Comparison of evaporative inlet air cooling systems to enhance the gas turbine generated power
Content
INTERNATIONAL JOURNAL OF ENERGY RESEARCH
Int. J. Energy Res. 2007; 31:1483–1503
Published online 7 March 2007 in Wiley InterScience
(www.interscience.wiley.com) DOI: 10.1002/er.1315
Comparison of evaporative inlet air cooling systems to enhance
the gas turbine generated power
Mohammad Ameri*,y, H. R. Shahbazian and M. Nabizadeh
Combined Heat & Power Specialized Unit (CHP), Energy Engineering Department, Power & Water University
of Technology, PO Box 16765-1719 Tehran, Islamic Republic of Iran
SUMMARY
The gas turbine performance is highly sensitive to the compressor inlet temperature. The output of gas
turbine falls to a value that is less than the rated output under high temperature conditions. In fact increase
in inlet air temperature by 18C will decrease the output power by 0.7% approximately. The solution of this
problem is very important because the peak demand season also happens in the summer. One of the
convenient methods of inlet air cooling is evaporating cooling which is appropriate for warm and dry
weather. As most of the gas turbines in Iran are installed in such ambient conditions regions, therefore this
method can be used to enhance the performance of the gas turbines.
In this paper, an overview of technical and economic comparison of media system and fog system is
given. The performance test results show that the mean output power of Frame-9 gas turbines is increased
by 11 MW (14.5%) by the application of media cooling system in Fars power plant and 8.1 MW (8.9%)
and 9.5 MW (11%) by the application of fog cooling system in Ghom and Shahid Rajaie power plants,
respectively. The total enhanced power generation in the summer of 2004 was 2970, 1701 and 1340 MWh
for the Fars, Ghom and Shahid Rajaie power plants, respectively.
The economical studies show that the payback periods are estimated to be around 2 and 3 years for fog
and media systems, respectively. This study has shown that both methods are suitable for the dry and hot
areas for gas turbine power augmentation. Copyright # 2007 John Wiley & Sons, Ltd.
KEY WORDS:
gas turbine; inlet air cooling; evaporating cooling; media; fog; power augmentation
1. INTRODUCTION
Gas turbines are used widely in power generation, gas transfer stations and petrochemical
industries (GE Energy, 2006). The site ambient conditions, especially the temperature, have
great influence on gas turbines performance (Brook, 1998). Since the air density is decreased
during warm days, the mass flow rate through the turbine is decreased. Therefore, it causes a
*Correspondence to: Mohammad Ameri, Combined Heat & Power Specialized Unit (CHP), Energy Engineering
Department, Power & Water University of Technology, PO Box 16765-1719 Tehran, Islamic Republic of Iran.
y
E-mail: ameri [email protected]
Copyright # 2007 John Wiley & Sons, Ltd.
Received 29 April 2006
Revised 28 December 2006
Accepted 31 January 2007
1484
M. AMERI, H. R. SHAHBAZIAN AND M. NABIZADEH
Figure 1. T–S diagram for a hot day.
drop in the output power. Moreover, the compressor work increases due to the divergence of
constant pressure lines in T–S diagram (Figure 1). On the other hand, the compressor final
pressure decreases. Given the fact that the turbine inlet temperature is constant, it will reduce
the turbine work. As a result the net output of gas turbine falls. Moreover, there is a peak
demand of electricity in summer. Therefore, the gas turbine inlet air cooling is one of the useful
methods which can be applied for the gas turbine power enhancement.
Kraneis et al. (2000) studied the effects of an evaporative cooler on the available power plant
capacity with a detailed outline of the climatic conditions prevailing on the various continents.
Nixdorf et al. (2002) investigated the economic benefits of some different ambient air
conditioning methods for reducing the gas turbine intake air temperature in order to enhance
the gas turbine power. Johnson (1998) discussed the theory of evaporative cooling and explained
the evaporative cooler design, installation, operation, feed water quality, and the causes and
prevention of water carry over. Kakaras et al. presented a computer simulation of the
integration of an evaporative cooler and of the air-cooling system and discussed the effect of
ambient air temperature variation on the power output and efficiency (Kakaras, 2004). Ameri
et al. (2004) have studied the installation of fog inlet air cooling system for six Frame-5 (25 MW)
gas turbines. The results of that study showed that the output power of each gas turbine was
increased by 3 MW. Also, they showed that the fog system was very cheap in comparison with
the installation of new gas turbines. McNeilly (1997) presented a new method for the test
correction error of evaporative coolers. Ameri et al. (2004) presented a good overview of an
intake air-cooling system that used a steam absorption chiller and an air cooler to increase the
power output of Frame-6 gas turbines.
The fog and media evaporative inlet cooling are the most economical alternatives for hot and
dry areas. In fact the other techniques such as the absorption and vapour compression chiller
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
COMPARISON OF EVAPORATIVE INLET AIR COOLING SYSTEMS
1485
methods are not suitable for that ambient condition as they are very expensive alternatives. The
purpose of this paper is to present the state-of-the-art application of gas turbine evaporative
inlet air cooling systems for three large gas turbine power plants (Shahid Rajaie, Ghom and
Fars) in Iran’s electrical grid and to compare the technical and economical aspects of fog and
media inlet air cooling systems. Although one may find various papers in literature regarding the
topic, however, there is neither many reliable actual power plant test data available for the large
gas turbines nor there exist many unbiased comparisons between different evaporative inlet air
cooling systems. In fact, most of the similar results are given by commercial companies which
have not been justified by an independent research. Each company claims that its system either
fog or media is the best.
Following the discussion on the different inlet air cooling methods, the actual performance
test results are presented. Finally, the economic benefits of the different cooling systems are
explained.
2. GAS TURBINE INLET AIR COOLING METHODS
There are several inlet air cooling methods available for gas turbine power augmentation
(Omidvar, 2001). They can be classified into three types:
(1) thermal energy storage systems;
(2) refrigerated cooling system (utilizing absorption or mechanical refrigeration);
(3) evaporative coolers (media and fog).
3. EVAPORATIVE COOLERS OVERVIEW
Using the evaporative coolers will cause the maximum reduction in inlet air temperature if the
inlet air dry bulb temperature approaches the wet bulb temperature (relative humidity of 100%).
This can take place for dry and hot weather. There are two types of evaporating cooling:
3.1. Media evaporating cooling
Media surfaces are flexuous and consist of beehive-shaped cells which make up an evaporating
cooler. The inlet air can be cooled with surface evaporating by spraying water on these cells and
humidifying them. Increasing the contact area between water and air will cause the surface
evaporation to be more and faster. The flexuous and beehive-shaped cells increase the contact
area between water and air.
The following important points should be considered for the selection of a media system:
(a) Pressure drop in this system is more than other evaporating cooling systems. However,
this pressure drop has not much influence on the gas turbine output.
(b) Power consumption of this system is less than other systems.
(c) This system does not need demineralized water and can use the raw water. However, it is
better to use distilled water.
(d) Since this system needs periodical replacement of media (every 3 or 4 years), its
maintenance cost is more than other systems.
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
1486
M. AMERI, H. R. SHAHBAZIAN AND M. NABIZADEH
(e) The installation cost of this system is more expensive.
(f) If the system is installed in filter room, the shutdown period of the unit is considerable.
This system which uses a media surface to evaporate water is widely used for gas turbines
especially in dry and hot areas.
The performance of this system is based on water evaporation, which consumes thermal
energy by the latent heat of vaporization value and reduces the ambient temperature.
Figure 2 shows the schematic of media evaporating cooling system. Media system equipment
includes piping, programmable logic controller (PLC) control system and measuring equipment,
water distributing headers, media surfaces, mist eliminator, movable wall in the media case,
circulation pump, water tank, and blow down system.
3.1.1. Media surfaces. The material of these surfaces is cellulose fibre. A media evaporating
cooler, which consists of these surfaces, is like beehive. Figure 3 shows the media
surface. Water is spread over the media area, so the ratio of vaporization area (m2) to the
media volume is increased and the air cooling is improved. These surfaces are covered
with special chemical material to prevent corrosion. Since the material of media surfaces is
cellulose, they are flammable. However, they can be made from fibre glass surfaces which are
not flammable.
Each of these surfaces is installed with a certain angle (the angles are 15 and 458 with respect
to the horizontal line) to increase the contact area between water and air (Figure 4). The airflow
is parallel to the horizontal plane (AAF Power & Industrial, 2002).
3.1.2. Media evaporating cooler function. In this system the water is pumped from the tank
below the cooler to the distributing header above the cooler (Figure 2). This water is distributed
over the surfaces and humidifies them. The outlet water is gathered below the cooler and
drained to the water tank. These surfaces which are called media surfaces have a thickness
of about 20 cm or more and cover the entire cross section of inlet air duct or air room. The
compressor inlet air passes through the media surfaces and evaporates the water to the
Figure 2. Media evaporating cooling system.
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
COMPARISON OF EVAPORATIVE INLET AIR COOLING SYSTEMS
1487
Figure 3. Evaporative cooler media surfaces (Munters Co., 2001)
Figure 4. Certain angles of media surface installation (AAF Power & Industrial, 2002).
saturation limit. The remaining water is used for continuous purging and discharging of
the other materials from the media surfaces. The amount of circulating water should be at least
2–3 times of evaporated water.
To prevent the removing of water droplets from media surfaces and damaging the compressor
blades due to the high velocity of air, the air velocity should be limited. The cooler air is passed
through a mist eliminator after the evaporating cooler and the water droplets are removed.
Figure 5 shows the reduction of saturation efficiency of the evaporating cooler at high velocities
for Celdek media surfaces with various thicknesses (Munters Co., 2001). On the other hand as
the inlet air velocity increases the pressure drop in the system increases as well. Also Figure 5
shows the pressure drop in Celdek media surfaces with various thicknesses versus the inlet air
velocity. Based on these facts, it is obvious that if the inlet air velocity is high, the evaporating
cooler saturation efficiency is reduced and the pressure drop is increased. Therefore, a diffuser is
used at the inlet air duct of evaporative cooler to reduce the inlet air velocity. However, it should
be noted that the pressure drop is very small.
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
1488
M. AMERI, H. R. SHAHBAZIAN AND M. NABIZADEH
Figure 5. The media cooler saturation efficiency and its pressure drop versus air velocity for different media
width (width of media surfaces are 75; 100; 150; 200 and 300 mm) (Munters Co., 2001).
3.2. Fog system
High-pressure fogging of gas turbine inlets has been applied for 15 years (Cyrus et al., 2000,
2002). In essence, it generates droplets of sizes 5–20 mm which are injected into the air stream
where they evaporate and provide air cooling. The test data have shown that this process can be
100% effective (i.e. wet bulb temperature can be reached) even in humid regions. Fog is
generated by the application of high-pressure demineralized water between 70 and 200 bar to an
array of specially designed fog nozzles. A typical fog system consists of a series of high-pressure
pumps that are mounted on a skid, PLC-based control system with temperature and humidity
sensors, and array of fog nozzles installed in the inlet air duct (Figure 6).
The nozzle is usually made of 316 stainless steel (SS) and consists of a small orifice from 127
to 178 mm for gas turbine applications. The water emanating from this orifice impacts a specially
designed impaction pin that breaks up the jet into billions of micro fine fog droplets, whose sizes
are between 10 and 40 mm. The rate of evaporation of the droplets essentially depends on the
surface area of water exposed to the air.
Figure 7 shows the distribution of droplets diameter in a specific nozzle. Typically, the
plunger type pumps are used to produce 130–200 bar pressures for gas turbine inlet air fogging
systems. They are positive displacement ceramic-plunger stainless steel pumps with stainless
steel heads. The pumps can be turned on sequentially to control the amount of cooling. For
example, 8.48C drop in temperature may be managed in three 2.88C increments.
3.2.1. Cooling system control. The control system incorporates a PLC, which is typically
mounted on the high-pressure pump skid. Sensors are provided to measure relative humidity
and dry bulb temperature. Special programming codes use these measured parameters to
compute the ambient wet bulb temperature and the wet bulb depression (i.e. the difference
between the dry bulb and wet bulb temperatures). They quantify and control the amount of
evaporative cooling that is possible with the ambient conditions. The system turns on or off fog
cooling stages to match the ability of the ambient conditions to absorb water vapour. The
control system also monitors pump skid operating parameters such as water flow rates and
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
COMPARISON OF EVAPORATIVE INLET AIR COOLING SYSTEMS
1489
Figure 6. Schematic of fog inlet air cooling system.
Figure 7. Distribution of droplet diameter for a typical nozzle (Omidvar, 2001).
operating pressure, and provides alarms when these parameters are outside acceptable ranges.
Table I indicates the typical cooling control system stages.
3.2.2. Fog nozzles position in inlet duct. There are two main options for installing the inlet
fogging system, i.e. locating them either upstream or downstream of the filters.
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
1490
M. AMERI, H. R. SHAHBAZIAN AND M. NABIZADEH
Table I. Typical fog cooling control stages (four stages of 2.88C each).
Time
Dry bulb (8C)
Wet bulb (8C)
Difference (8C)
Stage ‘ON’
Cooling (8C)
9:00
10:00
11:00
12:00
21.1
23.9
26.7
30
20
20.6
21.1
21.1
1.1
3.3
5.6
8.9
None
1
2
3
}
2.8
5.6
8.4
(a) Upstream of the inlet filters: One advantage of positioning the fog nozzle manifold
upstream of the air filters is that the installation can be accomplished without gas turbine outage.
In this case, a fog droplet mist eliminator filter must be added downstream of the fog nozzle
manifold to remove any unevaporated droplet. By definition, the droplet filter would not allow
any fog intercooling. Typically about half the water droplets by the fog nozzles is captured by the
droplet mist eliminator and drained away. This type of system while used on some early
installations is rarely applied to gas turbine installations today. It requires more fog nozzles
and more water and it is generally more expensive to operate and install. However the turbine
operators, who have experienced excessive loading of inlet air filters, might find this option
a cost-effective one.
(b) Downstream of the inlet filters: The most common location for the high-pressure fog nozzle
manifold is downstream of the air filters and upstream of silencers and trash screens. Installation
of the fog system in this location requires an outage of 1–2 days and calls for only minor
modifications to the turbine inlet structure. This type of installation allows fog intercooling.
While the fog nozzle manifolds can be also installed downstream of the silencers, it is generally
considered best to locate them upstream of the silencers, as this would allow more residence time
for the fog droplets to evaporate. Fog nozzle manifolds are usually installed upstream of the
trash screens to avoid any possibility of foreign object damage (FOD).
4. COMPARISON OF MEDIA AND FOG INLET AIR COOLING SYSTEMS
Comparisons of media and fog system for cooling air of gas turbine are shown in Table II
(Grance et al., 2001). Care should be taken for using these cooling systems as they may change
the performance curve of the gas turbine towards surge point. Moreover if the humidity of air is
high, the power increase is limited due to the low efficiency of the evaporative system. Therefore,
the required cooling is not achieved.
5. THE EFFECT OF EVAPORATING COOLER ON THE COMPRESSOR INLET AIR
COOLING AND THE ENHANCEMENT OF GAS TURBINE OUTPUT POWER
The media and fog cooling systems enhance the output power by reducing the inlet air
temperature as follows:
(1) The density and the flow rate of air passing through the gas turbine are increased
by reducing the compressor inlet air temperature. Therefore, the output power is
enhanced.
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
COMPARISON OF EVAPORATIVE INLET AIR COOLING SYSTEMS
1491
Table II. Comparisons of media and fog system for cooling air of gas turbine (Grance et al., 2004).
Main parameter
Efficiency of humidifying (%)
Capital cost (US$ kW1 installed)
Quality of consumption water
Pressure drop (Pa)
Droplet diameter (mm)
Time of unit exit for
installation (days)
Change of unit structure
Another advantage
Media
Fog
85–90
45–55
Raw water
200
Less than 100
5–7
90–100
35–45
Fully demineralized
Very low
Less than 20
1–2
Need to change of filter room
Installation in filter room
Increasing the life of filter,
reducing the amount of NOx
Ability to produce 100% humidity,
lower consumption of water
(2) The consumption power of compressor is decreased as its inlet air temperature is reduced
(The consumption power of compressor is proportional to the inlet air temperature
directly). Therefore, the net power output of turbine increases.
(3) If the inlet air temperature is reduced, the exhaust flue gas temperature is reduced. This is
due to the fact that if all other parameters are kept constant, the energy balance causes
the flue gas temperature to reduce. Therefore, the error signal which is the difference
between exhaust gas temperature and its set point temperature is reduced. The turbine
control system increases the fuel flow rate and therefore the output power of gas turbine
is enhanced. The flue gas temperature returns to its set point value.
6. THE CHARACTERISTICS OF GAS TURBINE UNITS AND INSTALLED COOLING
SYSTEMS FOR FARS, SHAHID RAJAIE AND GHOM POWER PLANTS
There are many gas turbines in Iran favoured for coping with the peak electricity demand of the
utilities due to their specific properties. However due to the high ambient temperature in the hot
seasons, the output power and efficiency of gas turbines decrease considerably.
The power output of the gas turbine is as low as 70% of its’ rated output when the
temperature increases in the summer and the electricity is most needed for air conditioning.
Each of these facilities serves as a vital equipment for the Iran electricity grid. For existing
installations the possible output enhancement can be determined by comparing the turbine
output at the design conditions with the output recorded at the desired inlet air temperature.
Essentially the inlet air cooling gives approximately winter performance during hot climatic
conditions. Two Frame-9 (PG9171E) gas turbines with the rated output of 99.3 MW in Fars,
two Frame-9 (PG9171E) gas turbines with the rated power of 98 MW in Shahid Rajaie and two
gas turbines (701D) with the rated output of 100 MW in Ghom power plants are selected for
the inlet air cooling system installation. Table III shows the characteristics of gas turbine units
and installed cooling systems. The detailed specifications of those gas turbines are given in the
technical documents of Fars, Shahid Rajaie and Ghom combined cycle power plants and are
presented by Shahbazian and Hoseinzadeh (2004) and Nabizadeh and Keshtgar (2004). All gas
turbines use the natural gas as their fuel. However, the heating values of their fuels are different.
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
1492
M. AMERI, H. R. SHAHBAZIAN AND M. NABIZADEH
Table III. Characteristics of gas turbine units and installed cooling systems.
Power plant
Fars
Shahid Rajaie
Ghom
No. of
case study
units
2
2
2
Power output at Min. & Max.
site conditions Temperatures
(MW)
(8C)
99.3
98
100
14 ! 43
10 ! 40
12 ! 45
Type of installed
cooling system
Design condition
for installed
cooling system
Media
Fog
Fog
388C and j=20%
408C and j=10%
458C and j=8%
7. TECHNICAL AND ECONOMIC EVALUATION OF APPLICATION OF MEDIA
AND FOG COOLING SYSTEM FOR GAS TURBINE INLET AIR COOLING
7.1. Prediction of gas turbine power enhancement using evaporative cooling system performance at
different ambient conditions
The evaporative cooling system is designed for a specific ambient condition. The design point is
not usually the maximum ambient temperature and minimum relative humidity temperature at
which the evaporative cooler has the maximum capacity. In fact as the power enhancement is
most required at the peak load, the design point is selected either based on this point or the
combination of this point and the maximum capacity point. Although it would be useful to test
the system at the design ambient conditions, however, it is impossible due to practical reasons.
Therefore, it is necessary to predict the evaporative cooler performance at the different ambient
conditions. This is usually important to check the guarantee requirements. In this section of the
paper, a method is presented to predict the gas turbine power enhancement.
The function of evaporating cooling system is cooling air by humidifying it in an adiabatic
process. During this process the wet bulb of air is constant. Figure 8 shows this process on the
psychometrics chart.
The saturation (or humidifying) efficiency of the system is defined as follows (Ameri et al.,
2004; AAF, 2002):
T1db T2db
Zhumidifying ¼
ð1Þ
T1db T1wb
The amount of required water in this system is determined from the following equation:
m
’ water ¼ m
’ air ðo2 o1 Þ
ð2Þ
where m
’ air is the inlet mass flow rate of dry air into the gas turbine.
As an example, the maximum power recovery under Fars power plant condition is determined
as follows.
At T1db ¼ 358C and relative humidity of j1 ¼ 20%; the specific humidity is o1 ¼ 0:008464:
Assuming maximum saturation efficiency (i.e. 100%, the best conditions for the evaporating
cooling system performance) the minimum achievable temperature is the wet bulb temperature.
Therefore, T2db ¼ T1wb ¼ 17:868C and the relative humidity and specific humidity are j1 ¼
100% and o2 ¼ 0:015568; respectively. Since the altitude of this power plant is 1530 m from the
sea level, the above values should be corrected. According to the tables of altitude correction
(ASHRAE, 1992), one can estimate Do to be 0.007104. The mass flow rate of inlet air to the
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
COMPARISON OF EVAPORATIVE INLET AIR COOLING SYSTEMS
1493
Figure 8. Cooling process on the psychometric diagram.
Comp. Inlet Tdb
Ambient Pressure
Evap.
Ambient Temperature
Cooler
Comp. Inlet Twb
Relative Humidity
System
Comp. Inlet Pressure
Gas Turbine
Power Output
Exhaust Gas Temp.
Figure 9. The schematic for the prediction of gas turbine power enhancement by evaporative cooling
system at different ambient conditions.
compressor is 406 kg s1 under design condition. Therefore, the required water mass flow rate is
m
’ water ¼ 2:88 kg s1 : According to the above values and the diagram of gas turbine output
power versus the inlet temperature for GE-F9 units of Fars power plant, the augmented power is
calculated (Shahbazian and Hoseinzadeh, 2004). The output power of gas turbine at the design
point is 79.13 MW and after the operation of media system it is estimated to be 90.18 MW.
Therefore, the maximum recovered power is 11.05 MW or 13.96%. For other power plants and
different ambient conditions the maximum power enhancement is determined by this method as
well (Nabizadeh and Keshtgar, 2004). It should be noted that the evaporative cooler efficiency is
estimated from the test results. Figure 9 shows the schematic for the prediction of gas turbine
power enhancement using the evaporative cooling system. The equations of evaporative cooler
system box are based on thermodynamical properties and evaporative cooler saturation
efficiency. The evaporative cooler efficiency is calculated using the test point results. The
equations of gas turbine box are based on the characteristics curve of the specific gas turbine.
A simple computer code has been prepared to calculate the temperature reduction and power
enhancement for different cases (Shahbazian and Hoseinzadeh, 2004; Nabizadeh and Keshtgar,
2004).
7.2. Performance test results and discussion
Since the evaporative systems performance test is an important issue in the feasibility study, the
complete performance tests have been done on all units. The main parameters in those tests were
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
1494
M. AMERI, H. R. SHAHBAZIAN AND M. NABIZADEH
the increase in output power and also the temperature and the humidity ratio at the compressor
inlet. Regarding these parameters, measuring points were specified. Since this system is used at
different times both day and night (especially in the summer nights when the ambient
temperatures are high enough that it may make the use of evaporating cooling units necessary)
the performance tests at night were also carried out to evaluate the performance of the units at
different conditions.
The correct functioning of sensors installed on PLC should be tested. Therefore, the
temperature and the air humidity data were also measured by another instrument and then the
results were compared to the data of PLC. The test procedure is as follows:
*
*
*
*
Starting up gas turbines and turning their automatic control on, letting each unit to reach
its full load (base load) (at this time 15-min waiting period was regarded for assurance that
the system had reached the steady-state operating conditions).
Recording parameters which were needed for performance test.
Turning on the cooling system and waiting for 20 min until the operating conditions
approach steady-state conditions.
Controlling and recording necessary parameters.
Table IV presents the results of performance tests of media evaporative cooling system for
units 1 and 2 of Fars power plant in August 2004.
Also, Tables V and VI present the results of performance tests of fogging cooling system for
units 1 and 2 of Shahid Rajaie and Ghom power plants in July and June 2004, respectively. It
should be noted the maximum calibration error for the output power measurement is 0.1%.
Table IV. The performance test results of media for units 1, 2 of Fars power plant (August 2004)
(Shahbazian and Hosseinzadeh, 2004).
GE (Frame-9) gas turbine unit 1
Parameters
Ambient
temp. (8C)
Relative
humidity (%)
Ambient
pressure (kPa)
Comp. inlet
air temp. (8C)
Comp. output
air temp. (8C)
Comp. output
air pressure
(bar)
Exhaust gas
temp. (8C)
Power
output (MW)
GE (Frame-9) gas turbine unit 2
After
Before
After
Before
operation operation
operation operation
Percent
Percent of media of media
of media of media
system Variation variation
system Variation variation system
system
38.17
38.27
8.3
8.2
0.08
1
83.85
83.82
0.023
40.6
22.66
371
8.23
559.6
76.6
347
8.92
549
87.71
Copyright # 2007 John Wiley & Sons, Ltd.
0.103
38.37
38.60
0.23
0.074
8.03
8.33
0.3
3.73
0.028
83.95
83.90
0.05
17.93
5.72
38
20
24
3.73
0.069
10.66
11.11
0.033
352
1.22
14.5
373.66
348.66
8.87
9.42
551
81.48
540
92.29
0.06
18
5.78
25
3.87
0.65
11
10.81
7.45
1.33
13.27
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
1495
COMPARISON OF EVAPORATIVE INLET AIR COOLING SYSTEMS
Table V. The performance test results of fog for units 1 & 2 of Shahid Rajaie power plant (July 2004)
(Nabizadeh and Keshtgar, 2004).
GE (Frame-9) gas turbine unit 1
Parameters
Ambient
temp. (8C)
Relative
humidity (%)
Ambient
pressure (mbar)
Comp. inlet
air temp. (8C)
Comp. output
air temp. (8C)
Comp. output
air pressure (bar)
Exhaust gas
temp. (8C)
Consumption
fuel rate
(m3 min1)
Power
output (MW)
GE (Frame-9) gas turbine unit 2
After
Before
After
Before
operation operation
operation operation
of fog
Percent
of fog
of fog
Percent
of fog
system Variation variation
system Variation variation system
system
35
35.33
0.33
0.107
34.66
35
0.33
0.107
11
11
0.0
0.0
11
11
0.0
0.0
868.6
868.7
0.1
0.012
868.6
868.7
0.1
0.012
33
19
32.66
16
16.66
0.054
363.33
345
18.33
2.88
361.33
341.33
0.48
5.44
8.75
9.26
0.92
8.823
9.303
558.33
550.66
7.66
463.5
498.06
34
86.55
95.03
8.48
7.44
9.79
14
0.046
20
3.15
0.51
5.86
560
552
8
480.1
519.53
39.36
8.19
92.22
9.2
11.08
83.02
0.96
Figures 10–14 show the gas turbine inlet air temperature reduction, power increase and
output power prediction for the different ambient conditions using the media evaporative cooler
or fog systems. The test point and the design points are also shown for reference. The maximum
error which is the difference between the actual (i.e. test) measured power and the predicted
power value is estimated to be around 2%. According to Figure 10 for media evaporating
system in Fars power plant, the inlet temperature drop increases due to the air relative humidity
decrease and temperature increase. For example, if the ambient temperature is 388C and the
relative humidity is 8%, the inlet air temperature is decreased by at least 178C using media
cooling system. Also according to Figures 11 and 12 the output power is enhanced by
11.11 MW. On the other hand, according to Figure 13 for fog cooling system in Ghom power
plant (for the ambient temperature of 338C and the relative humidity 16%) the inlet air
temperature is decreased by at least 168C. The output power is enhanced by 8.1 MW (Figures 14
and 15).
It should be noted that the power augmentation are clearly the results at the time of testing. If
an entire summer season or an entire year is considered, the overall power production increase
may be estimated. Figure 16 presents the total power generation for those three power plants in
the summer of 2004. It shows that the maximum power enhancement for the Fars power plant is
2970 MWh which is much more than the other two power plants i.e. 1701 MWh for Ghom and
1340 MWh for Shahid Rajaie power plants. It is 74.1 and 121.6% larger than those two other
power plants. In fact although the media evaporative cooler efficiency (85%) is generally a little
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
1496
M. AMERI, H. R. SHAHBAZIAN AND M. NABIZADEH
Table VI. The performance test results of fog for units 1 & 2 of Ghom power plant (June 2004)
(Nabizadeh and Keshtgar, 2004).
MW701D gas turbine unit 1
Parameters
Ambient
temp. (8C)
Relative
humidity (%)
Ambient
pressure (kPa)
Comp. inlet
air temp. (8C)
Comp. output
air temp. (8C)
Comp. output
air pressure
(kPa)
Exhaust gas
temp. (8C)
Fuel consumption
rate (kN m3 h1)
Power
output (MW)
MW-701D gas turbine unit 2
After
Before
After
Before
operation operation
operation operation
of fog
Percent
of fog
of fog
Percent
of fog
system Variation variation
system Variation variation system
system
31.96
32.66
16.66
15.66
88.87
88.87
0.00
31.8
16.58
415.66
404.06
1080
0.7
32
33
1
0.33
16
16
0.0
0.00
0.00
88.87
88.87
0.0
0.00
15.22
4.99
31.83
15.76
11
1.6
413.66
40
3.7
1
1120
528.93
523.7
0.23
6
1068
16.06
5.27
13.26
1.93
1113.33
45.33
4.24
400.4
5.22
0.65
528.76
523.63
5.13
0.64
29.45
30.76
1.31
4.44
28.60
30.14
1.53
5.37
94.93
100.53
5.60
5.89
90.1
98.2
8.1
8.99
Gas Turbine 1 in Fars Power Plant
Temperature Decrease (C)
26
Test Point
23
=5%
20
=8.3%
Design Point
17
=15%
14
=20%
11
=25%
8
=30%
5
15
20
25
30
35
40
45
50
55
Inlet Air Temperature (C)
Figure 10. Temperature decrease prediction using media system at various ambient conditions.
bit less than the fog system (90%), however, the climate in the Fars is more favourable for
installation of the evaporative cooling system. The high temperature, less humidity and more
hot hours per each day are the main reasons for higher power augmentation at the Fars site.
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
COMPARISON OF EVAPORATIVE INLET AIR COOLING SYSTEMS
1497
Gas Turbine 1 in Fars Power Plant
16
Power Increase (MW)
Test Point
14
ϕ = 5%
12
ϕ = 8.3%
Design Point
10
ϕ = 15%
ϕ = 20%
8
ϕ = 25%
6
ϕ = 30%
4
15
20
25 30 35 40 45
Inlet Air Temperature (C)
50
55
Figure 11. Power increase prediction using media system at various ambient conditions.
Gas Turbine 1 in Fars Power Plant
100
=5%
Design Point
96
Power (MW)
=8.3%
Test Point
92
=15%
88
=20%
84
=25%
=30%
80
76
15
20
25
30
35
40
45
50
55
Inlet Air Temperature (C)
Figure 12. Gas turbine output power prediction using media system at various ambient conditions.
Moreover in order to compare the media and fog system performances, Figure 17 presents the
temperature decrease and percent of power augmentation prediction due to installation of
media and fog system at three power plants (i.e. Fars, Rajaie and Ghom) for different ambient
conditions. It shows that the power augmentation for the media system (Fars) and the fog
system (Ghom and Rajaie) are almost the same at the same ambient conditions. In fact the
evaporative cooler efficiencies for the media and fog system are almost the same. These results
are quite different from the MEE Industries (2002) report (i.e. one of the major fog system
manufacturer) for a Frame 7111EA gas turbine. In fact, MEE has assumed an efficiency of
80–85% for the media evaporative cooling system and up to 100% efficiency for the fog system.
Therefore, MEE concluded that the percent of power boost attained by the use of fogger over
media type cooler power was up to 2.2% for different ambient temperatures and humidity
ratios. The reason for this difference can be due to the fact that for the fog system, the design of
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
1498
M. AMERI, H. R. SHAHBAZIAN AND M. NABIZADEH
Gas Turbine 2 in Ghom Power Plant
Temperature Decrease (C)
29
Design Point
= 5%
26
= 8%
Test Point
23
=10%
20
=16%
17
=20%
14
=25%
11
8
15
20
25
30
35
40
45
50
55
Inlet Air Temperature (C)
Figure 13. Temperature decrease prediction using fog system at various ambient conditions.
Gas Turbine 2 in Ghom Power Plant
18
Design Point
Power Increase (MW)
16
= 5%
14
Test Point
= 8%
12
=10%
10
=16%
8
=20%
6
=25%
4
15
20
25 30 35 40 45
Inlet Air Temperature (C)
50
55
Figure 14. Power increase prediction using fog system at various ambient conditions.
the nozzle manifold, type of nozzle and inlet duct length from the manifold to the compressor
inlet are the critical factors which affects the evaporative cooling system performance. In fact,
our fog systems have obtained the efficiency of 90% compared with the excellent MEE fog
system with the efficiency of up to 100%. Moreover, our media evaporative cooler system has
a very good design. Therefore, it has been able to achieve an efficiency of approximately 90%
(i.e. the same as our fog system efficiency).
7.3. Economic evaluation and discussion
The cost of an inlet cooling system is often evaluated in terms of US$ kW1. This can be
misleading because the output enhancement as a result of inlet air cooling varies with the
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
COMPARISON OF EVAPORATIVE INLET AIR COOLING SYSTEMS
Gas Turbine 2 in Ghom Power Plant
108
=5%
Test Point
105
Power (MW)
1499
Test Point
102
Design Point
=8%
=10%
=16%
99
=20%
96
=25%
93
90
15
20
25 30 35 40 45
Inlet Air Temperature (C)
50
55
Figure 15. Gas turbine output power prediction using fog system at various ambient conditions.
Figure 16. The comparison of gas turbine power augmentation (kWh) using the evaporative cooling
systems for the Fars, Ghom and Rajaie power plants during summer 2004.
ambient temperature. A better way of evaluating the economic feasibility of a cooling system is
through cost-benefit analysis in which the additional revenues are calculated as a result of cost
of electricity (COE) and rate of return (ROR) for additional MWh (Jones and Jacobs, 2002;
Grance et al., 2001).
The economic evaluation criterion may vary from one power producer to another. For some
producers, it may be revenues from enhanced capacity. For others, it may be the bonus for
meeting or exceeding capacity or avoiding any penalties for not meeting capacity. All these
factors contribute to total revenues and should be properly accounted for in economic
evaluation.
Table VII shows the capital cost for media and fog inlet air cooling system. They include the
initial installation investment cost, annual O&M cost, consumption water cost and consumption
fuel cost.
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
1500
M. AMERI, H. R. SHAHBAZIAN AND M. NABIZADEH
30
=10 %
T= 40°C and
T= 45°C and =5 %
25
20
T= 30°C and
=10 %
15
10
5
0
Fars
Rajaei
Ghom
Fars
Rajaei
Ghom
Fars
Rajaei
Temperature Decrease (Deg C)
14.45
16.92
16.79
19.05
23.68
23.54
20.97
26.94
25.87
Percent of Power Augmentation
10.63
11.81
9.98
16.93
18.46
16.39
18.59
20.44
17.93
Ghom
Figure 17. The comparison of temperature decrease and percent of gas turbine power
augmentation prediction for the media (Fars) and fog systems (Rajaie and Ghom)
installations at various ambient conditions.
Table VII. Final economical results for media evaporating system and fog system cases.
Case
Fars
Mean power increase (MW)
Annual generated power due to using cooling system
for 8 h per day for 4 months (kWh year1)
Annual power decrease due to using cooling system
(kWh year1)
Net power increase (kW year1)
Annual consumption of fuel due to using cooling
system (kN m3 year1)
Annual consumption of water due to using cooling
system (m3 year1)
Initial investment costs (US$)
Annual O&M costs (US$ year1)
Consumption fuel costs (US$ year1)
Consumption water costs (US$ year1)
Shahid Rajaie
11
10 912 000
10
9 920 000
2 102 400
Ghom
8
7 936 000
}
}
8 809 600
2281.6
9 920 000
2083.2
7 936 000
1517.7
10277.1
7285.2
9285.1
605 000
24 200
22 816
51 385
450 000
18 000
20 832
36 426
360 000
14 400
15 177
46 425
Payback period (year)
Cost of electricity (COE)
3 Cents kWh1
4 Cents kWh1
5 Cents kWh1
6.14
3.30
2.26
3.80
2.29
1.67
4.52
2.64
1.88
Rate of return (ROR, %)
24.37
36.90
31.24
For the economic calculations, the following assumptions have been made:
*
The investment cost is assumed to be 55 US$ kW1 and 45 US$ kW1 for media and fog
systems, respectively.
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
COMPARISON OF EVAPORATIVE INLET AIR COOLING SYSTEMS
*
*
*
*
*
1501
The O&M cost is assumed to be 3–5% of initial investment cost.
The installation cost of one purification plant for production of demineralized water for
fog system with capacity 100 m3 day1 is assumed to be 125 000 US$ and cost of water
consumption is assumed to be 5 US$ m3.
The actual fuel consumption cost in Iran is 0.02 US$ l1 for gas oil and 0.01 US$ m3 for
natural gas.
The current electricity price in Iran is 0.04 US$ kWh1.
For economical calculation we have considered 17% for domestic interest rate and 7% for
the foreign interest rate and 10 years for the equipment life.
Since the cooling system is operated in summer and under peak demand condition, its daily
operation time is assumed to be 8 h a day (although it’s actual operation time was much less
than this value due to some technical problems). If it is assumed that this system is operating for
4 months per year (May–September), the number of operation hours will be a 992 h year1.
Also the power decrease in gas turbines power due to the pressure drop (200 Pa) of media
system is considered in economical calculation.
Based on the economic analysis (Table VII), it is clear that both media evaporative and fog
systems are economical as their rates of return (24.37–36%) are higher than the domestic and
foreign rates of return (RORs) (i.e. 17 and 7%, respectively). However, the fog systems are more
economical in comparison with media evaporating system. There are some reasons for this
conclusion:
*
*
*
The ROR for media and fog systems are more than domestic interest rate but the ROR for
fog system is absolutely more than the ROR for media system.
The capital cost for media system is more than the capital cost for fog system. Therefore,
fog systems are more attractive than the media system at the first view.
The payback periods for fog system are shorter than media evaporating system for various
electricity costs.
Therefore, based on the economic evaluation, the best alternatives for the gas turbine power
augmentation can be both the fog and media inlet air cooling systems. However, one should
note that the media evaporating system is safer than the fog system due to the possibility of large
fog droplets entering the compressor if the fog nozzles types and their orientations are not
designed very well.
8. CONCLUSION
According to the feasibility study test results, using the media evaporating cooler in Fars
combined cycle power plant increased the gas turbine output power by 11 MW (or 14.5%).
Using the fogging system in Shahid Rajaie and Ghom combined cycle power plants enhanced
the gas turbine output power by 9.2 MW (or 11%) and 8.1 MW (8.9%). However, the actual
average power enhancement in the summer of 2004 was 11, 8.5 and 6.6 MW for Fars, Ghom and
Shahid Rajaie power plants, respectively. The results reveal that the media evaporative cooler
efficiency is the same as the fog evaporative system efficiency. However due to the fact that the
Fars climate conditions are more favourable, it has generated more enhanced power.
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
1502
M. AMERI, H. R. SHAHBAZIAN AND M. NABIZADEH
The results of technical and economic evaluation show that using the evaporative systems is
more cost effective than using new gas turbines for generating more power. Therefore, these
systems are very suitable for the dry central regions of Iran such as Fars, Ghom, Yazd, etc.
where the air temperature is high and the humidity ratio is low.
Also the capital costs are very cheap in comparison with the installation of the new gas
turbines (300 US$ kW1). The payback period for the application of the evaporative systems is
from 2 to 3 years.
NOMENCLATURE
m
’ air
m
’ water
Tdb
Twb
Zhumidifying
o
=air mass flow rate (kg s1)
=water consumption (kg s1)
=dry bulb temperature (8C)
=wet bulb temperature ( 8C)
=evaporative cooler saturation efficiency
=specific humidity (kgwater/kgdryair)
REFERENCES
AAF Power & Industrial. 2002. Gas Turbine Inlet Cooling. http://www.aafintl.com
Ameri M, Hejazi SH. 2004. The study of capacity enhancement of the Chabahar gas turbine installation using an
absorption chiller. Applied Thermal Engineering Journal 24(1):59–68.
Ameri M, Keshtgar A, Nabati H. 2004. Gas turbine power augmentation using fog inlet air-cooling system. Proceedings
of the ASME 7th Biennial Conference on Engineering Systems Design and Analysis, vol. 1, Manchester, U.K., 2004;
73–78.
ASHRAE Handbook (IP edn). 1992. HVAC Systems & Equipment.
Brook FJ. 1998. GE gas turbine performance characteristic. Report NO. GER-356H, GE power system.
Cyrus B, Meher-Homji CB, Mee III TR. 2000. Inlet fogging of gas turbine engines, part A: theory, psychometrics and
fog generation. Proceedings of the ASME Turbo Expo. 2000, Paper 2000GT-308, Munich.
Cyrus B, Meher-Homji CB, Mee III TR. 2002. Inlet fogging of gas turbine engines, part B: fog droplet sizing analysis
nozzle types, measurement and testing. Proceedings of the ASME Turbo Expo. 2002, Paper No. 2002 GT-30563,
Amsterdam, The Netherlands.
GE Energy. 2006. Library of Standard Gas Turbine. http://www.gepower.com
Grance D, Boncimino G, Stenzel W. 2001. Combined cycle plant optimization studies. Proceedings of the JPGC’01,
International Joint Power Generation Conference, The Netherlands, Paper No. JPGC2001/PWR-1930, New Orleans,
Louisiana.
Johnson RS. 1998. Theory and operation of evaporative coolers for industrial gas turbine installations. Presented at The
Gas Turbine and Aero Engine Congress, Amsterdam, Netherlands, ASME, 88-GT-41.
Jones C, Jacobs III JA. 2000. Economic and Technical Considerations for Combined-Cycle Performance–Enhancement
Options, GE Power Systems, GER-4200, Schenectady.
Kakaras E, Doukelis S, Karellas S. 2004. Compressor intake-air cooling in gas turbine plants. Energy Journal (Special
Issue) 29(12–15):2347–2358.
Kraneis W. 2000. Increased importance of evaporative coolers for gas turbine and combined-cycle power plants. VGB
Power Tech 80(9):22–25.
McNeilly JD. 1997. Test correction error for evaporative cooler equipped gas turbine power plants. Proceedings of the
1997 International Joint Power Generation Conference, Denver, U.S.A., Part 2 (of 2), vol. 32, 399–406.
MEE Industries. 2002. Economic benefits of replacing gas turbine media based evaporative cooling with inlet fogging
systems. AN-GT-205, MeeFog Technical Application Note, Mee Industries Inc.
Munters Co. 2001. CELdek 7060-15 Evaporative Cooling Pad. http://www.munters.com
Nabizadeh M, Keshtgar AR. 2004. The study of fog inlet air cooling system for the output power enhancement
of gas turbines. B.Sc. Thesis, Energy Engineering Department, Power & Water University of Technology,
Tehran, Iran.
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
COMPARISON OF EVAPORATIVE INLET AIR COOLING SYSTEMS
1503
Nixdorf M, Prelipceanu A, Hein D. 2002. Thermo-economic analysis of inlet air conditioning methods of a cogeneration
gas turbine plant. Proceedings of the ASME TURBO EXPO 2002: Ceramics, Industrial and Cogeneration Structures
and Dynamics, Amsterdam, Netherlands, IGTI, vol. 4 A, 403–412.
Omidvar B. 2001. Gas turbine inlet air cooling system. The 3rd Annual Australian Gas Turbine Conference, Melbourne,
Australia.
Shahbazian HR, Hoseinzadeh H. 2004. The study of media evaporating cooler for the output power enhancement
of fars gas turbines. B.Sc. Thesis, Energy Engineering Department, Power & Water University of Technology,
Tehran, Iran.
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
Int. J. Energy Res. 2007; 31:1483–1503
Published online 7 March 2007 in Wiley InterScience
(www.interscience.wiley.com) DOI: 10.1002/er.1315
Comparison of evaporative inlet air cooling systems to enhance
the gas turbine generated power
Mohammad Ameri*,y, H. R. Shahbazian and M. Nabizadeh
Combined Heat & Power Specialized Unit (CHP), Energy Engineering Department, Power & Water University
of Technology, PO Box 16765-1719 Tehran, Islamic Republic of Iran
SUMMARY
The gas turbine performance is highly sensitive to the compressor inlet temperature. The output of gas
turbine falls to a value that is less than the rated output under high temperature conditions. In fact increase
in inlet air temperature by 18C will decrease the output power by 0.7% approximately. The solution of this
problem is very important because the peak demand season also happens in the summer. One of the
convenient methods of inlet air cooling is evaporating cooling which is appropriate for warm and dry
weather. As most of the gas turbines in Iran are installed in such ambient conditions regions, therefore this
method can be used to enhance the performance of the gas turbines.
In this paper, an overview of technical and economic comparison of media system and fog system is
given. The performance test results show that the mean output power of Frame-9 gas turbines is increased
by 11 MW (14.5%) by the application of media cooling system in Fars power plant and 8.1 MW (8.9%)
and 9.5 MW (11%) by the application of fog cooling system in Ghom and Shahid Rajaie power plants,
respectively. The total enhanced power generation in the summer of 2004 was 2970, 1701 and 1340 MWh
for the Fars, Ghom and Shahid Rajaie power plants, respectively.
The economical studies show that the payback periods are estimated to be around 2 and 3 years for fog
and media systems, respectively. This study has shown that both methods are suitable for the dry and hot
areas for gas turbine power augmentation. Copyright # 2007 John Wiley & Sons, Ltd.
KEY WORDS:
gas turbine; inlet air cooling; evaporating cooling; media; fog; power augmentation
1. INTRODUCTION
Gas turbines are used widely in power generation, gas transfer stations and petrochemical
industries (GE Energy, 2006). The site ambient conditions, especially the temperature, have
great influence on gas turbines performance (Brook, 1998). Since the air density is decreased
during warm days, the mass flow rate through the turbine is decreased. Therefore, it causes a
*Correspondence to: Mohammad Ameri, Combined Heat & Power Specialized Unit (CHP), Energy Engineering
Department, Power & Water University of Technology, PO Box 16765-1719 Tehran, Islamic Republic of Iran.
y
E-mail: ameri [email protected]
Copyright # 2007 John Wiley & Sons, Ltd.
Received 29 April 2006
Revised 28 December 2006
Accepted 31 January 2007
1484
M. AMERI, H. R. SHAHBAZIAN AND M. NABIZADEH
Figure 1. T–S diagram for a hot day.
drop in the output power. Moreover, the compressor work increases due to the divergence of
constant pressure lines in T–S diagram (Figure 1). On the other hand, the compressor final
pressure decreases. Given the fact that the turbine inlet temperature is constant, it will reduce
the turbine work. As a result the net output of gas turbine falls. Moreover, there is a peak
demand of electricity in summer. Therefore, the gas turbine inlet air cooling is one of the useful
methods which can be applied for the gas turbine power enhancement.
Kraneis et al. (2000) studied the effects of an evaporative cooler on the available power plant
capacity with a detailed outline of the climatic conditions prevailing on the various continents.
Nixdorf et al. (2002) investigated the economic benefits of some different ambient air
conditioning methods for reducing the gas turbine intake air temperature in order to enhance
the gas turbine power. Johnson (1998) discussed the theory of evaporative cooling and explained
the evaporative cooler design, installation, operation, feed water quality, and the causes and
prevention of water carry over. Kakaras et al. presented a computer simulation of the
integration of an evaporative cooler and of the air-cooling system and discussed the effect of
ambient air temperature variation on the power output and efficiency (Kakaras, 2004). Ameri
et al. (2004) have studied the installation of fog inlet air cooling system for six Frame-5 (25 MW)
gas turbines. The results of that study showed that the output power of each gas turbine was
increased by 3 MW. Also, they showed that the fog system was very cheap in comparison with
the installation of new gas turbines. McNeilly (1997) presented a new method for the test
correction error of evaporative coolers. Ameri et al. (2004) presented a good overview of an
intake air-cooling system that used a steam absorption chiller and an air cooler to increase the
power output of Frame-6 gas turbines.
The fog and media evaporative inlet cooling are the most economical alternatives for hot and
dry areas. In fact the other techniques such as the absorption and vapour compression chiller
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
COMPARISON OF EVAPORATIVE INLET AIR COOLING SYSTEMS
1485
methods are not suitable for that ambient condition as they are very expensive alternatives. The
purpose of this paper is to present the state-of-the-art application of gas turbine evaporative
inlet air cooling systems for three large gas turbine power plants (Shahid Rajaie, Ghom and
Fars) in Iran’s electrical grid and to compare the technical and economical aspects of fog and
media inlet air cooling systems. Although one may find various papers in literature regarding the
topic, however, there is neither many reliable actual power plant test data available for the large
gas turbines nor there exist many unbiased comparisons between different evaporative inlet air
cooling systems. In fact, most of the similar results are given by commercial companies which
have not been justified by an independent research. Each company claims that its system either
fog or media is the best.
Following the discussion on the different inlet air cooling methods, the actual performance
test results are presented. Finally, the economic benefits of the different cooling systems are
explained.
2. GAS TURBINE INLET AIR COOLING METHODS
There are several inlet air cooling methods available for gas turbine power augmentation
(Omidvar, 2001). They can be classified into three types:
(1) thermal energy storage systems;
(2) refrigerated cooling system (utilizing absorption or mechanical refrigeration);
(3) evaporative coolers (media and fog).
3. EVAPORATIVE COOLERS OVERVIEW
Using the evaporative coolers will cause the maximum reduction in inlet air temperature if the
inlet air dry bulb temperature approaches the wet bulb temperature (relative humidity of 100%).
This can take place for dry and hot weather. There are two types of evaporating cooling:
3.1. Media evaporating cooling
Media surfaces are flexuous and consist of beehive-shaped cells which make up an evaporating
cooler. The inlet air can be cooled with surface evaporating by spraying water on these cells and
humidifying them. Increasing the contact area between water and air will cause the surface
evaporation to be more and faster. The flexuous and beehive-shaped cells increase the contact
area between water and air.
The following important points should be considered for the selection of a media system:
(a) Pressure drop in this system is more than other evaporating cooling systems. However,
this pressure drop has not much influence on the gas turbine output.
(b) Power consumption of this system is less than other systems.
(c) This system does not need demineralized water and can use the raw water. However, it is
better to use distilled water.
(d) Since this system needs periodical replacement of media (every 3 or 4 years), its
maintenance cost is more than other systems.
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
1486
M. AMERI, H. R. SHAHBAZIAN AND M. NABIZADEH
(e) The installation cost of this system is more expensive.
(f) If the system is installed in filter room, the shutdown period of the unit is considerable.
This system which uses a media surface to evaporate water is widely used for gas turbines
especially in dry and hot areas.
The performance of this system is based on water evaporation, which consumes thermal
energy by the latent heat of vaporization value and reduces the ambient temperature.
Figure 2 shows the schematic of media evaporating cooling system. Media system equipment
includes piping, programmable logic controller (PLC) control system and measuring equipment,
water distributing headers, media surfaces, mist eliminator, movable wall in the media case,
circulation pump, water tank, and blow down system.
3.1.1. Media surfaces. The material of these surfaces is cellulose fibre. A media evaporating
cooler, which consists of these surfaces, is like beehive. Figure 3 shows the media
surface. Water is spread over the media area, so the ratio of vaporization area (m2) to the
media volume is increased and the air cooling is improved. These surfaces are covered
with special chemical material to prevent corrosion. Since the material of media surfaces is
cellulose, they are flammable. However, they can be made from fibre glass surfaces which are
not flammable.
Each of these surfaces is installed with a certain angle (the angles are 15 and 458 with respect
to the horizontal line) to increase the contact area between water and air (Figure 4). The airflow
is parallel to the horizontal plane (AAF Power & Industrial, 2002).
3.1.2. Media evaporating cooler function. In this system the water is pumped from the tank
below the cooler to the distributing header above the cooler (Figure 2). This water is distributed
over the surfaces and humidifies them. The outlet water is gathered below the cooler and
drained to the water tank. These surfaces which are called media surfaces have a thickness
of about 20 cm or more and cover the entire cross section of inlet air duct or air room. The
compressor inlet air passes through the media surfaces and evaporates the water to the
Figure 2. Media evaporating cooling system.
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
COMPARISON OF EVAPORATIVE INLET AIR COOLING SYSTEMS
1487
Figure 3. Evaporative cooler media surfaces (Munters Co., 2001)
Figure 4. Certain angles of media surface installation (AAF Power & Industrial, 2002).
saturation limit. The remaining water is used for continuous purging and discharging of
the other materials from the media surfaces. The amount of circulating water should be at least
2–3 times of evaporated water.
To prevent the removing of water droplets from media surfaces and damaging the compressor
blades due to the high velocity of air, the air velocity should be limited. The cooler air is passed
through a mist eliminator after the evaporating cooler and the water droplets are removed.
Figure 5 shows the reduction of saturation efficiency of the evaporating cooler at high velocities
for Celdek media surfaces with various thicknesses (Munters Co., 2001). On the other hand as
the inlet air velocity increases the pressure drop in the system increases as well. Also Figure 5
shows the pressure drop in Celdek media surfaces with various thicknesses versus the inlet air
velocity. Based on these facts, it is obvious that if the inlet air velocity is high, the evaporating
cooler saturation efficiency is reduced and the pressure drop is increased. Therefore, a diffuser is
used at the inlet air duct of evaporative cooler to reduce the inlet air velocity. However, it should
be noted that the pressure drop is very small.
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
1488
M. AMERI, H. R. SHAHBAZIAN AND M. NABIZADEH
Figure 5. The media cooler saturation efficiency and its pressure drop versus air velocity for different media
width (width of media surfaces are 75; 100; 150; 200 and 300 mm) (Munters Co., 2001).
3.2. Fog system
High-pressure fogging of gas turbine inlets has been applied for 15 years (Cyrus et al., 2000,
2002). In essence, it generates droplets of sizes 5–20 mm which are injected into the air stream
where they evaporate and provide air cooling. The test data have shown that this process can be
100% effective (i.e. wet bulb temperature can be reached) even in humid regions. Fog is
generated by the application of high-pressure demineralized water between 70 and 200 bar to an
array of specially designed fog nozzles. A typical fog system consists of a series of high-pressure
pumps that are mounted on a skid, PLC-based control system with temperature and humidity
sensors, and array of fog nozzles installed in the inlet air duct (Figure 6).
The nozzle is usually made of 316 stainless steel (SS) and consists of a small orifice from 127
to 178 mm for gas turbine applications. The water emanating from this orifice impacts a specially
designed impaction pin that breaks up the jet into billions of micro fine fog droplets, whose sizes
are between 10 and 40 mm. The rate of evaporation of the droplets essentially depends on the
surface area of water exposed to the air.
Figure 7 shows the distribution of droplets diameter in a specific nozzle. Typically, the
plunger type pumps are used to produce 130–200 bar pressures for gas turbine inlet air fogging
systems. They are positive displacement ceramic-plunger stainless steel pumps with stainless
steel heads. The pumps can be turned on sequentially to control the amount of cooling. For
example, 8.48C drop in temperature may be managed in three 2.88C increments.
3.2.1. Cooling system control. The control system incorporates a PLC, which is typically
mounted on the high-pressure pump skid. Sensors are provided to measure relative humidity
and dry bulb temperature. Special programming codes use these measured parameters to
compute the ambient wet bulb temperature and the wet bulb depression (i.e. the difference
between the dry bulb and wet bulb temperatures). They quantify and control the amount of
evaporative cooling that is possible with the ambient conditions. The system turns on or off fog
cooling stages to match the ability of the ambient conditions to absorb water vapour. The
control system also monitors pump skid operating parameters such as water flow rates and
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
COMPARISON OF EVAPORATIVE INLET AIR COOLING SYSTEMS
1489
Figure 6. Schematic of fog inlet air cooling system.
Figure 7. Distribution of droplet diameter for a typical nozzle (Omidvar, 2001).
operating pressure, and provides alarms when these parameters are outside acceptable ranges.
Table I indicates the typical cooling control system stages.
3.2.2. Fog nozzles position in inlet duct. There are two main options for installing the inlet
fogging system, i.e. locating them either upstream or downstream of the filters.
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
1490
M. AMERI, H. R. SHAHBAZIAN AND M. NABIZADEH
Table I. Typical fog cooling control stages (four stages of 2.88C each).
Time
Dry bulb (8C)
Wet bulb (8C)
Difference (8C)
Stage ‘ON’
Cooling (8C)
9:00
10:00
11:00
12:00
21.1
23.9
26.7
30
20
20.6
21.1
21.1
1.1
3.3
5.6
8.9
None
1
2
3
}
2.8
5.6
8.4
(a) Upstream of the inlet filters: One advantage of positioning the fog nozzle manifold
upstream of the air filters is that the installation can be accomplished without gas turbine outage.
In this case, a fog droplet mist eliminator filter must be added downstream of the fog nozzle
manifold to remove any unevaporated droplet. By definition, the droplet filter would not allow
any fog intercooling. Typically about half the water droplets by the fog nozzles is captured by the
droplet mist eliminator and drained away. This type of system while used on some early
installations is rarely applied to gas turbine installations today. It requires more fog nozzles
and more water and it is generally more expensive to operate and install. However the turbine
operators, who have experienced excessive loading of inlet air filters, might find this option
a cost-effective one.
(b) Downstream of the inlet filters: The most common location for the high-pressure fog nozzle
manifold is downstream of the air filters and upstream of silencers and trash screens. Installation
of the fog system in this location requires an outage of 1–2 days and calls for only minor
modifications to the turbine inlet structure. This type of installation allows fog intercooling.
While the fog nozzle manifolds can be also installed downstream of the silencers, it is generally
considered best to locate them upstream of the silencers, as this would allow more residence time
for the fog droplets to evaporate. Fog nozzle manifolds are usually installed upstream of the
trash screens to avoid any possibility of foreign object damage (FOD).
4. COMPARISON OF MEDIA AND FOG INLET AIR COOLING SYSTEMS
Comparisons of media and fog system for cooling air of gas turbine are shown in Table II
(Grance et al., 2001). Care should be taken for using these cooling systems as they may change
the performance curve of the gas turbine towards surge point. Moreover if the humidity of air is
high, the power increase is limited due to the low efficiency of the evaporative system. Therefore,
the required cooling is not achieved.
5. THE EFFECT OF EVAPORATING COOLER ON THE COMPRESSOR INLET AIR
COOLING AND THE ENHANCEMENT OF GAS TURBINE OUTPUT POWER
The media and fog cooling systems enhance the output power by reducing the inlet air
temperature as follows:
(1) The density and the flow rate of air passing through the gas turbine are increased
by reducing the compressor inlet air temperature. Therefore, the output power is
enhanced.
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
COMPARISON OF EVAPORATIVE INLET AIR COOLING SYSTEMS
1491
Table II. Comparisons of media and fog system for cooling air of gas turbine (Grance et al., 2004).
Main parameter
Efficiency of humidifying (%)
Capital cost (US$ kW1 installed)
Quality of consumption water
Pressure drop (Pa)
Droplet diameter (mm)
Time of unit exit for
installation (days)
Change of unit structure
Another advantage
Media
Fog
85–90
45–55
Raw water
200
Less than 100
5–7
90–100
35–45
Fully demineralized
Very low
Less than 20
1–2
Need to change of filter room
Installation in filter room
Increasing the life of filter,
reducing the amount of NOx
Ability to produce 100% humidity,
lower consumption of water
(2) The consumption power of compressor is decreased as its inlet air temperature is reduced
(The consumption power of compressor is proportional to the inlet air temperature
directly). Therefore, the net power output of turbine increases.
(3) If the inlet air temperature is reduced, the exhaust flue gas temperature is reduced. This is
due to the fact that if all other parameters are kept constant, the energy balance causes
the flue gas temperature to reduce. Therefore, the error signal which is the difference
between exhaust gas temperature and its set point temperature is reduced. The turbine
control system increases the fuel flow rate and therefore the output power of gas turbine
is enhanced. The flue gas temperature returns to its set point value.
6. THE CHARACTERISTICS OF GAS TURBINE UNITS AND INSTALLED COOLING
SYSTEMS FOR FARS, SHAHID RAJAIE AND GHOM POWER PLANTS
There are many gas turbines in Iran favoured for coping with the peak electricity demand of the
utilities due to their specific properties. However due to the high ambient temperature in the hot
seasons, the output power and efficiency of gas turbines decrease considerably.
The power output of the gas turbine is as low as 70% of its’ rated output when the
temperature increases in the summer and the electricity is most needed for air conditioning.
Each of these facilities serves as a vital equipment for the Iran electricity grid. For existing
installations the possible output enhancement can be determined by comparing the turbine
output at the design conditions with the output recorded at the desired inlet air temperature.
Essentially the inlet air cooling gives approximately winter performance during hot climatic
conditions. Two Frame-9 (PG9171E) gas turbines with the rated output of 99.3 MW in Fars,
two Frame-9 (PG9171E) gas turbines with the rated power of 98 MW in Shahid Rajaie and two
gas turbines (701D) with the rated output of 100 MW in Ghom power plants are selected for
the inlet air cooling system installation. Table III shows the characteristics of gas turbine units
and installed cooling systems. The detailed specifications of those gas turbines are given in the
technical documents of Fars, Shahid Rajaie and Ghom combined cycle power plants and are
presented by Shahbazian and Hoseinzadeh (2004) and Nabizadeh and Keshtgar (2004). All gas
turbines use the natural gas as their fuel. However, the heating values of their fuels are different.
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
1492
M. AMERI, H. R. SHAHBAZIAN AND M. NABIZADEH
Table III. Characteristics of gas turbine units and installed cooling systems.
Power plant
Fars
Shahid Rajaie
Ghom
No. of
case study
units
2
2
2
Power output at Min. & Max.
site conditions Temperatures
(MW)
(8C)
99.3
98
100
14 ! 43
10 ! 40
12 ! 45
Type of installed
cooling system
Design condition
for installed
cooling system
Media
Fog
Fog
388C and j=20%
408C and j=10%
458C and j=8%
7. TECHNICAL AND ECONOMIC EVALUATION OF APPLICATION OF MEDIA
AND FOG COOLING SYSTEM FOR GAS TURBINE INLET AIR COOLING
7.1. Prediction of gas turbine power enhancement using evaporative cooling system performance at
different ambient conditions
The evaporative cooling system is designed for a specific ambient condition. The design point is
not usually the maximum ambient temperature and minimum relative humidity temperature at
which the evaporative cooler has the maximum capacity. In fact as the power enhancement is
most required at the peak load, the design point is selected either based on this point or the
combination of this point and the maximum capacity point. Although it would be useful to test
the system at the design ambient conditions, however, it is impossible due to practical reasons.
Therefore, it is necessary to predict the evaporative cooler performance at the different ambient
conditions. This is usually important to check the guarantee requirements. In this section of the
paper, a method is presented to predict the gas turbine power enhancement.
The function of evaporating cooling system is cooling air by humidifying it in an adiabatic
process. During this process the wet bulb of air is constant. Figure 8 shows this process on the
psychometrics chart.
The saturation (or humidifying) efficiency of the system is defined as follows (Ameri et al.,
2004; AAF, 2002):
T1db T2db
Zhumidifying ¼
ð1Þ
T1db T1wb
The amount of required water in this system is determined from the following equation:
m
’ water ¼ m
’ air ðo2 o1 Þ
ð2Þ
where m
’ air is the inlet mass flow rate of dry air into the gas turbine.
As an example, the maximum power recovery under Fars power plant condition is determined
as follows.
At T1db ¼ 358C and relative humidity of j1 ¼ 20%; the specific humidity is o1 ¼ 0:008464:
Assuming maximum saturation efficiency (i.e. 100%, the best conditions for the evaporating
cooling system performance) the minimum achievable temperature is the wet bulb temperature.
Therefore, T2db ¼ T1wb ¼ 17:868C and the relative humidity and specific humidity are j1 ¼
100% and o2 ¼ 0:015568; respectively. Since the altitude of this power plant is 1530 m from the
sea level, the above values should be corrected. According to the tables of altitude correction
(ASHRAE, 1992), one can estimate Do to be 0.007104. The mass flow rate of inlet air to the
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
COMPARISON OF EVAPORATIVE INLET AIR COOLING SYSTEMS
1493
Figure 8. Cooling process on the psychometric diagram.
Comp. Inlet Tdb
Ambient Pressure
Evap.
Ambient Temperature
Cooler
Comp. Inlet Twb
Relative Humidity
System
Comp. Inlet Pressure
Gas Turbine
Power Output
Exhaust Gas Temp.
Figure 9. The schematic for the prediction of gas turbine power enhancement by evaporative cooling
system at different ambient conditions.
compressor is 406 kg s1 under design condition. Therefore, the required water mass flow rate is
m
’ water ¼ 2:88 kg s1 : According to the above values and the diagram of gas turbine output
power versus the inlet temperature for GE-F9 units of Fars power plant, the augmented power is
calculated (Shahbazian and Hoseinzadeh, 2004). The output power of gas turbine at the design
point is 79.13 MW and after the operation of media system it is estimated to be 90.18 MW.
Therefore, the maximum recovered power is 11.05 MW or 13.96%. For other power plants and
different ambient conditions the maximum power enhancement is determined by this method as
well (Nabizadeh and Keshtgar, 2004). It should be noted that the evaporative cooler efficiency is
estimated from the test results. Figure 9 shows the schematic for the prediction of gas turbine
power enhancement using the evaporative cooling system. The equations of evaporative cooler
system box are based on thermodynamical properties and evaporative cooler saturation
efficiency. The evaporative cooler efficiency is calculated using the test point results. The
equations of gas turbine box are based on the characteristics curve of the specific gas turbine.
A simple computer code has been prepared to calculate the temperature reduction and power
enhancement for different cases (Shahbazian and Hoseinzadeh, 2004; Nabizadeh and Keshtgar,
2004).
7.2. Performance test results and discussion
Since the evaporative systems performance test is an important issue in the feasibility study, the
complete performance tests have been done on all units. The main parameters in those tests were
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
1494
M. AMERI, H. R. SHAHBAZIAN AND M. NABIZADEH
the increase in output power and also the temperature and the humidity ratio at the compressor
inlet. Regarding these parameters, measuring points were specified. Since this system is used at
different times both day and night (especially in the summer nights when the ambient
temperatures are high enough that it may make the use of evaporating cooling units necessary)
the performance tests at night were also carried out to evaluate the performance of the units at
different conditions.
The correct functioning of sensors installed on PLC should be tested. Therefore, the
temperature and the air humidity data were also measured by another instrument and then the
results were compared to the data of PLC. The test procedure is as follows:
*
*
*
*
Starting up gas turbines and turning their automatic control on, letting each unit to reach
its full load (base load) (at this time 15-min waiting period was regarded for assurance that
the system had reached the steady-state operating conditions).
Recording parameters which were needed for performance test.
Turning on the cooling system and waiting for 20 min until the operating conditions
approach steady-state conditions.
Controlling and recording necessary parameters.
Table IV presents the results of performance tests of media evaporative cooling system for
units 1 and 2 of Fars power plant in August 2004.
Also, Tables V and VI present the results of performance tests of fogging cooling system for
units 1 and 2 of Shahid Rajaie and Ghom power plants in July and June 2004, respectively. It
should be noted the maximum calibration error for the output power measurement is 0.1%.
Table IV. The performance test results of media for units 1, 2 of Fars power plant (August 2004)
(Shahbazian and Hosseinzadeh, 2004).
GE (Frame-9) gas turbine unit 1
Parameters
Ambient
temp. (8C)
Relative
humidity (%)
Ambient
pressure (kPa)
Comp. inlet
air temp. (8C)
Comp. output
air temp. (8C)
Comp. output
air pressure
(bar)
Exhaust gas
temp. (8C)
Power
output (MW)
GE (Frame-9) gas turbine unit 2
After
Before
After
Before
operation operation
operation operation
Percent
Percent of media of media
of media of media
system Variation variation
system Variation variation system
system
38.17
38.27
8.3
8.2
0.08
1
83.85
83.82
0.023
40.6
22.66
371
8.23
559.6
76.6
347
8.92
549
87.71
Copyright # 2007 John Wiley & Sons, Ltd.
0.103
38.37
38.60
0.23
0.074
8.03
8.33
0.3
3.73
0.028
83.95
83.90
0.05
17.93
5.72
38
20
24
3.73
0.069
10.66
11.11
0.033
352
1.22
14.5
373.66
348.66
8.87
9.42
551
81.48
540
92.29
0.06
18
5.78
25
3.87
0.65
11
10.81
7.45
1.33
13.27
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
1495
COMPARISON OF EVAPORATIVE INLET AIR COOLING SYSTEMS
Table V. The performance test results of fog for units 1 & 2 of Shahid Rajaie power plant (July 2004)
(Nabizadeh and Keshtgar, 2004).
GE (Frame-9) gas turbine unit 1
Parameters
Ambient
temp. (8C)
Relative
humidity (%)
Ambient
pressure (mbar)
Comp. inlet
air temp. (8C)
Comp. output
air temp. (8C)
Comp. output
air pressure (bar)
Exhaust gas
temp. (8C)
Consumption
fuel rate
(m3 min1)
Power
output (MW)
GE (Frame-9) gas turbine unit 2
After
Before
After
Before
operation operation
operation operation
of fog
Percent
of fog
of fog
Percent
of fog
system Variation variation
system Variation variation system
system
35
35.33
0.33
0.107
34.66
35
0.33
0.107
11
11
0.0
0.0
11
11
0.0
0.0
868.6
868.7
0.1
0.012
868.6
868.7
0.1
0.012
33
19
32.66
16
16.66
0.054
363.33
345
18.33
2.88
361.33
341.33
0.48
5.44
8.75
9.26
0.92
8.823
9.303
558.33
550.66
7.66
463.5
498.06
34
86.55
95.03
8.48
7.44
9.79
14
0.046
20
3.15
0.51
5.86
560
552
8
480.1
519.53
39.36
8.19
92.22
9.2
11.08
83.02
0.96
Figures 10–14 show the gas turbine inlet air temperature reduction, power increase and
output power prediction for the different ambient conditions using the media evaporative cooler
or fog systems. The test point and the design points are also shown for reference. The maximum
error which is the difference between the actual (i.e. test) measured power and the predicted
power value is estimated to be around 2%. According to Figure 10 for media evaporating
system in Fars power plant, the inlet temperature drop increases due to the air relative humidity
decrease and temperature increase. For example, if the ambient temperature is 388C and the
relative humidity is 8%, the inlet air temperature is decreased by at least 178C using media
cooling system. Also according to Figures 11 and 12 the output power is enhanced by
11.11 MW. On the other hand, according to Figure 13 for fog cooling system in Ghom power
plant (for the ambient temperature of 338C and the relative humidity 16%) the inlet air
temperature is decreased by at least 168C. The output power is enhanced by 8.1 MW (Figures 14
and 15).
It should be noted that the power augmentation are clearly the results at the time of testing. If
an entire summer season or an entire year is considered, the overall power production increase
may be estimated. Figure 16 presents the total power generation for those three power plants in
the summer of 2004. It shows that the maximum power enhancement for the Fars power plant is
2970 MWh which is much more than the other two power plants i.e. 1701 MWh for Ghom and
1340 MWh for Shahid Rajaie power plants. It is 74.1 and 121.6% larger than those two other
power plants. In fact although the media evaporative cooler efficiency (85%) is generally a little
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
1496
M. AMERI, H. R. SHAHBAZIAN AND M. NABIZADEH
Table VI. The performance test results of fog for units 1 & 2 of Ghom power plant (June 2004)
(Nabizadeh and Keshtgar, 2004).
MW701D gas turbine unit 1
Parameters
Ambient
temp. (8C)
Relative
humidity (%)
Ambient
pressure (kPa)
Comp. inlet
air temp. (8C)
Comp. output
air temp. (8C)
Comp. output
air pressure
(kPa)
Exhaust gas
temp. (8C)
Fuel consumption
rate (kN m3 h1)
Power
output (MW)
MW-701D gas turbine unit 2
After
Before
After
Before
operation operation
operation operation
of fog
Percent
of fog
of fog
Percent
of fog
system Variation variation
system Variation variation system
system
31.96
32.66
16.66
15.66
88.87
88.87
0.00
31.8
16.58
415.66
404.06
1080
0.7
32
33
1
0.33
16
16
0.0
0.00
0.00
88.87
88.87
0.0
0.00
15.22
4.99
31.83
15.76
11
1.6
413.66
40
3.7
1
1120
528.93
523.7
0.23
6
1068
16.06
5.27
13.26
1.93
1113.33
45.33
4.24
400.4
5.22
0.65
528.76
523.63
5.13
0.64
29.45
30.76
1.31
4.44
28.60
30.14
1.53
5.37
94.93
100.53
5.60
5.89
90.1
98.2
8.1
8.99
Gas Turbine 1 in Fars Power Plant
Temperature Decrease (C)
26
Test Point
23
=5%
20
=8.3%
Design Point
17
=15%
14
=20%
11
=25%
8
=30%
5
15
20
25
30
35
40
45
50
55
Inlet Air Temperature (C)
Figure 10. Temperature decrease prediction using media system at various ambient conditions.
bit less than the fog system (90%), however, the climate in the Fars is more favourable for
installation of the evaporative cooling system. The high temperature, less humidity and more
hot hours per each day are the main reasons for higher power augmentation at the Fars site.
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
COMPARISON OF EVAPORATIVE INLET AIR COOLING SYSTEMS
1497
Gas Turbine 1 in Fars Power Plant
16
Power Increase (MW)
Test Point
14
ϕ = 5%
12
ϕ = 8.3%
Design Point
10
ϕ = 15%
ϕ = 20%
8
ϕ = 25%
6
ϕ = 30%
4
15
20
25 30 35 40 45
Inlet Air Temperature (C)
50
55
Figure 11. Power increase prediction using media system at various ambient conditions.
Gas Turbine 1 in Fars Power Plant
100
=5%
Design Point
96
Power (MW)
=8.3%
Test Point
92
=15%
88
=20%
84
=25%
=30%
80
76
15
20
25
30
35
40
45
50
55
Inlet Air Temperature (C)
Figure 12. Gas turbine output power prediction using media system at various ambient conditions.
Moreover in order to compare the media and fog system performances, Figure 17 presents the
temperature decrease and percent of power augmentation prediction due to installation of
media and fog system at three power plants (i.e. Fars, Rajaie and Ghom) for different ambient
conditions. It shows that the power augmentation for the media system (Fars) and the fog
system (Ghom and Rajaie) are almost the same at the same ambient conditions. In fact the
evaporative cooler efficiencies for the media and fog system are almost the same. These results
are quite different from the MEE Industries (2002) report (i.e. one of the major fog system
manufacturer) for a Frame 7111EA gas turbine. In fact, MEE has assumed an efficiency of
80–85% for the media evaporative cooling system and up to 100% efficiency for the fog system.
Therefore, MEE concluded that the percent of power boost attained by the use of fogger over
media type cooler power was up to 2.2% for different ambient temperatures and humidity
ratios. The reason for this difference can be due to the fact that for the fog system, the design of
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
1498
M. AMERI, H. R. SHAHBAZIAN AND M. NABIZADEH
Gas Turbine 2 in Ghom Power Plant
Temperature Decrease (C)
29
Design Point
= 5%
26
= 8%
Test Point
23
=10%
20
=16%
17
=20%
14
=25%
11
8
15
20
25
30
35
40
45
50
55
Inlet Air Temperature (C)
Figure 13. Temperature decrease prediction using fog system at various ambient conditions.
Gas Turbine 2 in Ghom Power Plant
18
Design Point
Power Increase (MW)
16
= 5%
14
Test Point
= 8%
12
=10%
10
=16%
8
=20%
6
=25%
4
15
20
25 30 35 40 45
Inlet Air Temperature (C)
50
55
Figure 14. Power increase prediction using fog system at various ambient conditions.
the nozzle manifold, type of nozzle and inlet duct length from the manifold to the compressor
inlet are the critical factors which affects the evaporative cooling system performance. In fact,
our fog systems have obtained the efficiency of 90% compared with the excellent MEE fog
system with the efficiency of up to 100%. Moreover, our media evaporative cooler system has
a very good design. Therefore, it has been able to achieve an efficiency of approximately 90%
(i.e. the same as our fog system efficiency).
7.3. Economic evaluation and discussion
The cost of an inlet cooling system is often evaluated in terms of US$ kW1. This can be
misleading because the output enhancement as a result of inlet air cooling varies with the
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
COMPARISON OF EVAPORATIVE INLET AIR COOLING SYSTEMS
Gas Turbine 2 in Ghom Power Plant
108
=5%
Test Point
105
Power (MW)
1499
Test Point
102
Design Point
=8%
=10%
=16%
99
=20%
96
=25%
93
90
15
20
25 30 35 40 45
Inlet Air Temperature (C)
50
55
Figure 15. Gas turbine output power prediction using fog system at various ambient conditions.
Figure 16. The comparison of gas turbine power augmentation (kWh) using the evaporative cooling
systems for the Fars, Ghom and Rajaie power plants during summer 2004.
ambient temperature. A better way of evaluating the economic feasibility of a cooling system is
through cost-benefit analysis in which the additional revenues are calculated as a result of cost
of electricity (COE) and rate of return (ROR) for additional MWh (Jones and Jacobs, 2002;
Grance et al., 2001).
The economic evaluation criterion may vary from one power producer to another. For some
producers, it may be revenues from enhanced capacity. For others, it may be the bonus for
meeting or exceeding capacity or avoiding any penalties for not meeting capacity. All these
factors contribute to total revenues and should be properly accounted for in economic
evaluation.
Table VII shows the capital cost for media and fog inlet air cooling system. They include the
initial installation investment cost, annual O&M cost, consumption water cost and consumption
fuel cost.
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
1500
M. AMERI, H. R. SHAHBAZIAN AND M. NABIZADEH
30
=10 %
T= 40°C and
T= 45°C and =5 %
25
20
T= 30°C and
=10 %
15
10
5
0
Fars
Rajaei
Ghom
Fars
Rajaei
Ghom
Fars
Rajaei
Temperature Decrease (Deg C)
14.45
16.92
16.79
19.05
23.68
23.54
20.97
26.94
25.87
Percent of Power Augmentation
10.63
11.81
9.98
16.93
18.46
16.39
18.59
20.44
17.93
Ghom
Figure 17. The comparison of temperature decrease and percent of gas turbine power
augmentation prediction for the media (Fars) and fog systems (Rajaie and Ghom)
installations at various ambient conditions.
Table VII. Final economical results for media evaporating system and fog system cases.
Case
Fars
Mean power increase (MW)
Annual generated power due to using cooling system
for 8 h per day for 4 months (kWh year1)
Annual power decrease due to using cooling system
(kWh year1)
Net power increase (kW year1)
Annual consumption of fuel due to using cooling
system (kN m3 year1)
Annual consumption of water due to using cooling
system (m3 year1)
Initial investment costs (US$)
Annual O&M costs (US$ year1)
Consumption fuel costs (US$ year1)
Consumption water costs (US$ year1)
Shahid Rajaie
11
10 912 000
10
9 920 000
2 102 400
Ghom
8
7 936 000
}
}
8 809 600
2281.6
9 920 000
2083.2
7 936 000
1517.7
10277.1
7285.2
9285.1
605 000
24 200
22 816
51 385
450 000
18 000
20 832
36 426
360 000
14 400
15 177
46 425
Payback period (year)
Cost of electricity (COE)
3 Cents kWh1
4 Cents kWh1
5 Cents kWh1
6.14
3.30
2.26
3.80
2.29
1.67
4.52
2.64
1.88
Rate of return (ROR, %)
24.37
36.90
31.24
For the economic calculations, the following assumptions have been made:
*
The investment cost is assumed to be 55 US$ kW1 and 45 US$ kW1 for media and fog
systems, respectively.
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
COMPARISON OF EVAPORATIVE INLET AIR COOLING SYSTEMS
*
*
*
*
*
1501
The O&M cost is assumed to be 3–5% of initial investment cost.
The installation cost of one purification plant for production of demineralized water for
fog system with capacity 100 m3 day1 is assumed to be 125 000 US$ and cost of water
consumption is assumed to be 5 US$ m3.
The actual fuel consumption cost in Iran is 0.02 US$ l1 for gas oil and 0.01 US$ m3 for
natural gas.
The current electricity price in Iran is 0.04 US$ kWh1.
For economical calculation we have considered 17% for domestic interest rate and 7% for
the foreign interest rate and 10 years for the equipment life.
Since the cooling system is operated in summer and under peak demand condition, its daily
operation time is assumed to be 8 h a day (although it’s actual operation time was much less
than this value due to some technical problems). If it is assumed that this system is operating for
4 months per year (May–September), the number of operation hours will be a 992 h year1.
Also the power decrease in gas turbines power due to the pressure drop (200 Pa) of media
system is considered in economical calculation.
Based on the economic analysis (Table VII), it is clear that both media evaporative and fog
systems are economical as their rates of return (24.37–36%) are higher than the domestic and
foreign rates of return (RORs) (i.e. 17 and 7%, respectively). However, the fog systems are more
economical in comparison with media evaporating system. There are some reasons for this
conclusion:
*
*
*
The ROR for media and fog systems are more than domestic interest rate but the ROR for
fog system is absolutely more than the ROR for media system.
The capital cost for media system is more than the capital cost for fog system. Therefore,
fog systems are more attractive than the media system at the first view.
The payback periods for fog system are shorter than media evaporating system for various
electricity costs.
Therefore, based on the economic evaluation, the best alternatives for the gas turbine power
augmentation can be both the fog and media inlet air cooling systems. However, one should
note that the media evaporating system is safer than the fog system due to the possibility of large
fog droplets entering the compressor if the fog nozzles types and their orientations are not
designed very well.
8. CONCLUSION
According to the feasibility study test results, using the media evaporating cooler in Fars
combined cycle power plant increased the gas turbine output power by 11 MW (or 14.5%).
Using the fogging system in Shahid Rajaie and Ghom combined cycle power plants enhanced
the gas turbine output power by 9.2 MW (or 11%) and 8.1 MW (8.9%). However, the actual
average power enhancement in the summer of 2004 was 11, 8.5 and 6.6 MW for Fars, Ghom and
Shahid Rajaie power plants, respectively. The results reveal that the media evaporative cooler
efficiency is the same as the fog evaporative system efficiency. However due to the fact that the
Fars climate conditions are more favourable, it has generated more enhanced power.
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
1502
M. AMERI, H. R. SHAHBAZIAN AND M. NABIZADEH
The results of technical and economic evaluation show that using the evaporative systems is
more cost effective than using new gas turbines for generating more power. Therefore, these
systems are very suitable for the dry central regions of Iran such as Fars, Ghom, Yazd, etc.
where the air temperature is high and the humidity ratio is low.
Also the capital costs are very cheap in comparison with the installation of the new gas
turbines (300 US$ kW1). The payback period for the application of the evaporative systems is
from 2 to 3 years.
NOMENCLATURE
m
’ air
m
’ water
Tdb
Twb
Zhumidifying
o
=air mass flow rate (kg s1)
=water consumption (kg s1)
=dry bulb temperature (8C)
=wet bulb temperature ( 8C)
=evaporative cooler saturation efficiency
=specific humidity (kgwater/kgdryair)
REFERENCES
AAF Power & Industrial. 2002. Gas Turbine Inlet Cooling. http://www.aafintl.com
Ameri M, Hejazi SH. 2004. The study of capacity enhancement of the Chabahar gas turbine installation using an
absorption chiller. Applied Thermal Engineering Journal 24(1):59–68.
Ameri M, Keshtgar A, Nabati H. 2004. Gas turbine power augmentation using fog inlet air-cooling system. Proceedings
of the ASME 7th Biennial Conference on Engineering Systems Design and Analysis, vol. 1, Manchester, U.K., 2004;
73–78.
ASHRAE Handbook (IP edn). 1992. HVAC Systems & Equipment.
Brook FJ. 1998. GE gas turbine performance characteristic. Report NO. GER-356H, GE power system.
Cyrus B, Meher-Homji CB, Mee III TR. 2000. Inlet fogging of gas turbine engines, part A: theory, psychometrics and
fog generation. Proceedings of the ASME Turbo Expo. 2000, Paper 2000GT-308, Munich.
Cyrus B, Meher-Homji CB, Mee III TR. 2002. Inlet fogging of gas turbine engines, part B: fog droplet sizing analysis
nozzle types, measurement and testing. Proceedings of the ASME Turbo Expo. 2002, Paper No. 2002 GT-30563,
Amsterdam, The Netherlands.
GE Energy. 2006. Library of Standard Gas Turbine. http://www.gepower.com
Grance D, Boncimino G, Stenzel W. 2001. Combined cycle plant optimization studies. Proceedings of the JPGC’01,
International Joint Power Generation Conference, The Netherlands, Paper No. JPGC2001/PWR-1930, New Orleans,
Louisiana.
Johnson RS. 1998. Theory and operation of evaporative coolers for industrial gas turbine installations. Presented at The
Gas Turbine and Aero Engine Congress, Amsterdam, Netherlands, ASME, 88-GT-41.
Jones C, Jacobs III JA. 2000. Economic and Technical Considerations for Combined-Cycle Performance–Enhancement
Options, GE Power Systems, GER-4200, Schenectady.
Kakaras E, Doukelis S, Karellas S. 2004. Compressor intake-air cooling in gas turbine plants. Energy Journal (Special
Issue) 29(12–15):2347–2358.
Kraneis W. 2000. Increased importance of evaporative coolers for gas turbine and combined-cycle power plants. VGB
Power Tech 80(9):22–25.
McNeilly JD. 1997. Test correction error for evaporative cooler equipped gas turbine power plants. Proceedings of the
1997 International Joint Power Generation Conference, Denver, U.S.A., Part 2 (of 2), vol. 32, 399–406.
MEE Industries. 2002. Economic benefits of replacing gas turbine media based evaporative cooling with inlet fogging
systems. AN-GT-205, MeeFog Technical Application Note, Mee Industries Inc.
Munters Co. 2001. CELdek 7060-15 Evaporative Cooling Pad. http://www.munters.com
Nabizadeh M, Keshtgar AR. 2004. The study of fog inlet air cooling system for the output power enhancement
of gas turbines. B.Sc. Thesis, Energy Engineering Department, Power & Water University of Technology,
Tehran, Iran.
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
COMPARISON OF EVAPORATIVE INLET AIR COOLING SYSTEMS
1503
Nixdorf M, Prelipceanu A, Hein D. 2002. Thermo-economic analysis of inlet air conditioning methods of a cogeneration
gas turbine plant. Proceedings of the ASME TURBO EXPO 2002: Ceramics, Industrial and Cogeneration Structures
and Dynamics, Amsterdam, Netherlands, IGTI, vol. 4 A, 403–412.
Omidvar B. 2001. Gas turbine inlet air cooling system. The 3rd Annual Australian Gas Turbine Conference, Melbourne,
Australia.
Shahbazian HR, Hoseinzadeh H. 2004. The study of media evaporating cooler for the output power enhancement
of fars gas turbines. B.Sc. Thesis, Energy Engineering Department, Power & Water University of Technology,
Tehran, Iran.
Copyright # 2007 John Wiley & Sons, Ltd.
Int. J. Energy Res. 2007; 31:1483–1503
DOI: 10.1002/er
