International Journal of Thermal Sciences 86 (2014) 284e291

Contents lists available at ScienceDirect

International Journal of Thermal Sciences

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

Experimental study regarding the evolution of temperature proles

inside wet cooling tower under crosswind conditions
Ming Gao*, Feng-zhong Sun, A. Turan
a
b

School of Energy and Power Engineering, Shandong University, Jinan 250061, China
School of Mechanical, Aerospace and Civil Engineering, University of Manchester, P.O. Box 88, Manchester M60 1QD, UK

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 10 February 2014
Received in revised form
18 July 2014
Accepted 18 July 2014
Available online 20 August 2014

Based on similarity theory, this research details a thermal-state model experiment, concerning the
evolution of the air/water temperature proles inside a Natural Draft Wet Cooling Tower (NDWCT) under
windless and crosswind conditions. Prior studies have shown that the air/water temperature distribution
is fairly uniform and stable under windless (stagnant) conditions, but the uniformity is destroyed in the
presence of windy conditions, and the air/water temperature of different points displays a large variation
subject to the same crosswind velocity. Generally speaking, the highest air/water temperature values
inside the whole tower lie on the windward and leeward direction, but the highest air temperature at the
tower outlet appears near the leeward side zone, rather than exactly on the leeward side. Based on this
research, the air/water temperature proles regarding measurement of values can be obtained accurately
under windless and crosswind conditions, a fact that can help conrm the specic location of vortex on
the windward and leeward side. All of above ndings can provide an important theoretical foundation
concerning further research, specically for energy-saving aspects NDWCTs.
2014 Elsevier Masson SAS. All rights reserved.

Keywords:
Wet cooling tower
Thermal-state model experiment
Evolution of temperature proles
Crosswind

1. Introduction
As a primary component of the cool-end system in thermal
power plants including some nuclear power plants, the Natural
Draft Wet Cooling Towers (NDWCTs) play an important role to cool
the circulating water from the condenser, and its efciency has a
great impact on the total cycle efciency of power plants. Therefore,
it is extremely important and necessary to study the heat and mass
transfer performance under crosswind conditions both from an
academic as well as an industrial point of view.
Many studies in the past had concentrated on the thermal
performance of NDWCTs from several aspects under windless
conditions, including mathematical model development [1e4],
exergy analysis [5,6], the thermal performance of the rain zone and
lling [7,11], water distribution system [12], application of articial
neural network technology [13], and so on. Although some of the
fundamental aspects of the thermouid phenomena were derived
from such studies, the heat and mass transfer performance of the
cooling tower, including the natural draft dry-cooling towers
(NDDCTs) and the natural draft wet cooling towers (NDWCTs), is

* Corresponding author.
E-mail address: gm@sdu.edu.cn (M. Gao).
http://dx.doi.org/10.1016/j.ijthermalsci.2014.07.010
1290-0729/ 2014 Elsevier Masson SAS. All rights reserved.

actually inuenced by environmental conditions, especially environmental crosswind [14e17], and such processes have not been
studied extensively. Hence, the thermal performance of a generic
cooling tower under crosswind conditions is worth researching,
providing useful outcome both at an academic and applied level.
In the literature several studies can be found related to heat and
mass transfer performance of NDDCTs under crosswind conditions.
A. Jahangiri et al. [18] pointed out that environmental conditions
strictly affect Heller air-cooling tower's performance; thus it was
proposed to inject steam generator's ue gas into the tower for
better air suction. Results showed that the ue gas injection would
help improve the performance of the cooling tower to some extent.
A. Hossein et al. [19] developed a 3-D model to investigate the effect
of water spray and crosswind on the effectiveness of the NDDCTs.
This computational uid dynamics (CFD) study discovered that
NDDCTs operated most effectively at the crosswind velocity of 3 m/
s, and as the wind velocity continues to rise to more than 3 m/s up
to 12 m/s, the tower efciency would decrease by approximately
18%, based on windless condition. One CFD model had been carried
out by Y. Lu et al. [20] to numerically analyze the heat transfer
performance of a 15 m-high small NDDCT under different crosswind velocity. Simulations showed that, at certain crosswind velocity, the crosswind signicantly degrades the cooling
performance. However, the negative effect of the crosswind could

M. Gao et al. / International Journal of Thermal Sciences 86 (2014) 284e291

be turned into advantage by introducing windbreak walls. Likewise, R. Al-Waked and M. Behnia [21e22] also established the CFD
model to study the performance of NDDCTs, and put forward the
proposal to use windbreak walls to improve the thermal performance of NDDCTs. In addition, the effect of crosswind on an indirect
dry cooling tower with or without windbreaks had been studied by
H. Reshadatjoo et al. [23], the phenomena responsible for the
reduction of air intake ow rate (AIFR) under crosswind conditions
were identied, and results obtained had been compared with the
available experimental results.
Compared with the NDDCTs, the crosswind has less effect on
NDWCTs, but it still cannot be neglected during the actual operation. What's more, in the conventional thermal design of NDWCTs,
the impact of crosswind, which actually exists in most cases, has
not been adequately analyzed and paid appropriate attention.
T.J. Bender et al. [24e26] studied the inuence of crosswind on
the heat transfer performance of wet cooling towers by means of
wind tunnel experiment and numerical computation, but these
studies did not agree with the basic approaches of geometric similarity and dynamic similarity between the model and prototype
tower. Based on the comparison of the calculated results with
experimental data [27,28], A.I. Petrichik et al. [29] concluded that
the efciency of evaporative heat transfer in a cooling tower was
inuenced by the wind that may rise near the cooling tower.
R. Al-Waked and M. Behnia [30e32] developed a CFD model
for NDWCTs, and found that environmental crosswind affected
seriously the cooling efciency. The CFD studies revealed that
the circulating water temperature differences were found to be
less than 1 K for the whole span of crosswind velocities. M. Gao
et al. [33,34] also studied the effect of crosswind on the thermal
performance of NDWCTs by thermal-state model experiments
and articial neural networks technology, and arrived at some
similar conclusions. Y. L. Chen et al. [35] conducted experimental
studies on the cross walls effect on the thermal performance of
NDWCTs under crosswind conditions. Experimental researches
showed that, at all crosswind velocities, the cross wall at a
setting angle of 0
resulted in higher performance than that at
45
, regardless of the cross wall shapes. Moreover, the cross wall
at 45
degraded the thermal performance under high crosswind
velocity conditions. K. Wang et al. [36] had reported experimental research on the guiding channel effects regarding the
thermal performance of NDWCTs subject to crosswinds, and
found that although the guiding channels with 70
setting angle
led to better cooling performance, they may actually be caused
by more circulating water consumption. M. Gao et al. [37]
concluded that the unsymmetrical circumferential inow air
and vortex under crosswind conditions destroyed the air dynamic eld inside cooling tower, affecting seriously the whole
airow rate, which in turn deteriorated the heat and mass
transfer performance.
As can be seen from the above brief review, previous studies
demonstrated that environmental crosswind leads to adverse effects on the thermal performance of NDWCTs, and in fact prior
researches revealed the inuence mechanism, circumferential
inow air proles, vortex distribution and prevention measures,
attributes that are more valuable to NDWCT research; however,
previous studies did not analyze the vortex's specic location and
inuence range, and also failed to discuss the air/water temperature evolution patterns inside the whole wet cooling tower, aspects
that are crucial to the further energy-saving research.
Therefore, in this paper studies are conducted regarding the
development of the air/water temperature elds inside the wet
cooling tower under crosswind conditions via basic thermal-state
model experiments to reveal the relevant fundamental heat and
mass transfer processes.

285

2. Experimental study design

2.1. Experimental objectives
The vortices inside NDWCTs under crosswind conditions deteriorate the heat and mass transfer performance of lling and
raining zone [37], and hence give rise to substantial temperature
variations at different points inside tower. But the specic temperature values at different points had not so far been considered
and/or discussed in previous experimental studies. This is a stumbling block for the development of experimental research for
NDWCTs.
For that reason, an appropriately designed thermal-state model
experiment regarding the development of air/water temperature
proles is conducted in this paper to conrm the temperature
values at different points inside a generic NDWCT. During this
experiment, the air/water temperature at different points is
monitored in real-time by a suitable data collection system and
measured by copper-constantan thermocouples under different
operating conditions. The conrmation of air/water temperature
patterns can lay the foundation on researching for accurately accounting for the inuence of the crosswind with a view to
enhancing heat and mass transfer performance.
2.2. Experimental setup
The experimental setup is displayed via the schematic diagram
depicted in Fig. 1. The model tower adopted in this experiment is
designed to simulate a typical NDWCT in large-scale power plants
in terms of the engineering similarity theory [15,33,34,37]. i.e., the
relevant scale measure of model to prototype tower is 1:100. In
addition, the details of the primary measurement setup and instruments are listed in Table 1.
The whole experimental activity hence simulates the actual
working process of a typical NDWCT in current thermal power
plants, and the crosswind velocity is controlled by frequency conversion fans, including the upper and lower fan shown in Fig. 1.
Prior to the start up of the experiments, the circulating water is
heated up to required temperature by several electric heaters, and
then transported to the upper tank by the circulating water pump.
During the experiments, the circulating water enters the model
tower and goes through the llings from top to bottom, while the
dry air ows through the llings from bottom to top. The heat and
mass transfer are conducted in the presence of cross ow.
2.3. Similarity criterion
The dimensions of the model cooling tower is given as
37 cm  68 cm  85 cm (top outlet diameter  bottom
diameter  height), and the height of the tower inlet is 50 mm. The
test process also complies with the dynamic similarity, kinematic
similarity and thermodynamic similarity besides the geometric
(scale) similarity, including the Froude number and wind velocity
scale.
Under actual operating conditions, it is difcult to implement Re
and Fr number similarity simultaneously. The velocity in Re number
varies inversely with the model scale in terms of Eq. (1) and (2),
while the velocity in Fr number varies directly with the square root
of model scale in terms of Eq. (3). Therefore, the Re and Fr number
similarity cannot be satised simultaneously in one model experiment. In this thermal model experiment, the driving force of
buoyancy and the inertial force of crosswind are the main factors to
be considered, while the viscous force is less important. Therefore,
the density Fr number similarity, which is dened by Eq. (3), should
have the priority to be satised over the Re number [38].

286

M. Gao et al. / International Journal of Thermal Sciences 86 (2014) 284e291

Fig. 1. Schematic diagram of experimental cooling tower.

Reout

Vout Dout
y

(1)

Retop

Vtop Dout
y

(2)

"
Fr Vout

"
,s#
,s#
Dr
Dr
gL Vout
gL
rout
rout
P

(3)

where subscript P represents prototype tower and subscript M

denotes model tower, rout is the density of the outlet air, and Dr is
the density difference between the inlet and outlet air, and L is the
characteristic size. According to the Fr similarity (as can be seen in
Eq. (3)), the ratio of experimental velocity to actual velocity is 1:10.
Besides the density Fr, another very important parameter is the
ratio of wind velocity, especially for the case of windy conditions.
The wind velocity scale between the model and prototype tower
must also be equal according to kinematic similarity, and is given as
[15,35e38],

vout
vtop

vout
vtop


(4)
M

where vout is the top outlet wind velocity and vtop is the top level
wind velocity of model tower.

shown in Fig. 2, and the detailed drawing of thermocouple layout at

any layer is shown in Fig. 3 which is the planform of Fig. 2.
In Fig. 2, there are four layers of thermocouples which lie on the
surface of water basin (the rst layer), above the water distribution
system (the second layer), in the middle of tower (the third later)
and at the tower outlet (the fourth layer), respectively. Every layer
includes three laps and there are twelve measuring points in every
lap, so there are altogether 37 measuring points in every layer
including one measuring point in the center. In total, there are 148
measuring points (37 measuring points per layer  4 layers) in the
whole model tower.
The intersection points of any two lines are the measuring
points depicted as black points in Fig. 3. In the following discussion,
No. 1 represents the measurement point for the windward side in
any lap, then the measuring points are successively ordered as 2, 3,
4, 511, 12 in a clockwise manner. Furthermore, No. 7 indicates
the measurement point for the leeward side. In addition, the
outermost lap is called the rst lap, the innermost lap is known as
the third lap, and the remaining is the second lap.
2.5. Experimental operating conditions
In this study, the inlet wind velocity at the windward side of
tower produced by the lower fan, is 0 m/s for the windless state,
while 0.1 m/s, 0.2 m/s, 0.3 m/s, 0.4 m/s, 0.5 m/s, 0.6 m/s, 0.8 m/s and

2.4. Thermocouple layout inside the model tower

In order to study the development of air/water temperature
proles, there are a series of thermocouples placed inside the
model tower. The schematic diagram of thermocouple layout is
Table 1
Monitored parameters and measurement instruments.
Items

Measuring instruments

Accuracy

Atmospheric pressure
Crosswind velocity
Inlet dry and wet bulb temperature
Outlet air temperature

Hot-wire manometer (KA31)

Hot-wire anemoscope (KA31)
Psychrometer
Copper-constantan
thermocouple
Mercury thermometer
Hygrometer (HI8564)
Rotameter

3%
0.1 m/s
0.1
C
0.3
C

Inlet and outlet water temperature

Air humidity
Circulating water ow rate

0.1
C
2%
1.5%

Fig. 2. Schematic diagram of thermocouple layout inside model cooling tower.

M. Gao et al. / International Journal of Thermal Sciences 86 (2014) 284e291

287

Fig. 3. Detail drawing of thermocouple layout in any layer (planform of Fig. 2).

1.0 m/s represent crosswind values. As given by the natural wind

velocity distribution proles above ground (Seen in Eq. (5)) and the
vertical distance between the two fans in this experiment (seen in
Fig. 1), the top level wind velocity which is produced by the upper
fan, is about 0 m/s for the windless state, while 0.2 m/s, 0.4 m/s,
0.6 m/s, 0.8 m/s, 1.0 m/s, 1.2 m/s, 1.6 m/s and 2.0 m/s represent
crosswind conditions respectively.

vz
vz;ref

!0:2
(5)

zref

Here, vz represents the environmental crosswind velocity at zhigh position, and vz,ref represents the reference wind velocity at
the reference high (zref) position. Another thing to note is that the
whole thermal state model experiment complies with the similarity criterions of wind velocity scale based on the Eq. (4) and Eq.
(5).
Regarding the large number of inherent variables considered
and the long experimental durations required for the study, the
practice of design of orthogonal experiment is both a requirement
as well as an effective tool to arrive at an engineering interpretation
of results. In the light of research requirements, there are a total of
270 experimental operating conditions, are listed in Table 2.
3. Aspects of the development of temperature proles inside
the tower under windless (stagnant) conditions

Fig. 4. Water temperature proles of the rst layer under windless conditions
(t1 55
C; Q 0.36 m3/h).

measurement points is lower than that of the third lap measuring

points, indicating that the heat and mass transfer performance in
the second lap zone is stronger than that in the third lap zone. Dry
air absorbs heat and moisture continuously in the course of
crossing the cooling tower. When air arrives near the center of
tower, both air temperature and humidity rise higher, so the performance of heat and moisture absorption decreases, leading to
degraded heat and mass transfer performance. Therefore, the water
temperature of the third lap measuring points is higher due to the
higher air temperature in this zone.
It also can be observed in Figs. 4 and 5 that the water temperature of the rst lap measuring points is not the lowest, the reason
being that there exist small vortices next to the inner wall of cooling
tower under windless conditions according to research by Gao et al.
[37]. These vortices deteriorate the heat and mass transfer performance in the rst lap zone, a fact that leads to the observed higher
water temperature of the rst lap zone.

3.1. Development of water temperature on the surface of water

basin under windless conditions
Figs. 4 and 5 depict the water temperature proles of the rst
layer measurement points under windless conditions when the
circulating water inlet temperature is 55
C, and the ow rate is
0.36 m3/h and 0.60 m3/h, respectively.
On the basis of Figs. 4 and 5, the water temperature has a
relatively small change at the same lap under windless conditions.
Furthermore, the prevailing water temperature for the second lap

Table 2
Experimental operating conditions.
Items

Symbol Unit

Circulating water ow rate

Circulating water inlet
temperature
Crosswind velocity
(from the lower fan)
Crosswind velocity
(from the upper fan)

Q
t1

m3/h 0.12, 0.24, 0.36, 0.48, 0.60, 0.72

C
40, 45, 50, 55, 60

Value

v0

m/s

0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.8, 1.0

vtop

m/s

0, 0.2, 0.4, 0.6, 0.8, 1.0, 1.2, 1.6, 2.0

Fig. 5. Water temperature proles of the rst layer under windless conditions
(t1 55
C; Q 0.60 m3/h).

288

M. Gao et al. / International Journal of Thermal Sciences 86 (2014) 284e291

circulating water ow rate increases, the water-spraying density

rises, and the water distribution is well-proportioned, leading to
uniform and stable heat and mass transfer performance for every
section of the lling. Consequently, the air temperature changes
very little under the larger circulating water ow rate.
Fig. 7 illustrates the air temperature proles for all the measurement points in the second layer when the circulating water
inlet temperature and ow rate are 55
C and 0.72 m3/h, respectively. It is shown in Fig. 7 that the air temperature of the second lap
measurement points is lower than that of the third lap. The reason
is the same as the analysis in Part 3.1. In fact, the air temperature
development patterns expressed in Fig. 7 are almost the same as
those in Fig. 6.
According to the above analysis, as the circumferential inow air
is well-proportioned under windless conditions, the water temperature (the rst layer measurement points) and air temperature
(the second layer measurement points) display a relatively uniform
relationship, so the corresponding heat and mass transfer performance in different parts of the volume is uniform under windless
conditions.

Fig. 6. Air temperature proles of the rst lap in the second layer under windless
conditions (t1 55
C, Q1, Q2, Q3, Q4 0.36, 0.48, 0.60, 0.72 m3/h).

3.2. Air temperature development above the llings under windless

conditions
Figs. 6 and 7 display the typical proles of air temperature under
windless conditions when the circulating water inlet temperature
is 55
C, the corresponding water ow rate Q1, Q2, Q3 and Q4 are
0.36 m3/h, 0.48 m3/h, 0.60 m3/h and 0.72 m3/h in Fig. 6, and that is
0.72 m3/h in Fig. 7.
It can be seen in Fig. 6 that the air temperature of the same lap
displays no observable change at the same circulating water ow
rate and the inlet water temperature. The circumferential inow air
is well-proportioned under windless conditions, so the air dynamic
eld inside tower is uniform and stable, leading to small/negligible
temperature variations. What is more, the larger the circulating
water ow rate is, the less the air temperature changes. As the

Fig. 7. Air temperature proles of the second layer under windless conditions
(t1 55
C; Q 0.72 m3/h).

4. On the development of the temperature eld under

crosswind conditions
On the basis of the studies by Gao et al. [37], there exist a
number of vortices inside tower due to the unsymmetrical
circumferential inow air under crosswind conditions. What is
more, the temperature will develop to be more nonuniform as the
crosswind velocity increases in magnitude. In this section, the
specic temperature development processes inside the cooling
tower, including air and water temperature, are studied under
different crosswind conditions.
It is reported that the vortex did not appear in the center of
cooling tower [37], implying that the vortex has a little inuence on
the air/water temperature of center zone, so the prevailing air/
water temperature development transport processes for the second
lap is primarily discussed under crosswind conditions in this paper.
4.1. Development of water temperature on the surface of water
basin under crosswind conditions
Fig. 8 presents the water temperature proles of the second lap
in the rst layer measurement points under different crosswind
velocities, when the circulating water inlet temperature and ow
rate are 55
C and 0.60 m3/h, respectively.
In Fig. 8, The water temperature at any point can be observed
under different crosswind velocity, it also can be concluded in Fig. 8
that the basic change tendency of water temperature is almost the
same under all of the investigated velocity values. Certainly, for the
case of 0.2 m/s, there are tiny differences compared to other
investigated speeds, which might have potentially resulted from
the experimental errors based on the effort and data processing. In
addition, The No. 7 measuring point stands almost as a symmetrical point for all the other measurement points, and the water
temperature of No. 1 and No. 7 points is higher than that of other
points at the same crosswind velocity. That effectively implies that
the inuence region of crosswind lies on near the second lap zone
under crosswind conditions, i.e., the vortices on the windward and
leeward side are located on the second lap zone, leading to the
unsymmetrical water temperature distribution.
In addition, the water temperature increases gradually when the
crosswind velocity changes from 0.8 m/s to 0.4 m/s, it reaches a
maximum at 0.4 m/s wind velocity conditions according to Fig. 8, so
the 0.4 m/s wind velocity is a turning point. The water temperature
by inference represents the heat and mass transfer performance, so

M. Gao et al. / International Journal of Thermal Sciences 86 (2014) 284e291

Fig. 8. Water temperature proles of the second lap in the rst layer under crosswind
conditions (t1 55
C; Q 0.60 m3/h).

the heat and mass transfer performance enhances gradually as the

crosswind velocity increases when the crosswind velocity is more
than 0.4 m/s.

289

the driving force of moist air, thus the moist air concentrates
around the vortices in the lling zone, which leads to the higher
temperatures of the No. 1 and No. 7 measuring points. Obviously,
this fact proves once again that the inuence region of vortices lies
on the second lap zone, resulting in very unsymmetrical air temperature distribution in the second lap.
Likewise, the air temperature decreases gradually when the
crosswind velocity changes from 0.8 m/s to 0.4 m/s, and it reaches a
minimum under 0.4 m/s wind velocity conditions according to
Fig. 9. The lower the air temperature is, the weaker the heat exchange is. So this also explains that the heat and mass transfer
performance enhances gradually as the crosswind velocity increases when the crosswind velocity is more than 0.4 m/s.
On the basis of above analysis, the water temperature evolution
of the rst layer and the air temperature distribution of the second
layer are not uniform and stable under crosswind conditions, and
the air/water temperature of twelve measuring points in the same
lap is changed substantially due to the variation of the crosswind
velocity, especially around the lling zone (the second layer).
Experimental research has also revealed that the vortices on the
windward and leeward side appear near the second lap zone, and
the air/water temperature of any measurement points can be obtained reliably under crosswind conditions. These conclusions can
lay the important theoretical foundation for further research
regarding energy-saving considerations for NDWCTs.
4.3. Air temperature prole development at the tower outlet under
crosswind conditions

The air temperature proles for the second lap in the second
layer under crosswind conditions are shown in Fig. 9. In this case,
the circulating water inlet temperature and ow rate are 55
C and
0.60 m3/h, respectively.
In Fig. 9, the higher air temperature appears in the No. 1 and No.
7 measurement points at the same crosswind velocity, and the
other points have a symmetric distribution centered on the No. 7.
Under crosswind conditions, there are two vortices on the windward side and leeward side. The appearance of the vortex reduces

Figs. 10e12 explain the air temperature prole development for

all the measurement points in the fourth layer under crosswind
conditions when the circulating water inlet temperature is 60
C
and ow rate is 0.12 m3/h, respectively. In addition, the crosswind
velocity is 0.2 m/s, 0.4 m/s and 0.8 m/s in Figs. 10e12.
It can be seen in Fig. 10 that the air temperature uctuation
range of the rst and third lap measuring points is relatively less
than that of the second lap. This aspect points to the fact that the
crosswind has great inuence on the second lap measuring points,
that is, the vortex at the tower outlet lies near the second lap zone.
In addition, it also can be observed in Fig. 10 that the highest
temperature of any lap appears near the No. 6 measuring point,

Fig. 9. Air temperature proles of the second lap in the second layer under crosswind
conditions (t1 55
C; Q 0.60 m3/h).

Fig. 10. Air temperature proles of the fourth layer (v0 0.2 m/s) (t1 60
C;
Q 0.12 m3/h).

4.2. Air temperature proles above the lling under crosswind

conditions

290

M. Gao et al. / International Journal of Thermal Sciences 86 (2014) 284e291

5. Conclusions

Fig. 11. Air temperature proles of the fourth layer (v0 0.4 m/s) (t1 60
C;
Q 0.12 m3/h).

(1) The water temperature on the surface of water basin and air
temperature above the water distribution system evolve
relatively symmetrically under windless conditions, so the
corresponding heat and mass transfer performance for the
volume is uniform and stable at different points.
(2) On the contrary, the water temperature on the surface of
water basin and air temperature above the water distribution
system are no longer uniform and stable under crosswind
conditions, and the air/water temperature of different points
in the same lap is changed a great deal under the same
crosswind velocity, especially the lling zone.
(3) At the tower outlet, the highest temperature of any lap appears near the leeward side zone, rather than exactly on the
leeward side, implying that the vortex on the tower outlet is
not located exactly on the leeward side, but displaced.
(4) Experimental research in this paper has revealed that the
vortices on the windward and leeward side appear near the
second lap zone. In addition, the air/water temperature of
any measurement point can be obtained under crosswind
conditions based on the results reported in this paper. All of
above conclusions can lay an important theoretical foundation for further research concerned with energy-saving for
NDWCTs.
Acknowledgment
This paper is supported by National Natural Science Foundation
of China (No. 51106092), Youth foundation of Shandong Natural
Science Foundation of China (No. ZR2011EEQ018) and Ji'nan University Institute Innovation Plan (No. 201303077).
References

Fig. 12. Air temperature proles of the third layer (v0 0.8 m/s) (t1 60
Q 0.12 m3/h).

C;

rather than the No. 7 measuring point. On the one hand, there is a
ow vortex near the leeward side zone due to the top level crosswind inuence, so the hot-air cannot timely drop away, leading to
the higher air temperature observed near the leeward side zone. On
the other hand, the combined effect of both crosswind on the tower
top and the driving force (hot-air buoyancy) inside cooling tower
deects the outlet airow. Therefore, the vortex at the tower outlet
is not located exactly on the leeward side under crosswind conditions, but displaced. The development of air temperature proles
shown in Figs. 11 and 12 are almost the same as those in Fig. 10.
Figs. 10e12 show that the air temperature is lowest when the
crosswind velocity is 0.4 m/s, so the heat and mass transfer performance is substantially degraded when the crosswind velocity is
0.4 m/s.
According to the analysis for the fourth layer measurement
points, one can build up the evolution patterns for the air temperature at different points and reveal the vortex location at the
tower outlet, a fact that is substantially benecial for future
research.

[1] S.P. Fisenko, A.I. Petruchik, A.D. Solodukhin, Evaporative cooling of water in a
natural draft cooling tower, Int. J. Heat Mass Transf. 45 (23) (2002)
4683e4694.
[2] N. Williamson, S. Armeld, M. Behnia, Numerical simulation of ow in a
natural draft wet cooling tower-the effect of radial thermouid elds, Appl.
Therm. Eng. 28 (2e3) (2008) 178e189.
ger, A critical investigation into the heat and mass
[3] J.C. Kloppers, D.G. Kro
transfer analysis of counterow wet-cooling towers, Int. J. Heat Mass Transf.
48 (3e4) (2005) 765e777.
[4] W. Asvapoositkul, S. Treeutok, A simplied method on thermal performance
capacity evaluation of counter ow cooling tower, Appl. Therm. Eng. 38
(2012) 160e167.
[5] L. Wang, N.P. Li, Exergy transfer and parametric study of counter ow wet
cooling towers, Appl. Therm. Eng. 31 (2011) 954e960.
[6] T. Muangnoi, W. Asvapoositkul, S. Wongwises, An exergy analysis on the
performance of a counter ow wet cooling tower, Appl. Therm. Eng. 27 (5e6)
(2007) 910e917.
ger, Drop size distribution below
[7] R. Terblanche, H.C.R. Reuter, D.G. Kro
different wet-cooling tower lls, Appl. Therm. Eng. 29 (2009) 552e560.
[8] Bilal A. Qureshi, Syed M. Zubair, A complete model of wet cooling towers with
fouling in lls, Appl. Therm. Eng. 26 (2006) 1982e1989.
[9] P.J. Grobbelaar, H.C.R. Reuter, T.P. Bertrand, Performance characteristics of a
trickle ll in cross-and counter-ow conguration in a wet-cooling tower,
Appl. Therm. Eng. 50 (2013) 475e484.
[10] M. Lemouari, M. Boumaza, Experimental investigation of the performance
characteristics of a counter-ow wet cooling tower, Int. J. Therm. Sci. 49
(2010) 2049e2056.
[11] J.C. Kloppers, D.G. Kroger, Loss coefcient correlation for wet-cooling tower
lls, Appl. Therm. Eng. 23 (2003) 2201e2211.
[12] J. Smrekar, J. Oman, B. Sirok, Improving the efciency of natural draft cooling
towers, Energy Convers. Manag. 47 (9e10) (2006) 1086e1100.
[13] M. Hosoz, H.M. Ertunc, H. Bulgurcu, Performance prediction of a cooling tower
using articial neural network, Energy Convers. Manag. 48 (4) (2007)
1349e1359.
[14] A.F. Preez, The Inuence of Cross-wind on the Performance of Natural Draft
Dry-cooling Towers (Dissertation of PhD), University of Stellenbosch, 1992.
[15] M. Gao, F. Zh. Sun, K. Wang, et al., Experimental research of heat transfer
performance on natural draft counter ow wet cooling tower under crosswind conditions, Int. J. Therm. Sci. 47 (2008) 935e941.

M. Gao et al. / International Journal of Thermal Sciences 86 (2014) 284e291

[16] E. Ding (Ed.), Air Cooling Techniques in Power Plants, Water and Electric
Power Press, Beijing, 1992.
[17] A. Pertuchik, S. Fisenko, N. Stolovich, A. Solodukhin, Express evaluation of the
performance of natural draft cooling tower, in: 16# IAHR Cooling Tower and
Air-cooled Heat Exchanger Conference, Belarus, 2013, p. 10.
[18] A. Jahangiri, A. Golnesham, Feasibility study of the thermal performance
improvement of natural draft dry cooling towers due to ue gas injection,
World Appl. Sci. 12 (2011) 568e575.
[19] A. Hossein, S. Mohsen, G. Ehsan, Simultaneous effects of water spray and
crosswind on performance of natural draft dry cooling tower, Therm. Sci. 17
(2013) 443e449.
[20] Y. Lu, Z. Guan, H. Gurgenci, Z. Zou, Windbreak walls reverse the negative effect
of crosswind in short natural draft dry cooling towers into a performance
enhancement, Int. J. Heat Mass Transf. 63 (2013) 162e170.
[21] R. Al-Waked, M. Behnia, The performance of natural draft dry cooling towers
under crosswind: CFD study, Int. J. Energy Res. 28 (2) (2004) 147e161.
[22] R. Al-Waked, M. Behnia, The effect of windbreak walls on the thermal performance of natural draft dry cooling towers, Heat Transf. Eng. 26 (8) (2005)
50e62.
[23] H. Reshadatjoo, N. Pormahmod, S.Y. Moltagh, The effect of cross-winds on an
indirect dry cooling tower with or without windbreaks, J. Power Energy 225
(2011) 635e646.
[24] D.D. Derksen, T.J. Bender, D.J. Bergstrom, K.S. Rezkallah, A study on the effects
of wind on the air intake ow rate of a cooling tower: part 1. Wind tunnel
study, J. Wind Eng. Ind. Aerod. 64 (1996) 47e59.
[25] T.J. Bender, D.J. Bergstrom, K.S. Rezkallah, A study on the effects of wind on the
air intake ow rate of a cooling tower: part 2. Wind all study, J. Wind Eng. Ind.
Aerod. 64 (1996) 61e72.
[26] T.J. Bender, D.J. Bergstrom, K.S. Rezkallah, A study on the effects of wind on the
air intake ow rate of a cooling tower: part 3. Numerical study, J. Wind Eng.
Ind. Aerod. 64 (1996) 73e88.

291

[27] A.I. Petruchik, A.D. Solodukhin, S.P. Fisenko, Simulation of cooling of water
droplets and lm ows in large natural wet cooling towers, J. Eng. Phys.
Thermophys. 74 (1) (2001) 62e68.
[28] V.V. Antonik, A.I. Petruchik, A.D. Solodukhin, N.N. Stolovich, S.P. Fisenko, Proc.
IV Minsk Int. Forum 5 (2000) 314e324 (in Russian).
[29] A.I. Petruchik, A.D. Solodukhin, N.N. Stolovich, S.P. Fisenko, Toward the
analysis of experimental data on thermal efciency of evaporative cooling
tower, Appl. Energy Russ. J. Fuel Power Heat Syst. 37 (6) (2000) 142e149.
[30] R. Al-Waked, M. Behnia, CFD simulation of wet cooling towers, Appl. Therm.
Eng. 26 (4) (2006) 382e395.
[31] R. Al-Waked, M. Behnia, Enhancing performance of wet cooling towers, Energy Convers. Manag. 48 (10) (2007) 2638e2648.
[32] R. Al-Waked, Crosswinds effect on the performance of natural draft wet
cooling towers, Int. J. Therm. Sci. 49 (1) (2010) 218e224.
[33] M. Gao, F.Zh. Sun, Sh. J. Zhou, et al., Performance prediction of wet cooling
tower using articial neural network under cross-wind conditions, Int. J.
Therm. Sci. 48 (2009) 583e589.
[34] M. Gao, Y.T. Shi, N.N. Wang, et al., Articial neural network model research on
effects of cross-wind to performance parameters of wet cooling tower based
on level Froude number, Appl. Therm. Eng. 51 (2013) 1226e1234.
[35] Y.L. Chen, F.Zh. Sun, H.G. Wang, et al., Experimental research of the cross walls
effect on the thermal performance of wet cooling towers under crosswind
conditions, Appl. Therm. Eng. 31 (2011) 4007e4013.
[36] K. Wang, F.Zh Sun, Y.B. Zhao, et al., Experimental research of the guiding
channels effect on the thermal performance of wet cooling towers subjected
to crosswinds, Appl. Therm. Eng. 30 (2010) 533e538.
[37] M. Gao, F.Zh Sun, N.N. Wang, et al., Experimental research on circumferential
inow air and vortex distribution for wet cooling tower under crosswind
conditions, Appl. Therm. Eng. 64 (2014) 93e100.
[38] Zh. G. Zhao, Cooling Tower, China Water Conservancy and Hydropower Press,
Beijing, 1997 (in Chinese).