Professional Documents
Culture Documents
Qureshi 2014
Qureshi 2014
Heat exchanger inventory cost optimization for power cycles with one
feedwater heater
Bilal Ahmed Qureshi, Mohamed A. Antar, Syed M. Zubair
Mechanical Engineering Department, KFUPM Box # 1474, King Fahd University of Petroleum & Minerals, Dhahran 31261, Saudi Arabia
a r t i c l e
i n f o
Article history:
Received 6 March 2014
Accepted 7 May 2014
Keywords:
Inventory
Heat exchanger
Thermoeconomic
Optimization
Feedwater heater
a b s t r a c t
Cost optimization of heat exchanger inventory in power cycles with one open feedwater heater is undertaken. In this regard, thermoeconomic analysis for an endoreversible power cycle with an open feedwater
heater is shown. The scenarios of constant heat rejection and addition rates, power as well as rate of heat
transfer in the open feedwater heater are studied. All cost functions displayed minima with respect to the
high-side absolute temperature ratio (h1). In this case, the effect of the Carnot temperature ratio (U1),
absolute temperature ratio (n) and the phase-change absolute temperature ratio for the feedwater heater
(U2) are qualitatively the same. Furthermore, the constant heat addition scenario resulted in the lowest
value of the cost function. For variation of all cost functions, the smaller the value of the phase-change
absolute temperature ratio for the feedwater heater (U2), lower the cost at the minima. As feedwater
heater to hot end unit cost ratio decreases, the minimum total conductance required increases.
2014 Elsevier Ltd. All rights reserved.
1. Introduction
Using feedwater heaters, also called regenerators, to enhance
efciency of steam power plants is a standard practice in industry
and, therefore, thermoeconomic analysis of such systems is
important. Feedwater heaters can either be open or closed type.
When the heat is transferred from the steam (bled from the turbine) to the feedwater by mixing, it is considered an open type.
The advantages offered include a higher efciency of the power
plant due to a rise in the average temperature of heat addition.
Next, it helps to prevent boiler corrosion by providing an easy
way to remove air that leaks into the condenser. Also, it lowers
the volume ow in the nal turbine stages. For further details,
the work of Babcock & Wilcox [1] may be consulted along with
textbooks on Thermodynamics such as Cengel and Boles [2].
Authors such as Bejan [3,4] addressed the issue of heat exchanger inventory allocation for different situations such as maximizing
the efciency. The Carnot model developed by Bejan [4] was used
by Antar and Zubair [5] to study cost optimization of power plant
heat exchanger inventory for a specied power output. The total
inventory reached a minimum when the unit cost ratio attained
unity. Sahin and Kodal [6] carried out thermoeconomic optimization of endoreversible heat engines using a new thermoeconomic
optimization criterion i.e. power output per unit total cost.
Corresponding author. Tel.: +966 3 860 3135.
E-mail address: smzubair@kfupm.edu.sa (S.M. Zubair).
http://dx.doi.org/10.1016/j.enconman.2014.05.028
0196-8904/ 2014 Elsevier Ltd. All rights reserved.
380
Nomenclature
A
F
G
K
k1
_
m
Q_
s
T
U
area (m2)
non-dimensional cost function ()
unit cost conductance ratio ()
non-dimensional quantity dened by Eq. (15f) ()
ratio of feedwater heater to condenser entropy
change ()
mass ow rate (kg s1)
rate of heat transfer (kW)
specic entropy (kJ kg1 K1)
absolute temperature (K)
overall heat transfer coefcient (kW m2 K1)
Greek
c
U1
U2
h1
h2
n
Subscripts
01
at condenser
02
at phase-change in feedwater heater
a
constant rate of work
b
constant rate of heat rejection from the condenser
C
reversible compartment
c
constant rate of heat addition in the boiler
d
constant rate of heat transfer in the open feedwater
heater
H
hot end
L
cold end
min
minimum
OFH
open feedwater heater
tot
total
evaluation purpose. Therefore, this paper aims to develop the relevant endoreversible models and then perform thermoeconomic
analysis of this system. The scenarios of constant heat addition
and rejection rates, power as well as rate of heat transfer in the
open feedwater heater will be studied.
2. Mathematical framework
Following the methodology of Bejan [4] and Antar and Zubair
[5], the endoreversible form of the power cycle with one open
feedwater heater is now considered. The schematic of system
under consideration is shown in Fig. 1(a) while Fig. 1(b) shows
its T-s diagram. The purpose of the current study is to determine
the minimum of the total cost of conductance (UA) for the following scenarios: constant rate of heat addition, power, heat transfer
in the preheater and heat rejection. The Heat Exchanger Inventory
Cost Equation (HEICE) can be written in terms of heat exchanger
unit cost parameters as [5]:
where cH, cL, and cOFH are unit cost of conductance for the boiler,
condenser and preheater, respectively, such that C has units of
dollars. Next,
Q_ H UAH T H T HC
Q_ L UAL T 01 T L
4a
T OFH T 02 DT OFH;av g
4b
C cH
Q_ H
Q_ L
Q_ OFH
cL
cOFH
T H T HC
T 01 T L
DT OFH;av g
cH
c
c
Q_ H
Q_ L
Q_ OFH
L
OFH
T H T HC cH T 01 T L
cH DT OFH;av g
381
_ 02 T 02 s6 s3
Q_ OFH m
_ 01 T 01 s7 s1
m
Q_ L
10
We see from Fig. 1(b) that, in general, (s6 s3) = k1(s7 s1) where k1
can be any number less than one. Now, Eq. (10) becomes
_ 02 T 02
m
Q_ OFH
k1
_ 01 T 01
m
Q_ L
11
Q_ H
Q_ L
;
T HC T 01
12a
Q_ L
T 01
Q_ H T HC
12b
_ 02 T 02
m
Q_ OFH
k1
_
_
m
01 T HC
QH
13
Putting Eqs. (12b) and (13) into Eq. (8), we get after dividing the
right hand side by TH/TH,
2
3
_ 02 T 02
T 01
m
k
1
Q_ H 4 1
_
T
m T
GL T HC T GOFH DT 01 HC 5
01
OFH;av g
cH T H 1 TTHC
TL
T
14
TH
h1
T HC
TH
15a
h2
DT OFH;av g
TH
15b
U1
T 01
T HC
15c
U2
T 02
T HC
15d
Fig. 1. Endoreversible power cycle with an open feedwater heater: (a) Schematic
and (b) T s diagram.
Let
cL
cH
GL and
cOFH
cH
Q_ H
Q_ L
Q_ OFH
GL
GOFH
cH T H T HC
T 01 T L
DT OFH;av g
cH
"
Q_ H
1
Q_ L =Q_ H
Q_ OFH =Q_ H
GL
GOFH
T H T HC
T 01 T L
DT OFH;av g
15e
_ 02
m
_ 01
m
15f
Putting Eqs. (15a), (15b), (15c), (15d), (15e), (15f) into Eq. (14), and
multiplying both sides by TH gives us
cH
T H Q_ H
1
U1
K U2
GL
GOFH
1 h1
U1 h1 n
h2
16
Beginning from Eq. (16), the next section discusses the following
scenarios: constant power, heat addition and rejection capacities as
well as heat transfer rate in the open feedwater heater.
K k1
C
7
TL
TH
#
8
_ 01 T OFH s3 s2
Q_ OFH m
9a
_ 02 T 02 s6 s3
Q_ 36 m
9b
_ 01 T 01 s7 s1
Q_ L m
9c
But Q_ OFH Q_ 36 since they are exchanging heat with each other
exclusively in the feedwater heater. Thus,
382
Fa
cH W_
TH
1
U1
K U2
Q_ H
GL
GOFH
_ 1 h1
U1 h1 n
h2
W
and
using
the
_ Q_ H Q_ L
W
First
17
Law
of
18
_
W
Q_ L
1
Q_ H
Q_ H
19
_
W
T 01
1
_
T HC
QH
20a
1
Q_ H
01
_
1 TTHC
W
20b
Combining Eqs. (15c) and (20b) and putting them into Eq. (17)
results in
Fa
1
1
U1
K U2
GL
GOFH
1 U1 1 h1
U1 h1 n
h2
21
@F a
1
1
U1
K U2
GL
GOFH
@ U1 1 U1 1 h1
U1 h1 n
h2
"
#
n
GL
U1 h1 n2
0
22a
or
U1;min
cn
p
n1 c cn
1 ch1
22b
where
1
K U2 1
GOFH
1 h1
h2 GL
Fig. 2(a) shows a plot of the cost function Fa against U1 for different
values of n. It is found that the minimum value of U1 as well as the
cost rise as n increases. The reason for the increasing cost is evident
from Eq. (21) where n is seen in the second term inside the brackets
only. As n increases, this term increases in value and, thus, Fa as well.
Regarding the minimum value of U1 (that provides a minimum cost),
it can be seen from Eq. (22b) that this is due to the fact that U1,min is
directly proportional to n. Furthermore, this shows that a lower ambient temperature would result in a lower cost if all other quantities
remained the same. Fig. 2(b) shows a plot of the cost function Fa
against U1 for different values of h1. It is found that the minimum
value of U1 decreases as h1 increases but the cost rises. The reason
for the increasing cost is understood from Eq. (21) where h1 is seen
in the rst and second terms inside the brackets. The rst term containing h1 is the dominant term, therefore, as h1 increases, this term
increases in value and, thus, Fa as well. Regarding the minimum value
Fig. 2. Dimensionless HEICE for specied power versus U1: (a) effect of varying n,
(b) effect of varying h1 and (c) effect of varying U2.
of U1, it can be seen from Eq. (22b) that this is due to the fact that
U1,min is inversely proportional to h1. Also, the (1 + c) term, which
contains h1 as well, dominates the quantities containing c in the
numerator. Furthermore, this shows that a higher furnace temperature may result in a lower cost if all other quantities remained the
same. Fig. 2(c) shows a plot of the cost function Fa against U1 for different values of U2. It was noted that lower values of U2 produced
lower costs while the minimum value of U1 was not affected by
change in U2. The reason for the decreasing cost is understood from
Eq. (21) where U2 is seen in the last term inside the bracket only
and directly proportional to Fa. Regarding the minimum value of U1
not changing signicantly, it can be explained from the fact that
the effect of variation in U2 is very small on c. Furthermore, this
shows that a lower feedwater heater temperature (T02) may result
in a lower cost if all other quantities remained the same.
383
h1;min
p
p
1 n GL 1 U1 GL GL
GL 1
U1
23
It should be noted that Eq. (23) is identical to the respective derivative found by Antar and Zubair [5] though they did not plot it.
Fig. 3(a) shows the effect of different values of n as Fa varies against
h1. It is found that the optimum point of h1 shifts to a greater value
as n increases as well as the cost. This behavior is similar to that found
in Fig. 2(a). As can be seen from Eq. (23), this is simply due to the fact
that n is directly proportional to h1,min. Fig. 3(b) shows the effect of different values of U1 as Fa varies against h1. It is seen that the cost function and h1,min shift to a lower value as U1 increases. This is because
Fb
cH Q_ L
TH
1
U1
K U2
Q_ H
GL
GOFH
U1 h1 n
h2
Q_ L 1 h1
24
Fb
U1
1
U1
K U2
GL
GOFH
1 h1
U1 h1 n
h2
25
@F b
1
1
U1
K U2
GL
GOFH
@ U1 U1 1 h1
U1 h1 n
h2
"
#
n
GL
U1 h1 n2
0
26
In the above equation, all terms are positive and, therefore, a practical minimum is not possible. The result of taking the derivative of
Fb with respect to h1, in this case, is identical to Eq. (23). Fig. 4(a)(c)
are plotted for conditions identical to those of Fig. 3(a)(c). It is
found that the behavior for this cost function is qualitatively the
same as Fa and the only difference is in the values. It can be
explained from the fact that the terms inside the brackets for both
the cost functions are identical while there is a minor difference in
the term outside it for both the cost functions are identical while
there is a minor difference in the second bracketed term.
3.3. Constant heat addition rate
The non-dimensional cost equation resulting from dividing Eq.
(16) by the boiler heat transfer rate is:
Fc
cH Q_ H
TH
1
U1
K U2
GL
GOFH
1 h1
U 1 h1 n
h2
27
@F c
n
0
@ U1 n U1 h1 2
28
384
Fig. 4. Dimensionless HEICE for specied heat rejection rate versus h1: (a) effect of
varying n, (b) effect of varying U1 and (c) effect of varying U2.
As all terms in the equation are clearly positive on one side, therefore, no minimum exists. The result of taking the derivative of Fc
with respect to h1, in this case, is identical to Eq. (23). Fig. 5(a)(c)
are plotted for conditions identical to those of Fig. 3(a)(c). It is
found that the behavior for this cost function is qualitatively the
same as Fa and the only difference is in the values. The reason for
this is that the quantities inside the brackets for both the cost
functions (i.e. Fa and Fc) are identical while there is a small
difference outside it which is a multiplying factor only.
Fig. 5. Dimensionless HEICE for specied heat addition rate versus h1: (a) effect of
varying n, (b) effect of varying U1 and (c) effect of varying U2.
Fd
cH Q_ OFH
TH
1
K U2
1
U1
K U2
GL
GOFH
1 h1
U1 h1 n
h2
29b
@F d
1
1
1
U1
K U2
GOFH
G
L
OFH
h2 U2 K U22 1 h1
@ U2
U1 h1 n
h2
Fd
cH Q_ OFH
TH
Q_ H
1
U1
K U2
GL
GOFH
U1 h1 n
h2
Q_ OFH 1 h1
29a
0
or, after simplication, gives
30a
385
@F d
1
U1
GL
0
@ U2 1 h1
U1 h1 n
30b
As all terms in the equation are clearly positive, therefore, no minimum exists. Finally, oFd/oh1was found to be the same as in Eq. (23).
Fig. 6(a)(c) are plotted for conditions identical to those of Fig. 3(a)
(c). It is found that the behavior for this cost function is qualitatively
the same as Fa and the only difference is in the values. This is due to
the fact that the quantities inside the brackets for both the cost
functions (i.e. Fa and Fd) are the same and, although the term outside it is different, it only acts a multiplying factor resulting in a
change of value but not the behavior.
cH UAH
C
cL UAL
3.5. Effect of unit cost ratios
1
U1 1 h1
K U2 1 h1
1 GL
GOFH
h2
U1 h1 n
1 U1 h1 n
GOFH K U2 U1 h1 n
1
GL U1 1 h1
GL h2
U1
31
1
32
Fig. 6. Dimensionless HEICE for specied heat transfer rate in the feedwater heater
versus h1: (a) effect of varying n, (b) effect of varying U1 and (c) effect of varying U2.
Fig. 7. Example of all conductances versus unit cost ratio of cold to hot end: (a) at
GOFH = 1, (b) at GOFH = 0.5 and (c) at GOFH = 0.1.
386
Table 1
Comparison of Antar and Zubair [5] with current work for total conductance required.
a
b
GL ccL
H
UAtota
UAtotb (GOFH = 1)
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
2
3
4
5
6
7
8
9
10
26.783
22.895
21.380
20.599
20.148
19.875
19.711
19.616
19.568
19.555
20.148
21.067
21.999
22.895
23.748
24.559
25.333
26.073
26.783
79.403
73.261
70.878
69.654
68.949
68.525
68.270
68.123
68.050
68.029
68.922
70.272
71.610
72.870
74.047
75.149
76.183
77.160
78.084
76.757
71.932
70.175
69.345
68.922
68.713
68.630
68.625
68.672
68.756
70.220
71.808
73.263
74.580
75.783
76.890
77.918
78.880
79.785
78.084
74.580
73.390
72.876
72.650
72.572
72.577
72.632
72.719
72.828
74.218
75.579
76.800
77.904
78.915
79.851
80.726
81.550
82.331
While taking into account the effect of the unit cost ratios, the
following conclusions can be made:
Ratios of the conductance costs of each heat exchanger to the
total cost were attained.
If cold and hot end heat exchanger unit costs are the same i.e.
GL = 1, the total conductance was minimized at GOFH = 1.
At optimum conditions, the total conductance is not equally
divided among the heat exchangers. But when all unit cost
ratios are unity, the hot and cold end conductances are equal
in value.
cOFH UAOFH
C
GOFH
h2
GL h2
U1
1
K U2 1 h1 GOFH K U2 U1 h1 n
For variation of all F-values with respect to h1 and U1, the smaller the value of U2, lower the cost at the minima.
For variation of all F-values with respect to h1, the effect of different values of n is qualitatively the same. This is found to be
respectively true for the effect of different values of U1 and
U2. Furthermore, in these cases, the constant heat addition scenario resulted in the lowest value cost function.
For the system investigated, the minimum with respect to U1
exists only for the scenario of specied power cost function
i.e. Fa.
No minima were found with respect to n, h2 and U2.
1
33
Acknowledgements
The authors acknowledge the support provided by King Fahd
University of Petroleum & Minerals through the project IN121042.
References
[1] The Babcock & Wilcox. Steam: its generation and use. 41st ed. Babcock &
Wilcox Company, 2005.
[2] Cengel YA, Boles MA. Thermodynamics: an engineering approach. 7th
ed. McGraw-Hill Inc.; 2010.
[3] Bejan A. Theory of heat transfer-irreversible power plants. Int J Heat Mass
Transf 1988;31:12119.
[4] Bejan A. Power and refrigeration plants for minimum heat exchanger
inventory. J Energy Resour Technol 1993;115:14850.
[5] Antar MA, Zubair SM. Thermoeconomic considerations in the optimum
allocation of heat exchanger inventory for a power plant. Energy Convers
Manag 2001;42:116979.
[6] Sahin B, Kodal A. Performance analysis of an endoreversible heat engine based
on a new thermoeconomic optimization criterion. Energy Convers Manag
2001;42:108593.
[7] Kodal A, Sahin B. Finite size thermoeconomic optimization for irreversible heat
engines. Int J Therm Sci 2003;42:77782.
[8] Chen L, Sun F, Wu C. Maximum-prot performance for generalized irreversible
Carnot-engines. Appl Energy 2004;79:1525.
[9] Bandyopadhyay S, Bera NC, Bhattacharyya S. Thermoeconomic optimization of
combined cycle power plants. Energy Convers Manag 2001;42:35971.
[10] Rovira A, Snchez C, Muoz M, Valds M, Durn MD. Thermoeconomic
optimisation of heat recovery steam generators of combined cycle gas turbine
power plants considering off-design operation. Energy Convers Manag
2011;52:18409.
[11] Baghernejad A, Yaghoubi M. Exergoeconomic analysis and optimization of an
Integrated Solar Combined Cycle System (ISCCS) using genetic algorithm.
Energy Convers Manag 2011;52:2193203.
[12] Ahmadi P, Dincer I. Thermodynamic analysis and thermoeconomic
optimization of a dual pressure combined cycle power plant with a
supplementary ring unit. Energy Convers Manag 2011;52:2296308.
[13] Kaviri AG, Jaafar MNM, Lazim TM. Modeling and multi-objective exergy based
optimization of a combined cycle power plant using a genetic algorithm.
Energy Convers Manag 2012;58:94103.
[14] Silveira JL, Tuna CE. Thermoeconomic analysis method for optimization of
combined heat and power systemspart II. Prog Energy Combust Sci
2004;30:6738.
[15] Al-Sulaiman FA, Dincer I, Hamdullahpur F. Thermoeconomic optimization of
three trigeneration systems using organic Rankine cycles: Part I
Formulations. Energy Convers Manag 2013;69:199208.
[16] Al-Sulaiman FA, Dincer I, Hamdullahpur F. Thermoeconomic optimization of
three trigeneration systems using organic Rankine cycles: Part II
Applications. Energy Convers Manag 2013;69:20916.
387
[21] Silveira JL, Tuna CE, de Queiroz Lamas W, Aparecida de Castro Villela I. A
contribution for thermoeconomic modelling: a methodology proposal. Appl
Therm Eng 2010;30:173440.
[22] Seyyedi SM, Ajam H, Farahat S. A new approach for optimization of thermal
power plant based on the exergoeconomic analysis and structural
optimization method: application to the CGAM problem. Energy Convers
Manag 2010;51:220211.
[23] Durmayaz A, Sogut OS, Sahin B, Yavuz H. Optimization of thermal systems
based on nite-time thermodynamics and thermoeconomics. Prog Energy
Combust Sci 2004;30:175217.