Sensitivity and Optimization Analysis of Design Wind

Angle for Air-Cooled Power Plant

Weifeng He, Ph.D. 1; Dong Han 2; Chen Yue, Ph.D. 3; and Wenhao Pu 4
Downloaded from ascelibrary.org by University of California, San Diego on 03/03/16. Copyright ASCE. For personal use only; all rights reserved.

Abstract: The layout of an air-cooled steam condenser (ACSC) has a great impact on a power plant’s operation. This is owed to the fact that
the performance of the condenser and subsequently of the whole power plant is strongly affected by meteorological conditions. Hence, it is of
great significance to determine the optimal design of wind angle that gives a higher performance for an air-cooled power plant. This study
applies an existing numerical model in order to investigate the wind angle’s impact on the performance of an air-cooled steam condenser, as
well as on the whole power plant. Moreover, the optimal design of wind angle is obtained based on the corresponding sensitivity analysis.
Finally, the benefit to the power plant’s performance at the optimal-design wind angle is calculated through the comparison with the actual
design values at a wind angle of β ¼ 22.5°. The obtained results show that the net power and thermal efficiency can be improved by am-
plitudes of 1.2 and 0.37%, respectively, at the wind angle of β ¼ 90°, which should be proposed as the design wind angle for air-cooled power
plants. DOI: 10.1061/(ASCE)EY.1943-7897.0000347. © 2016 American Society of Civil Engineers.
Author keywords: Air-cooled steam condenser; Design wind angle; Numerical model; Net power; Thermal efficiency.

Introduction arising from the types for the heat exchangers and arrangement
of the power plants were evaluated. Van Rooyen and Kroger
Air-cooled steam condensers have become widely popular to reject (2008) found the phenomenon of flow distortions and the corre-
waste heat from the turbine exhaust in arid regions (Moore et al. sponding low-pressure region under the windward fans, which con-
2014; Yang et al. 2014). Turbine exhaust condenses in the finned tributed to the performance reduction of the air-cooled steam
tube exchangers while ambient air flow outside takes away the rel- condenser as the wind speed increases. Mathematical models for
evant latent heat. In view of the physical properties of air fluid, it is flow and heat transfer of the air-side fluid in air-cooled condensers
impossible to absorb the tremendous heat load from the turbine ex- were established by Yang et al. (2011a, b), and the boundary con-
haust by natural convection in large-scale power plants, and forced- ditions by radiator type was first introduced to describe the finned
draft fans must be utilized to induce enough air to cool the turbine tube bundles. The volume flow rates as well as the inlet air temper-
exhaust. However, the power consumption of such equipped axial ature and heat rejection for different air-cooled steam condensers
fans should be considered in the forced-draft thermal system. were obtained. Hu et al. (2014) proposed a reduced-order model
Due to the influence from the inflow and outflow situations of based on proper orthogonal decomposition (POD) to rapidly and
the fans under uncertain wind conditions, the characteristics of the accurately predict the flow and heat-transfer characteristics of an
air-cooled steam condenser will be altered (Thiart 1990). As a re- air-cooled steam condenser under the influence of natural wind,
sult, the performance of the power plant will deteriorate during ad- and the velocity and temperature fields of four typical test cases
verse wind conditions, such as strong wind and high-temperature were estimated with a maximum relative errors of 1.75 and 0.19%,
conditions. It is significant to investigate the influence principles respectively, while the degree of freedom was reduced by 105 com-
from the wind conditions on the air-cooled steam condenser as well pared to calculations using computational fluid dynamics. User-
as the whole power plant. An air-cooled steam condenser was first defined functions based on steam properties were utilized to
numerically modeled by Van Staden (1995), and the impacts from simulate the condensation of the turbine exhaust in the finned tube
ambient conditions were investigated. Performance differences heat exchangers (He et al. 2013; He and Dai 2012), and a numerical
1
model close to the true status was established. Gao et al. (2009)
Nanjing Univ. of Aeronautics and Astronautics, Jiangsu Province simulated the effect from the wind conditions as well as the height
Key Laboratory of Aerospace Power System, No. 29 Yudao St., Qinhuai
of the air-cooled steam condenser on the relevant condenser per-
District, Nanjing 210016, China (corresponding author). E-mail: wfhe@
nuaa.edu.cn; [email protected] formance, modeling the simplified model with a cubic fluid zone
2
Professor, Nanjing Univ. of Aeronautics and Astronautics, Jiangsu regarded as the condenser cell. It was found that the condenser per-
Province Key Laboratory of Aerospace Power System, No. 29 Yudao St., formance decreases with an increase of wind speed and increases as
Qinhuai District, Nanjing 210016, China. E-mail: [email protected] the platform height is elevated. Owen (1992) developed an efficient
3
Nanjing Univ. of Aeronautics and Astronautics, Jiangsu Province Key and reliable method to evaluate the performance of an air-cooled
Laboratory of Aerospace Power System, No. 29 Yudao St., Qinhuai steam condenser under windy conditions, and the pressure jump
District, Nanjing 210016, China. E-mail: [email protected] fan model was attempted to minimize the computational expense
4
Professor, Nanjing Univ. of Aeronautics and Astronautics, Jiangsu of the performance evaluation. The accuracy of the numerical
Province Key Laboratory of Aerospace Power System, No. 29 Yudao St.,
model was validated through the test data obtained during the
Qinhuai District, Nanjing 210016, China. E-mail: [email protected]
Note. This manuscript was submitted on June 25, 2015; approved on
full-scale tests in an air-cooled steam condenser.
November 24, 2015; published online on February 18, 2016. Discussion All the literatures mentioned earlier discusses the coupling
period open until July 18, 2016; separate discussions must be submitted relation between the wind conditions and performance of the
for individual papers. This paper is part of the Journal of Energy Engi- air-cooled steam condenser (ACSC). In addition, effect of the
neering, © ASCE, ISSN 0733-9402. configuration of the ACSC was also the research focus (Yang et al.

© ASCE 04016010-1 J. Energy Eng.

J. Energy Eng., 04016010

 2011a, b; Zhang et al. 2015). Zhang and Chen (2015) investigated
the influences of the windbreak mesh on the flow and heat-transfer
characteristics of an air-cooled steam condenser under windy con-
ditions. The windbreak mesh was arranged below the ACSC plat-
form and outside the ACSC steel supporting structure. It was found
that compared with grid and cross type configurations, the wind-
break mesh of a rectangle type was beneficial to protect the periph-
ery fans, avoiding suffering from ambient wind to the greatest
extent, allowing the best performance in general to appear. Gu et al.
(2015) completed a numerical study on the effect of roof windbreak
structures in an air-cooled system, and the results showed that the
vertical and inclined length, as well as inclination angle of
Downloaded from ascelibrary.org by University of California, San Diego on 03/03/16. Copyright ASCE. For personal use only; all rights reserved.

the roof’s windbreak line screen, will affect the performance

of the air-cooled system significantly. He et al. (2013) connected
the wind conditions and performance of the air-cooled power plant
through the predicted stable backpressures, and the length of the
inner walkway was optimized for a better performance of the
air-cooled thermal system.
Due to the demonstrated influences from the wind conditions,
the layout of the ACSC based on local meteorological conditions is Fig. 1. Schematic diagram of the 2 × 600 MW air-cooled power plant
actually significant to the corresponding performance of the whole
power plant. Thus, it is meaningful to obtain the characteristics of
the air-cooled power plant at different wind angles to provide
homologous references for the determination of the design wind ∂ðρϕÞ ∂ðρvx ϕÞ ∂ðρvy ϕÞ ∂ðρvz ϕÞ
angle, while quantitative analyses are scarce in the previous liter- þ þ þ
∂t ∂x ∂y ∂z
ature. In this paper, an existing numerical model (He et al. 2013) is



∂ ∂ϕ ∂ ∂ϕ ∂ ∂ϕ
utilized to achieve the sensitivity and optimization analysis of the ¼ Γ þ Γ þ Γ þS ð1Þ
design wind angle for a 2 × 600 MW air-cooled power plant. The ∂x ∂x ∂y ∂y ∂z ∂z
performance of the air-cooled steam condenser as well as the power
where Φ = different variables; Γ = corresponding generalized dif-
plant at different wind conditions is numerically investigated, and
fusion coefficient; and S = source term. The specific expression of
the optimal-design wind angle with the best performance is ob-
the governing equations corresponding to these variables is given
tained. Furthermore, the net output power and thermal efficiency
out by Wang (2004) with the buoyancy effect considered, and the
of the closed Ranking cycle at the optimal-design wind angle is
standard k-e turbulence model is used during the simulation.
compared to the values for the current design scheme. The present
The conservation equations are discretized utilizing the finite-
paper provides the references to optimize the design of the wind
volume method, and semi-implicit method for pressure linked
angle for a much better performance by the air-cooled power plant.
equations (SIMPLE) algorithm is implemented for a steady-state
solution during the numerical simulation with the FLUENT 6.3
Numerical Model and Computing Method platform. In view of the symmetrical relation between the ambient
wind and the power plant, five wind angles from β ¼ 0° to β ¼
180° are designated to investigate the influence from the wind angle
Geometry and Computational Grid on the performance of the air-cooled power plant, and five increas-
Performance of the air-cooled steam condenser as well as the whole ing wind speeds are also considered for each wind angle during the
power plant is numerically investigated at different wind conditions simulation. Boundary conditions of the whole computational model
based on the established geometry model (He et al. 2013), for a are shown in Fig. 2. It is found that a no-slip wall is appointed to the
power plant located in the north of China in Shaanxi Province, ground with the ambient temperature. The entrances of the domain are
as presented in Fig. 1. For the air-cooled steam condenser of the 2 × treated as velocity inlets with one inlet for the front wind case and two
600 MW power plant, the fan array consists of 112 axial fans, and inlets for the skew wind case, while the surfaces opposite the inlets are
finned tube exchangers are laid over the fan array. Hence, ambient set as outlets with a static pressure. Furthermore, the two lateral sur-
air from the fans will carry out the waste heat from the turbine ex- faces are regarded as the symmetry boundaries for the front wind
haust to maintain the power plant’s operations. The computing fluid cases. The exponential wind speed distribution is demonstrated based
domain, including the geometry of a 2 × 600 MW air-cooled on VGB-131 Me (Almmer et al. 1997) as follows:
power plant has the scale of 500 × 1,000 × 300 m. Mesh genera-
0.2
Hw
tion for the whole computational model is achieved with the help vw ¼ vm ð2Þ
of Gambit, and mesh inside the fan drum and exchanger zones is Hm
refined. The local mesh of the ACSC as well as the fan is also pre-
sented in Fig. 1. where vm = wind speed at the height of the measuring location;
vw = wind speed above the distributing tube with a height of Hw ,
where Hw ¼ 56 m according to the guide from VGB-131 Me; and
Computing Method and Boundary Conditions Hm = height of the measuring location, where H m ¼ 10 m.
The phenomenon of fluid flow is dominated by conservation laws
including those for mass, momentum, energy, and extra turbulence
Numerical Model Validation
transportation, and governing equations are the mathematical ex-
pressions of the conservation laws. The general governing equation In view of the importance of the regions of fan and heat exchangers
is as follows (Tao 2002): for the performance of the whole power plant, different mesh

© ASCE 04016010-2 J. Energy Eng.

J. Energy Eng., 04016010

Downloaded from ascelibrary.org by University of California, San Diego on 03/03/16. Copyright ASCE. For personal use only; all rights reserved.

Fig. 2. Boundary conditions of the numerical model for the air-cooled power plant: (a) skew wind; (b) front wind

2000 35

1800 4,120,000 4,680,000 5,170,000

30
1600
Design back pressure
Heat transfer rate/MW

1400 Predicted back pressure

Back pressure/kPa 25
1200

1000 20

800
15
600

400
10
200

0 5
TRL THA TMCR VWO CP TRL THA TMCR VWO CP

(a) Design mode (b) Design mode

Fig. 3. Grid independence test and validation of the numerical model: (a) grid independence test; (b) validation with the corresponding design values

densities at such regions are achieved, and thus three sets of grids, the volumetric effectiveness of the fan array and the heat transfer
namely 4,120,000, 4,680,000, and 5,170,000, are gained. A grid performance of the finned tube exchangers are obtained for differ-
independence test is verified with a maximum deviation of the heat ent wind conditions. Based on the coupling effect between the wind
transfer rate of 2.1%, shown in Fig. 3(a). As a result, the middle conditions and condenser performance, new stable backpressures
grid consisting of 4,680,000 cells is used to ensure the accuracy of are iterated utilizing the stated user defined function (UDF) method
the simulation results and short the calculation period simultane- by He et al. (2013), and then the connection between the backpres-
ously. Furthermore, the stable back pressures of the air-cooled sure of the air-cooled steam condenser and other thermal parame-
steam condenser in the designed wind conditions, including the ters in the closed Ranking cycle is exploited to calculate the
current-design wind angle, β ¼ 22.5° and the wind speed, characteristics of the whole power plant. During the investigation,
vw ¼ 3 m=s, is validated through the five typical design cases, in- the parameters of the turbine exhaust in the TRL case in Table 1 are
cluding turbine rated load (TRL), turbine heat acceptance (THA), considered, while the ambient temperature is 306 K and the pres-
turbine maximum continues rate (TMCR), valve wide open sure is 101,325 Pa.
(VWO), and choking pressure (CP). It is found that the predicted
backpressures are almost consistent with the design values, as Performance of the Air-Cooled Steam Condenser
shown in Fig. 3(b), which verifies the precision of the numeri-
cal model. In the air-cooled steam condenser, the designed fan performance is
determined to calculated the face velocity through the finned tube
exchangers corresponding to the heat load of the turbine exhaust.
Results and Discussion For an axial fan, the fan performance is obtained for specific wind
conditions, and it will differ at off-design wind speeds and wind
Due to the great influence from wind conditions, the performance angles. The pressure rise and power of the axial fan in the air-
of the air-cooled steam condenser is numerically investigated, and cooled steam condenser with the volume flow rate through the rotor

© ASCE 04016010-3 J. Energy Eng.

J. Energy Eng., 04016010

 Table 1. Steam Parameters Flowing into the Air-Cooled Steam Condenser V d ¼ 467 m3 · s−1 . Corresponding to the total volume flow rate of
for the TRL Case the fan array, the average volumetric effectiveness of the fan array is
Item Value calculated in Eq. (4) as follows:
ms (kg=s) 738.5 Pn
ps (kPa) 29 i¼1 V i
x (%) 96.4
eav ¼ ðn ¼ 112Þ ð4Þ
nV d
Δps (Pa) 820

where eav = average volumetric effectiveness of the fan array; and

n = fan number in the fan array.
150 75 Fig. 5 shows the absolute and relative performance of the fan
array with the wind angle at different wind speeds, respectively.
Downloaded from ascelibrary.org by University of California, San Diego on 03/03/16. Copyright ASCE. For personal use only; all rights reserved.

140 Based on the adverse effect from the wind conditions, it is found
70
that a reduction of the volume flow rate for the fan array appears
130 with an increase in the wind speed. It is inferred that the correlation

Fan shaft power/kW

of the wind speed and fan array’s performance differs with the wind
Pressure rise/Pa

65
120 angle. The maximum decrease in amplitude at a wind angle of 180°
is 15,264.6 m3 · s−1 from vm ¼ 2 m · s−1 to vm ¼ 10 m · s−1 . Fur-
110
60 thermore, the fan array performance rises gently until reaching its
Δ pf peak value at the wind angle of 90° for all the wind speeds, and then
100
drops sharply with an increase in the wind angle. It is also seen that
N
55 the decreased amplitude of the volume flow rate with the wind an-
90
gle is more obvious for high-wind-speed cases. Taking the case of
vm ¼ 10 m · s−1 for example, the flow rate difference reaches
80 50
300 350 400 450 500 550 600 650 6,637.4 m3 · s−1 from β ¼ 0° to β ¼ 180°, while the correspond-
Volume fan flow rate/m3·s-1
ing value at vm ¼ 2 m · s−1 is only 1,079.9 m3 · s−1.
Regardless of the performance difference for the axial fans, the
Fig. 4. Design performance of the axial forced-draft fan average volumetric effectiveness of the fan array is also calculated
and presented in Fig. 5(b). It is found that the average volumetric
effectiveness of the fan array is close to 1 at vm ¼ 2 m · s−1 for all
the wind angles, which illustrates the performance of the fan array
is presented in Fig. 4. In the current paper, the influence of wind as it approaches the design point, and the worst conditions appear
conditions on the performance of the total fan array is investigated. at vm ¼ 10 m · s−1 and β ¼ 180° with only 70% of the design
Volumetric effectiveness, which is defined as the ratio of the capacity.
calculated flow rate and the design value in Eq. (3), is applied With the existing turbine exhaust parameters at the TRL mode
to characterize the actual performance of the axial fan under differ- and wind conditions, the heat transfer rate of the finned tube ex-
ent wind conditions changers in the air-cooled steam condenser is obtained, which is
presented in Fig. 6(a). Obviously, the sensitivity of fan performance
V to the wind conditions will result in alternations to the heat transfer
e¼ ð3Þ
Vd capacity. Corresponding to the trend of the fan performance
with wind conditions, the maximum heat transfer rate appears at
where e = volumetric effectiveness; V = actual calculated fan flow β ¼ 90° for all the wind-speed cases, which is reduced from
rate; and V d = design flow rate of the axial fan in this paper, where 1,719.9 MW at vm ¼ 2 m · s−1 to 1,454.6 MW at vm ¼ 2 m · s−1 .

1.05
54000
1.00
52000

50000 0.95
Volumetric effectiveness/%
Volume flow rate/m3· s-1

48000
0.90
46000
0.85
44000
vm=2m/s
vm=2m/s
0.80
42000
vm=4m/s vm=4m/s
40000 vm=6m/s 0.75 vm=6m/s
vm=8m/s vm=8m/s
38000
0.70 vm=10m/s
vm=10m/s
36000
0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180
(a) Wind angle/° (b) Wind angle/°

Fig. 5. Performance of the fan array with wind angles at different wind speeds: (a) volume flow rate; (b) volumetric effectiveness

© ASCE 04016010-4 J. Energy Eng.

J. Energy Eng., 04016010

 70
1700 vm=2m/s
vm=4m/s
60 vm=6m/s
1600
Heat transfer rate/MW

vm=8m/s

Back pressure/kPa
50 vm=10m/s
1500

1400 vm=2m/s 40
vm=4m/s
vm=6m/s
Downloaded from ascelibrary.org by University of California, San Diego on 03/03/16. Copyright ASCE. For personal use only; all rights reserved.

1300 30
vm=8m/s
vm=10m/s
1200 20
0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180

(a) Wind angle/ ° (b) Wind angle/°

Fig. 6. Performance of the air-cooled steam condenser with wind angles at different wind speeds: (a) heat transfer rate; (b) backpressure

The ideal heat rejection of the turbine exhaust, which is defined

9400
as the heat rejection during condensation corresponding to each ps=8kPa
pressure if the exhaust steam is not supercooled, is expressed as 9200 ps=15kPa
(He et al. 2013) ps=30kPa
-1 9000
0
Qid ¼ ms ðh − h Þ 00
ð5Þ Heat rate/kJ kW h ps=45kPa
-1

8800
where Qid = ideal heat rejection; ms = steam flow rate; h 0 = exhaust
8600
steam enthalpy; and h 0 0 = enthalpy of the saturated water. Accord-
ing to the parameters of the TRL case, the ideal heat rejection is 8400
calculated as 1,665.6 MW.
The running point with a significant difference between the ac- 8200
tual heat-transfer rate and the ideal heat rejection is virtual due to
8000
the heat imbalance for the air-cooled steam condenser, and then an
iterative method will be utilized to obtain the final operating mode 7800
with a stable backpressure (He et al. 2013), which is shown in 350 400 450 500 550 600 650 700
Fig. 6(b). It is found that the lowest backpressure for each wind Steam turbine power/MW
speed at β ¼ 90° results from the best heat-transfer status between
the ambient and turbine exhaust, and the backpressure value of Fig. 7. Heat rate of the power plant with turbine power at different
25.6 kPa is obtained at 2 m · s−1 while the highest backpressure backpressures (data from He et al. 2014, © ASCE)
with a value of 70.9 kPa emerges at vm ¼ 10 m · s−1 and β ¼ 180°
due to the poor heat-transfer environment. However, control action
is not taken into consideration during the numerical study. A higher
Based on the correlation between the exhaust heat load and
backpressure will contribute to the performance reduction, and the
turbine power at different backpressures as shown in Fig. 7
trip can detonate in serious cases. Generally speaking, a pressure
(He et al. 2014), the benefit output at different wind conditions
limit of 45 kPa is allowed in a 2 × 600-MW air-cooled power plant.
corresponding to stable backpressure is interpolated and presented
As a result, all the calculated cases with a backpressure more than
in Fig. 8(a). In comparison with the design output power of
the value of 45 kPa will be avoided.
1,200 MW, an excess value of 65.7 MW is observed at vm ¼ 2 m ·
s−1 and β ¼ 90°, while the worst value of 976.9 MW only fulfills
Performance of the Air-Cooled Power Plant the 81.4% of the design value. The fan array’s power is also pre-
sented in Fig. 8(b) according to the demonstration in Fig. 4. It is
In an air-cooled power plant, the superiority of water conservation found that a value of about 7,462 kW, which is close to the design
is acquired at the cost of tremendous power consumption for the fan fan power, is consumed by the axial fan array at vm ¼ 2 m · s−1 .
array, and the turbine output as well as the fan power should be However, under the effect of flow distortions at the fan inlets, es-
emphasized. The characteristics of the air-cooled steam condenser, pecially for the windward fans, the fan power is also reduced sig-
including the performance of the fan array and finned tube ex- nificantly, and the fan array has a power consumption of 6,419 kW
changers, are obtained considering the coupling effect from the at vm ¼ 10 m · s−1 and β ¼ 180° with a decreased amplitude of
wind conditions and the backpressure, which indicates that the per- 14% compared to the design value.
formance connection between the air-cooled steam condenser and Furthermore, the final net power of the air-cooled power plant is
the whole power plant is also achieved through the proposed iter- expressed as
ative method. It is necessary to calculate the power plant perfor-
mance, thus establishing the standard to judge the economy of the
whole thermal system. Pnet ¼ P − N=1000 ð6Þ

© ASCE 04016010-5 J. Energy Eng.

J. Energy Eng., 04016010

 1300 7600

1250 7400

1200 7200

Fan shaft power/kW

Plant power/MW

1150 7000

1100 6800 vm=2m/s

vm=2m/s
vm=4m/s
vm=4m/s
1050 6600 vm=6m/s
vm=6m/s
vm=8m/s
Downloaded from ascelibrary.org by University of California, San Diego on 03/03/16. Copyright ASCE. For personal use only; all rights reserved.

vm=8m/s
1000 6400 vm=10m/s
vm=10m/s

950 6200
0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180

(a) Wind angle/° (b) Wind angle/°

Fig. 8. Performance of the air-cooled power plant with wind angles at different wind speeds: (a) plant power; (b) fan shaft power

44 where P = output power of the generator; and N = power consump-

tion of the fan array. Despite the fan array’s consumption, it can be
43 verified that the trend of the net power with the wind conditions
remains the same as to the turbine output with a maximum value
Ranking cycle efficiencey/%

42
of 1,258.3 MW.
Turbine exhaust in the air-cooled power plants condenses
in the finned tube exchangers when ambient air flows outside
41
and the thermal cycle in the power plant is completely closed. After
the illustration of turbine power distributions in different wind con-
40 vm=2m/s
ditions, the input heat from the steam generator can be calculated in
vm=4m/s combination with the relation between the steam turbine’s power
39 vm=6m/s and the relevant heat rate at different backpressures shown in Fig. 7;
vm=8m/s in addition, the total thermal efficiency of the closed Ranking cycle,
38 vm=10m/s which is listed in Fig. 9, can be calculated as the ratio of the benefit
power and the heat input in the steam generator.
37 With an increase in wind speed from vm ¼ 2 m · s−1 to
0 20 40 60 80 100 120 140 160 180 vm ¼ 10 m · s−1 , the thermal efficiency of the air-cooled power
Wind angle/° plant rises until the wind angle of β ¼ 90°, and then drops contin-
uously. The efficiency values at vm ¼ 2 m · s−1 implies that the
Fig. 9. Ranking cycle efficiency of the air-cooled power plant with
design thermal efficiency is close to 43%, and the optimal effi-
wind angles at different wind speeds
ciency corresponding to each wind speed is reduced from 43.2

1280 43.5

1260
43.0
Ranking cycle efficiency/%

1240

42.5
1220
Net plant power/MW

1200
42.0
1180
41.5
1160
β =22.5°
1140 β =22.5°
41.0 β =90 °
β =90°
1120
40.5
1100

1080 40.0
2 4 6 8 10 2 4 6 8 10

(a) Wind speed/m· s-1 (b) Wind speed/m· s-1

Fig. 10. Comparison of the power plant performance for the optimized wind angle and current design: (a) net power; (b) thermal efficiency

© ASCE 04016010-6 J. Energy Eng.

J. Energy Eng., 04016010

 to 41.0%, while the minimum value is 37.7% at vm ¼ 10 m · s−1 P =
power (MW);
and β ¼ 180° under great influence from wind conditions. p =
pressure (Pa);
S source term (N · m−3 );
=
Comparison of the Ranking Cycle Efficiency for the T =
temperature (K);
Optimized-Design Wind Angle and the Current Design t =
time (s);
v velocity (m · s−1 );
=
In accordance with the simulation results for different wind condi- x; y; z =
coordinates (m), steam quality;
tions, it is concluded that the characteristics of the power plant at ρ density (kg · m−3 );
=
β ¼ 90° are always more predominant compared to those for other ε turbulent kinetic energy dissipation rate (m2 · s−3 ), heat
=
wind angles. The net power and thermal efficiency for the air- transfer effectiveness;
cooled power plant at the optimized wind angle, β ¼ 90°, and de- Φ = general variable; and
sign case, β ¼ 22.5°, are listed in Fig. 10. It is found that the Γ = diffusion coefficient.
Downloaded from ascelibrary.org by University of California, San Diego on 03/03/16. Copyright ASCE. For personal use only; all rights reserved.

increasing magnitude for both net power and thermal efficiency

is significant, from 13.6 MW and 0.14% at vm ¼ 2 m · s−1 to Subscripts
28.08 MW to 0.57% at vm ¼ 10 m · s−1 , respectively. It is inferred
that the improvement effect from the optimal-design wind angle is a = average;
more obvious in high-wind-speed cases due to the deterioration in d = design;
system performance. Moreover, taking the design wind speed, f = fan;
vm ¼ 2.1 m · s−1 , for example, illustrated by the dotted line in id = ideal;
the figure, the beneficial net power and efficiency is 13.8 MW m = measure;
and 0.16% with a raised amplitude of 1.2 and 0.37% based on n = net;
the design values, which implies a more-economical design scheme s = steam;
for the air-cooled power plant is possible if the design’s wind angle t = turbulence; and
is fixed at β ¼ 90° under the same designed wind speed conditions. w = wind.

Conclusions
References
In the present paper, an existing numerical model is applied in order
to investigate the characteristics of a 2 × 600 MW air-cooled power Almmer, A., Bakran, B., and Hoxtermann, L. (1997). “VGB guideline
acceptance test beasurements and operation monitoring of air-cooled
plant. The effect of wind speed and angle on the performance of the
condensers under vacuum.” Vgb Technische Vereinigung Der
air-cooled steam condenser is first investigated, and then the rel- Grosskraftwerksbetreiber, EV, Germany.
evant impact on the whole power plant is presented in detail FLUENT version 6. 3 [Computer software]. Ansys, Canonsburg, PA.
through the connection effect of the predicted stable backpressures. Gao, X. F., Zhang, C. W., Wei, J. J., and Yu, B. (2009). “Numerical sim-
The design’s wind angle is optimized and determined at the peak ulation of heat transfer performance of an air-cooled steam condenser in
point of the power plant’s performance. Finally, a comparison be- a thermal power plant.” Heat Mass Transf., 45(11), 1423–1433.
tween the optimal and current designs of wind angle for both the Gu, H. F., Zhe, Z., Wang, H. J., and Qi, C. (2015). “A numerical study on
net power and thermal efficiency is undertaken. the effect of roof windbreak structures in an air-cooled system.” Appl.
It is found that performance of the air-cooled steam condenser is Thermal Eng., 90, 684–693.
quite sensitive to the wind conditions, and the extent of this influ- He, W. F., and Dai, Y. P. (2012). “Pressure forecast of an air-cooled steam
condenser under wind speeds.” Int. Conf. on Manufacturing Science and
ence corresponding to the wind angle differs with the wind’s speed.
Technology, Advanced Materials Research, Switzerland, 6187–6193.
However, it is observed that the most predominant performance al- He, W. F., Dai, Y. P., Wang, J. F., Li, M. Q., and Ma, Q. Z. (2013). “Per-
ways arises at the wind angle of β ¼ 90° for all wind-speed cases. formance prediction of an air-cooled steam condenser using UDF
Compared to the current design cases at vm ¼ 2.1 m · s−1 and method.” Appl. Thermal Eng., 50(1), 1339–1350.
β ¼ 22.5°, it is found that the net power and thermal efficiency He, W. F., Han, D., Yue, C., and Pu, W. H. (2014). “Influence mechanism
can be improved by an amplitude of 1.2%, 13.8 MW, and 0.37, from the inner walkway on the performance of an air-cooled power
and 0.16%, respectively. It is strongly recommended that the wind plant.” J. Energy Eng., 04014052(9), 1–9.
angle of β ¼ 90° should be fully considered during the design Hu, H. M., Du, X. Z., Yang, L. J., Zhou, Y. Y., and Yang, Y. P. (2014). “POD
period to optimize the performance of the air-cooled steam con- based modeling on flow and heat transfer of air-cooled condenser influ-
denser as well as the whole power plant. enced by natural wind.” Int. J. Heat Mass Transfer, 74, 431–440.
Moore, J., Grimes, R., Walsh, E., and O’Donovana, A. (2014). “Modelling
the thermodynamic performance of a concentrated solar power plant
Acknowledgments with a novel modular air-cooled condenser.” Energy, 69, 378–391.
Owen, M. T. F. (1992). “A numerical investigation of air-cooled steam con-
The authors gratefully acknowledge the financial support by the denser performance under windy conditions.” Ph.D. thesis, Univ. of
Stellenbosch, Stellenbosch, South Africa.
National Natural Science Foundation of China (Grant No. 51406081).
Tao, W. Q. (2002). Numerical heat transfer, Xi’an Jiaotong University
Press, Xi’an, China.
Thiart, G. D. (1990). “A numerical procedure for predicting the effects of
Notation distorted inflow conditions on the performance of axial flow fans.”
Ph.D. dissertation, Stellenbosch Univ., Stellenbosch, South Africa.
The following symbols are used in this paper:
Van Rooyen, J. A., and Kroger, D. G. (2008). “Performance trends of an air-
cp = specific heat (J · kg−1 · K−1 ); cooled steam condenser under wind conditions.” J. Eng. Gas Turbine
H = height (m); Power, 130(2), 023006.
k = turbulent kinetic energy (m2 · s−2 ); Van Staden, M. P. (1995). “Numerical modeling of the effects of ambient
N = fan power consumption (kW); conditions on large power station air-cooled steam condensers.” Am.
n = fan number; Soc. Mech. Eng., 221, 145–150.

© ASCE 04016010-7 J. Energy Eng.

J. Energy Eng., 04016010

 Wang, F. J. (2004). Computational fluid dynamics analysis, Tsinghua Yang, L. J., Zhao, X. L., Du, X. Z., and Yang, Y. P. (2014). “Heat load
University Press, Beijing. capability matching principle and its applications to anti-freezing of
Yang, L. J., Du, X. Z., and Yang, Y. P. (2011a). “Influences of wind-break air-cooled condenser.” Appl. Energy, 127, 34–43.
wall configurations upon flow and heat transfer characteristics of air- Zhang, X. L., and Chen, H. P. (2015). “Effects of windbreak mesh on
cooled condensers in a power plant.” Int. J. Thermal Sci., 50(10), thermo-flow characteristics of air-cooled steam condenser under windy
2050–2061. conditions.” Appl. Thermal Eng., 85, 21–32.
Yang, L. J., Du, X. Z., and Yang, Y. P. (2011b). “Space characteristics of the Zhang, Z. Y., Yang, J. G., and Wang, Y. Y. (2015). “A favorable face
thermal performance for air-cooled condensers at ambient winds.” Int. velocity distribution and a V-frame cell for power plant air-cooled
J. Heat Mass Transfer, 54(15-16), 3109–3119. condensers.” Appl. Thermal Eng., 87, 1–9.
Downloaded from ascelibrary.org by University of California, San Diego on 03/03/16. Copyright ASCE. For personal use only; all rights reserved.

© ASCE 04016010-8 J. Energy Eng.

J. Energy Eng., 04016010