- Email: [email protected]
Energy saving assessment of a desiccant-enhanced evaporative cooling system in variable air volume applications
Accepted Manuscript Energy saving assessment of a desiccant-enhanced evaporative cooling system in variable air volume applications Sung-Joon Lee, Hui-Jeong Kim, Hye-Won Dong, Jae-Weon Jeong PII: DOI: Reference:
S1359-4311(17)30758-5 http://dx.doi.org/10.1016/j.applthermaleng.2017.02.007 ATE 9887
To appear in:
Applied Thermal Engineering
Received Date: Revised Date: Accepted Date:
10 June 2016 18 October 2016 4 February 2017
Please cite this article as: S-J. Lee, H-J. Kim, H-W. Dong, J-W. Jeong, Energy saving assessment of a desiccantenhanced evaporative cooling system in variable air volume applications, Applied Thermal Engineering (2017), doi: http://dx.doi.org/10.1016/j.applthermaleng.2017.02.007
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.
Energy saving assessment of a desiccant-enhanced evaporative cooling system in variable air volume applications
Sung-Joon Lee, Hui-Jeong Kim, Hye-Won Dong, Jae-Weon Jeong* Department of Architectural Engineering, College of Engineering, Hanyang University, 222 Wangsimni-Ro, Seongdong-Gu, Seoul 04763, Republic of Korea
Abstract Desiccant-enhanced evaporative (DEVap) cooling system have been recently introduced as an alternative to conventional air conditioning systems. This system consists of a liquid desiccant dehumidifier and a dew point evaporative cooler. In this research, workable operation strategies for the DEVap system were suggested and applied to an energy simulation for estimating the energy saving potential of the system over the conventional variable air volume (VAV) system. Four different modes of operation determined according to outdoor air conditions were suggested for annual operation of the DEVap system. Energy simulations for both DEVap and VAV systems were performed for a model office building. The results showed that 2-3% of annual primary energy could be saved by using the DEVap cooling system compared with the conventional VAV system.
Keywords: Desiccant-enhanced evaporative cooler, non-vapor-compression system, variable air volume, DEVap cooling system, operation strategy.
Nomenclature
*
Corresponding author. Tel.: +82-2-2220-2370; Fax: +82-2-2220-1945. E-mail address: [email protected] (J.-W. Jeong) 1
Face area of IEC (
)
Face area in the reference condition (m/s) Heat capacity rate (
)
Specific heat of moist air ( Specific heat of vapor (
) )
Minimum specific heat between water and the solution (
)
Maximum specific heat between water and the solution (
)
Experimental constant (-) Experimental constant (-) Enthalpy (
)
Enthalpy of vaporization at 0
(
)
Ratio of enthalpy changes to wet-bulb changes in IEC ( Mass flow rate (kg/s) Regeneration rate (
)
Experimental constant (-) NTU
Number of transfer unit (-) Part load ratio (-) Power ( Load (
) )
Liquid to gas ratio (-) Temperature ( ) Hot water temperature ( ) Desiccant solution temperature ( ) 2
)
Overall heat transfer coefficient times heat transfer area of IEC (
) Volume flow rate (
)
Wet-bulb ( ) Solution concentration (-)
Greek Symbols
Pressure rise in pump and fan ( Effectiveness (-) Efficiency (-) Fluid density (kg/m3) Humidity ratio (
)
Subscripts
Air design
Design condition
fan
Fan
fuel
Natural gas Hot water Hot water supply Inlet
max
Maximum value
min
Minimum value 3
)
outlet p
Primary channel
pump
Pump
reg
Reference IEC
reg
Regenerator
reg,T
Regenerator, thermal
req
Required
sec
Secondary Solution in liquid desiccant system Solution equivalent humidity ratio
tank
Solution tank Water
Abbreviations
BEC
Boiler efficiency curve
CA
Conditioned air
CAPFT
Chiller capacity as a function of temperature
CC
Cooling coil
CWR
Cooling water return
CWS
Cooling water supply
DEC
Direct evaporative cooler
DEVap
Desiccant-enhanced evaporative cooler
DP-IEC
Dew point indirect evaporative cooler
EA
Exhaust air 4
EIRFT
Energy input to cooling output ratio as a function of temperature
EIRFPLR
Energy input to cooling output ratio as a function of the part load ratio
HX
Heat exchanger
IEC
Indirect evaporative cooler
LD
Liquid desiccant
OA
Outdoor air
RA
Return air
SA
Supply air
SHE
Sensible heat exchanger
VAV
Variable air volume
1. Introduction Recently, non-vapor-compression heating, ventilating, and air conditioning (HVAC) technologies are attracting more interest in the building sector because of the desirable characteristic of not using refrigerants unlike conventional vapor-compression systems. Liquid desiccant (LD) systems are a promising technology capable of latent cooling without involving the vapor-compression cycle. Compared with solid desiccant systems, LD systems have several advantages such as lower regeneration temperature, higher dehumidification capacity, and less pressure drop in the process air side [1]. Furthermore, LD systems have become popular in hot and humid climates because of their ability to dehumidify outdoor air using relatively low-grade heat or renewable energy sources such as solar thermal energy and fuel cells [2]. There are two common types of LD systems: adiabatic and internally cooled types. Increases in the temperature of the desiccant solution during dehumidification process can be avoided using the latter type but not the former type. Consequently, internally cooled 5
LD systems provide higher latent cooling capacity than adiabatic type systems [3]. Indirect evaporative coolers (IEC) represent another promising non-vapor-compression technology that uses the latent heat of water vaporization to cool the process air. Recently, the dew point IEC (DP-IEC) has become prominent because of its ability to cool the process air close to the dew point temperature by redirecting a portion of the process air leaving the DPIEC (i.e., dehumidified air) into the wet channel [4–8]. The sensible cooling efficiency of the DP-IEC is highly affected by the dew point temperature of the process air [9]. Regarding the combination of LD and IEC systems, Kim et al. [10] introduced LD and indirect and direct evaporative cooling-assisted 100% outdoor air system (LD-IDECOAS), and showed their system can reduce the annual energy consumption by up to 51% compared with conventional variable air volume (VAV) systems. Kozubal et al. [11] suggested a desiccant-enhanced evaporative (DEVap) cooler that comprises a LD and a DP-IEC; the LD improves the performance of the DP-IEC by redirecting a portion of dehumidified supply air to the wet channel of the DP-IEC. Because of the characteristics of liquid desiccant unit and evaporative cooler, it is hard to sensitively control the condition of supply air, and be applied to highly humid climate. Boranian [12] suggested several operation modes of the DEVap cooler: dehumidification mode, indirect evaporative cooling mode, and standard mode. These operation modes are determined according to whether the operation of the liquid desiccant and the DP-IEC. However, these modes cannot be directly integrated into the air handling system yet because it should be considered that exhaust air in the DP-IEC is required to be supplemented by introducing adequate amount of outdoor air. As such, in this research, a variable air volume system integrated with the DEVap cooler and an additional cooling coil, which is called the DEVap cooling system is suggested as a HVAC system serving an office building. The detailed modes of operation for the proposed system 6
are also provided. The DEVap cooling system accommodates thermal loads of the building by varying the supply air flow rates like a conventional VAV system. The annual energy consumption of the DEVap system was estimated via detailed energy simulations, and then compared with that for a conventional VAV system.
2. DEVap cooling system
2.1 DEVap cooler
The DEVap cooler cools and dehumidifies the process air via two thermodynamic stages connected in series. The first stage is the dehumidification process in the LD section and the second stage is the sensible cooling process in the DP-IEC section (Fig. 1). For the dehumidification stage, the LD section comprised a stack of process air channels and working air channels. The process air channels were composed of hydrophobic microporous membranes covered with LD films (Fig. 1(b)). The working air channels were separated from the process air channels by plastic sheets for preventing unnecessary mass transfer between the channels. Water was sprayed into the working air channels for removing sensible heat transferred from the process air channels via evaporative cooling (Fig. 1(a)). A strong desiccant solution and water were supplied to the top of the process air channels and the working air channels, respectively, and drained under gravity. The mixed air (MA), mixture of outdoor air (OA), and return air (RA) entered the process air channels (i.e., State 1) at a certain ratio, and then, they were dehumidified by the LD, while releasing latent heat of vaporization. The heat released from this exothermic reaction was transferred through the plastic sheet to the working air cooled by the direct evaporative cooling process in the working air channels. The weak desiccant solution leaving the LD 7
section after dehumidifying the process air was regenerated in the regenerator; this is not shown in Figure 1.
(a) Exterior configuration
(b) Internal configuration Figure 1. Illustration of the two-stage DEVap cooler
The DP-IEC used in the second stage consisted of dry channels (i.e., primary channels) and wet channels (i.e., secondary channels). When the process air leaving the LD section passed through the dry channel, indirect evaporative cooling caused by evaporation in the wet channel cooled the process air in the dry channel. A portion of the process air leaving the dry channels was redirected to the wet channels for enhancing evaporative cooling (i.e., State 5 in Figure 1(b)). The exhaust air fraction (EAF) of the second stage, which is the ratio of the redirected air flow rate to the mixed air flow rate in the second state was assumed to be 0.3 based on open literature [6].
8
2.2 Configuration of DEVap cooling system
As shown in Figure 2, the DEVap cooling system consisted of a DEVap cooler, a cooling coil (CC), and a terminal electric reheating coil (RHC) with a VAV box. As a pair of a liquid desiccant system, a regenerator is installed in this system. During the dehumidification process, the regeneration energy occupies significant portion of the whole operating energy consumption in liquid desiccant-assisted systems. Because it is difficult to control the DEVap cooler to meet the SA temperature setpoint by modulating its dehumidification and sensible cooling efficiency, CC and RHC were installed downstream of the DEVap cooler for additional cooling or reheating to the SA stream. The proposed system could also be categorized into VAV systems because the SA flow rate was modulated by the cooling demand of the conditioned zone; however, it requires more OA intake flow and additional components compared with conventional VAV systems. The DEVap cooler was operated during a hot and humid season for cooling and dehumidifying the SA as much as possible without operating the chiller. During off-cooling seasons, the SA bypassed the deactivated DEVap cooler, such that the proposed system became a conventional VAV system. In conventional VAV systems, the cooling coil accommodates sensible and latent cooling loads simultaneously, whereas, in the DEVap cooling system, sensible and latent cooling are decoupled by the LD and DP-IEC, resulting in significant reduction of the CC loads. During the operation of the DEVap cooler, a regenerator maintained the target concentration of the desiccant solution.
9
Figure 2. Configuration of the DEVap cooling system
2.3 Operation modes of the DEVap cooling system
2.3.1 Determination of OA intake flow rate When the DEVap cooler was operated during the cooling season, the mass balance at the airside (Fig. 3) can be expressed as Equation 1. The OA intake flow ( with the recirculation air (
), and then the mixed air (
the DEVap cooler. A portion of
) was initially mixed
) was dehumidified and cooled by
leaving the DEVap cooler was redirected to the wet
channel of the DP-IEC, and then exhausted outside (
). Finally, the remaining SA (
)
was supplied to the conditioned space, and the same amount of air was extracted from the room ( recirculated ( (
). Similar to the conventional air-handling unit, a portion of
was
) for energy saving purposes, and the rest of the air was exhausted outside
).
(1) 10
Figure 3. Schematic of air mass flow rate in the DEVap cooling system
Figure 4 represents the determination of OA intake flow of the DEVap cooling system based on the OA condition, which is suggested in this research for satisfying ventilation requirement and minimizing energy consumption in conditioning supply air. The OA intake determination is divided in two cases in Region 1 as shown in Figure 4; that is, Case1 and Case2. Those two cases are different in terms of introducing minimum ventilation rate for the space.
11
Figure 4. Outdoor air intake strategies for the DEVap cooling system on a psychrometric chart
1) OA intake flow in Region 1 a) Case1 In the initial strategy for the OA intake determination of the DEVap cooler, the OA intake flow (
) should be equal to the exhausted air (
) to maintain the mass balance in air
side. Region 1 shown in Figure 4 represents OA conditions where OA enthalpy ( than room air enthalpy ( temperature (
) or OA dry-bulb temperature (
) was higher than room air
). In this region, minimum OA intake was required to decrease the OA
conditioning load. The OA intake flow ( ventilation rate (
) was higher
) should meet the recommended minimum
) and replenish the exhausted air ( 12
). Based on open literature [5], it
is recommended that
in the DEVap cooler should be 30% of the
at least (Eq. 2).
Then Equation 1 can be rewritten as Equation 3. Moreover, basically the recirculated air ( was same as the SA air flow rate (
), then the OA intake should be determined with
Equation 4. If the exhausted air flow rate ( rate (
)
) was lower than the minimum ventilation
), the OA intake should meet the minimum ventilation rate (Eq. 5)
(2) (3) (4) (5)
b) Case2 The OA intake flow (
) should meet the required minimum ventilation rate (
)
recommended by existing ventilation standards [13]. However, in the proposed system, when the DEVap cooler was operated, flow for exhausting
was required to be higher than the required ventilation
outside (Eq. 6).
(6)
Consequently, the OA intake flow rate (
of the DEVap cooling system is highly affected
by the DEVap cooler’s mode of operation determined by the OA condition. As shown in Figure 4,
could be determined by locating the OA condition on the psychrometric chart
divided by four regions. Region 1 shown in Figure 4 represents OA conditions where OA enthalpy (
13
) was higher
than room air enthalpy ( temperature (
) or OA dry-bulb temperature (
) was higher than room air
). In this region, the enthalpy or the dry-bulb temperature of the mixed air
was always lower than that of the OA, which decreased the OA conditioning load. The in Region 1 is the sum of
and
(Eq. 6).
According to open literature [5], it is recommended that be 30% of the
in the DEVap cooler should
at least (Eq. 2). Consequently, by replacing
in Equation 6 with
Equation 2 and performing mass balance manipulation at the airside, Equation 6 can be rewritten to Equation 7. The mixed air flow rate (Eq. 1) can also be expressed by Equation 8.
(7) (8)
2) OA intake flow in Region 2 Region 2 represents OA conditions where OA enthalpy was lower than room air enthalpy (
), and OA dry-bulb temperature was between room air temperature (
temperature setpoint (
) and the SA
). In this region, the DEVap cooling system should operate at 100%
OA mode to avoid increases in enthalpy and dry bulb temperature of process air by mixing with the RA, which increases cooling energy consumption; that is, the mass flow rate of recirculated air (
is equal to zero, and 100% OA is introduced (Eq. 9) to the system for
air conditioning. The introduced OA was cooled by the DP-IEC to meet the SA setpoint ( operation, and
was required to be 30% of
) without LD
(Eq. 10). Therefore, Equation 11 can
express the OA intake flow rate in Region 2. It is valid when the required SA flow rate (
was higher than the minimum required ventilation rate (
was lower than or equal to
,
was required to be set to 14
. When the required for maintaining the
minimum ventilation rate in the conditioned zone, and the OA intake flow rate can be determined by Equation 12.
(9) (10) (when
)
(11) (when
)
(12)
3) OA intake flow in Region 3 In Region 3, when the OA dry bulb temperature (
) was lower than the SA setpoint (
)
and higher than the lower limit of the OA temperature, the DEVap cooler was required to be deactivated and bypassed. It means that the proposed system becomes a conventional VAV system, and operates at the air-side economizer mode. The OA and RA mixing dampers were modulated to meet the SA temperature setpoint by mixing both air flows (Eq. 13). Equation 14 can be used to determine the OA intake flow rate in Region 3. The lower limit temperature (
) of the OA is the OA dry bulb temperature where the SA setpoint temperature can be
obtained by mixing the minimum OA ventilation flow (i.e., the required ventilation rate (
)) with the RA flow (Eq. 15).
(13) (14) (15)
4) OA intake flow in Region 4 15
When the OA dry-bulb temperature (
) was lower than
not less than the minimum ventilation rate (
, an OA intake flow (
of
was required, which is identical to the
conventional VAV system operation (Eq. 16).
(16)
(2) Operating modes of DEVap cooling system Four different modes of operation (i.e., Modes A, B, C, and D), which were determined by considering the mixed air condition on the psychrometric chart with respect to the SA setpoint condition (Fig. 5), for the DEVap cooling system are suggested in this paper.
Figure 5. Operation strategies for the DEVap cooling system on a psychrometric chart
16
a) Mode A When the dry-bulb temperature (
and the humidity ratio of mixed air (
are higher
than those at the SA setpoint (i.e., Region A in Fig. 5), the DEVap cooler should be activated to cool and dehumidify the mixed air. If the SA temperature after the DEVap cooler is lower than the SA temperature setpoint (e.g. 15°C), one may activate the reheat coil (RHC) at the VAV box terminal. However, if the humidity ratio of the mixed air is too high for the LD in the DEVap cooler to sufficiently dehumidify the introduced mixed air, the SA temperature at the DP-IEC outlet may be higher than its setpoint. In this case, one may need to activate the cooling coil at the VAV box terminal to meet the SA temperature setpoint.
b) Mode B When the mixed air temperature ( and the humidity ratio of mixed air (
is higher than the SA setpoint temperature ( is lower than that of the SA setpoint (
, (i.e.,
Region B in Fig. 5), the process air need not be dehumidified in the DEVap unit, although sensible cooling is required. Consequently, LD in the DEVap unit is deactivated, and the introduced mixed air is cooled by the DP-IEC. If the outlet temperature of the DEVap unit is lower than or higher than the SA temperature setpoint, reheating or additional cooling may be required at the VAV box terminal.
c) Mode C When the dry-bulb temperature and the humidity ratio of mixed air are lower than those of the SA setpoint (i.e., Region C in Fig. 5), the DEVap cooler should be switched off and bypassed. Only reheating of the mixed air is required at the terminal VAV Box to meet the SA temperature setpoint (
). 17
d) Mode D If the mixed air temperature ( the humidity ratio of the mixed air (
is lower than the SA temperature setpoint ( is higher than that at the SA setpoint (
while (i.e.,
Region D in Fig. 5), the mixed air introduced to the DEVap cooler should be dehumidified by LD. Sensible cooling of the process air is not required, so the DP-IEC is deactivated. If the SA temperature leaving the DEVap unit is lower than the SA setpoint temperature, the reheating coil in the terminal unit should be activated to get the SA setpoint temperature (
.
3. Energy simulation
3.1 Simulation condition
In order to evaluate the energy performance of the DEVap cooling system, energy simulations considering a typical office space were performed using TRNSYS 17 [14] and a commercial equation solver program (i.e., EES) [15]. Then, simulation results for the DEVap cooling system were compared with those for the conventional VAV system. Hourly thermal loads of the model office space were estimated using TRNSYS 17. Based on the acquired hourly thermal load profile for the model space and weather data, hourly energy consumption by both DEVap cooling and VAV systems were estimated by integrating existing models of each system component using the EES program. Simulation conditions are summarized in Table 1.
18
Table 1. Simulation conditions Location
Seoul, Republic of Korea
Weather data
TMY2 weather data
Building dimension
type (width
and Office building, Single zone (10 m× 10 m × 3 m) ×
depth × height) U-values [14]
0.630 W/m2·K (Roof), 0.468 W/m2·K (Wall), 0.952 W/m2·K (Floor)
Internal heat gain
Electronics
140 W/work station
Occupants
Loads
75
W/person
(sensible),
75
W/person
(latent) Density
5 people/100 m2
Schedules
Office occupants and HVAC schedules in ASHRAE 90.1
Room set point
Summer season
Temperature
26°C
(June to August)
Relative Humidity
50%
Intermediate season
Temperature
20–24°C
(March to May, September Relative Humidity
50%
to November)
Supply air set point
Winter season
Temperature
24°C
(December to February)
Relative Humidity
60%
Summer season
Temperature
15°C
(June to August)
Relative Humidity
80%
Intermediate season
Temperature
15°C
Temperature
20°C
(March to May, September to November) Winter season (December to February)
3.2 System models
19
3.2.1 Liquid desiccant model For estimating the conditions of air leaving the LD, an LD model was derived by performing linear regression of experimental data provided by Kozubal et al. [11]. The proposed LD model (Eq. 17) predicts the humidity ratio of the process air at the LD outlet (
) as a
function of four independent variables; that is, process air temperature at the LD inlet (
), humidity ratio of the air at the LD inlet (
desiccant solution (
), and liquid-to-gas ratio (
variable is presented in Table 2.
), concentration of strong
). The valid range of each independent
-value of the model was 0.89 and Figure 6 represents the
validation of LD model data against experimental data. In contrast, when any independent variable was out of their valid ranges, the derived LD model could not be used. In this case, the
is predicted by Equation 18 with the
assumption that the dehumidification efficiency of the LD at normal operation (
) is around
70% [11].
(17) (18)
Table 2 Ranges of variables for regression equation
Inlet Temperature [°C] Inlet humidity ratio [kg/kg] Concentration of liquid desiccant [%] Liquid to gas ratio
20
Minimum
Maximum
24.5
35
0.0128
0.0182
35.1
43.9
0.054091
0.223065
Figure 6 Validation of LD model against empirical data
3.2.2 Dew-point indirect evaporative cooler model The effectiveness of the DP-IEC (
) shown in Figure 7 can be expressed by Equation 19, a
simplified ε-NTU (effectiveness-number of transfer units) model [4,16] predicting the heat and mass transfer performances of DP-IEC by assuming Lewis number as unity. Once
is
known, the amount of heat exchanged in DP-IEC can be determined by Equation 20. Consequently, the SA temperature at the DP-IEC outlet (
) can be predicted by
Equation 21.
Figure 7 Schematic of the dew-point indirect evaporative cooler
(19)
21
where,
(20) (21)
When this simplified
-NTU model is used to predict
, the secondary air inlet
temperature should be known. However, in the DP-IEC, the inlet temperature of the secondary air was identical to the primary air temperature at the DP-IEC outlet ( which was not yet known. To solve this problem, temperature of the primary air at the primary channel inlet ( iterative process,
was calculated. To determine
)
was initially set to the wet-bulb ), and then through the (i.e., the UA value of DP-
IEC), the empirical equation suggested by Taylor et al. [16] is used in this paper (Eq. 22).
(22)
To obtain the values of experimental constants, the empirical data from open literature [11] 22
was used with the Levenberg–Marquardt method [17]. The values of experimental constants were
,
,
, and
based on
experimental data [11]. The ratio of the primary air to the secondary air mass flow rates (
) was set to 0.3
based on the characteristics of DP-IEC operation. The range of the mass flow rate of the primary air used in the energy simulation was 0.062–0.154 kg/s, based on experimental data [11]. Figure 8 shows that the predicted primary air temperature at the DP-IEC outlet ( agrees well with existing experimental data.
Figure 8. Validation of DP-IEC model data against empirical data
3.2.3 Regenerator model In the DEVap air conditioner, the double-effect regenerator was used for solution regeneration [11]; however, it is not commonly used yet. In this study, to evaluate required energy of the regenerator, the performance of commercialized regenerator was modelled with 23
)
assumptions. Regeneration heat was delivered to the weak desiccant solution through a counter-flow hot water heat exchanger and outdoor air was used for the regeneration air. The regenerator inlet solution temperature was set to 55˚C which is commonly used in regenerator operation. The regeneration rate of the regenerator was assumed to be equal to the dehumidification rate of LD, and the mass flow rate of solution inlet and outlet of the absorber and the regenerator were set to be equal for the mass balance of solution. The concentration of the desiccant solution in the storage tank was assumed to be maintained at 38%, and it was possible because of the osmosis membrane in the middle of solution tank. The regeneration efficiency was determined by Equation 23 and the regeneration efficiency was within 20-30% ranges, which is fit into typical regeneration efficiency [18,19]. The efficiency of the counter-flow heat exchanger (
is defined by Equation 24 [20].
Consequently, the amount of heat supplied for regeneration (
) through the heat
exchanger can be calculated by Equation 25.
(23) (24) where,
(25)
3.2.4 Hot water boiler model Based on the required heat energy in the regenerator, the natural gas consumption was estimated with using the simple hot water boiler model mentioned in Energy Plus Engineering Reference [23]. The required fuel ( 24
) (Eq. 26) can be calculated considering
the boiler efficiency cure (BEC) (Eq. 27), the part load ratio (PLR) (Eq. 28) and the nominal thermal efficiency of the boiler ( constants were used as follows,
) which was set to be 100% [23]. In Equation 27, the =0.626428326,
=0.645643582,
=-0.77720685 and
=0.313806701.
(26) (27) (28)
3.2.5 Fans and pumps As shown in Figure 2, the DEVap cooling system had five fans: supply, return, LD, DP-IEC, and regeneration air fans. The regeneration air fan was a constant air volume type, and other fans were variable air volume types. In the conventional VAV system, two variable air volume fans (supply and return fans) were used. The power of the variable air volume fan (
can be estimated by a general
variable-air-volume fan model (Eq. 29) suggested by ASHRAE Standard 90.1 [20], where the part load ratio of the variable flow fan ( (
) is the ratio of the current fan airflow
to the design airflow of the fan (
design air flow (
(Eq. 30). The fan power at the
) and for the constant air volume fan can be determined by
Equation 31. The fan pressure at the design air flow ( 3. The efficiency of each fan (
) in each fan is presented in Table
) was assumed to be 50% [11].
(29)
25
(30) (31)
Table 3. Fan pressures at design airflow Components DEVap SA fan
Design
Specifications
768 Pa
DEVap cooler [11] and balance of system
VAV
RA fan
200 Pa
Return dampers, ducts, etc.
LD fan
112 Pa
LD [11]
IEC fan
98 Pa
DP-IEC [11]
SA fan
750 Pa
Air handling unit (AHU), dampers, ducts and balance of system.[22]
RA fan
200 Pa
Return dampers, ducts, etc.
For the chilled water pump in both proposed and VAV systems, Equation 32 is used to calculate the pump power (
. In both systems, the efficiency of the pump (
assumed as 60%, and the pump head (
was
was set to 20 m [20].
(32)
3.2.6 Chiller model The DOE-2 air-cooled electric chiller model [22-23] was used in this research for estimating energy consumption by the chiller. The coefficient of performance (COP) in reference 26
condition is set to be 2.93.
This model consists of three performance curves: CAPFT,
EIRFT, and EIRFPLR. CAPFT (Eq. 33) represents the available capacity of an air-cooled chiller. EIRFT (Eq. 34) provides the full-load efficiency of a chiller as a function of evaporator and condenser temperatures. EIRFPLR (Eq. 35) corresponds to the part-load efficiency of an air-cooled chiller (Eq.36). The required power of the chiller is defined by Equation 37.
(33)
(34) (35) (36) (37)
The energy performance of the VAV system with the air-side economizer control serving the model office building was estimated through a detailed energy simulation, and then compared with that for the DEVap system. The DOE-2 air-cooled chiller model was also used in the VAV system simulation, and an electric heating coil was considered as the reheat coil.
4. Simulation results
4.1 Psychrometrics of the DEVap system Figure 9 shows a typical thermodynamic behavior of the DEVap system on a psychrometric 27
chart at the peak time on a peak summer day for the determination of OA intake case1 and case2. Since OA enthalpy ( enthalpy (
) and dry bulb temperature (
) and temperature (
) were higher than the return air
) (Region 1 in Figure 4), the OA intake flow rate (
) of
the DEVap system is determined by Equation 4 and 7 for case1 and case2 respectively. In the determination of OA intake case1, the mixed air temperature ( SA temperature setpoint (
) was higher than the
), so the LD and DP-IEC cooled and dehumidified the
mixed air initially. The LD dehumidifies air from 0.01316 kg/kg to 0.009823 kg/kg of humidity ratio and the DP-IEC cooled the dehumidified mixed air down to 17.85˚C. The cooling coil was operated to meet the SA temperature setpoint (e.g.,15˚C). In the determination of OA intake case2, the mixed air temperature ( SA temperature setpoint (
) was higher than the
), so the LD and DP-IEC cooled and dehumidified the
mixed air initially. The humidity ratio of the mixed air changed from 0.01429 kg/kg to 0.009958 kg/kg by the LD, and the DP-IEC cooled the dehumidified mixed air down to 17.85˚C. The cooling coil was operated to meet the SA temperature setpoint (e.g.,15˚C). Compared with the OA intake case1, more OA was introduced and that resulted higher temperature and humidity ratio of the MA2 than those of the MA1. As a result, the outlet air condition of the DEVap cooler with case1 showed lower temperature and humidity ratio than those of case2.
28
(a) Determination of OA intake – Case1
(b) Determination of OA intake – Case2 Figure 9. Thermodynamic behavior on a psychrometric chart (July 26th 15:00)
Figure 10 shows variations in hourly temperature and humidity ratio of the process air after 29
each component of the DEVap system on a peak summer day with the determination of OA intake case1 and case2 respectively, when the system was operated under Mode A. The SA setpoint temperature (e.g.,15˚C) was maintained by the cooling coil whenever required, while the temperature and the humidity ratio of the process air decreased through each system component.
(a) Determination of OA intake case1
(b) Determination of OA intake case2 Figure 10. Temperature and humidity ratio of the DEVap system on a peak summer day
4.2 Comparison of humidity ratios of SA during cooling season
In conventional VAV systems, the cooling coil performs latent cooling and sensible cooling 30
simultaneously by decreasing SA temperature until the SA reaches its target condition (e.g., 15°C dry bulb and 80% relative humidity). Terminal reheat may be required if the SA temperature is lower than its setpoint. On the other hand, in the DEVap cooling system, latent loads of introduced OA and conditioned space are accommodated by the LD independently from the sensible cooling device (i.e., DP-IEC). Figure 11 shows the humidity ratio variations of the SA in both VAV and DEVap systems predicted for a week of operation during the cooling season. The SA humidity ratio in the DEVap cooling system (
) varied with the dehumidification capacity of the LD
affected by the OA condition, while the humidity ratio of the SA in the VAV system (
)
was maintained at the target condition by the cooling coil. If the OA was highly humid, the humidity ratio of the SA in the DEVap cooling system could be higher than the target condition, although it may not be significant.
Figure 11. Profiles of humidity ratio of each state for both systems during one week
4.3 Amount of outdoor air intake Figure 12 shows hourly comparisons of the OA ratio in the SA flow between the conventional 31
VAV and the DEVap systems on a peak summer day. Because the OA intake ratio was set to be 0.3
when the supply air flow rate was higher than the minimum ventilation rate in the
determination of OA intake case1, the outdoor air intake ratio showed constant (i.e., 10 to 22).It might result that the OA intake is not enough to dilute the mixing air when the outdoor air intake ratio is lower than that of VAV system (i.e., 10 to 13 and 20 to 22). Because additional OA intake flow was required for DP-IEC operation in the DEVap cooling system with the determination of OA intake case2, the OA intake ratio of the DEVap cooling system was always higher than that of the conventional VAV system. The SA flow was set to the minimum ventilation rate during the system warming up hours (i.e., 7 to 10), so the OA ratio in both systems was 1.0.
Figure 12. Outdoor air mixing ratio on a peak summer day
4.4 Comparison of energy consumption Annual primary energy consumptions of the DEVap cooling system and the conventional VAV system were estimated via detailed energy simulations in this research. The primary 32
energy conversion factors were set to 2.67 for electricity, and 1.0 for natural gas based on a local guideline [24]. One should understand that simulation results be strongly affected by several assumptions (e.g. COP of chiller, redirection air ratio, etc.) defined initially. However, each operating condition assumed for the simulation was based on the values recommended by open literatures.
(a) Energy consumption comparison for operation during cooling season Figure 13 shows a comparison of energy consumptions by both DEVap and VAV systems estimated for the cooling season (i.e., June to August). In cooling season, the energy consumption of the DEVap cooling system could represent the trend of energy consumption in the application of hot and humid climate. Because of the additional fans and higher OA intake flow rate, 46% and 50% more fan energy was consumed in the DEVap cooling system in case1 and case2 respectively compared with the VAV system. The DEVap cooling system also requires natural gas for the regeneration of the desiccant solution, unlike in the VAV system. Because the case1 introduced less outdoor air compared with the case2, the case1 saved 5% of regeneration energy. However, the DEVap cooling system could provide both sensible and latent cooling with negligible cooling coil usage, which resulted in a significant (i.e., 98% in case1 and 92% in case2) reduction in energy consumption by the chiller compared with the VAV system. Consequently, in terms of the primary energy, the DEVap cooling system showed 18% and 10% primary energy saving in the case1 and case2 over the conventional VAV system during summer.
33
1500 2500
Electricity [kWh]
1000
1500
1000 500
500
0
Natural gas and Primary Energy [kWh]
2000
0 Electricity Natural gas Primary
Electricity Natural gas Primary
Electricity Natural gas Primary
DEVap(CASE1)
DEVap(CASE2)
VAV
P_fan_LD
P_fan_IEC
P_fan_RA
P_fan_SA
P_chiller
P_re-heating
P_pump_cw
Q_parallel
Q_regeneration
Primary_Fuel
P_pp_Reg,sol
P_pp_Reg,hw
Primary_Electricity P_fan_reg
Figure 13. Energy performance during summer
(b) Energy consumption of DEVap cooling system during intermediate season Figure 14 shows the operating energy consumptions of the DEVap system and the VAV system estimated for the intermediate season (i.e., March to May and September to October). In intermediate season, the energy consumption of the DEVap cooling system could represent the tendency of energy consumption in the application of hot and dry climate. The DEVap cooling system with the determination of OA intake case1 and case2 consumed 41% and 39% more energy regarding the fans and 97% and 96% less energy regarding the chiller compared 34
with the VAV system. The DEVap cooling system consumed natural gas for the regeneration of weak solution. Consequently, in terms of primary energy consumption, the DEVap cooling system with the determination of OA intake case1 and case2 showed 2.4% and 3.1% primary energy saving compared with VAV system during the intermediate season.
8000 7500
7000
2500
6500
Electricity [kWh]
6000 5500
2000
5000 4500
1500
4000 3500 3000
1000
2500 2000 1500
500
1000 500
0
0 Electricity
Natural gas
Primary Electricity
DEVap_case1
Natural gas
Primary Electricity
Natural gas
DEVap_case2
Primary
VAV
P_fan_LD
P_fan_IEC
P_fan_RA
P_fan_SA
P_chiller
P_re-heating
P_pump_cw
Q_parallel
Q_regeneration
Primary_Fuel
P_pp_Reg,sol
P_pp_Reg,hw
Primary_Electricity P_fan_reg
Natural gas and Primary Energy [kWh]
3000
Figure 14. Energy performance during intermediate season
(c) Energy consumption of DEVap cooling system during heating season Figure 15 represents the energy consumptions of the DEVap cooling and VAV systems estimated during the heating season (i.e., December to February). The DEVap cooling system operated as a VAV system during the heating season. As expected, the energy consumption of both systems were similar. 35
5000 12000
4500 4000
10000
8000
3000 2500
6000
2000 4000
1500
1000 2000 500 0
0 Electricity
Natural gas
Primary Electricity
DEVap_case1
Natural gas
Primary Electricity
DEVap_case2
Natural gas
Primary
Natural gas and Primary Energy [kWh]
Electricity [kWh]
3500
VAV
P_fan_LD
P_fan_IEC
P_fan_RA
P_fan_SA
P_chiller
P_re-heating
P_pump_cw
Q_parallel
Q_regeneration
Primary_Natural gas
Primary_Electricity
P_fan_reg
P_pp_Reg,sol
P_pp_Reg,hw
Figure 15. Energy performance during heating season
(d) Annual energy consumption of the DEVap cooling system Figure 16 shows a comparison of the annual energy consumption of the DEVap cooling system with that of the conventional VAV system. The DEVap cooling system consumed less electrical energy regarding the chiller (i.e., 12.2% in the case1 and 11.8% in the case2) but 39% and 35% more energy regarding the fans in the case1 and case2 respectively because of higher air-side pressure drop and airflow rate compared with the VAV system. The DEVap cooling system also requires natural gas for regeneration of the desiccant solution. Overall, 3%
36
and 2% primary energy saving was achieved by using the DEVap cooling system with the determination of OA intake case1 and case2 respectively over the conventional VAV system.
9000
22000 20000
7500
18000
Electricity [kWh]
14000 12000
4500
10000 8000
3000
6000 4000
1500
2000 0
0 Electricity
Natural gas
Primary Electricity
DEVap_case1
Natural gas
Primary Electricity
DEVap_case2
Natural gas
Primary
VAV
P_fan_LD
P_fan_IEC
P_fan_RA
P_fan_SA
P_chiller
P_re-heating
P_pump_cw
Q_parallel
Q_regeneration
Primary_Fuel
P_pp_Reg,sol
P_pp_Reg,hw
Primary_Electricity P_fan_reg
Natural gas and Primary Energy [kWh]
16000 6000
Figure 16. Annual energy consumption comparison
5. Conclusion
The DEVap cooling system suggested in this study is the integration of the DEVap cooler with the conventional VAV system for reducing the dependence of cooling coil on the vapor compression refrigeration cycle. In addition, annual operation strategies of the DEVap cooling system were suggested. From the energy simulation for the DEVap cooling system, significant energy savings regarding the chiller was observed, while this energy saving was 37
offset instantaneously by the heating energy consumed in the regenerator. A primary energy saving of only 3% and 2% in the case1 and case2 was observed in the energy simulation of the DEVap cooling system over the conventional VAV system. However, a significant reduction (over 90%) of energy consumed in the operation of chillers in comparison with the conventional VAV system can be expected. This implies that the DEVap cooling system shows a possibility of realizing a non-vapor compression HVAC system without energy penalty. Critical energy consumption of desiccant-assisted HVAC systems occurs in the regeneration process. It means that the energy saving potential of the proposed system is highly affected by the amount of heat for the regeneration that is proportionally increased with the level of dehumidification. This research performed for the highly humid climate zone during the summer; that is, relative humidity of outdoor air is over 70% for 1665 hours out of 2208 hours. If the proposed system is used in the less humid climate or dry climate zones, one can expect much enhanced energy saving potential from the proposed system. Additional empirical research may be required for generalizing the energy performance of the proposed system.
Acknowledgements This work was supported by a National Research Foundation (NRF) of Korea (No. 2015R1A2A1A05001726), the Korea Agency for Infrastructure Technology Advancement (KAIA) grant (16CTAP-C116268-01), and the Korea Institute of Energy Technology Evaluation and Planning (KETEP) (No. 20164010200860).
Reference [1]
Liu XH, Qu KY, Jiang Y. Empirical correlations to predict the performance of the dehumidifier using liquid desiccant in heat and mass transfer. Renew Energy 2006;31:1627–39.
38
doi:10.1016/j.renene.2005.08.029. [2]
Gommed K, Grossman G. Experimental investigation of a liquid desiccant system for solar cooling and dehumidification. Sol Energy 2007;81:131–8. doi:10.1016/j.solener.2006.05.006.
[3]
Yin Y, Zhang X, Peng D, Li X. Model validation and case study on internally cooled/heated dehumidifier/regenerator of liquid desiccant systems. Int J Therm Sci 2009;48:1664–71. doi:10.1016/j.ijthermalsci.2008.12.017.
[4]
Hasan A. Going below the wet-bulb temperature by indirect evaporative cooling: Analysis using a modified ε-NTU method. Appl Energy 2012;89:237–45. doi:10.1016/j.apenergy.2011.07.005.
[5]
Lee J, Lee DY. Experimental study of a counter flow regenerative evaporative cooler with finned channels. Int J Heat Mass Transf 2013;65:173–9. doi:10.1016/j.ijheatmasstransfer.2013.05.069.
[6]
Bruno F. On-site experimental testing of a novel dew point evaporative cooler. Energy Build 2011;43:3475–83. doi:10.1016/j.enbuild.2011.09.013.
[7]
Jradi M, Riffat S. Experimental and numerical investigation of a dew-point cooling system for thermal comfort in buildings. Appl Energy 2014;132:524–35. doi:10.1016/j.apenergy.2014.07.040.
[8]
Zhan C, Duan Z, Zhao X, Smith S, Jin H, Riffat S. Comparative study of the performance of the M-cycle counter-flow and cross-flow heat exchangers for indirect evaporative cooling Paving the path toward sustainable cooling of buildings. Energy 2011;36:6790–805. doi:10.1016/j.energy.2011.10.019.
[9]
Jradi M, Riffat S. Experimental investigation of a biomass-fuelled micro-scale tri-generation system with an organic Rankine cycle and liquid desiccant cooling unit. Energy 2014;71:80– 93. doi:10.1016/j.energy.2014.04.077.
[10]
Kim MH, Park JS, Jeong JW. Energy saving potential of liquid desiccant in evaporative-
39
cooling-assisted 100% outdoor air system. Energy 2013;59:726–36. doi:10.1016/j.energy.2013.07.018. [11]
Kozubal E, Woods J, Judkoff R. Development and Analysis of Desiccant Enhanced Evaporative Air Conditioner Prototype. National Renewable Energy Laboratory 2012.
[12]
Boranian AP. An Investigation of Optimal Control of Desiccant-Enhanced Evaporative Air Conditioning. Univ Color 2012.
[13]
ASHRAE. ANSI/ASHRAE Standard 62.1-2013, Ventilation for Acceptable Indoor Air Quality Atlanta: American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc 2013.
[14]
TRNSYS 17. TRNSYS 17 Mathematical Reference. Massachusetts Institute of Technology 2009.
[15]
Engineering Equation Solver (EES) for Microsoft Windows Operating Systems, F-Chart Software 2007.
[16]
Taylor P, Liu Z, Allen W, Modera M. Simplified thermal modeling of indirect evaporative heat exchangers 2013:37–41. doi:10.1080/10789669.2013.763653.
[17]
MathWorks. MATLAB 2014.
[18]
Yonggao Yin and Xiaosong Zhang. Comparative study on internally heated and adiabatic regenerators in liquid desiccant air conditioning system. Building and Environment 2010:1799-1807. doi: 10.1016/j.buildenv.2010.02.008
[19]
Lisa Crofoot. Experimental evaluation and modeling of a solar liquid desiccant air conditioner. Queen’s University. 2012.
[20]
ASHRAE. ANSI/ASHRAE/IES Standard 90.1-2013 Energy Standard for Buildings Except Low-Rise Residential Buildings. vol. 2013. Atlanta: American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc; 2013.
[21]
Company PG and E company. Design Guidelines: Advanced Variable Air Volume (VAV)
40
Systems. Energy Design Resources; 2009. [22]
EnergyPlus. EnergyPlus Input Output Reference. U.S. Department of Energy, Building Technologies Program; 2013.
[23]
Energy UDO. EnergyPlus Engineering Reference: The Reference to EnergyPlus Calculations. US Dep Energy 2010:1051. doi:citeulike-article-id:10579266.
[24]
Ministry of Trade Industry and Energy. Energy Caloric Value Conversion Standard, Energy Act Article 5. 2011.
41
Research highlights
-
Energy saving potential of the DEVap cooling system was evaluated.
-
Workable modes of operation for the DEVap cooling system were suggested.
-
The proposed system provides 2-3% annual energy saving over the conventional VAV system.
-
The proposed system showed 10-18% primary energy saving potentials over the conventional VAV system in summer.
-
The DEVap cooling system can be an energy conservative option for hot and humid climate zones.
42
S1359-4311(17)30758-5 http://dx.doi.org/10.1016/j.applthermaleng.2017.02.007 ATE 9887
To appear in:
Applied Thermal Engineering
Received Date: Revised Date: Accepted Date:
10 June 2016 18 October 2016 4 February 2017
Please cite this article as: S-J. Lee, H-J. Kim, H-W. Dong, J-W. Jeong, Energy saving assessment of a desiccantenhanced evaporative cooling system in variable air volume applications, Applied Thermal Engineering (2017), doi: http://dx.doi.org/10.1016/j.applthermaleng.2017.02.007
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.
Energy saving assessment of a desiccant-enhanced evaporative cooling system in variable air volume applications
Sung-Joon Lee, Hui-Jeong Kim, Hye-Won Dong, Jae-Weon Jeong* Department of Architectural Engineering, College of Engineering, Hanyang University, 222 Wangsimni-Ro, Seongdong-Gu, Seoul 04763, Republic of Korea
Abstract Desiccant-enhanced evaporative (DEVap) cooling system have been recently introduced as an alternative to conventional air conditioning systems. This system consists of a liquid desiccant dehumidifier and a dew point evaporative cooler. In this research, workable operation strategies for the DEVap system were suggested and applied to an energy simulation for estimating the energy saving potential of the system over the conventional variable air volume (VAV) system. Four different modes of operation determined according to outdoor air conditions were suggested for annual operation of the DEVap system. Energy simulations for both DEVap and VAV systems were performed for a model office building. The results showed that 2-3% of annual primary energy could be saved by using the DEVap cooling system compared with the conventional VAV system.
Keywords: Desiccant-enhanced evaporative cooler, non-vapor-compression system, variable air volume, DEVap cooling system, operation strategy.
Nomenclature
*
Corresponding author. Tel.: +82-2-2220-2370; Fax: +82-2-2220-1945. E-mail address: [email protected] (J.-W. Jeong) 1
Face area of IEC (
)
Face area in the reference condition (m/s) Heat capacity rate (
)
Specific heat of moist air ( Specific heat of vapor (
) )
Minimum specific heat between water and the solution (
)
Maximum specific heat between water and the solution (
)
Experimental constant (-) Experimental constant (-) Enthalpy (
)
Enthalpy of vaporization at 0
(
)
Ratio of enthalpy changes to wet-bulb changes in IEC ( Mass flow rate (kg/s) Regeneration rate (
)
Experimental constant (-) NTU
Number of transfer unit (-) Part load ratio (-) Power ( Load (
) )
Liquid to gas ratio (-) Temperature ( ) Hot water temperature ( ) Desiccant solution temperature ( ) 2
)
Overall heat transfer coefficient times heat transfer area of IEC (
) Volume flow rate (
)
Wet-bulb ( ) Solution concentration (-)
Greek Symbols
Pressure rise in pump and fan ( Effectiveness (-) Efficiency (-) Fluid density (kg/m3) Humidity ratio (
)
Subscripts
Air design
Design condition
fan
Fan
fuel
Natural gas Hot water Hot water supply Inlet
max
Maximum value
min
Minimum value 3
)
outlet p
Primary channel
pump
Pump
reg
Reference IEC
reg
Regenerator
reg,T
Regenerator, thermal
req
Required
sec
Secondary Solution in liquid desiccant system Solution equivalent humidity ratio
tank
Solution tank Water
Abbreviations
BEC
Boiler efficiency curve
CA
Conditioned air
CAPFT
Chiller capacity as a function of temperature
CC
Cooling coil
CWR
Cooling water return
CWS
Cooling water supply
DEC
Direct evaporative cooler
DEVap
Desiccant-enhanced evaporative cooler
DP-IEC
Dew point indirect evaporative cooler
EA
Exhaust air 4
EIRFT
Energy input to cooling output ratio as a function of temperature
EIRFPLR
Energy input to cooling output ratio as a function of the part load ratio
HX
Heat exchanger
IEC
Indirect evaporative cooler
LD
Liquid desiccant
OA
Outdoor air
RA
Return air
SA
Supply air
SHE
Sensible heat exchanger
VAV
Variable air volume
1. Introduction Recently, non-vapor-compression heating, ventilating, and air conditioning (HVAC) technologies are attracting more interest in the building sector because of the desirable characteristic of not using refrigerants unlike conventional vapor-compression systems. Liquid desiccant (LD) systems are a promising technology capable of latent cooling without involving the vapor-compression cycle. Compared with solid desiccant systems, LD systems have several advantages such as lower regeneration temperature, higher dehumidification capacity, and less pressure drop in the process air side [1]. Furthermore, LD systems have become popular in hot and humid climates because of their ability to dehumidify outdoor air using relatively low-grade heat or renewable energy sources such as solar thermal energy and fuel cells [2]. There are two common types of LD systems: adiabatic and internally cooled types. Increases in the temperature of the desiccant solution during dehumidification process can be avoided using the latter type but not the former type. Consequently, internally cooled 5
LD systems provide higher latent cooling capacity than adiabatic type systems [3]. Indirect evaporative coolers (IEC) represent another promising non-vapor-compression technology that uses the latent heat of water vaporization to cool the process air. Recently, the dew point IEC (DP-IEC) has become prominent because of its ability to cool the process air close to the dew point temperature by redirecting a portion of the process air leaving the DPIEC (i.e., dehumidified air) into the wet channel [4–8]. The sensible cooling efficiency of the DP-IEC is highly affected by the dew point temperature of the process air [9]. Regarding the combination of LD and IEC systems, Kim et al. [10] introduced LD and indirect and direct evaporative cooling-assisted 100% outdoor air system (LD-IDECOAS), and showed their system can reduce the annual energy consumption by up to 51% compared with conventional variable air volume (VAV) systems. Kozubal et al. [11] suggested a desiccant-enhanced evaporative (DEVap) cooler that comprises a LD and a DP-IEC; the LD improves the performance of the DP-IEC by redirecting a portion of dehumidified supply air to the wet channel of the DP-IEC. Because of the characteristics of liquid desiccant unit and evaporative cooler, it is hard to sensitively control the condition of supply air, and be applied to highly humid climate. Boranian [12] suggested several operation modes of the DEVap cooler: dehumidification mode, indirect evaporative cooling mode, and standard mode. These operation modes are determined according to whether the operation of the liquid desiccant and the DP-IEC. However, these modes cannot be directly integrated into the air handling system yet because it should be considered that exhaust air in the DP-IEC is required to be supplemented by introducing adequate amount of outdoor air. As such, in this research, a variable air volume system integrated with the DEVap cooler and an additional cooling coil, which is called the DEVap cooling system is suggested as a HVAC system serving an office building. The detailed modes of operation for the proposed system 6
are also provided. The DEVap cooling system accommodates thermal loads of the building by varying the supply air flow rates like a conventional VAV system. The annual energy consumption of the DEVap system was estimated via detailed energy simulations, and then compared with that for a conventional VAV system.
2. DEVap cooling system
2.1 DEVap cooler
The DEVap cooler cools and dehumidifies the process air via two thermodynamic stages connected in series. The first stage is the dehumidification process in the LD section and the second stage is the sensible cooling process in the DP-IEC section (Fig. 1). For the dehumidification stage, the LD section comprised a stack of process air channels and working air channels. The process air channels were composed of hydrophobic microporous membranes covered with LD films (Fig. 1(b)). The working air channels were separated from the process air channels by plastic sheets for preventing unnecessary mass transfer between the channels. Water was sprayed into the working air channels for removing sensible heat transferred from the process air channels via evaporative cooling (Fig. 1(a)). A strong desiccant solution and water were supplied to the top of the process air channels and the working air channels, respectively, and drained under gravity. The mixed air (MA), mixture of outdoor air (OA), and return air (RA) entered the process air channels (i.e., State 1) at a certain ratio, and then, they were dehumidified by the LD, while releasing latent heat of vaporization. The heat released from this exothermic reaction was transferred through the plastic sheet to the working air cooled by the direct evaporative cooling process in the working air channels. The weak desiccant solution leaving the LD 7
section after dehumidifying the process air was regenerated in the regenerator; this is not shown in Figure 1.
(a) Exterior configuration
(b) Internal configuration Figure 1. Illustration of the two-stage DEVap cooler
The DP-IEC used in the second stage consisted of dry channels (i.e., primary channels) and wet channels (i.e., secondary channels). When the process air leaving the LD section passed through the dry channel, indirect evaporative cooling caused by evaporation in the wet channel cooled the process air in the dry channel. A portion of the process air leaving the dry channels was redirected to the wet channels for enhancing evaporative cooling (i.e., State 5 in Figure 1(b)). The exhaust air fraction (EAF) of the second stage, which is the ratio of the redirected air flow rate to the mixed air flow rate in the second state was assumed to be 0.3 based on open literature [6].
8
2.2 Configuration of DEVap cooling system
As shown in Figure 2, the DEVap cooling system consisted of a DEVap cooler, a cooling coil (CC), and a terminal electric reheating coil (RHC) with a VAV box. As a pair of a liquid desiccant system, a regenerator is installed in this system. During the dehumidification process, the regeneration energy occupies significant portion of the whole operating energy consumption in liquid desiccant-assisted systems. Because it is difficult to control the DEVap cooler to meet the SA temperature setpoint by modulating its dehumidification and sensible cooling efficiency, CC and RHC were installed downstream of the DEVap cooler for additional cooling or reheating to the SA stream. The proposed system could also be categorized into VAV systems because the SA flow rate was modulated by the cooling demand of the conditioned zone; however, it requires more OA intake flow and additional components compared with conventional VAV systems. The DEVap cooler was operated during a hot and humid season for cooling and dehumidifying the SA as much as possible without operating the chiller. During off-cooling seasons, the SA bypassed the deactivated DEVap cooler, such that the proposed system became a conventional VAV system. In conventional VAV systems, the cooling coil accommodates sensible and latent cooling loads simultaneously, whereas, in the DEVap cooling system, sensible and latent cooling are decoupled by the LD and DP-IEC, resulting in significant reduction of the CC loads. During the operation of the DEVap cooler, a regenerator maintained the target concentration of the desiccant solution.
9
Figure 2. Configuration of the DEVap cooling system
2.3 Operation modes of the DEVap cooling system
2.3.1 Determination of OA intake flow rate When the DEVap cooler was operated during the cooling season, the mass balance at the airside (Fig. 3) can be expressed as Equation 1. The OA intake flow ( with the recirculation air (
), and then the mixed air (
the DEVap cooler. A portion of
) was initially mixed
) was dehumidified and cooled by
leaving the DEVap cooler was redirected to the wet
channel of the DP-IEC, and then exhausted outside (
). Finally, the remaining SA (
)
was supplied to the conditioned space, and the same amount of air was extracted from the room ( recirculated ( (
). Similar to the conventional air-handling unit, a portion of
was
) for energy saving purposes, and the rest of the air was exhausted outside
).
(1) 10
Figure 3. Schematic of air mass flow rate in the DEVap cooling system
Figure 4 represents the determination of OA intake flow of the DEVap cooling system based on the OA condition, which is suggested in this research for satisfying ventilation requirement and minimizing energy consumption in conditioning supply air. The OA intake determination is divided in two cases in Region 1 as shown in Figure 4; that is, Case1 and Case2. Those two cases are different in terms of introducing minimum ventilation rate for the space.
11
Figure 4. Outdoor air intake strategies for the DEVap cooling system on a psychrometric chart
1) OA intake flow in Region 1 a) Case1 In the initial strategy for the OA intake determination of the DEVap cooler, the OA intake flow (
) should be equal to the exhausted air (
) to maintain the mass balance in air
side. Region 1 shown in Figure 4 represents OA conditions where OA enthalpy ( than room air enthalpy ( temperature (
) or OA dry-bulb temperature (
) was higher than room air
). In this region, minimum OA intake was required to decrease the OA
conditioning load. The OA intake flow ( ventilation rate (
) was higher
) should meet the recommended minimum
) and replenish the exhausted air ( 12
). Based on open literature [5], it
is recommended that
in the DEVap cooler should be 30% of the
at least (Eq. 2).
Then Equation 1 can be rewritten as Equation 3. Moreover, basically the recirculated air ( was same as the SA air flow rate (
), then the OA intake should be determined with
Equation 4. If the exhausted air flow rate ( rate (
)
) was lower than the minimum ventilation
), the OA intake should meet the minimum ventilation rate (Eq. 5)
(2) (3) (4) (5)
b) Case2 The OA intake flow (
) should meet the required minimum ventilation rate (
)
recommended by existing ventilation standards [13]. However, in the proposed system, when the DEVap cooler was operated, flow for exhausting
was required to be higher than the required ventilation
outside (Eq. 6).
(6)
Consequently, the OA intake flow rate (
of the DEVap cooling system is highly affected
by the DEVap cooler’s mode of operation determined by the OA condition. As shown in Figure 4,
could be determined by locating the OA condition on the psychrometric chart
divided by four regions. Region 1 shown in Figure 4 represents OA conditions where OA enthalpy (
13
) was higher
than room air enthalpy ( temperature (
) or OA dry-bulb temperature (
) was higher than room air
). In this region, the enthalpy or the dry-bulb temperature of the mixed air
was always lower than that of the OA, which decreased the OA conditioning load. The in Region 1 is the sum of
and
(Eq. 6).
According to open literature [5], it is recommended that be 30% of the
in the DEVap cooler should
at least (Eq. 2). Consequently, by replacing
in Equation 6 with
Equation 2 and performing mass balance manipulation at the airside, Equation 6 can be rewritten to Equation 7. The mixed air flow rate (Eq. 1) can also be expressed by Equation 8.
(7) (8)
2) OA intake flow in Region 2 Region 2 represents OA conditions where OA enthalpy was lower than room air enthalpy (
), and OA dry-bulb temperature was between room air temperature (
temperature setpoint (
) and the SA
). In this region, the DEVap cooling system should operate at 100%
OA mode to avoid increases in enthalpy and dry bulb temperature of process air by mixing with the RA, which increases cooling energy consumption; that is, the mass flow rate of recirculated air (
is equal to zero, and 100% OA is introduced (Eq. 9) to the system for
air conditioning. The introduced OA was cooled by the DP-IEC to meet the SA setpoint ( operation, and
was required to be 30% of
) without LD
(Eq. 10). Therefore, Equation 11 can
express the OA intake flow rate in Region 2. It is valid when the required SA flow rate (
was higher than the minimum required ventilation rate (
was lower than or equal to
,
was required to be set to 14
. When the required for maintaining the
minimum ventilation rate in the conditioned zone, and the OA intake flow rate can be determined by Equation 12.
(9) (10) (when
)
(11) (when
)
(12)
3) OA intake flow in Region 3 In Region 3, when the OA dry bulb temperature (
) was lower than the SA setpoint (
)
and higher than the lower limit of the OA temperature, the DEVap cooler was required to be deactivated and bypassed. It means that the proposed system becomes a conventional VAV system, and operates at the air-side economizer mode. The OA and RA mixing dampers were modulated to meet the SA temperature setpoint by mixing both air flows (Eq. 13). Equation 14 can be used to determine the OA intake flow rate in Region 3. The lower limit temperature (
) of the OA is the OA dry bulb temperature where the SA setpoint temperature can be
obtained by mixing the minimum OA ventilation flow (i.e., the required ventilation rate (
)) with the RA flow (Eq. 15).
(13) (14) (15)
4) OA intake flow in Region 4 15
When the OA dry-bulb temperature (
) was lower than
not less than the minimum ventilation rate (
, an OA intake flow (
of
was required, which is identical to the
conventional VAV system operation (Eq. 16).
(16)
(2) Operating modes of DEVap cooling system Four different modes of operation (i.e., Modes A, B, C, and D), which were determined by considering the mixed air condition on the psychrometric chart with respect to the SA setpoint condition (Fig. 5), for the DEVap cooling system are suggested in this paper.
Figure 5. Operation strategies for the DEVap cooling system on a psychrometric chart
16
a) Mode A When the dry-bulb temperature (
and the humidity ratio of mixed air (
are higher
than those at the SA setpoint (i.e., Region A in Fig. 5), the DEVap cooler should be activated to cool and dehumidify the mixed air. If the SA temperature after the DEVap cooler is lower than the SA temperature setpoint (e.g. 15°C), one may activate the reheat coil (RHC) at the VAV box terminal. However, if the humidity ratio of the mixed air is too high for the LD in the DEVap cooler to sufficiently dehumidify the introduced mixed air, the SA temperature at the DP-IEC outlet may be higher than its setpoint. In this case, one may need to activate the cooling coil at the VAV box terminal to meet the SA temperature setpoint.
b) Mode B When the mixed air temperature ( and the humidity ratio of mixed air (
is higher than the SA setpoint temperature ( is lower than that of the SA setpoint (
, (i.e.,
Region B in Fig. 5), the process air need not be dehumidified in the DEVap unit, although sensible cooling is required. Consequently, LD in the DEVap unit is deactivated, and the introduced mixed air is cooled by the DP-IEC. If the outlet temperature of the DEVap unit is lower than or higher than the SA temperature setpoint, reheating or additional cooling may be required at the VAV box terminal.
c) Mode C When the dry-bulb temperature and the humidity ratio of mixed air are lower than those of the SA setpoint (i.e., Region C in Fig. 5), the DEVap cooler should be switched off and bypassed. Only reheating of the mixed air is required at the terminal VAV Box to meet the SA temperature setpoint (
). 17
d) Mode D If the mixed air temperature ( the humidity ratio of the mixed air (
is lower than the SA temperature setpoint ( is higher than that at the SA setpoint (
while (i.e.,
Region D in Fig. 5), the mixed air introduced to the DEVap cooler should be dehumidified by LD. Sensible cooling of the process air is not required, so the DP-IEC is deactivated. If the SA temperature leaving the DEVap unit is lower than the SA setpoint temperature, the reheating coil in the terminal unit should be activated to get the SA setpoint temperature (
.
3. Energy simulation
3.1 Simulation condition
In order to evaluate the energy performance of the DEVap cooling system, energy simulations considering a typical office space were performed using TRNSYS 17 [14] and a commercial equation solver program (i.e., EES) [15]. Then, simulation results for the DEVap cooling system were compared with those for the conventional VAV system. Hourly thermal loads of the model office space were estimated using TRNSYS 17. Based on the acquired hourly thermal load profile for the model space and weather data, hourly energy consumption by both DEVap cooling and VAV systems were estimated by integrating existing models of each system component using the EES program. Simulation conditions are summarized in Table 1.
18
Table 1. Simulation conditions Location
Seoul, Republic of Korea
Weather data
TMY2 weather data
Building dimension
type (width
and Office building, Single zone (10 m× 10 m × 3 m) ×
depth × height) U-values [14]
0.630 W/m2·K (Roof), 0.468 W/m2·K (Wall), 0.952 W/m2·K (Floor)
Internal heat gain
Electronics
140 W/work station
Occupants
Loads
75
W/person
(sensible),
75
W/person
(latent) Density
5 people/100 m2
Schedules
Office occupants and HVAC schedules in ASHRAE 90.1
Room set point
Summer season
Temperature
26°C
(June to August)
Relative Humidity
50%
Intermediate season
Temperature
20–24°C
(March to May, September Relative Humidity
50%
to November)
Supply air set point
Winter season
Temperature
24°C
(December to February)
Relative Humidity
60%
Summer season
Temperature
15°C
(June to August)
Relative Humidity
80%
Intermediate season
Temperature
15°C
Temperature
20°C
(March to May, September to November) Winter season (December to February)
3.2 System models
19
3.2.1 Liquid desiccant model For estimating the conditions of air leaving the LD, an LD model was derived by performing linear regression of experimental data provided by Kozubal et al. [11]. The proposed LD model (Eq. 17) predicts the humidity ratio of the process air at the LD outlet (
) as a
function of four independent variables; that is, process air temperature at the LD inlet (
), humidity ratio of the air at the LD inlet (
desiccant solution (
), and liquid-to-gas ratio (
variable is presented in Table 2.
), concentration of strong
). The valid range of each independent
-value of the model was 0.89 and Figure 6 represents the
validation of LD model data against experimental data. In contrast, when any independent variable was out of their valid ranges, the derived LD model could not be used. In this case, the
is predicted by Equation 18 with the
assumption that the dehumidification efficiency of the LD at normal operation (
) is around
70% [11].
(17) (18)
Table 2 Ranges of variables for regression equation
Inlet Temperature [°C] Inlet humidity ratio [kg/kg] Concentration of liquid desiccant [%] Liquid to gas ratio
20
Minimum
Maximum
24.5
35
0.0128
0.0182
35.1
43.9
0.054091
0.223065
Figure 6 Validation of LD model against empirical data
3.2.2 Dew-point indirect evaporative cooler model The effectiveness of the DP-IEC (
) shown in Figure 7 can be expressed by Equation 19, a
simplified ε-NTU (effectiveness-number of transfer units) model [4,16] predicting the heat and mass transfer performances of DP-IEC by assuming Lewis number as unity. Once
is
known, the amount of heat exchanged in DP-IEC can be determined by Equation 20. Consequently, the SA temperature at the DP-IEC outlet (
) can be predicted by
Equation 21.
Figure 7 Schematic of the dew-point indirect evaporative cooler
(19)
21
where,
(20) (21)
When this simplified
-NTU model is used to predict
, the secondary air inlet
temperature should be known. However, in the DP-IEC, the inlet temperature of the secondary air was identical to the primary air temperature at the DP-IEC outlet ( which was not yet known. To solve this problem, temperature of the primary air at the primary channel inlet ( iterative process,
was calculated. To determine
)
was initially set to the wet-bulb ), and then through the (i.e., the UA value of DP-
IEC), the empirical equation suggested by Taylor et al. [16] is used in this paper (Eq. 22).
(22)
To obtain the values of experimental constants, the empirical data from open literature [11] 22
was used with the Levenberg–Marquardt method [17]. The values of experimental constants were
,
,
, and
based on
experimental data [11]. The ratio of the primary air to the secondary air mass flow rates (
) was set to 0.3
based on the characteristics of DP-IEC operation. The range of the mass flow rate of the primary air used in the energy simulation was 0.062–0.154 kg/s, based on experimental data [11]. Figure 8 shows that the predicted primary air temperature at the DP-IEC outlet ( agrees well with existing experimental data.
Figure 8. Validation of DP-IEC model data against empirical data
3.2.3 Regenerator model In the DEVap air conditioner, the double-effect regenerator was used for solution regeneration [11]; however, it is not commonly used yet. In this study, to evaluate required energy of the regenerator, the performance of commercialized regenerator was modelled with 23
)
assumptions. Regeneration heat was delivered to the weak desiccant solution through a counter-flow hot water heat exchanger and outdoor air was used for the regeneration air. The regenerator inlet solution temperature was set to 55˚C which is commonly used in regenerator operation. The regeneration rate of the regenerator was assumed to be equal to the dehumidification rate of LD, and the mass flow rate of solution inlet and outlet of the absorber and the regenerator were set to be equal for the mass balance of solution. The concentration of the desiccant solution in the storage tank was assumed to be maintained at 38%, and it was possible because of the osmosis membrane in the middle of solution tank. The regeneration efficiency was determined by Equation 23 and the regeneration efficiency was within 20-30% ranges, which is fit into typical regeneration efficiency [18,19]. The efficiency of the counter-flow heat exchanger (
is defined by Equation 24 [20].
Consequently, the amount of heat supplied for regeneration (
) through the heat
exchanger can be calculated by Equation 25.
(23) (24) where,
(25)
3.2.4 Hot water boiler model Based on the required heat energy in the regenerator, the natural gas consumption was estimated with using the simple hot water boiler model mentioned in Energy Plus Engineering Reference [23]. The required fuel ( 24
) (Eq. 26) can be calculated considering
the boiler efficiency cure (BEC) (Eq. 27), the part load ratio (PLR) (Eq. 28) and the nominal thermal efficiency of the boiler ( constants were used as follows,
) which was set to be 100% [23]. In Equation 27, the =0.626428326,
=0.645643582,
=-0.77720685 and
=0.313806701.
(26) (27) (28)
3.2.5 Fans and pumps As shown in Figure 2, the DEVap cooling system had five fans: supply, return, LD, DP-IEC, and regeneration air fans. The regeneration air fan was a constant air volume type, and other fans were variable air volume types. In the conventional VAV system, two variable air volume fans (supply and return fans) were used. The power of the variable air volume fan (
can be estimated by a general
variable-air-volume fan model (Eq. 29) suggested by ASHRAE Standard 90.1 [20], where the part load ratio of the variable flow fan ( (
) is the ratio of the current fan airflow
to the design airflow of the fan (
design air flow (
(Eq. 30). The fan power at the
) and for the constant air volume fan can be determined by
Equation 31. The fan pressure at the design air flow ( 3. The efficiency of each fan (
) in each fan is presented in Table
) was assumed to be 50% [11].
(29)
25
(30) (31)
Table 3. Fan pressures at design airflow Components DEVap SA fan
Design
Specifications
768 Pa
DEVap cooler [11] and balance of system
VAV
RA fan
200 Pa
Return dampers, ducts, etc.
LD fan
112 Pa
LD [11]
IEC fan
98 Pa
DP-IEC [11]
SA fan
750 Pa
Air handling unit (AHU), dampers, ducts and balance of system.[22]
RA fan
200 Pa
Return dampers, ducts, etc.
For the chilled water pump in both proposed and VAV systems, Equation 32 is used to calculate the pump power (
. In both systems, the efficiency of the pump (
assumed as 60%, and the pump head (
was
was set to 20 m [20].
(32)
3.2.6 Chiller model The DOE-2 air-cooled electric chiller model [22-23] was used in this research for estimating energy consumption by the chiller. The coefficient of performance (COP) in reference 26
condition is set to be 2.93.
This model consists of three performance curves: CAPFT,
EIRFT, and EIRFPLR. CAPFT (Eq. 33) represents the available capacity of an air-cooled chiller. EIRFT (Eq. 34) provides the full-load efficiency of a chiller as a function of evaporator and condenser temperatures. EIRFPLR (Eq. 35) corresponds to the part-load efficiency of an air-cooled chiller (Eq.36). The required power of the chiller is defined by Equation 37.
(33)
(34) (35) (36) (37)
The energy performance of the VAV system with the air-side economizer control serving the model office building was estimated through a detailed energy simulation, and then compared with that for the DEVap system. The DOE-2 air-cooled chiller model was also used in the VAV system simulation, and an electric heating coil was considered as the reheat coil.
4. Simulation results
4.1 Psychrometrics of the DEVap system Figure 9 shows a typical thermodynamic behavior of the DEVap system on a psychrometric 27
chart at the peak time on a peak summer day for the determination of OA intake case1 and case2. Since OA enthalpy ( enthalpy (
) and dry bulb temperature (
) and temperature (
) were higher than the return air
) (Region 1 in Figure 4), the OA intake flow rate (
) of
the DEVap system is determined by Equation 4 and 7 for case1 and case2 respectively. In the determination of OA intake case1, the mixed air temperature ( SA temperature setpoint (
) was higher than the
), so the LD and DP-IEC cooled and dehumidified the
mixed air initially. The LD dehumidifies air from 0.01316 kg/kg to 0.009823 kg/kg of humidity ratio and the DP-IEC cooled the dehumidified mixed air down to 17.85˚C. The cooling coil was operated to meet the SA temperature setpoint (e.g.,15˚C). In the determination of OA intake case2, the mixed air temperature ( SA temperature setpoint (
) was higher than the
), so the LD and DP-IEC cooled and dehumidified the
mixed air initially. The humidity ratio of the mixed air changed from 0.01429 kg/kg to 0.009958 kg/kg by the LD, and the DP-IEC cooled the dehumidified mixed air down to 17.85˚C. The cooling coil was operated to meet the SA temperature setpoint (e.g.,15˚C). Compared with the OA intake case1, more OA was introduced and that resulted higher temperature and humidity ratio of the MA2 than those of the MA1. As a result, the outlet air condition of the DEVap cooler with case1 showed lower temperature and humidity ratio than those of case2.
28
(a) Determination of OA intake – Case1
(b) Determination of OA intake – Case2 Figure 9. Thermodynamic behavior on a psychrometric chart (July 26th 15:00)
Figure 10 shows variations in hourly temperature and humidity ratio of the process air after 29
each component of the DEVap system on a peak summer day with the determination of OA intake case1 and case2 respectively, when the system was operated under Mode A. The SA setpoint temperature (e.g.,15˚C) was maintained by the cooling coil whenever required, while the temperature and the humidity ratio of the process air decreased through each system component.
(a) Determination of OA intake case1
(b) Determination of OA intake case2 Figure 10. Temperature and humidity ratio of the DEVap system on a peak summer day
4.2 Comparison of humidity ratios of SA during cooling season
In conventional VAV systems, the cooling coil performs latent cooling and sensible cooling 30
simultaneously by decreasing SA temperature until the SA reaches its target condition (e.g., 15°C dry bulb and 80% relative humidity). Terminal reheat may be required if the SA temperature is lower than its setpoint. On the other hand, in the DEVap cooling system, latent loads of introduced OA and conditioned space are accommodated by the LD independently from the sensible cooling device (i.e., DP-IEC). Figure 11 shows the humidity ratio variations of the SA in both VAV and DEVap systems predicted for a week of operation during the cooling season. The SA humidity ratio in the DEVap cooling system (
) varied with the dehumidification capacity of the LD
affected by the OA condition, while the humidity ratio of the SA in the VAV system (
)
was maintained at the target condition by the cooling coil. If the OA was highly humid, the humidity ratio of the SA in the DEVap cooling system could be higher than the target condition, although it may not be significant.
Figure 11. Profiles of humidity ratio of each state for both systems during one week
4.3 Amount of outdoor air intake Figure 12 shows hourly comparisons of the OA ratio in the SA flow between the conventional 31
VAV and the DEVap systems on a peak summer day. Because the OA intake ratio was set to be 0.3
when the supply air flow rate was higher than the minimum ventilation rate in the
determination of OA intake case1, the outdoor air intake ratio showed constant (i.e., 10 to 22).It might result that the OA intake is not enough to dilute the mixing air when the outdoor air intake ratio is lower than that of VAV system (i.e., 10 to 13 and 20 to 22). Because additional OA intake flow was required for DP-IEC operation in the DEVap cooling system with the determination of OA intake case2, the OA intake ratio of the DEVap cooling system was always higher than that of the conventional VAV system. The SA flow was set to the minimum ventilation rate during the system warming up hours (i.e., 7 to 10), so the OA ratio in both systems was 1.0.
Figure 12. Outdoor air mixing ratio on a peak summer day
4.4 Comparison of energy consumption Annual primary energy consumptions of the DEVap cooling system and the conventional VAV system were estimated via detailed energy simulations in this research. The primary 32
energy conversion factors were set to 2.67 for electricity, and 1.0 for natural gas based on a local guideline [24]. One should understand that simulation results be strongly affected by several assumptions (e.g. COP of chiller, redirection air ratio, etc.) defined initially. However, each operating condition assumed for the simulation was based on the values recommended by open literatures.
(a) Energy consumption comparison for operation during cooling season Figure 13 shows a comparison of energy consumptions by both DEVap and VAV systems estimated for the cooling season (i.e., June to August). In cooling season, the energy consumption of the DEVap cooling system could represent the trend of energy consumption in the application of hot and humid climate. Because of the additional fans and higher OA intake flow rate, 46% and 50% more fan energy was consumed in the DEVap cooling system in case1 and case2 respectively compared with the VAV system. The DEVap cooling system also requires natural gas for the regeneration of the desiccant solution, unlike in the VAV system. Because the case1 introduced less outdoor air compared with the case2, the case1 saved 5% of regeneration energy. However, the DEVap cooling system could provide both sensible and latent cooling with negligible cooling coil usage, which resulted in a significant (i.e., 98% in case1 and 92% in case2) reduction in energy consumption by the chiller compared with the VAV system. Consequently, in terms of the primary energy, the DEVap cooling system showed 18% and 10% primary energy saving in the case1 and case2 over the conventional VAV system during summer.
33
1500 2500
Electricity [kWh]
1000
1500
1000 500
500
0
Natural gas and Primary Energy [kWh]
2000
0 Electricity Natural gas Primary
Electricity Natural gas Primary
Electricity Natural gas Primary
DEVap(CASE1)
DEVap(CASE2)
VAV
P_fan_LD
P_fan_IEC
P_fan_RA
P_fan_SA
P_chiller
P_re-heating
P_pump_cw
Q_parallel
Q_regeneration
Primary_Fuel
P_pp_Reg,sol
P_pp_Reg,hw
Primary_Electricity P_fan_reg
Figure 13. Energy performance during summer
(b) Energy consumption of DEVap cooling system during intermediate season Figure 14 shows the operating energy consumptions of the DEVap system and the VAV system estimated for the intermediate season (i.e., March to May and September to October). In intermediate season, the energy consumption of the DEVap cooling system could represent the tendency of energy consumption in the application of hot and dry climate. The DEVap cooling system with the determination of OA intake case1 and case2 consumed 41% and 39% more energy regarding the fans and 97% and 96% less energy regarding the chiller compared 34
with the VAV system. The DEVap cooling system consumed natural gas for the regeneration of weak solution. Consequently, in terms of primary energy consumption, the DEVap cooling system with the determination of OA intake case1 and case2 showed 2.4% and 3.1% primary energy saving compared with VAV system during the intermediate season.
8000 7500
7000
2500
6500
Electricity [kWh]
6000 5500
2000
5000 4500
1500
4000 3500 3000
1000
2500 2000 1500
500
1000 500
0
0 Electricity
Natural gas
Primary Electricity
DEVap_case1
Natural gas
Primary Electricity
Natural gas
DEVap_case2
Primary
VAV
P_fan_LD
P_fan_IEC
P_fan_RA
P_fan_SA
P_chiller
P_re-heating
P_pump_cw
Q_parallel
Q_regeneration
Primary_Fuel
P_pp_Reg,sol
P_pp_Reg,hw
Primary_Electricity P_fan_reg
Natural gas and Primary Energy [kWh]
3000
Figure 14. Energy performance during intermediate season
(c) Energy consumption of DEVap cooling system during heating season Figure 15 represents the energy consumptions of the DEVap cooling and VAV systems estimated during the heating season (i.e., December to February). The DEVap cooling system operated as a VAV system during the heating season. As expected, the energy consumption of both systems were similar. 35
5000 12000
4500 4000
10000
8000
3000 2500
6000
2000 4000
1500
1000 2000 500 0
0 Electricity
Natural gas
Primary Electricity
DEVap_case1
Natural gas
Primary Electricity
DEVap_case2
Natural gas
Primary
Natural gas and Primary Energy [kWh]
Electricity [kWh]
3500
VAV
P_fan_LD
P_fan_IEC
P_fan_RA
P_fan_SA
P_chiller
P_re-heating
P_pump_cw
Q_parallel
Q_regeneration
Primary_Natural gas
Primary_Electricity
P_fan_reg
P_pp_Reg,sol
P_pp_Reg,hw
Figure 15. Energy performance during heating season
(d) Annual energy consumption of the DEVap cooling system Figure 16 shows a comparison of the annual energy consumption of the DEVap cooling system with that of the conventional VAV system. The DEVap cooling system consumed less electrical energy regarding the chiller (i.e., 12.2% in the case1 and 11.8% in the case2) but 39% and 35% more energy regarding the fans in the case1 and case2 respectively because of higher air-side pressure drop and airflow rate compared with the VAV system. The DEVap cooling system also requires natural gas for regeneration of the desiccant solution. Overall, 3%
36
and 2% primary energy saving was achieved by using the DEVap cooling system with the determination of OA intake case1 and case2 respectively over the conventional VAV system.
9000
22000 20000
7500
18000
Electricity [kWh]
14000 12000
4500
10000 8000
3000
6000 4000
1500
2000 0
0 Electricity
Natural gas
Primary Electricity
DEVap_case1
Natural gas
Primary Electricity
DEVap_case2
Natural gas
Primary
VAV
P_fan_LD
P_fan_IEC
P_fan_RA
P_fan_SA
P_chiller
P_re-heating
P_pump_cw
Q_parallel
Q_regeneration
Primary_Fuel
P_pp_Reg,sol
P_pp_Reg,hw
Primary_Electricity P_fan_reg
Natural gas and Primary Energy [kWh]
16000 6000
Figure 16. Annual energy consumption comparison
5. Conclusion
The DEVap cooling system suggested in this study is the integration of the DEVap cooler with the conventional VAV system for reducing the dependence of cooling coil on the vapor compression refrigeration cycle. In addition, annual operation strategies of the DEVap cooling system were suggested. From the energy simulation for the DEVap cooling system, significant energy savings regarding the chiller was observed, while this energy saving was 37
offset instantaneously by the heating energy consumed in the regenerator. A primary energy saving of only 3% and 2% in the case1 and case2 was observed in the energy simulation of the DEVap cooling system over the conventional VAV system. However, a significant reduction (over 90%) of energy consumed in the operation of chillers in comparison with the conventional VAV system can be expected. This implies that the DEVap cooling system shows a possibility of realizing a non-vapor compression HVAC system without energy penalty. Critical energy consumption of desiccant-assisted HVAC systems occurs in the regeneration process. It means that the energy saving potential of the proposed system is highly affected by the amount of heat for the regeneration that is proportionally increased with the level of dehumidification. This research performed for the highly humid climate zone during the summer; that is, relative humidity of outdoor air is over 70% for 1665 hours out of 2208 hours. If the proposed system is used in the less humid climate or dry climate zones, one can expect much enhanced energy saving potential from the proposed system. Additional empirical research may be required for generalizing the energy performance of the proposed system.
Acknowledgements This work was supported by a National Research Foundation (NRF) of Korea (No. 2015R1A2A1A05001726), the Korea Agency for Infrastructure Technology Advancement (KAIA) grant (16CTAP-C116268-01), and the Korea Institute of Energy Technology Evaluation and Planning (KETEP) (No. 20164010200860).
Reference [1]
Liu XH, Qu KY, Jiang Y. Empirical correlations to predict the performance of the dehumidifier using liquid desiccant in heat and mass transfer. Renew Energy 2006;31:1627–39.
38
doi:10.1016/j.renene.2005.08.029. [2]
Gommed K, Grossman G. Experimental investigation of a liquid desiccant system for solar cooling and dehumidification. Sol Energy 2007;81:131–8. doi:10.1016/j.solener.2006.05.006.
[3]
Yin Y, Zhang X, Peng D, Li X. Model validation and case study on internally cooled/heated dehumidifier/regenerator of liquid desiccant systems. Int J Therm Sci 2009;48:1664–71. doi:10.1016/j.ijthermalsci.2008.12.017.
[4]
Hasan A. Going below the wet-bulb temperature by indirect evaporative cooling: Analysis using a modified ε-NTU method. Appl Energy 2012;89:237–45. doi:10.1016/j.apenergy.2011.07.005.
[5]
Lee J, Lee DY. Experimental study of a counter flow regenerative evaporative cooler with finned channels. Int J Heat Mass Transf 2013;65:173–9. doi:10.1016/j.ijheatmasstransfer.2013.05.069.
[6]
Bruno F. On-site experimental testing of a novel dew point evaporative cooler. Energy Build 2011;43:3475–83. doi:10.1016/j.enbuild.2011.09.013.
[7]
Jradi M, Riffat S. Experimental and numerical investigation of a dew-point cooling system for thermal comfort in buildings. Appl Energy 2014;132:524–35. doi:10.1016/j.apenergy.2014.07.040.
[8]
Zhan C, Duan Z, Zhao X, Smith S, Jin H, Riffat S. Comparative study of the performance of the M-cycle counter-flow and cross-flow heat exchangers for indirect evaporative cooling Paving the path toward sustainable cooling of buildings. Energy 2011;36:6790–805. doi:10.1016/j.energy.2011.10.019.
[9]
Jradi M, Riffat S. Experimental investigation of a biomass-fuelled micro-scale tri-generation system with an organic Rankine cycle and liquid desiccant cooling unit. Energy 2014;71:80– 93. doi:10.1016/j.energy.2014.04.077.
[10]
Kim MH, Park JS, Jeong JW. Energy saving potential of liquid desiccant in evaporative-
39
cooling-assisted 100% outdoor air system. Energy 2013;59:726–36. doi:10.1016/j.energy.2013.07.018. [11]
Kozubal E, Woods J, Judkoff R. Development and Analysis of Desiccant Enhanced Evaporative Air Conditioner Prototype. National Renewable Energy Laboratory 2012.
[12]
Boranian AP. An Investigation of Optimal Control of Desiccant-Enhanced Evaporative Air Conditioning. Univ Color 2012.
[13]
ASHRAE. ANSI/ASHRAE Standard 62.1-2013, Ventilation for Acceptable Indoor Air Quality Atlanta: American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc 2013.
[14]
TRNSYS 17. TRNSYS 17 Mathematical Reference. Massachusetts Institute of Technology 2009.
[15]
Engineering Equation Solver (EES) for Microsoft Windows Operating Systems, F-Chart Software 2007.
[16]
Taylor P, Liu Z, Allen W, Modera M. Simplified thermal modeling of indirect evaporative heat exchangers 2013:37–41. doi:10.1080/10789669.2013.763653.
[17]
MathWorks. MATLAB 2014.
[18]
Yonggao Yin and Xiaosong Zhang. Comparative study on internally heated and adiabatic regenerators in liquid desiccant air conditioning system. Building and Environment 2010:1799-1807. doi: 10.1016/j.buildenv.2010.02.008
[19]
Lisa Crofoot. Experimental evaluation and modeling of a solar liquid desiccant air conditioner. Queen’s University. 2012.
[20]
ASHRAE. ANSI/ASHRAE/IES Standard 90.1-2013 Energy Standard for Buildings Except Low-Rise Residential Buildings. vol. 2013. Atlanta: American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc; 2013.
[21]
Company PG and E company. Design Guidelines: Advanced Variable Air Volume (VAV)
40
Systems. Energy Design Resources; 2009. [22]
EnergyPlus. EnergyPlus Input Output Reference. U.S. Department of Energy, Building Technologies Program; 2013.
[23]
Energy UDO. EnergyPlus Engineering Reference: The Reference to EnergyPlus Calculations. US Dep Energy 2010:1051. doi:citeulike-article-id:10579266.
[24]
Ministry of Trade Industry and Energy. Energy Caloric Value Conversion Standard, Energy Act Article 5. 2011.
41
Research highlights
-
Energy saving potential of the DEVap cooling system was evaluated.
-
Workable modes of operation for the DEVap cooling system were suggested.
-
The proposed system provides 2-3% annual energy saving over the conventional VAV system.
-
The proposed system showed 10-18% primary energy saving potentials over the conventional VAV system in summer.
-
The DEVap cooling system can be an energy conservative option for hot and humid climate zones.
42