Kalina Exergy Analysis of A Dual Level Binary Geothermal Power Plant 2002 Geothermics
Kalina Exergy Analysis of A Dual Level Binary Geothermal Power Plant 2002 Geothermics
www.elsevier.com/locate/geothermics
Abstract
Exergy analysis of a 12.4 MW existing binary geothermal power plant is performed using
actual plant data to assess the plant performance and pinpoint sites of primary exergy
destruction. Exergy destruction throughout the plant is quantified and illustrated using an
exergy flow diagram, and compared to the energy flow diagram. The causes of exergy
destruction in the plant include the exergy of the working fluid lost in the condenser, the
exergy of the brine reinjected, the turbine-pump losses, and the preheater–vaporizer losses.
The exergy destruction at these sites accounts for 22.6, 14.8, 13.9, and 13.0% of the total
exergy input to the plant, respectively. Exergetic efficiencies of major plant components are
determined in an attempt to assess their individual performances. The exergetic efficiency of
the plant is determined to be 29.1% based on the exergy of the geothermal fluid at the
vaporizer inlet, and 34.2% based on the exergy drop of the brine across the vaporizer–pre-
heater system (i.e. exergy input to the Rankine cycle). For comparison, the corresponding
thermal efficiencies for the plant are calculated to be 5.8 and 8.9%, respectively.
# 2002 CNR. Published by Elsevier Science Ltd. All rights reserved.
Keywords: Geothermal; Power; Binary plant; Exergy; Exergetic efficiency; Nevada; USA
1. Introduction
Most of the world’s geothermal power plants were built in the 1970s and 1980s
following the 1973 oil crisis. The urgency to generate electricity from alternative
energy sources and the fact that geothermal energy was essentially free adversely
affected careful designs of power plants to maximize their efficiency for a given
0375-6505/02/$22.00 # 2002 CNR. Published by Elsevier Science Ltd. All rights reserved.
PII: S0375-6505(02)00032-9
710 M. Kanoglu / Geothermics 31 (2002) 709–724
Nomenclature
Subscripts
0 dead state
I level I
II level II
cond condenser
pp pinch-point
pre preheater
turb turbine
vap vaporizer
geothermal resource. Binary plants were introduced in the last two decades while
combined flash (or steam) processes coupled to binary plants have essentially
appeared only in the last decade (Phair, 1994). Although new geothermal power
plants are currently being built with advanced technologies, not many more geo-
thermal power plants are expected to be built soon. The thermal efficiencies of con-
ventional combustion-based power plants have increased significantly in recent years
with the help of the combined cycle. Power generation from geothermal energy has
become less and less cost-competitive with combustion-based power generation
technologies as a result. There is, however, a tremendous potential to improve the
performance of many existing geothermal power plants by optimizing the operating
conditions, upgrading inefficient components, retro-fitting, and incorporating co-gen-
eration into the existing power plant. Examples of these studies can be found in Kano-
glu and Cengel (1999a, b), and Kanoglu et al. (1998). Other authors such as Vasil’ev et
al. (1996), Yuan and Michaelides (1993), Paloso and Mohanty (1993), Subbiah and
Natarejen (1988), and Milora and Tester (1976), present studies on the thermodynamic
M. Kanoglu / Geothermics 31 (2002) 709–724 711
2. Plant operation
The geothermal power plant analyzed is a binary design plant that generates 12.4
MWe net electricity from seven identical paired units. Full power production started in
April 1989. The plant operates in a closed loop with no environmental discharge and
100% reinjection of geothermal fluid. The modular power plant operates on a liquid-
dominated resource at 163 C. It utilizes dry-air condensers to condense the working
fluid, so no fresh water is consumed. The geothermal field includes four production
wells and three reinjection wells. The plant uses isopentane as the working (binary)
fluid. Isopentane circulates in a closed cycle, which is based on the Rankine cycle.
The plant is designed to operate with seven paired units of Levels I and II energy
converters. The schematic layout of the plant in Fig. 1 shows only one representative
unit. The heat source for the plant is the flow of geothermal water (brine) entering
the plant at 163 C with a total mass flow rate of 338.94 kg/s. Geothermal fluid
remains as a liquid throughout the plant. This flow is fed equally to the seven
712 M. Kanoglu / Geothermics 31 (2002) 709–724
vaporizers of Level I. Therefore, each unit receives 48.42 kg/s mass flow rate. The
brine then exits Level I vaporizers at approximately 131 C and is directly fed to the
paired Level II vaporizers where the brine is cooled to 100 C. The brine is then
divided equally and flows in parallel to the preheaters of Levels I and II. Subse-
quently, the Levels I and II preheaters extract additional heat from the brine, drop-
ping the temperature of the brine to 68 and 65 C, respectively. The brine leaving the
preheaters is directed to the reinjection wells where it is reinjected back into the
ground.
M. Kanoglu / Geothermics 31 (2002) 709–724 713
In Level I, 19.89 kg/s of working fluid circulates through the cycle. The working
fluid enters the preheater at 32 C and leaves at about 98 C. It then enters the
vaporizer where it is evaporated at 133 C and superheated to 136 C. The working
fluid then passes through the turbine. It exhausts to an air-cooled condenser at
about 85 C where it condenses to a temperature of 31 C. Approximately 530 kg/s
air at an ambient temperature of 13 C is required to absorb the heat yielded by the
working fluid. This raises the air temperature to 29 C. The working fluid is pumped
to preheater pressure to complete the Rankine cycle. The Level I isopentane cycle on
a T-s diagram is shown in Fig. 2.
In Level II, 21.92 kg/s of working fluid cycles through the loop. The working fluid
enters the preheater at 27 C and leaves at 94 C. It then enters the vaporizer where
it is evaporated at 98 C and slightly superheated to 99 C. The working fluid passes
through the turbine, and then exhausts to the condenser at about 65 C where it
condenses to a temperature of 27 C. Approximately 666 kg/s air enters the con-
denser at 13 C, and leaves at 26 C. The Level II isopentane cycle on a T-s diagram
is shown in Fig. 3. It can be noted in Figs. 2 and 3 that the saturated vapor line of
isopentane has a positive slope ensuring superheated vapor state at the turbine out-
let. Thus, no moisture is involved in the turbine operation. This is one reason iso-
pentane is a suitable working fluid in binary geothermal power plants. Isopentane
has other suitable thermophysical properties such as a relatively low boiling tem-
perature that provides a good match with the brine in the heat exchange system, and
a relatively high heat capacity. It is a safe fluid to handle, it is not corrosive, and it is
not poisonous.
The heat exchange process between the geothermal brine and isopentane is shown
in Figs. 4 and 5 for Levels I and II, respectively. An energy balance can be written
from Fig. 4 for part of the heat exchange taking place in the vaporizer of Level I as
Fig. 4. Diagram showing the heat exchange process between the geothermal brine and the working fluid
isopentane in Level I. The states refer to Fig. 1. The x-axis represents the path of the fluid flow in the heat
exchanger.
: :
m 1 hpp h2 ¼ m9 ðhf h9 Þ ð1Þ
where hf is the saturated liquid enthalpy of isopentane at the saturation (i.e. vapor-
ization) temperature, 133.1 C, and hpp is the enthalpy of brine at the pinch-point
temperature of the brine. Solving this equation for hpp, we determine the corre-
sponding brine pinch-point temperature, Tpp, to be 140.5 C. The pinch-point tem-
perature difference Tpp is simply the difference between brine pinch-point
M. Kanoglu / Geothermics 31 (2002) 709–724 715
Fig. 5. Diagram showing the heat exchange process between the geothermal brine and the working fluid
isopentane in Level II. The states refer to Fig. 1. x-axis as in Fig. 4.
3. Exergy analysis
Disregarding kinetic and potential energy changes, the specific flow exergy of
geothermal fluid at any state (plant location) can be calculated from
e ¼ h h0 T0 ðs s0 Þ ð3Þ
where T0 is the environment (dead state) temperature, h and s are the enthalpy and
the entropy of the geothermal fluid at the specified state, and h0 and s0 are the cor-
responding properties at the restricted dead state. Multiplying specific exergy by the
mass flow rate of the geothermal fluid gives the exergy rate
: :
E ¼ me ð4Þ
Table 1
Exergy rates and other properties at various plant locations for one representative unit. State numbers
refer to Fig. 1
State Fluid Phase Temperature Pressure, Enthalpy, Entropy, s Mass flow Specific Exergy .
.
no T ( C) P (bar abs) h (kJ/kg) (kJ/kg C) rate, m exergy rate E
(kg/s) e (kJ/kg) (kW)
Table 2
Some exergetic and energetic performance data provided for one representative unit of the plant. I and II
indicates whether the component belongs to Level I or II
In Table 1, temperature, pressure, and mass flow rate data for geothermal fluid,
working fluid, and air are given according to their state numbers specified in Fig. 1.
Exergy rates are calculated for each state, and listed in Table 1. State 0 and 00 are the
restricted dead states for the geothermal and working fluids, respectively. They cor-
respond to an environment temperature of 12.8 C and an atmospheric pressure of
84 kPa, which were the values measured at the time when the plant data were
obtained. For geothermal fluid, the thermodynamic properties of water are used. By
doing so, the effects of any salts and non-condensable gases that might be present in
the geothermal brine are neglected. This should not cause any significant error in
calculations since their fractions are estimated by the plant management to be small.
The thermodynamic properties of working fluid, isopentane, are obtained from a
thermodynamic property evaluation software (Friend, 1992).
Preheaters, vaporizers, and condensers in the plant are essentially heat exchangers
designed to perform different tasks. The exergetic efficiency of a heat exchanger may
be measured by the increase in the exergy of the cold stream divided by the decrease
in the exergy of the hot stream (Wark, 1995). Applying this definition to Level I
vaporizer, we obtain
: :
E10 E9
"vap I ¼ : : ð5Þ
E1 E2
718 M. Kanoglu / Geothermics 31 (2002) 709–724
where the exergy rates are given in Table 1. The difference between the numerator
and denominator in Eq. (3) is the exergy destruction in the heat exchanger. That is,
: : : : :
I vap I ¼ E1 E2 E10 E9 ð6Þ
Because of the complex appearance of the entire heat exchange system, we write
the exergetic efficiency and exergy destruction relations for Level I vaporizer-pre-
heater system as
: :
E10 E8
"vap-pre I ¼ : : : : ð7Þ
E1 E2 þ E3 E4
: : : : : : :
I vap-pre I ¼ E1 þ E3 þ E8 E2 þ E4 þ E10 ð8Þ
Fig. 6. Exergy flow diagram. Given as the percentages of brine exergy input.
M. Kanoglu / Geothermics 31 (2002) 709–724 719
The exergetic efficiency of a turbine is defined as a measure of how well the stream
exergy of the fluid is converted into actual turbine work output. Applying this to the
Level I turbine, we obtain
:
Wturb I
"turb I ¼ : : ð9Þ
E10 E11
The difference between the numerator and denominator in Eq. (9) is simply the
exergy destruction in the turbine.
: : : :
I turb I ¼ E10 E11 Wturb I ð10Þ
The exergetic efficiency and exergy destruction for the Level I pump are calculated
from the following relations:
: :
E8 E7
"pump I ¼ : ð11Þ
Wpump I
: : : :
I pump I ¼ Wpump I E8 E7 ð12Þ
The data provided by the plant management do not include the heat and pressure
losses in the transmitting pipes and valves. We believe that their effect would not
have been significant.
The exergetic efficiency of Level I isopentane cycle can be determined from
:
Wnet I
"level I ¼ : : : : ð13Þ
E1 E2 þ E3 E4
where given in the denominator is the exergy drop of brine across the Level I
vaporizer-preheater (i.e. exergy input to Level I). The net power of Level I is the
difference between the turbine power output and the pump power input. The total
exergy lost in Level I cycle is determined from
: : : : : :
I level I ¼ Ipump I þ Ivap I þ Ipre I þ Iturb I þ I cond I ð14Þ
The exergetic efficiency of the plant, based on total brine exergy drops across the
vaporizer-preheater systems of Level I and Level II cycles (i.e. total exergy inputs to
Level I and Level II cycles), can be expressed as
:
Wnet plant
"plant;a ¼: : : : : : : : : ð15Þ
E1 E2 þ E3 E4 þ E2 E3 E5 þ E5 E6
720 M. Kanoglu / Geothermics 31 (2002) 709–724
where given in the numerator is the net power output from the plant obtained by
subtracting total parasitic power, 390 kWe, from the total net power output from
the Level I and Level II cycles, 2159 kW. Eq. (15) may also be viewed as the exer-
getic efficiency of the Rankine cycle. Therefore, it is only meaningful for binary
geothermal power plants and it allows a meaningful comparison between different
binary geothermal power plants.
The exergetic efficiency of the plant can also be calculated based on the brine
exergy input to the plant (i.e. exergy of the brine at the Level I vaporizer inlet). That
is:
:
Wnet plant
"plant;b ¼ : ð16Þ
E1
When using Eq. (16), the exergy input to the plant is sometimes taken as the
exergy of geothermal fluid in the reservoir. Those who prefer this approach argue
that a realistic and most meaningful comparison between geothermal power plants
can only be performed considering the methods of harvesting the geothermal fluid.
However, others argue that taking the reservoir as the starting point is not proper
for geothermal power plants since conventional power plants are evaluated on the
basis of the exergy of the fuel burned at the plant site (DiPippo and Marcille,
1984).
The total exergy destruction in the plant is the difference between the brine exergy
at the vaporizer inlet and the net power outputs from Levels I and II cycles:
: : : :
I plant ¼ E1 Wnet I þ Wnet II ð17Þ
This includes various exergy losses in the plant components as well as the exergy
of brine leaving Levels I and II preheaters. One could argue that the exergy of used
brine is a recovered exergy, and should not be considered as part of the exergy loss.
However, the used brine is reinjected back into the ground without any attempt to
make use of it.
The exergetic efficiencies and exergy destructions of major plant components and
the entire plant are calculated as explained in this section, and listed in Table 2. All
values are for one representative unit. To pinpoint the sites of exergy destruction
and quantify these losses, an exergy flow diagram is given in Fig. 6.
For comparison, certain First Law performance data are provided in Table 2. The
rates of heat transfer for vaporizers, preheaters, and condensers are specified while
powers are specified for turbines, pumps, Levels I and II cycles, and the overall
plant. The isentropic efficiencies of the turbines and pumps, the thermal efficiencies
of Level I and II cycles, and the overall plant are also given. The thermal efficiency
of Level I cycle is calculated from the ratio of the net power output from Level I
cycle to the heat input to Level I cycle (i.e. the total heat transfer in Level I vapor-
izer-preheater). The First Law thermal efficiency of the plant, based on the energy
input to the plant, is calculated from
M. Kanoglu / Geothermics 31 (2002) 709–724 721
:
Wnet plant
plant;a ¼ : : : : ð18Þ
m1 ðh1 h2 Þ þ m 3 ðh3 h4 Þ þ m2 ðh2 h5 Þ þ m5 ðh5 h6 Þ
where the terms given in the denominator are heat transfer rates in vaporizer I,
preheater I, vaporizer II, and preheater II, respectively. The First Law efficiency of
the plant may be expressed in another form:
:
Wnet plant
plant;b ¼ : ð19Þ
m 1 ð h1 h0 Þ
where h0 is the dead state enthalpy of brine specified in Table 1. The entire term in the
denominator is the energy of the brine at the Level I vaporizer inlet. An energy flow
diagram is given in Fig. 7 to provide a comparison with the exergy flow diagram.
An investigation of the exergy flow diagram given in Fig. 6 shows that 64.5% of
the exergy entering the plant is lost. The remaining 35.5% is converted to power and
18.1% of this power is used for parasitic load in the plant. The exergetic efficiency of
the plant is determined at 34.2%, based on the exergy input to the isopentane Rankine
cycles (i.e. exergy drops of the brine in the vaporizer and preheater), and 29.1% based
on the exergy input to the plant (i.e. exergy of the brine at the Level I vaporizer inlet)
(Table 2). Bodvarsson and Eggers (1972) give the exergetic efficiencies of a single-flash
and a double-flash cycle as 38.7 and 49.0%, respectively, based on 250 C resource
water temperature and 40 C sink temperature. Both values are significantly greater
Fig. 7. Energy flow diagram. Given as the percentages of brine energy input.
722 M. Kanoglu / Geothermics 31 (2002) 709–724
than the value calculated for the binary plant analyzed in this paper. This is to be
expected, since additional exergy destruction occurs during the heat exchange
between the geothermal and the working fluids in binary plants. DiPippo and Mar-
cille (1984) calculated the exergetic efficiency of an actual binary power plant, using
a 140 C resource and 10 C, as 20 and 33.5% based on the exergy input to the plant
and to the Rankine cycle, respectively. A binary geothermal power plant considered
by Kanoglu and Cengel (1999b), using a 158 C resource and 3 C sink, has an
exergetic efficiency of 22.6 and 34.8%, based on the exergy input to the plant and to
the Rankine cycle, respectively.
Using low-temperature resources, geothermal power plants generally have low
First Law efficiencies. Consequently, the First Law efficiency of the plant is calcu-
lated to be 5.8% based on the energy input to the plant and 8.9% based on the
energy input to the isopentane Rankine cycles. This means that more than 90% of
the energy of the brine is discarded as waste heat. There is a strong argument here
for the use of geothermal resources for direct applications such as district heating
instead of power generation, when economically feasible. A co-generation scheme
involving power generation and district heating could also be considered when used
brine is reinjected back into the ground at a relatively high temperature. The energy
flow diagram in Fig. 7 shows that 35.2% of the energy of the brine is reinjected,
57.8% of it is rejected in the condenser, and the remaining is converted to power.
Yet it provides no specific information on how the performance can be improved.
This shows the value of an exergy analysis.
The causes of exergy destruction in the plant include vaporizer-preheater losses,
turbine-pump losses, the exergy of the brine reinjected, and the exergy of isopentane
lost in the condenser. They represent 13.0, 13.9, 14.8, and 22.6% of the brine exergy
input, respectively (Fig. 6). The exergetic efficiencies of Level I vaporizer–preheater
and Level II vaporizer–preheater are 87 and 83%, respectively. These percentages
can be considered high, and indicate the high performance of the heat exchange
system. In binary geothermal power plants, heat exchangers are important compo-
nents and their individual performances have a significant effect on the overall per-
formance of the plant. The exergetic efficiency of the vaporizer is significantly
greater than that of the preheater, because the average temperature difference
between the brine and the working fluid is smaller in the vaporizer than in the pre-
heater.
The exergetic efficiencies of the turbines are 75 and 70% for Levels I and II,
respectively. These relatively low efficiencies, together with the significant exergy loss
associated with its operation, indicate that performance of the turbines can be
improved. In fact, a newly built binary geothermal power plant has a turbine oper-
ating with an exergetic efficiency of over 80% (Kanoglu et al., 1998). The same
observation can also be made for the relatively low turbine isentropic (First Law)
efficiencies (in the range of 65–70%) listed in Table 2. On the other hand, the pumps
seem to be performing well.
The exergetic efficiencies of the condensers are around 30%, making them the
least efficient components in the plant. This is primarily due to the high average
temperature difference between the isopentane and the cooling air. The brine is
M. Kanoglu / Geothermics 31 (2002) 709–724 723
reinjected back into the ground at about 65 C. In at least one binary plant using a
resource at about 160 C, the brine is reinjected at temperatures above 90 C
(Kanoglu et al., 1998). The percent of exergy loss associated with brine reinjection is
therefore comparatively low in this plant.
For binary geothermal power plants using air as the cooling medium, the con-
denser temperature varies as the ambient air temperature fluctuates throughout the
year and even throughout the day. As a result, the power output decreases by up to
50% from winter to summer (Kanoglu and Cengel, 1999b; Michaelides and Ryder,
1992). Consequently, the exergy destruction rates and percentages at various sites
change, this effect being most noticeable in the condenser.
5. Conclusion
The aim of the exergy analysis for a power plant is usually to identify and quantify
the sites of exergy destruction so as to pinpoint where we should attempt to improve
the performance. The exergy analysis served this purpose well in this paper. The
plant analyzed has a unique heat exchange design that appears to be performing
well. As expected, a significant fraction of the input brine exergy to the plant is lost
in the condensers. There seems to be potential for improving the turbine perfor-
mance. Brine reinjection is another form of major exergy loss, whose fraction
appears to be reasonable. A limited comparison with some other geothermal power
plants suggests that performance of this plant is satisfactory.
Acknowledgements
The author thanks Chuck Salo of plant management for providing plant oper-
ation data.
References
Bejan, A., 1988. Advanced Engineering Thermodynamics. John Wiley & Sons, New York.
Bodvarsson, G., Eggers, D.E., 1972. The exergy of thermal power. Geothermics 1, 93–95.
DiPippo, R., Marcille, D.F., 1984. Exergy analysis of geothermal power plants. Geothermal Resources
Council Transactions 8, 47–52.
DiPippo, R., 1994. Second law analysis of flash-binary and multilevel binary geothermal power plants.
Geothermal Resources Council Transactions 18, 505–510.
Friend, D.G., 1992. NIST Mixture Property Database. Version 9.08. Fluid Mixtures Data Center. Ther-
mophysics Division. National Institute of Standards and Technology, Boulder, CO 80303.
Gaggioli, R.A., 1998. Available energy and exergy. International Journal of Applied Thermodynamics 1,
1–8.
Kanoglu, M., Cengel, Y.A., 1999a. Retrofitting a geothermal power plant to optimize performance: a case
study. Transactions of the ASME, Journal of Energy Resources Technology 121/4, 295–301.
Kanoglu, M., Cengel, Y.A., 1999b. Improving the performance of an existing binary geothermal power
724 M. Kanoglu / Geothermics 31 (2002) 709–724
plant: a case study. Transactions of the ASME, Journal of Energy Resources Technology 121/3, 196–
202.
Kanoglu, M., Cengel, Y.A., Turner, R.H., 1998. Incorporating a district heating/cooling system into an
existing geothermal power plant. Transactions of the ASME, Journal of Energy Resources Technology
120/2, 179–184.
Kestin, J., 1978. Available Work in Geothermal Energy, US Department of Energy. Division of Geo-
thermal Energy. Washington, DC.
Lee, K.C., 2001. Classification of geothermal resources by exergy. Geothermics 30, 431–442.
Moran, M.J., 1982. Availability analysis: a guide to efficient energy use. Prentice-Hall, Englewood Cliffs,
NJ.
Michaelides, E.E., Ryder, J.K., 1992. The influence of seasonal and daily temperature fluctuations on the
work produced by geothermal power plants. International Journal of Energy Systems 12/2, 68–72.
Milora, S.L., Tester, J.F., 1976. Geothermal energy as a source of electric power, thermodynamic and
economic design criteria. The MIT Press, Cambridge, Massachusetts.
Paloso jr, G., Mohanty, B., 1993. A flashing binary combined cycle for geothermal power generation.
Energy-The International Journal 18/8, 803–814.
Phair, K.A., 1994. Getting the most out of geothermal power. ASME Mechanical Engineering 116/9, 76–
80.
Subbiah, S., Natarejen, R., 1988. Thermodynamic analysis of binary-fluid Rankine cycles for geothermal
power plants. Energy Conversion & Management 28/1, 47–52.
Vasil’ev, V.A., Krainov, A.V., Gevorkov, I.G., 1996. Calculation of the parameters of a unified geother-
mal power plant operating on a water-ammonia mixture. Thermal Engineering 43/5.
Wark, K.J., 1995. Advanced Thermodynamics for Engineers. McGraw-Hill, New York.
Yuan, Z., Michaelides, E., 1993. Binary-flashing geothermal power plants. Transactions of the ASME,
Journal of Energy Resources Technology 115/3, 232–236.