Available online at www.sciencedirect.

com

APPLIED
ENERGY
Applied Energy 85 (2008) 931950
www.elsevier.com/locate/apenergy

Optimization of water-cooled chiller system

with load-based speed control
F.W. Yu *, K.T. Chan
Department of Building Services Engineering, The Hong Kong Polytechnic University, Hung Hom, Hong Kong, China
Received 19 December 2007; received in revised form 15 February 2008; accepted 18 February 2008
Available online 2 April 2008

Abstract
This study investigates the energy performance of chiller and cooling tower systems integrated with variable condenser
water ow and optimal speed control for tower fans and condenser water pumps. Thermodynamic-behaviour chiller and
cooling tower models were developed to assess how dierent control methods of cooling towers and condenser water
pumps inuence the trade-o between the chiller power, pump power, fan power and water consumption under various
operating conditions. Load-based speed control is introduced for the tower fans and condenser water pumps to achieve
optimum system performance. With regard to an example chiller system serving an oce building, the optimal control
coupled with variable condenser water ow could reduce the annual system electricity use by 5.3% and operating cost
by 4.9% relative to the equivalent system using constant speed fans and pumps with a xed set point for cooling water
temperature control.
2008 Elsevier Ltd. All rights reserved.
Keywords: Water-cooled chiller; Coecient of performance; Electricity consumption; Operating costs

1. Introduction
Chiller and cooling tower systems are widely used to provide cooling energy for commercial and industrial
facilities but with considerable electricity and water consumption. There are many studies on how the control
of chillers and cooling towers can be optimized to enhance their performance [117], but most of them focus
either on electricity savings or on water savings without considering the likely trade-o between the two savings. Hartman [18] launched the equal marginal performance principle (EMPP) to assist in optimizing the
energy performance of chiller systems. The prerequisites for implementing EMPP are that each of the system
components should be driven by a variable speed drive (VSD) and that the mathematical relationship between
the system cooling energy output and the power demand of each component has to be identied. The EMPP
may be an extravagant application for most systems where VSDs are applied solely to secondary-loop pumps
and cooling tower fans.
*

Corresponding author. Tel.: +852 27664560; fax: +852 27657198.

E-mail address: [email protected] (F.W. Yu).

0306-2619/$ - see front matter 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.apenergy.2008.02.008

932

F.W. Yu, K.T. Chan / Applied Energy 85 (2008) 931950

Nomenclature
Aimp
AU
BCL
COP
Cpa
Cpaf
Cpw
Cprg
Cprl
E
Echr
Ectr
hawb
hawbl
hadbe
hadbl
hi
LMTD
ma
mar
mcdw
mctwl
mr
mw
ni
Ntu
P
Pi
PLR
Qcd
Qcl
Qcr
Qm
qrf
Sfan
Spump
T
Tadb
Tadbe
Tadbl
Tawb
Tcdwe
Tcdwl
Tctwe
Tctwl
Tchwr
Tchws
Tevsh
Tref

impeller outlet area (m2)

overall heat transfer coecient (kW C1)
building cooling load (kW)
coecient of performance of chiller
specic heat capacity of air (assumed to be 1.02 kJ kg1 C1)
ctitious specic heat capacity of saturation air (kJ kg1 C1)
specic heat capacity of water (assumed to be 4.19 kJ kg1 C1)
specic heat capacity of vapour refrigerant at evaporator (kJ kg1 C1)
specic heat capacity of liquid refrigerant at condenser (kJ kg1 C1)
power input (kW)
rated power input to chiller compressor (kW)
rated power consumption of the cooling tower fan (kW)
specic enthalpy of saturation air (kJ kg1)
specic enthalpy of saturation air leaving the cooling tower (kJ kg1)
specic enthalpy of air entering the cooling tower (kJ kg1)
specic enthalpy of air leaving the cooling tower (kJ kg1)
specic enthalpy of refrigerant at state point i (kJ kg1)
log mean temperature dierence (C)
mass ow rate of air entering the cooling tower (kg s1)
nominal mass ow rate of air entering the cooling tower (kg s1)
mass ow rate of water entering the condenser or cooling tower (kg s1)
mass ow rate of water leaving the cooling tower (kg s1)
refrigerant mass ow rate (kg s1)
mass ow rate of chilled water (kg s1)
index of reversible polytropic expansion
number of transfer units of the cooling tower
saturated refrigerant pressure of the refrigeration circuit, absolute kPa
refrigerant pressure at state point i, absolute kPa
chiller part load ratio given by Qcl/Qcr
heat rejection (kW)
cooling capacity (kW)
nominal cooling capacity (kW)
cooling capacity for motors heat dissipation (kW)
refrigeration eect (kJ kg1)
speed of cooling tower fan (rpm)
speed of condenser water pump (rpm)
temperature of saturated refrigerant within the refrigeration circuit (C)
dry bulb outdoor temperature (C)
dry bulb temperature of air entering the cooling tower (C)
dry bulb temperature of air leaving the cooling tower (C)
wet-bulb outdoor temperature (C)
condenser water entering temperature (equivalent of Tctwl) (C)
condenser water leaving temperature (equivalent of Tctwe) (C)
cooling water entering temperature (equivalent of Tcdwl) (C)
cooling water leaving temperature (equivalent of Tcdwe) (C)
temperature of return chilled water (C)
temperature of supply chilled water (C)
degree of superheat (C)
reference temperature of water (C)

F.W. Yu, K.T. Chan / Applied Energy 85 (2008) 931950

Uimp
Vr
vi
Win
gpol
gm
b
qa
p
pimp
d
ea
xadbe
xadbl
xawbe
xawbl

933

tip speed of impeller (m s1)

volumetric ow rate of refrigerant (m3 s1)
specic volume of refrigerant at state point i (m3 kg1)
Mechanical work input to compressor (kJ kg1)
eciency of polytropic compression
combined motor and transmission eciency of compressor
angle of impeller (rad)
air density (assumed to be 1.2 kg m3)
system pressure ratio
impeller pressure ratio
throttling rate
airside heat transfer eectiveness of the cooling tower
moisture content of air entering the cooling tower (kg kg1 dry air)
moisture content of air leaving the cooling tower (kg kg1 dry air)
moisture content of saturation air entering the cooling tower (kg kg1 dry air)
moisture content of saturation air leaving the cooling tower (kg kg1 dry air)

Subscripts
cc
compressor
cd
condenser
cdwp condenser water pump
ch
chiller
ct
cooling tower
ev
evaporator
imp
impeller
max
maximum
op
optimum
r
rated condition
Variable ow of chilled water and condenser water is increasingly used to reduce pumping energy in chilled
water systems. According to Bahneth and Peyer [19], the successful application of chilled water systems with
variable primary ow depends on how the ow and chiller capacity can be adjusted to match changing load
conditions. This application is specic rather than generic, and the airside cooling coils are required to be furnished with two-way control valves in order to allow the ow of chilled water to drop under the reduced load
conditions. For a constant air volume (CAV) system where the latent load removal capacity is a concern, the
potential of reducing chilled water ow rate under part load conditions is rather limited and so is the pump
energy savings from this application.
Gordon et al. [20] highlighted that the condenser water ow rate could be a control variable in improving
the energy performance of chiller systems. They established an analytic semi-empirical chiller model to study
variations of chiller COP (coecient of performance) at dierent condenser water ow rates. No analysis was
made on the system level involving the interaction between the compressor power, pump power and tower fan
power. The model serves well for fault detection and diagnosis purposes, but is incapable of accounting the
control variables of cooling towers. No control regime was generalized on how the condenser water ow
should be varied in response to various chiller load and wet-bulb conditions in order to achieve optimal energy
performance of the system.
Applying VSD to cooling tower fans can reduce their cycling frequency and allow better heat rejection control for any given chiller load while maintaining the cooling water temperature set point. This application constitutes a requirement for all tower fans with rated power exceeding 3.7 kW in ASHRAE Standard 90.1-2004
energy standard for buildings except low-rise residential buildings [21]. Yet the true benets of using variable
speed tower fans hinge on how to control their speed to achieve system optimization. For conventional control
of cooling water leaving temperature, a xed and high set point of 34 C above the designed wet-bulb

934

F.W. Yu, K.T. Chan / Applied Energy 85 (2008) 931950

temperature is often used to limit tower fan power, and hence the compressor needs to work at high condensing temperatures with a low chiller COP even when the chiller load or wet-bulb temperature drops. Some engineers tend to reduce the set point at the lowest possible level of around 15 C for all operating conditions to
increase the chiller COP by minimizing the compressor power with more fan power. A xed approach is an
alternative method in which the set point is adjusted by a xed value of 34 C above the changing wet-bulb
temperature. This method assumes that the chiller load would change linearly with the wet-bulb temperature
and that the tower fans can operate at lower speeds with power savings at the reduced wet-bulb temperatures.
None of these temperature controls is a proven technique to achieve minimum energy use of chillers and cooling towers.
Computer simulation is an expeditious means of analysing chiller system performance. Lots of chiller and
cooling tower models have been developed in various forms. Pure empirical models generally fail to perform
optimization studies as most control variables are absent. Transient models, on the other hand, are too
complicated for general engineers to master easily and robust numerical methods are required to solve the
dierential equations. Moreover, they are cumbersome for carrying out hour-by-hour energy analysis of multiple-chiller systems frequently encountered. A general search for and comprehensive surveys [22,23] on many
existing chiller and cooling tower models indicated that very few of them consider the interaction between
chillers and cooling towers. Furthermore, none of individual models is capable of assessing power relationships of chillers, condenser water pumps and cooling towers together with water consumption in the heat
rejection process, with regard to various control methods of cooling towers and condenser water pumps. Hydeman et al. [24] developed a modied DOE-2 regression-based chiller model for the variable condenser water
ow application. One major task along with the model development was to ascertain which modelling
approach would provide the most accurate simulation results with regard to the validation tests using
manufacturer-supplied data and eld-monitored data. Compared with other three public domain models
DOE-2 model [25], Gordon-Ng model [20] and ASHRAE Primary toolkit model [26], the modied model
provided more accurate prediction over the chiller power, especially under low load conditions with variable
condenser water ow or variable speed drives. The improvement in the accuracy of the modied model is due
to the inclusion of the condenser water temperature along with the chiller part load ratio in the EIRFPLR
function which expresses chiller eciency under part load conditions. No control algorithm is included in
the model to govern the cooling tower characteristics in terms of changes in the cooling water temperature
under various airow and water ow conditions for any given heat rejection.
Lu et al. [27] presented a model-based optimization strategy for the condenser water loop of a chiller system. The strategy involved using a genetic algorithm to minimize the total power of the chillers, condenser
water pumps and cooling tower fans. Regarding the physical constraints of the cooling towers, the interaction
between the varying air ow rate and water ow rate for a given heat rejection rate was considered. The inuence of condenser water entering temperature on both the chiller power consumption and cooling tower performance was considered. Yet they did not explain what set point should be used for the condenser water
entering temperature to control the fan speed for optimizing the system. Benton et al. [28] developed a regression model to represent the improved cooling tower simulation algorithm (CTSA). No details were given
about the mathematical expressions of the algorithm but the testing and comparison results were presented
regarding the accuracy of the model against ve CTSAs with the use of manufacturers performance data.
The algorithm considered the cooling tower approach (cooling water leaving temperature subtracted from
the wet-bulb temperature) as a dependent variable which is determined by the independent variables of the
wet-bulb temperature, range (temperature dierence of cooling water), condenser water ow and fan power.
It remains uncertain how the algorithm can be used to evaluate an optimal set point for cooling water temperature as the cooling temperature leaving temperature is an implicit variable in the evaluation of cooling
tower characteristics.
Graves [29] reformulated the Gordon-Ng model in order to study more specically the optimization of the
whole system including two chiller-pump pairs and one cooling tower satisfying the nominal heat rejection
capacity of the two chillers. To properly describe the chiller-tower interaction, a condenser load and condenser
water leaving temperature was involved in the mathematical expression of the compressor power. The chiller
model was coupled with a NTU-eectiveness model for evaluating the cooling tower performance. The modied model was capable of analysing how the system COP varied with the changing condenser water ow. Yet

F.W. Yu, K.T. Chan / Applied Energy 85 (2008) 931950

935

the tower model discounted the water loss due to evaporation, and a single NTU value was assumed to represent the tower performance at dierent water and air ow rates ranging from 50% to 100% of their nominal
levels. Two correlations were identied to facilitate near-optimal system operation: one is the linear relationship between the cooling water set point and the wet-bulb temperature (an analogy to the xed approach
method); another is the linear relationship between the tower fan speed and pump speed.
It is worth promoting variable speed control as a standard environmental-friendly feature for condenser
water pumps and cooling tower fans in water-cooled chiller systems and strengthening simulation techniques
for optimizing such application. The aim of this paper is to assess the energy savings and cost benets of
applying optimal control to the cooling tower fans and condenser water pumps of typical water-cooled chiller
systems. This paper rst presents how a thermodynamic chiller model interacts with a cooling tower model to
determine power relationships between the compressor, condenser water pump and tower fan under various
operating conditions. The tower model, coupled with real control of cooling water temperature, is capable of
simulating fan power and water loss by evaporation based on any given heat rejection and ambient condition.
An assessment will be made on the extent of the increase in the system COP resulting from the optimum speed
control of the tower fans and condenser water pumps. Discussion will be given on how the fan and pump
speed can be controlled directly based on the chiller load in order to achieve optimum system performance.
Following that, an example chiller system serving a local oce building will be considered to characterize
and compare annual electricity and water cost savings when applying various tower and pump control methods. The signicance of this study rests on providing more quantitative analysis to promote water-cooled chiller systems with optimal operating schemes in order to boost their environmental performance in terms of
annual electricity and water consumption, and, at the same time, to reduce their operating costs.
2. Description of the chiller and cooling tower models
The chiller and cooling tower models were developed using TRNSYS 15 [30]. Fig. 1 gives the vapour compression cycle of the water-cooled centrifugal chiller studied. The saturated vapour refrigerant leaves the evaporator at point 1. From point 1 to point 10 superheating occurs when the refrigerant absorbs heat from the
compressor motor. Point 2 indicates a pressure drop in the refrigerant entering the compressor when some
of the refrigerant is throttled through the partially opened inlet guide vanes at part load. After the polytropic
compression process the refrigerant at point 3 is desuperheated in the condenser to a saturated condition at
point 30 . From point 30 the refrigerant is further condensed to the saturated liquid state at point 4. The saturated liquid refrigerant expands isentropically to point 5 before entering the evaporator. No pressure loss in
the refrigerant pipelines was assumed, considering that changes in the condensing pressure and evaporating
pressure due to the loss do not cause an apparent variation in the chiller COP. Given this, P1, P 10 and P5
are equal to the evaporating pressure (Pev), and P3, P 30 and P4 are equal to the condensing pressure (Pcd).

Pressure (kPa)

Pcd

Pev

5
Throttling of
refrigerant at
part load

Polytropic
compression

Refrigerant specific enthalpy, h (kJ/kg)

Fig. 1. Vapour compression cycle of the chiller.

936

F.W. Yu, K.T. Chan / Applied Energy 85 (2008) 931950

Each operating condition comprises seven inputs: dry bulb outdoor temperature (Tadb), wet-bulb outdoor
temperature (Tawb), cooling capacity (Qcl), chilled water ow rate (mw), the temperature of supply chilled
water (Tchws), condenser water ow rate (mcdw) and the temperature of cooling water leaving the cooling tower
(Tctwl). The cooling capacity (Qcl), Tadb and the coincident Tawb would come from the hourly load prole of
chillers operating for a given set of building cooling load and weather conditions. The chilled water ow rate
(mw) was xed at its designed level according to the interlocking operation of one chiller with one chilled water
pump at constant speed. The temperature of supply chilled water (Tchws) was set to be 7 C in all operating
conditions. The reset of Tchws for chiller performance improvements is beyond this study, considering that
not all airside systems and chilled water distribution systems are compatible with this control.
Unlike many past studies considering a xed mcdw for all operating conditions, this analysis attempts to
ascertain how mcdw should be varied to minimize the sum of the compressor power, pump power and tower
fan power. There is no generic method to regulate mcdw under various operating conditions for optimum system performance. Based on the velocity limits of condenser water passing through the condenser, the acceptable range of varying mcdw would be 50100% of the nominal ow rate. One trial in this study is to vary mcdw
linearly with the chiller load in order to keep the temperature change of the condenser water roughly constant.
This could enhance the stability of the cooling water temperature. The implementation of this control calls for
the interlocking operation of each chiller with a variable speed condenser pump and a controller which regulates the pump speed as a function of the chiller load.
The outputs or operating variables of the chiller were determined by the following sets of algebraic equations through an iterative procedure.
2.1. Evaporator model
The cooling capacity (Qcl) of an evaporator is expressed by Eqs. (1)(4):
Qcl PLRQcr

Qcl mw C pw T chwr  T chws

Qcl mr qrf

2
3

Qcl AU ev LMTDev

where
qrf h1  h5
1
AU ev
0:8
c1 mw c2 Qcl0:745 c3
T chwr  T chws
LMTDev
lnT chwr  T ev =T chws  T ev

5
6
7

The method of log mean temperature dierence (LMTD) was used to model the heat transfer performance of
the shell-and-tube ooded type evaporator. The overall heat transfer coecient of the evaporator (AUev) is
described by a mechanistic relation in Eq. (6) [31], where c1, c2 and c3 are parameters to be evaluated based
on the performance data of the chiller at full load and part load conditions.
2.2. Compressor model
The chiller contained one refrigeration circuit and one hermetic centrifugal compressor within which inlet
guide vanes were used to modulate the cooling output at between 25% and 100% of the rated full load capacity
while controlling the temperature of supply chilled water at its set point. The volumetric ow rate of the refrigerant at the impeller outlet (Vr) can be evaluated from Eq. (8) [26,32]:




Aimp
ni
P 2 v2 pni1=ni  1  U 2imp
tanb
Vr
8
ni  1
U imp

F.W. Yu, K.T. Chan / Applied Energy 85 (2008) 931950

937

The angle (b), outlet area (Aimp) and tip speed (Uimp) of the impeller are constants with regard to the constant
speed compressor and they were determined by using the parameter identication program given by Bourdouxhe et al. [26]. The system pressure ratio p given by Eq. (9) accounts for the throttling rate d which is equal
to 1 at full load. Under part load conditions, d is less than 1, indicating an increase in the system pressure ratio,
p

P cd
dP ev

It is dicult to nd d at dierent part load ratios because the refrigerant state at the impeller inlet is not identied. To reduce the computation eort to determine d based on fundamental equations, it is possible to correlate mr with its maximum mr,max at full load condition by a ratio c, as shown in Eq. (10). The ratio c was
identied to be a function of chiller part load ratio (PLR), as shown in Eq. (13), where the constants a1 to
a3 were determined based on the modelled results of mr under various operating conditions,
mr cmr; max
V r 1=ni
mr; max pimp
v2
"
ni  1 1
pimp 1
2ni P 2 v2

10
11
U 2imp

V 2r
 2
Aimp sin2 b

c a1 PLR2 a2 PLR a3

!#ni=ni1
12
13

Regarding the computation of the actual compressor power (Ecc) given by Eq. (14), the polytropic compression eciency (gpol) can be expressed by an empirical polynomial of chiller part load ratio and was determined
by correlating the modelled results of the mechanical work input to the compressor (Win) with the performance data of Ecc under part load conditions. The combined motor and transmission eciency (gm) was taken
to be 0.8 for the constant speed compressor. The specic enthalpy of superheated refrigerant at the compressor
discharge (h3) is given by Eq. (17),
Ecc

W in
gpol gm

14

where

ni  ni1
p ni  1
ni  1
2
gpol a4 PLR a5 PLR a6
Ecc
h3 h2
mr
W in mr P 2 v2

15
16
17

2.3. Expansion valve model

An orice plate with a xed opening was used to throttle proper level of the refrigerant in the evaporator at
various load conditions. Assuming that the motors heat dissipation (Qm) accounts for 5% of Ecc as given by
Eq. (19), the degree of superheat (Tevsh) can then be calculated by using Eq. (20),
Qm mr h10  h1
Qm 0:05Ecc

18
19

h10 h1 C prg T evsh

20

938

F.W. Yu, K.T. Chan / Applied Energy 85 (2008) 931950

2.4. Condenser model

Heat rejection (Qcd) involves the energy and mass balance in the condenser and is given by Eqs. (21)(26).
The overall heat transfer coecient (AUcd) was predicted by Eq. (25) [31], where the constant parameters c4c6
were evaluated using the performance data of the chiller at full load and part load conditions. The condenser
water entering temperature (Tcdwe) is the temperature of cooling water leaving the cooling tower (Tctwl) of
which the set point is used to control the cooling tower fan while satisfying Qcd,
Qcd Qcl Ecc
Qcd mr h3  h4

21
22

Qcd mcdw C pw T cdwl  T cdwe

Qcd AU cd LMTDcd

23
24

where
AU cd

1
1=3

0:8
c4 mcdw

c5 Qcd c6
T cdwl  T cdwe
LMTDcd
lnT cd  T cdwe =T cd  T cdwl

25
26

2.5. General control of cooling tower fans by use of cooling water leaving temperature
A robust cooling tower model with actual control of cooling water leaving temperature is developed to
illustrate how all operating variables in the heat rejection process can be computed. Modelling equations from
an eectiveness model given by Braun [33] was adopted but the problem-solving approach was redened to
match the real fan control logic. The model is capable of evaluating the states of air passing through a cooling
tower, and thus water loss due to evaporation can be estimated. As Eq. (27) illustrates, the heat rejection
capacity of a cooling tower correlates with the airside heat transfer eectiveness (ea), the air mass ow rate
(ma) and the hypothetical dierence between the enthalpy of saturation air leaving the tower and the enthalpy
of air entering the tower. ea of a cross-ow cooling tower is given by Eq. (28), assuming that the Lewis number
is one. The variable Cpaf is the ctitious specic heat which is dened as the enthalpy dierence over the temperature dierence of the leaving and entering conditions of the saturation air. Regardless of the entering or
leaving conditions, the saturation enthalpy (hawb) can be determined from its wet-bulb temperature (Tawb) by
Eq. (31), where the constants k0 to k3 are 9.3625 kJ kg1, 1.7861 kJ kg1 C1, 0.01135 kJ kg1 C2 and
0.00098855 kJ kg1 C3, respectively.
The number of transfer units (Ntu) was used to determine the cooling tower performance, which is given by
Eq. (32) with the characteristic constants c and n. It varied with the air mass ow rate (ma) for a constant or
variable mass ow rate of water leaving from the condenser (mcdw).
Qcd ea ma hawbl  hadbe
1
ea  1  expm 1  expNtu
m

27
28

where
ma C paf
mcdw C pw
hawbe  hawbl

T awbe  T awbl
k 0 k 1 T awb k 2 T 2awb k 3 T 3awb

m

29

C paf

30

hawb

1n

Ntu cmcdw =ma

31
32

F.W. Yu, K.T. Chan / Applied Energy 85 (2008) 931950

939

The states (enthalpy hadbl and moisture content xadbl) of air leaving the cooling tower were determined by Eqs.
(33) and (34). Given xadbl, the mass ow rate of water leaving the cooling tower (mctwl) can be evaluated using
the mass balance in Eq. (35), where the water loss rate due to evaporation is given by (mcdw  mctwl). The cooling water leaving temperature (Tctwl) is related to the cooling water entering temperature (Tctwe) by Eq. (36).
Tref is the water reference temperature and 0 C was used,
hadbl hadbe ea hawbl  hadbe
xadbl xawbl xadbe  xawbl expNtu

33
34

mctwl mcdw  ma xadbl  xadbe

35

T ctwl

mcdw C pw T ctwe  T ref  Qcd

T ref
mctwl C pw

36

According to the conventional approach to evaluating cooling tower performance, the heat rejection airow
rate (ma) is one of the inputs required to calculate the temperatures of cooling water, states of ambient air, etc.
This is in contrary to the real situation where ma is a variable depending on how the tower fan is controlled by
a set point of cooling water leaving temperature (Tctwl) or condenser water entering temperature (Tctwe).
Where a constant speed fan is used, it will be cycled on and o to maintain Tctwl within the tolerable range
of the set point. For a variable speed tower fan, the air mass ow rate varies by modulating the fan speed
to satisfy the required Tctwl. Based on the aforementioned equations, an iteration procedure shown in
Fig. 2 is required to solve ma based on a given Tctwl.
For a given set of inputs (mcdw, Tctwl and Tawb) and Qcd computed by the condenser model, the range and
approach of the cooling tower can be evaluated. Tctwl has to be set at above Tawb to achieve a positive
approach (Tctwl  Tawb) in order to calculate a logical heat rejection airow rate (ma). The initial trial for
the hypothetical wet-bulb temperature of air leaving the cooling tower (Tawbl) is taken to be the cooling water
entering temperature (Tctwe) subtracted by 0.01 C. The ctitious specic heat Cpaf can then be determined,
given the enthalpies and temperatures of the saturation air calculated at the leaving and entering conditions.
The LMTD method is applied to describe the eective heat transfer coecient (AUct) of the cooling tower, and
ma is then computed by using the given relationship between Ntu and AUct. A new Tawbl can be calculated
based on the ma computed and checked with its previous value. If the dierence between the values of Tawbl
in successive calculations is within 0.005 C, then the convergent criterion of the solutions can be met and all
other variables can be determined explicitly. Otherwise Tawbl needs to be deducted by 0.01 C in each iteration
to repeat the calculation steps until the convergent criterion is satised. In cases where the computed ma is
greater than its nominal level mar, mar should be used to determine the smallest approach and all other variables for the given set of inputs.
The water evaporation rate of a cooling tower working under various operating conditions was computed
according to the mass ow rate of air entering the cooling tower and the dierence between the leaving and
entering moisture contents of that air, i.e. ma (xadbl  xadbe). The power of a constant speed fan (Ect) is the
rated power Ectr multiplied by the factor (ma/mar), considering that the average cycling period of the fan during an hour is directly proportional to the required ma and assuming that the fan power increases as a linear
function of the tower airow. If the fan operates continuously at its full ow (mar) throughout an hour, the
factor will be equal to one and its rated power will be taken up over that hour. While a variable speed fan
is used, the fan law is applied and the factor becomes (ma/mar)3 based on the fan law with the cube relationship
between the power and ow rate. In this analysis, the variable speed drive was assumed to account for 3% of
the power of the fan operating at any speed and the minimum ow rate delivered could be down to 10% of the
rated ow rate.
2.6. Evaluation of operating variables
The chiller and cooling tower models were used to evaluate all the operating variables based on any given
input data and constant parameters. The ow charts in Fig. 3 show the relationships between the component
models and how the operating variables were evaluated. The evaporator model started to compute PLR,
Tchwr, AUev and LMTDev based on the inputs: cooling capacity (Qcl), chilled water ow rate (mw) and the

940

F.W. Yu, K.T. Chan / Applied Energy 85 (2008) 931950

Fig. 2. Procedure for evaluating the required air mass ow rate under the actual control of cooling tower fans.

temperature of supply chilled water (Tchws). It then calculated the evaporating temperature (Tev) and pressure
(Pev), and refrigerant properties at the evaporator discharge (h1, v1). Given that the condensing temperature
(Tcd) linked the compressor and condenser models, the operating variables of these models had to be computed altogether at a specic accuracy through an iterative procedure. The iterative procedure started with
an initial condensing temperature (Tcdo) of 45 C in calculating the variables of the compressor model (the
case: ITERo = 0). A degree of superheat (Tevsho) of, say, 3 C was initially assumed in order to compute
the compressor power (Ecc). Iterations were carried out on each Tevsh calculated by the expansion valve model
until Tevsh converged to within a specied tolerance of 0.005 C.
In the condenser model, the inputs consisted of cooling water leaving temperature (Tctwl), compressor
power (Ecc), cooling capacity (Qcl) and the refrigerant mass ow rate (mr) satisfying the requirement of the

F.W. Yu, K.T. Chan / Applied Energy 85 (2008) 931950

941

Fig. 3. Procedure for determining the operating variables of the chiller model.

evaporator and compressor models. Based on the heat rejection (Qcd) computed by the condenser model, operating variables related to the cooling tower were then determined by solving the modelling equations and using
the control algorithm of operating the tower fan. If the dierence between the condensing temperature (Tcd)
evaluated from the condenser model and its previous value calculated from the compressor model was within
0.005 C, all operating variables would be calculated logically with the required accuracy; otherwise the next
value of Tcd would substitute for its previous one to perform the next iteration until the accuracy was met.
The COP of a chiller is dened as the cooling capacity (Qcl) divided by the compressor power (Ecc). In the
calculation of a system COP, the power input means the total power of the compressor (Ecc), condenser water
pump (Ecdwp) and cooling tower fan (Ect). When the condenser water pump operates at constant speed, Ecdwp
is xed at its rated value Ecdwp,r for the chiller operating. When a VSD is applied to the pump, Ecdwp is Ecdwp,r
multiplied by a factor (mcdw/mcdwr)3 based on the cube relationship between the power and ow rate. mcdw

942

F.W. Yu, K.T. Chan / Applied Energy 85 (2008) 931950

varies linearly with the chiller part load ratio (PLR), from 0.5 mcdwr at a PLR of 0.5 to mcdwr at a PLR of 1.
The variable speed drive was assumed to take up 3% of Ecdwp at any speed.
3. Results and discussion
3.1. Improved performance from optimal tower fan control with variable condenser water ow
Table 1 summarizes the details of the chiller system studied. The total capacity of the system was designed
to satisfy the peak cooling load of an oce building (see Section 3.3 for details). The chiller model was calibrated for the part load performance of the chiller shown in Fig. 4. Each performance curve represents only
the variation of the chiller COP at dierent part load ratios for a given condenser water entering temperature
(Tcdwe), irrespective of how Tcdwe or the cooling water leaving temperature (Tctwl) is actually controlled by the
cooling tower at any given wet-bulb temperature. The parameters of the modelled chiller given in Table 1 were
identied from the model validation exercise and the characteristic parameters (c and n) used for the cooling
tower are representative of typical cross-ow cooling towers.
The operating schemes shown in Table 2 were analysed with respect to dierent controls of the condenser
water pump and cooling tower. Schemes 14 represent a constant speed conguration of the cooling tower fan
and condenser water pump. Under this conguration, the tower fan is cycled on and o to deliver the heat
rejection airow required to meet a given Tctwl for any heat rejection. The condenser water pump is staged
continuously to provide the operating chiller with the rated condenser water ow rate for all loading conditions. For schemes 58, variable speed control is applied to the fan and pump. Under this control, the fan
Table 1
Details of the chiller system
Total plant cooling capacity (kW) (four identical sets of chillers, pumps and cooling towers)
For each chiller
Refrigerant type
Nominal cooling capacity (kW)
Nominal compressor power (kW)
COP at full load
Design chilled water supply/return temperature (C)
Design chilled water ow rate (l/s)
Design condenser water entering/leaving temp. (C)
Design condenser water ow rate (l/s)
Chiller model parameters
Evaporator (c1, c2, c3) (units omitted)
Compressor (Aimp, Uimp, b)
Compressor (a1, a2, a3 for c)
Compressor (a4, a5, a6 for gpol)
Condenser (c4, c5, c6) (units omitted)
For each cooling tower
Type
Heat rejection capacity (kW)
Design entering/leaving temperature (C)
Water ow rate (l/s)
Motor
Air volume ow rate (m3/s)
Air mass ow rate (kg/s)
Fan motor power (kW)
Drift loss (% of nominal ow)
Design water evaporation rate (% of nominal ow)
Design wet-bulb outdoor temperature (C)
Tower model parameters
c, n
Rated power of each chilled water pump (kW)
Rated power of each condenser water pump (kW)

6400
R134a
1600
280.7
5.7
7/12.5
72
33/38
87
0.1172, 0.0593, 0.0001
0.0059 m2, 200 m/s,
2.269 rad
0.0704, 0.931, 0.00006
0.8131, 1.537, 0.0057
0.0005, 0.3443, 0.0002
Cross-ow
2004
34.4/29.4
87
Staged or variable speed
63
75.6
22.8
0.2
0.53
28
2.3, 0.7
47.0
21.6

F.W. Yu, K.T. Chan / Applied Energy 85 (2008) 931950

943

T cdwe (oC)

Chiller COP

33
30
28
26
24
22
20
18

3
0

0.2

0.4
0.6
Chiller part load ratio

0.8

Fig. 4. Part load performance of the water-cooled centrifugal chiller.

Table 2
Operating schemes of dierent controls of condenser water pump and cooling tower
Scheme

Notation

Description

1 (baseline)
2
3
4
5
6
7
8

Tctwl29.4CS
Tctwl18.3CS
App4CS
Tctwl,opCS
Tctwl29.4VS
Tctwl18.3VS
App4VS
Tctwl,opVS

Tctwl set at 29.4 C with constant speed congurations

Tctwl set at 18.3 C with constant speed congurations
Approach (TctwlTawb) xed at 4 C with constant speed congurations
Optimum Tctwl for maximum system COP with constant speed congurations
Tctwl set at 29.4 C with variable speed congurations
Tctwl set at 18.3 C with variable speed congurations
Approach (TctwlTawb) xed at 4 C with variable speed congurations
Optimum Tctwl for maximum system COP with variable speed congurations

speed is modulated in response to a given Tctwl to continuously adjust the airow for any heat rejection with
fan power savings. The variable speed pump provides varying condenser water ow which changes in direct
proportion to the chiller load in order to maintain a constant temperature dierence across the condenser.
The set points of 29.4 C and 18.3 C used for Tctwl in the schemes 1, 2, 5 and 6 are based on the upper and
lower limits of part load rating conditions given in ARI standard 550/590 [34]. A xed approach of 4 C was
considered in the schemes 3 and 7, considering that it helps oer fan power savings when the chiller load drops
and that this control provides more energy savings compared with the xed set point control [12]. For schemes
4 and 8, the models contain an algorithm which searches for an optimal Tctwl to minimize the sum of compressor power, condenser water pump power and cooling tower fan power for any given operating condition.
Fig. 5 shows the part load performance of the system (one chiller dedicated with one condenser water pump
and one cooling tower) operating at dierent combinations of part load ratios and wet-bulb temperatures in
6
Wet bulb outdoor temperature (oC)

System COP

16
20
24
28

2
0.2

0.4

0.6
Chiller part load ratio

0.8

1.0

Fig. 5. System part load performance at dierent wet-bulb temperatures in scheme 1.

944

F.W. Yu, K.T. Chan / Applied Energy 85 (2008) 931950

20

Percentage change in system COP (%)

Percentage change in system COP (%)

the scheme 1 (the baseline). There is no signicant dierence in the trend of system COPs at part load with
dierent wet-bulb temperatures when the set point of Tctwl is xed at 29.4 C for all operating conditions.
For a given wet-bulb temperature, the system COP is maximized at a part load ratio of 0.80.9, instead of
at full load, and it drops considerably when the chiller load reduced from such part load ratios. This variation
in the chiller COP is governed by compressor eciency and the capacity control done by the inlet guide vanes.
Comparisons were made on how the system COP varied under various tower fan and pump controls. As
Fig. 6 illustrates, lowering Tctwl to 18.3 C could not bring an absolute increase in the system COP, though the
compressor power was lowest for most operating conditions among the other operating schemes. Indeed, the
system COP could drop by up to 10% when the chiller load reduced from a part load ratio of 0.8. This is
because the tower fans operated at full speed continuously and the increase of the fan power exceeded the
reduction of the compressor power. Except the 18.3 C Tctwl control, the system COP could be improved
by reducing the condenser water ow rate via modulating the pump speed in response to the chiller load. This
implies that keeping the temperature dierence of the condenser water at the design level is a viable means to
optimize the trade-o between the compressor power and condenser water pump power.
The system COP could increase by various degrees for all operating conditions when using the 4 C
approach or implementing the optimal Tctwl, regardless of whether the on/o or variable speed control was
used for the tower fans and condenser water pumps. When the wet-bulb temperature was below 28 C, the
percentage increase in the system COP was more noticeable at higher chiller loads when the 4 C approach
control was in place of the 29.4 C Tctwl control. The system COP increased by 1.416.1% by using the optimum Tctwl and VSDs for both the fans and pumps. Such an increase tends to be signicant at the minimum

Operating schemes
15

10

Tctwl18.3CS

App4CS

Tctwl,opCS

Tctwl29.4VS

Tctwl18.3VS

App4VS

Tctwl,opVS

-5
0.2

0.4

0.6

0.8

10

-5

-10

0.4

Chiller part load ratio

Tctwl29.4VS
App4VS

0.6

0.8

1.0

0.8

1.0

Chiller part load ratio

10

20

-5

Operating schemes
Tctwl18.3CS
Tctwl,opCS
Tctwl18.3VS
Tctwl,opVS

-10

App4CS
Tctwl29.4VS
App4VS

-15
0.2

0.4

0.6

Chiller part load ratio

0.8

1.0

Percentage change in system COP (%)

Percentage change in system COP (%)

App4CS

Tctwl,opCS
Tctwl18.3VS
Tctwl,opVS

-15
0.2

1.0

Operating schemes
Tctwl18.3CS

Operating schemes
15
10

Tctwl18.3CS

App4CS

Tctwl,opCS
Tctwl18.3VS

Tctwl29.4VS
App4VS

Tctwl,opVS

5
0
-5
-10
0.2

0.4

0.6

Chiller part load ratio

Fig. 6. Percentage change of system COP in schemes 28 in relation to the baseline.

F.W. Yu, K.T. Chan / Applied Energy 85 (2008) 931950

945

and maximum loading conditions and be moderate at a part load ratio of around 0.5 with a wet-bulb temperature of below 28 C.
3.2. Load-based speed control for system optimization
The simulation results showed that for a chiller operating at above half load, it is possible to adjust the
condenser water ow rate linearly with the part load ratio (PLR) to optimize the trade-o between the compressor power and condenser water pump power while maintaining the temperature dierence at the condenser side at its design level. To achieve such ow modulation, the variable speed drive should regulate
the pump speed (Spump,op) based on Eq. (37), given that speed is direct proportional to ow rate according
to the pump laws. Spump,full in Eq. (37) is the full speed of the pump,

PLRS pump;full if PLR > 0:5
S pump;op
37
0:5 S pump;full
if PLR 6 0:5
To perform optimum temperature control for cooling tower fans, it is essential to have a dedicated controller
which calculates the optimum set point (Tctwl,op) based on signals of wet-bulb temperatures (Tawb) and the
chiller load in terms of a part load ratio (PLR), and operates the fans at the right speed to meet that set point.
Eq. (38) gives the basic mathematical expression of Tctwl,op, where b0 to b2 are coecients to be determined
based on the power relationships between the chillers, tower fans and condenser water pumps for a given system design. In this analysis, b0, b1 and b2 were identied to be 2.9178, 4.8636 and 2.095, respectively, according to the plot of (Tctwl,opTawb) against PLR given in Fig. 7. The data in the plot refer to Tctwl,op coincided
with the maximum system COP at various combinations of PLR from 0.2 to 1 at 0.1 intervals and Tawb from
16 to 28 C at 4 C intervals. The Tctwl,op predicted by using Eq. (38) deviates from its actual value by 0.50 to
0.49 C only, based on its possible change from 20.0 to 33.2 C. This slight prediction error would bring no
signicant eect on attaining the maximum system COP,
T ctwl;op T awb b0 b1 PLR b2 PLR2

38

The calculation of PLR involves measurements of three variables: the ow rate (mw), supply temperature
(Tchws) and return temperature (Tchwr) of chilled water. Overall there are four measured variables aecting
the uncertainty of the calculated Tctwl,op, in addition to its prediction error. A high quality relative humidity
sensor should be used to limit the uncertainty of the measured Tawb. Furthermore, the measurements should
have response time which is short enough to trace the individual changing patterns of PLR and Tawb.
For most existing cooling water temperature control, it would be dicult to incorporate the reset algorithm
into the controllers. Alternatively, it is expected to have more simple and direct control for the tower fans in

(Tctwl,op - Tawb) at maximum system COP

8
7
6
5
4
y = -2.095x

3
2
0.2

+ 4.8636x + 2.9178

R2 = 0.8497

0.4

0.6
0.8
Chiller part load ratio

1.0

Fig. 7. Relationship between the optimum cooling water leaving temperature (Tctwl,op), wet-bulb temperature (Tawb) and chiller part load
ratio.

F.W. Yu, K.T. Chan / Applied Energy 85 (2008) 931950

% full speed of fans at maximum system COP

946

100

80

60

40

20

Near-optimum % fan speed = 72.808 PLR + 17.762

R2 = 0.9215

0
0.2

0.4

0.6
Chiller part load ratio

0.8

1.0

Fig. 8. Load-based speed control for cooling tower fans.

order to achieve maximum system COP for all operating conditions. As Fig. 8 illustrates, it is possible to correlate the fan speed (in % full speed) with the chiller part load ratio (PLR). The data set shown in Fig. 8 refers
to the optimum fan speed at which the maximum system COP took place for a set of operating conditions in
terms of various combinations of PLRs from 0.2 to 1 at 0.1 intervals and wet-bulb temperatures from 16 to
28 C at 4 C intervals. Based on the regression curve with the coecient of determination (R2) of 0.9215, the
near-optimum fan speed (Sfan,op) can be predicted by Eq. (39), where Sfan,full is the full speed of the fan,
S fan;op 0:7281PLR 0:1776S fan;full

39

Table 3 gives a comparison between the percentage increase of the system COP from baseline under the
load-based speed control and that under the optimum Tctwl control of the tower fans, with regard to various
operating conditions. For any given operating condition, the results of the two controls are quite comparable,
suggesting that it is viable to use the load-based speed control to achieve near-optimum operation of the system while eliminating sophisticated controllers and sensors for the cooling water temperature reset. When the
load-based speed control is recognized for the tower fans and condenser water pumps, the optimal control of
the whole system can be highly simplied. This is because the sequencing of chillers, pumps and tower fans and
their individual speed controls can be based entirely on the chiller load conditions only. Such load-based system control serves to emphasize the relevance of chiller load measurement and verication to system optimization. It is expected that such control is generic for all types of chiller systems with multiple chiller
arrangements and with full or partial use of VSDs for the system components.
3.3. Potential benets from the load-based speed control
The cooling load prole of a reference oce building in Hong Kong was considered in order to assess the
potential operating cost savings when the load-based speed control was applied to the chiller plant designed
Table 3
Comparison between the percentage increase in system COP from baseline under the load-based fan speed control (SC) and that under the
optimum cooling water temperature control (TC)
Wet-bulb (C)

Chiller part load ratio

0.2

16
20
24
28

0.4

0.6

0.8

SC

TC

SC

TC

SC

TC

SC

TC

SC

TC

10.21
8.19
6.89
10.52

10.30
8.21
7.12
11.00

5.27
3.79
2.99
6.48

5.28
3.83
3.06
6.57

5.44
3.11
1.54
3.52

5.52
3.29
1.82
3.95

9.61
5.60
2.46
2.31

9.71
5.86
2.87
2.93

16.01
9.51
4.03
1.26

16.08
9.73
4.43
1.81

F.W. Yu, K.T. Chan / Applied Energy 85 (2008) 931950

947

with four identical sets of the chillers described in Table 1. Detailed features of the building are given in Ref.
[35]. The building has 40 storeys and a total air-conditioned oor area of 42840 m2. Fig. 9 is a histogram showing the building cooling load prole. There are 2834 cooling hours in total which account for 90.5% of the
total oce hours (3131 h a year). The annual cooling energy for the building is 7,423,883 kW h based on local
weather conditions of an example year for building energy analysis. To meet the changing building cooling
loads, conventional chiller sequencing was considered, under which all the chillers are operating at the same
load, and no additional chillers start to operate until each of the running chillers is operating at its full load of
1600 kW. Within the chiller system, a single-loop pumping system with a dierential pressure by-pass pipe was
used to control the amount of chilled water owing from the operating chillers to cooling coils of the airside
equipment. There are four constant speed pumps, each dedicated to one chiller to provide a constant chilled
water ow of 72 l/s. Each pump has a rated power of 47 kW.
Drawing on the chiller and cooling tower models, calculations were made on the annual electricity consumption of the chiller plant with the eight operating schemes listed in Table 2. The load-based speed control
(L-VS) was considered as a substitute for the optimum Tctwl control in the scheme 8. The electricity consumption is normalized by the total air-conditioned oor area of the building in terms of kW h m2. Table 3 shows
the results of the normalized electricity consumption of the system components along with the annual water
consumption of the cooling towers. The water consumption is made up of three parts of water loss: evaporation, drift and bleed-o. The cooling tower model was used to estimate the loss rate by evaporation. The drift
rate and bleed-o rate were assumed to be 0.2% and 0.6%, respectively, of the cooling water circulation rate,
with regard to the use of traditional chlorination water treatment.
Table 4 shows that the annual electricity consumption of chilled water pumps is 6.6 kW h m2 for all the
operating schemes because the same sequencing of the chillers was applied. The use of variable speed control
for the condenser water pumps enables their annual electricity consumption to drop by 46.5%. A low set point
of the cooling water leaving temperature helps lower the annual chiller electricity consumption (as in schemes
2 and 6), but with much higher fan electricity consumption. Using the load-based speed control for the tower
fans and condenser water pumps, the trade-o between the chiller power, pump power and fan power was
optimized, resulting in a drop of 5.3% in the annual plant electricity consumption. The reduced condenser
water ow along with the advanced control could bring about a 12.8% reduction or 63.8 l m2 drop in the
annual water consumption.
To investigate the cost eectiveness of the eight operating schemes for the chiller system, the annual operating costs and life cycle costs were calculated and compared. Local tari structures were considered to evaluate the operating costs associated with the annual electricity and water consumption. Operating costs other
than the electricity and water charges were assumed to be the same value for all the schemes and they were

500

1 chiller operating

2 chillers operating

3 chillers operating

4 chillers operating

No. of operating hours

400

300

200

100

0
400

800 1200 1600 2000 2400 2800 3200 3600 4000 4400 4800 5200 5600 6000 6400

Building cooling load (kW)

Fig. 9. Building cooling load prole.

948

F.W. Yu, K.T. Chan / Applied Energy 85 (2008) 931950

Table 4
Annual chiller plant electricity and water consumption per unit of air-conditioned oor area of the building
Scheme

1 (baseline)

Normalized annual electricity consumption (kW h m2)

Chiller
30.6
29.0
Chilled water pump
6.6
6.6
Condenser water pump
3.1
3.1
Cooling tower fan
1.8
3.1
Total
42.0
41.8
Normalized annual water consumption (l m2)
497.4
598.5

8 (% saving from baseline)

30.1
6.6
3.1
2.0
41.7

29.9
6.6
3.1
2.1
41.7

31.4
6.6
1.6
0.7
40.4

29.8
6.6
1.6
3.0
41.2

31.0
6.6
1.6
0.7
39.9

513.0

519.7

413.3

519.6

424.5

30.8
6.6
1.6
0.7
39.8

(0.7%)
(0%)
(46.5%)
(57.6%)
(5.3%)

433.6 (12.8%)

Table 5
Annual operating costs and life cycle costs per unit of air-conditioned oor area of the building for the chiller plant with eight operating
schemes
Scheme

1 (baseline)

Normalized annual operating cost (HK$ m2)

Electricity
51.7
51.3
Water
2.6
3.1
Total
54.3
54.4

8 (saving from baseline)

51.4
2.7
54.0

51.3
2.7
54.0

49.9
2.1
52.1

50.9
2.6
53.5

49.6
2.2
51.8

49.5 (2.2)
2.2 (0.4)
51.7 (2.6)

Normalized life cycle operating cost (HK$ m2)

Electricity
393.2
390.5
390.7
Water
19.7
23.3
20.3
Total
413.0
413.8
411.0

389.9
20.5
410.4

379.9
16.3
396.2

387.1
20.0
407.1

377.0
16.7
393.7

376.4 (16.8)
17.0 (2.7)
393.4 (19.5)

excluded in this cost analysis. Regarding the calculation of the life cycle operating costs, a life span of 15 years
was assumed for the chiller plant and a discount rate of 10% was considered throughout the lifespan. Table 5
gives the calculation results. The annual operating cost could drop by HK$2.6 to HK$51.7 m2 by using the
load-based speed control for the condenser water pumps and cooling tower fans instead of the conventional
on/o control with a xed cooling water leaving temperature. The actual annual savings of HK$111,384 helps
decide how much investment cost the load-based speed control is worth taking. If the building owner accepts a
maximum payback of up to 2 years for a chiller plant improvement program, the ceiling on the investment
cost for the advanced control could be HK$222,768 which is expected to be sucient for purchasing eight
VSD controllers for the tower fans and condenser water pumps and the associated control system. After
the operating cost savings fully recoup the implementation cost, the building owner can enjoy prots brought
from the optimum control throughout the rest of the systems functional life. It is envisaged that the implementation of load-based speed control would be more economically attractive if the chiller plant needs to
operate at part load conditions for longer hours per year or if there is a subsidized scheme for purchasing
energy ecient equipment.
4. Conclusions
This paper presents the use of load-based speed control to enhance the energy performance of water-cooled
chiller systems. Thermodynamic-behaviour chiller and cooling tower models have been developed to investigate how the energy and water uses vary for a chiller system operating under various controls of condenser
water pumps and cooling tower fans. The optimum operation of the system can be achieved simply and
directly by the load-based speed control under which the speed of the tower fans and condenser water pumps
is regulated as a linear function of the chiller part load ratio. The superiority of such control rests on its coherence with typical sequencing of chillers based entirely on their load conditions and on eliminating the need of
high quality humidity sensors for the reset of cooling water temperature. The system COP under the optimal
control could increase by 1.416.1% relative to the equivalent system with xed temperature and ow rate

F.W. Yu, K.T. Chan / Applied Energy 85 (2008) 931950

949

controls for the cooling water leaving from cooling towers. A case study showed that the optimum control
could bring about a drop of 5.3% in the operating cost of a chiller plant running for the cooling load prole
of a local oce building. It is envisaged that the implementation of the optimum control is economically viable, as the operating cost savings would fully recoup the investment cost associated with the control in two
years or less. The ndings of this research highlight the need to widen the use of variable speed drives with
load-based speed control for water-cooled chiller systems serving air-conditioned buildings in order to
enhance their sustainability.
Acknowledgement
This study was supported by a grant from the central research grant of The Hong Kong Polytechnic University, Project A/C Code: G-U272.
References
[1] Braun JE, Klein SA, Mitchell JW, Beckman WA. Applications of optimal control to chilled water systems without storage. ASHRAE
Trans 1989;95(1):66375.
[2] Braun JE, Diderrich GT. Near-optimal control of cooling towers for chilled water systems. ASHRAE Trans 1990;SL-90:313.
[3] Tassou SA, Al-Nizari HO. Eect of refrigerant ow control on the thermodynamic performances of reciprocating chillers. Appl
Energy 1993;45(2):10116.
[4] Austin SB. Optimum chiller loading. ASHRAE J 1991;33(7):403.
[5] Kaya A. Improving eciency in existing chillers with optimization technology. ASHRAE J 1991;33(10):308.
[6] Beyene A. Performance evaluation of conventional chiller systems. ASHRAE J 1995;37(6):3644.
[7] Roper MA. Energy ecient chiller control. Technical Note TN 16/2000. Bracknell: Building Services Research and Information
Association; 2000.
[8] Ahn BC, Mitchell JW. Optimal control development for chilled water plants using a quadratic representation. Energy Build
2001;33:3718.
[9] Avery G. Improving the eciency of chilled water plants. ASHRAE J 2001;43(5):148.
[10] Hartman T. All-variable speed centrifugal chiller plants. ASHRAE J 2001;43(9):4353.
[11] Briley GC. Increasing operating eciency. ASHRAE J 2003;45(5):73.
[12] Crowther H, Furlong J. Optimizing chillers and towers. ASHRAE J 2004;46(7):3440.
[13] Chan CWH. Optimizing chiller plant control logic. ASHRAE J 2006;48(7):3842.
[14] Bhattacharjee K. Reducing energy cost in an industrial chilled water plant. Energy Eng 2007;107(4):4251.
[15] Hydeman M, Zhou G. Optimizing chilled water plant control. ASHRAE J 2007;49(6):4454.
[16] Kairouani L, Hassairi M, Tarek Z. Performance of cooling tower in south of Tunisia. Build Environ 2004;39:3515.
[17] Qureshi B, Zubair S. Prediction of evaporation losses in wet cooling towers. Heat Transfer Eng 2006;27(9):8692.
[18] Hartman T. Designing ecient systems with the equal marginal performance principle. ASHRAE J 2005;47(7):6470.
[19] Bahneth WP, Peyer EB. Varying views on variable-primary ow. HeatPipingAir Cond Eng 2004;76(3):S59.
[20] Gordon JM, Ng KC, Chua HT, Lim CK. How varying condenser coolant ow rate aects chiller performance: thermodynamic
modeling and experimental conrmation. Appl Therm Eng 2000;20:114959.
[21] ASHRAE. ANSI/ASHRAE/IESNA 90.1-2004, Energy standard for buildings except low-rise residential buildings. Atlanta:
American Society of Heating, Refrigeration, and Air-Conditioning Engineers, Inc.; 2004.
[22] Browne MW, Bansal PK. Challenges in modeling vapour-compression liquid chillers. ASHRAE Trans 1998;104(1):47486.
[23] Bourdouxhe JP, Grodent M, Lebrun JJ. Reference guide for dynamic models of HVAC equipment. Atlanta, Georgia: American
Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc.; 1998.
[24] Hydeman M, Sreedharan P, Webb N, Blanc S. Development and testing of a reformulated regression-based electric chiller model.
ASHRAE Trans 2002;108(2), paper no. HI-02-18-2.
[25] DOE (Department of Energy). DOE 2 reference manual, Part 1, version 2.1, Berkeley. CA: Lawrence Berkeley National Laboratories;
1980.
[26] Bourdouxhe JP, Grodent M, Lebrun JJ. A toolkit for primary HVAC system energy calculations [computer program]. Atlanta:
American Society of Heating, Refrigerating and Air-Conditioning Engineers; 1995.
[27] Lu L, Cai W, Soh YC, Xie L, Li S. HVAC system optimizationcondenser water loop. Energy Convers Manage 2004;45:61330.
[28] Benton DJ, Hydeman M, Bowman CF, Miller P. An improved cooling tower algorithm for the CoolToolsTM simulation model.
ASHRAE Trans 2002;108(1), paper no. AC-02-9-4.
[29] Graves RD. Thermodynamic modeling and optimization of a screw compressor chiller and cooling tower system. Msc thesis, The
Oce of Graduate Studies of Texas A&M University; 2003.
[30] Solar Energy Laboratory. TRNSYS 15: A transient system simulation program (reference manual). Madison, WI: University of
WisconsinMadison Press; 2000.

950

F.W. Yu, K.T. Chan / Applied Energy 85 (2008) 931950

[31] Wang SW. Dynamic simulation of a central chilling system and evaluation of EMCS on-line control strategies. Build Environ
1998;33(1):120.
[32] Browne MW, Bansal PK. Steady-state model of centrifugal liquid chillers. Int J Refrig 1998;21(5):34358.
[33] Braun JE. Methodologies for the design and control of chilled water systems. PhD thesis. Madison, WI: University of Wisconsin
Madison; 1988.
[34] Air-conditioning & Refrigeration Institute. Standard 550/590: performance rating of water chilling packages using the vapour
compression cycle. Arlington, Virginia; 2003.
[35] Yu FW, Chan KT. An alternative approach for the performance rating of air-cooled chillers used in air-conditioned buildings. Build
Environ 2006;41:172330.