it provides a higher COP by lowering the condensation temperature of the refrigerant uid.
The scheme of the new system is shown in
Figure 3. In order to undertake this design, it
is necessary (1) to install in the chiller a water
condenser in parallel with the current one of air,
Renedo et al.
Current System
Water distributed
into the Building
Water
Condenser
Air
Condenser
Chilled Water
Auxiliary
Water
cooler
Evaporator
Bypass
Fire
Water
Tank
(50 m3)
Closed-Loop Water
Compressor
Expansor
Pump
1.5 m
C.L.W.
C.L.W.
7.5 m
C.L.W.
4.5 m
C.L.W.
Figure 4. Scheme of the immersion heat exchanger in the fire water tank.
and (4) to install a water cooler to serve, if necessary, as auxiliary cooling of the tank.
The new waterwater exchanger could be a
simple bundle of tubes immersed in the tank
and for these tubes circulate the water consumed
by the building, Figure 4. The urban network
supplies water from a reservoir located about
10
40 m above the level of the building, and therefore the incoming water has sucient pressure
to overcome the pressure losses caused by the
installation of the heat exchanger before of the
building water network. In the case that a crack
appears in the heat exchanger, the urban network pressure would cause leakage from the network water into the tank, therefore the domestic
water distributed in the building will be never
polluted.
This heat exchanger can be sized using some
of the numerous methods for calculating the
heat transfer coecient. In this case, the ow
inside the pipes will lead to forced convection
while the external ow will lead to natural
convection.
The internal heat transfer coecient can be
determined from Dittus-Boelter correlation.
Taking into account de value of the Reynolds
number and the Prandtl number, the Nusselt
number can be calculated and then the convective coecient. In this case, hci is about 5103 W/
(m2K). The external heat transfer coecient can
be estimated from the Rayleigh number; in this
case, hce is about 5.5102 W/(m2K). Considering
the thinness of the pipes (1 mm) and the high
thermal conductivity of the copper (395 W/
(mK)), its thermal resistance is negligible compared to convection one.
Finally, the LMTD method (logarithmic
mean temperature dierence) can be used to
determine the eective surface of the heat
exchanger. In summary, the heat exchanger
shown in Figure 4 would consist of two layers
of 120 copper pipes 6/8 mm diameter (inside/
outside) and 4 m long. These results show the
technical feasibility of the proposed solution
since such a heat exchanger could be placed
into the existing re tank.
It could be also possible to use a plate heat
exchanger, through which circulate the domestic
water and the tank water. However, the purpose
of this paper is not the mechanical calculation of
this element, but studying the technical feasibility of the proposed system.
The water temperature in the tank must be
low enough in order to be available in case of
Mathematical modeling
In this section, the hourly consumption of NG is
modeled considering the chiller is cooled using
the re water tank. This requires knowing three
aspects:
. The hourly demand of air conditioning. It has
been determined taking into account: the
hourly demand prole, the NG consumption
and the chiller COP in the current air-cooled
system.
. The operating parameters of the chiller,
which have been obtained from the technical
data sheet of the machine.
. The re water temperature, that has been calculated taking into account the three following factors: The initial re water temperature,
the heat supplied by the chiller condenser and
the heat removed by the domestic water used
in the building.
The inuence of gains and losses to the environment has not been considered due to the fact
that gains are negligible compared to losses
when calculating the re water temperature.
According to Table 2 and taking into account
that the maximum expected re water temperature is about 21 C, gains from the environment
Renedo et al.
11
24
X
CDi
h1
15,0%
% Water
12,5%
% Refrigeration
10,0%
7,5%
5,0%
2,5%
23 to 24
22 to 23
21 to 22
20 to 21
19 to 20
18 to 19
17 to 18
16 to 17
15 to 16
14 to 15
13 to 14
12 to 13
11 to 12
10 to 11
9 to 10
0,0%
8 to 9
12
(3) temperature variations between the refrigerant and the water cooled in the evaporator
of 7 C.64
The adjustment of simulation parameters for
COP calculation, such as pressure losses in condenser, evaporator and suction line, superheat in
evaporator and in suction line, condenser subcooling, and compressor isoentropic eciency,
was performed considering a COP in the chiller
of 2.93, in nominal working conditions.57 These
are: 251 kW of cooling production, a NGC of
24 Nm3/h, an air temperature in the condenser
of 35 C, and 12 and 7 C as input and output
temperatures of the cooled water. This agrees
with equation (3).
COPjNominal
Cooling Capacity kWh
NGCNm3 NCVNG kWhNm3
Gas Engine Efficiency
3
To calculate the CD(d) value, the %CD(i)
value, Figure 5 has been taken into account.
This is an iterative process that, considering
the COPa(i), is conducted to verify that the
NGC(d) obtained coincides with that of Table 3.
Then, the hourly cooling demand was established for the building, CD(i), equation (4). For
this, CD(d) and %CD(i) have been considered,
Figure 5.
CDi CDd %CDi
The NGC with air condensation has been calculated each hour, NGCa(i), considering
NGCa(d), Table 2; the %CD(i), Figure 5; and
the COP simulated hourly, COPa(i), equation (5).
%CDi COPai
NGCai NGCad P24
i1 %CDi COPai
5
Renedo et al.
13
QCond wi kJ
Mw Tank kg cp kJ=kg C
7
CDi 1
1
COPw i
8
9
In this study, the maximum supply temperature of TW inside the building has been limited
to 25 C, so the TTW is calculated based on
(TTW in) as equation (11)
TTW C 25 C TTW in C
11
the COP obtained each hour with water condensation, COPW (i), NCVNG, and the eciency of
NG engine, equation (12), the daily NGC,
NGCW (d), is calculated with the corresponding
sum, equation (13).
NGC: wi
CDi
COPwi NCVNG Engine Efficiency
12
NGCwd
NGCwi
13
In summary, the calculation of NGC(d) follows several steps: (1) to estimate CD(1) and
TWD(1), (2) to know TTank (0), (3) to calculate
COPW (1), (4) to calculate QCond W (1), (5) to calculate TT_Cond (i), (6) to calculate TT TW, (7) to
calculate TTank (1), (8) to calculate NGC(1), (9) to
repeat the process for the following hours, and
(10) to make the sum of all the NGC(i) values.
In these calculations, the power consumption
caused by the new pumps installed in the water
condensation system was not considered,
because as initial estimate, it is assumed that
this is compensated by the reduction in consumption that occurs when fans of the chiller
air condensation system stop.
(a)
COPa
NGCw (Nm3)
NGCa (Nm3)
COPw
5,8
10,0
5,4
8,5
7,0
5,5
4,6
4,0
MAY
20 to 21
19 to 20
18 to 19
17 to 18
16 to 17
15 to 16
14 to 15
(b)
COPa
NGCw (Nm3)
NGCa (Nm3)
COPw
5,8
10,0
5,4
8,5
7,0
5,0
5,5
4,6
4,0
JULY
20 to 21
19 to 20
18 to 19
17 to 18
16 to 17
15 to 16
14 to 15
13 to 14
12 to 13
1,0
11 to 12
3,8
10 to 11
2,5
9 to 10
4,2
8 to 9
COP
13 to 14
12 to 13
11 to 12
1,0
10 to 11
3,8
9 to 10
2,5
8 to 9
4,2
NG (Nm3)
COP
5,0
NG (Nm3)
14
Figure 6. Hourly COP and NGC of air and water condensation for the two months of minimum and maximum
cooling demand.
Renedo et al.
15
May
June
July
August
September
October
22,0
TTank(i) (C)
19,5
17,0
14,5
22 to 23
21 to 22
20 to 21
19 to 20
18 to 19
17 to 18
16 to 17
15 to 16
14 to 15
13 to 14
12 to 13
11 to 12
10 to 11
9 to 10
8 to 9
12,0
Figure 7. Hourly temperature of the fire water tank for the six month with cooling demand.
June
July
Aug.
Sep.
Oct.
25,251
1148
20,209
919
23,309
1110
12,264
557
Building CD (kWthh)
Monthly
Daily
7566
344
14,256
679
Monthly
Daily
Monthly
Daily
459.8
20.9
375
17.0
907.7
43.2
753.5
35.8
Cost analysis
Considering that the cost of NG is about 0.095
E/kWh, the annual cost saving would be near
1300 E. Furthermore, it must be taken into
account that the cost of the NG has been
increasing along the years and is expected to
go on rising.
On the other hand, if this proposal would
have been taken into account during the design
phase of the building, the re water tank and the
Annual
1755.3
79.7
1446.9
65.7
1429.7
64.9
1134.2
51.5
1592.7
75.8
1291.1
61.4
770.3
35.0
628.1
28.5
6915.4
5628.4
16
Table 5. Sensitivity analysis of monthly NGC (Nm3) with varying: water demand (WD), inlet temperature of tap
water in the building (TTW in), and temperature of distribution tap water in the building (TTW db).
Air condensation
Water condensation
Water demand (10%)
Water demand (+10%)
TTW in (+1 C)
TTW in (1 C)
TTW db (1 C)
TTW db (+1 C)
May
June
July
Aug.
Sep.
Oct.
Annual
459.8
374.5
374.5
374.5
387.5
362.3
374.5
374.5
907.7
753.5
753.5
753.5
778.5
730.1
753.5
753.5
1755.3
1446.9
1484.2
1406.4
1539.8
1361.2
1492.4
1404.3
1429.7
1134.2
1158.6
1117.4
1201.9
1081.1
1163.9
1116.6
1592.7
1291.1
1309.7
1286.0
1362.6
1245.0
1319.2
1285.6
770.3
628.1
628.1
628.1
648.2
607.0
628.1
628.1
6915.4
5628.4
5708.7
5565.9
5918.6
5386.8
5731.7
5562.5
Sensitivity analysis
This study was completed by performing a sensitivity analysis that takes into account three
aspects, which are:
. change in building WD (10%);
. variation in urban water supply temperature,
TTW in, of (1 C); and
Renedo et al.
17
2.000
NGC (Nm3)
1.500
1.000
500
Oct.
Sep.
Aug.
July
June
May
Month
Figure 8. Monthly natural gas demand when considering a 10% increase in building thermal demand.
Conclusions
The study presented in this paper has considered
the use of the TW consumed in a university
building to cool the condenser of the chiller,
which provides the cooling used for air conditioning in the building. Due to the dierence
between cooling needs and the water consumption, the use of the re water tank has been
proposed for heat storage.
The results reveal that performing the condensation at low temperatures, the COP of the
chiller increases, which produces several benecial eects: (1) the annual energy consumption
of the chiller is reduced, in this case 18.6%, (2)
Funding
This research received no specic grant from any
funding agency in the public, commercial, or notfor-prot sectors.
References
1. Perez-Lombard L, Ortiz J and Pout C. A review on buildings energy consumption information. Energy Build 2008;
40: 394398.
18
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
descarga?file=/documentos_10540_Comentarios_RITE_
GT7_07_2200d691.pdf (accessed 02 June 2014).
Waqas A and Din ZU. Phase change material (PCM)
storage for free cooling of buildings A review. Renew
Sustain Energy Rev 2013; 18: 607625.
Santamouris M, Sfakianaki A and Pavlou K. On the
efficiency of night ventilation techniques applied to residential buildings. Energy Build 2010; 42: 13091313.
Kim MH, Kim JH, Kwon OH, et al. Energy conservation potential of an indirect and direct evaporative cooling assisted 100% outdoor air system. Build Service Eng
Res Technol 2011; 32: 345360.
Santamouris M and Kolokotsa D. Passive cooling
dissipation techniques for buildings and other
structures: The state of the art. Energy Build 2013; 57:
7494.
Qiun GQ and Riffat SB. Novel design and modelling of
an evaporative cooling system for buildings. Int J
Energy Res 2006; 30: 985999.
Liu X and Hong T. Comparison of energy efficiency
between variable refrigerant flow systems and ground
source heat pump systems. Energy Build 2010; 42:
584589.
Underwood CP and Spider JD. Analysis of vertical
ground heat exchangers applied to buildings in the
UK. Build Service Eng Res Technol 2007; 28: 133159.
Xiangguo X, Chao S and Shiming D. Using two parallel
connected compressors to implement a novel control
algorithm for improved indoor humidity at a low cost.
Build Service Eng Res Technol 2012; 34: 349354.
Ekren O, Sahin S and Isler Y. Comparison of different
controllers for variable speed compressor and electronic
expansion valve. Int J Refrigerat 2010; 33: 11611168.
Chinnaraj C, Govindarajan P and Vijayan R. Influence
of electronic expansion valve on the performance of
small window air conditioner retrofitted with R407C
and R290. Thermal Sci 2011; 15: 327339.
Zhou YP, Wu JY, Wang RZ, et al. Simulation and
experimental validation of the variable-refrigerantvolume (VRV) air-conditioning system in EnergyPlus.
Energy Build 2008; 40: 10411047.
Aynur TN. Variable refrigerant flow systems: A review.
Energy Build 2010; 42: 11061112.
Chinnaraj C, Vijayan R and Govindarajan P. Analysis
of eco friendly refrigerants usage in air-conditioner. Am
J Environ Sci 2011; 7: 510514.
Balghouthi M, Chahbani MH and Guisan A. Feasibility
of solar absorption air conditioning in Tunisia. Build
Environ 2008; 43: 14591470.
Hassan HZ and Mohamad AA. A review on solar cold
production through absorption technology. Renew
Sustain Energy Rev 2012; 16: 53315348.
Lu ZS, Wang RZ, Xia ZZ, et al. Study of a novel solar
adsorption cooling system and a solar absorption cooling system with new CPC collectors. Renew Energy
2013; 50: 299306.
Renedo et al.
19
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
Appendix 1
Notation
a
cp
CD
%CD
(d)
db
(i)
20
in
MW Tank
QCond w
T
Tank
TW
TTank
TTW
TTW
db
in
TTCond
TTTW
TTW
W