A METHODOLOGY FOR THE GEOMETRIC DESIGN OF HEAT

RECOVERY STEAM GENERATORS APPLYING GENETIC

ALGORITHMS

Mª. Dolores Durán*1, Manuel Valdés+, Antonio Rovira°, E. Rincón*

*Departamento de Ingeniería Mecánica, Facultad de Ingeniería, UAEM, Cerro de Coatepec s/n, Toluca,
Méx.; +Departamento de Ingeniería Energética y Fluidomecánica. Universidad Politécnica de Madrid.
E. T.S. Ingenieros Industriales. José Gutiérrez Abascal, 2. 28006. Madrid. Spain;.°Depto. de Ing.
Energética, Universidad Nacional de Educación a Distancia, C/ Juan del Rosal nº 12. 28040 Madrid,
Spain.

ABSTRACT
This paper shows how the geometric design of heat recovery steam generators (HRSG) can be
achieved. The method calculates the product of the overall heat transfer coefficient (U) by the area of the
heat exchange surface (A) as a function of certain thermodynamic design parameters of the HRSG. A
genetic algorithm is then applied to determine the best set of geometric parameters which comply with the
desired UA product and, at the same time, result in a small heat exchange area and low pressure losses in
the HRSG.
In order to test this method, the design was applied to the HRSG of an existing plant and the results
obtained were compared with the real exchange area of the steam generator. The findings show that the
methodology is sound and offers reliable results even for complex HRSG designs.

KEYWORDS: HRSG, CCGT, Thermal Design, Thermoeconomic Optimization, Genetic

Algorithms.

1. INTRODUCTION
The Combined Cycle Gas Turbine (CCGT) power plant efficiency is affected by the design of all its
components. Among them, the Heat Recovery Steam Generator (HRSG) is one of the most important,
since there the union of the gas and steam cycles takes place. The power generated by the steam cycle is
strongly dependent on the HRSG thermal efficiency. The power generated by the steam cycle is highly
dependent on the HRSG thermal efficiency. Therefore the thermodynamic HRSG design parameters must
be carefully selected in order to achieve the optimum performances in the combined cycle.
Many authors have directed their researches to the optimization of CCGT power plants and HRSG
[1,4,5 and 14].

NOMENCLATURE
A HRSG Area (m2)

1
Corresponding author. Email: mduran@fi.uaemex.mx, FAX: 722-2140855.
ap Approach point (K)
b HRSG width (m)
d0 Outer tube diameter (mm)
d2 Fin diameter (mm)
et Thickness of tube (mm)
e Thickness of fin (mm)
f Function
h Enthalpy (kJ/kg)
k Thermal conductivity (W/mK)
K Penalization constant
L Tube length (m)
lperp Tubes separation
lpar Row separation
η Thermal efficiency
Nt Number of tubes per row
Nprof Number of tubes per column
pp Pinch point (K)
Pen Penalization
ri Inner radius
ro Outer radius
t Temperature (K)
v Velocity (m/s)
U Overall heat transfer coefficient (W/m2K)
Wcc Combined cycle total power (kW)
x General variable
xi Restriction value of a general variable
z number of fins per unit length
T Gas to steam temperature difference at the superheater (K)
p Pressure loss (Pa)
 Density (kg/m3)
Subscripts
e Error
Ec Economizer
Ev Evaporator
f Fitness
HP High pressure
HRSG Heat recovery steam generator
i Inside conditions
IP Intermediate pressure
g Gas
Geom Obtained with the geometric parameters
Lim Limit
LP Low pressure
 o Outside conditions
obj Objective
sh Superheater
Therm Obtained with the thermodynamic parameters
v Vapour

Certain works are specifically concerned with the design of the HRSG, for instance, that of Franco
and Giannini [6] shows a method, based on hierarchical strategy, for the optimal design of the HRSG,
considering the maximization of the compactness index and the minimization of the pressure losses. On
the same topic, Manassaldi et. al. [10] proposed a methodology for the HRSG design. This methodology
applies a mixed nonlinear program model to optimize the design according to three criteria: net power
maximization, the ratio between net power and material weight maximization, and net heat transfer
maximization. The results of this paper are accurate but the economic optimization problem is not
discussed.
On the other hand, Rovira et. al. [11], Duran [5] and Valdés et. al. [15] (the latter is the preceding
work to this most recent one) made a thermoeconomic optimization which minimizes the generating cost
or maximizes the annual cash flow of the plant, considering as independent design parameters of the
HRSG the thermodynamic design variables: namely, drum pressure, pinch points, approach points and
temperature differences at the superheater exit. In this work, the selection of a given set of these
thermodynamic parameters led to the determination of the products (UA) of the overall heat transfer
coefficient (U) by the area (A) of the HRSG heat exchange surfaces. Nevertheless, various pairs of U and
A might lead to the same UA product. Thus, as the same thermal CCGT performances could be obtained
with different HRSG geometric designs, the HRSG geometric design was undetermined and there is still
room for improvement of the optimization process.
This more recent work applied the results obtained with the thermoeconomic optimization proposed
by Valdés et. al. [15] and Durán and Galindo [5] and presents a method that solves the uncertainty in the
determination of the HRSG geometric design parameters. The method proposed here uses a genetic
algorithm in order to find the geometric design of the HRSG which fulfills the desired UA product while,
at the same time, obtaining a small heat exchange area and low pressure losses.
2. HRSG DESCRIPTION
2.1. Geometric parameters of the HRSG.

A single pressure level HRSG has three different sections: economizer, evaporator and superheater
(see fig. 1)2. The flow conditions and the manufacturing materials differ from section to section, so the
pressure losses and the convective heat transfer coefficient are different too.
Each HRSG section could be considered a crossflow heat exchanger. Its corresponding arrangement
and some of the geometric design parameters are shown in fig. 2. Most of the geometric design
parameters for each section are different, but in this paper it will be considered that the length of the tubes

2
When the HRSG has more pressure levels the sections are the same but corresponding to each pressure level.
(L) and the width of each section (b) (economizer, evaporator, superheater) have the same value, in order
to have a HRSG with uniform transverse section.

2.2. Assumptions.

Some of the main assumed hypotheses to derive the mathematical model are listed as follows:

- Inlet gas turbine conditions are assumed as known (fuel mass flow rate, inlet temperature,
chemical composition of the fuel)

- Exhaust gas conditions (mass flow rate and temperature) are obtained during the simulation of
the gas turbine.

- Unfired HRSG equipment is considered.

- No cooling between sections is considered.

- Correlations taken from IPWS [18] are used.

- Solid fins are considered

- Inline tubes arrangement

2.3. Constraints of the HRSG design parameters.

Some of the design parameters should be restricted owing to manufacturing and operative reasons.
The limit values for some of these parameters were suggested by Deschamps [4] and Ganapathy [7] and
are described in table 1.

Other relevant geometric parameters have a fixed value, like the parallel gap (lpar) which is taken
equal to 0,1m in the design proposed in this work [4] and the thickness fin whose value is 0,8 mm.
In addition to these, there are some restrictions that must be taken into account in the HRSG design:
a) Perpendicular gap (lperp). Its value is restricted by the fin diameter and it is a function of the
section width (b) and the number of tubes per row (Nt) as it is shown in the following equation:

b−N t d o
l perp =
Nt
(1)

b) Gas side convective heat transfer coefficient. Its value depends strongly on the gas flow velocity
and thus on the HRSG width. A high value of this coefficient could lead to a very narrow HRSG and, as a
consequence, an increase in the gas side pressure losses. Some authors [9] recommend that this
convective heat transfer coefficient must be kept between 30 and 100 W/m2 K.
c) Gas side pressure drop. Its value depends on the cross section of the HRSG for a given gas flow.
Some authors [8] suggest that this value must be between 5 and 15 mbar.
d) Steam side pressure drop. If this value is high, too much energy will be spent to pump the water.
For this reason it is proposed [9] that it needs to be less than 10% of the drum pressure.
 The HRSG width (b) is the most influential parameter on the above constrains and, as it was already
mentioned, its value will be considered the same for all the HRSG sections. This parameter is calculated
using the equation (1) in an iterative process that is described in section 4.

3. OVERALL HEAT TRANSFER COEFFICIENT CALCULATION

The product of the overall heat transfer coefficient U by the HRSG area A should be calculated
applying the logarithmic temperature difference equation and the Thermodynamic First Law to each of
the one pressure level HRSG sections (see fig. 1)3:

η HRSG mg ( h3 −h 4 )
U ec A ec=
( t 3 −t d ) −( t 4 −t e )
t 3 −t d
ln
t 4 −t e

η HRSG m g ( h 2−h 3 )
U ev A ev =
( t 2−t 3 )
t 2−t c
ln
t 3−t c (2)
η HRSG m g ( h1 −h2 )
U sh A sh =
( t1 −t a ) −( t 2 −t b )
t 1 −t a
ln
t 2 −t b

Once the thermodynamic (or thermoeconomic) design has been done applying the methodology

proposed in [16], the parameters of the right hand side of equations 2 are known and the UA product
calculation is straightforward.
On the other hand, the HRSG geometric design that matches with the equations 2 needs to be
obtained, but it must be taken into account that the overall heat transfer coefficient U is also function of
the geometric design and the flow conditions, since it depends on the convective heat transfer coefficients
(see eq. 3).
ro
r o ln
1 1 ri ro
= + +
U ho k r i hi (3)
The convective heat transfer coefficients (hi and he) also depend on flow conditions and on the
geometric parameters of the steam generator. Their calculation is thoroughly described by Weir [19] and
Bejan and Kraus [2].

4. DESCRIPTION OF THE DESIGN TOOL

4.1. Description of the model

3
It was considered an efficiency of 0.95 for the HRSG.
 This section describes the mathematical model that assesses the possible designs of the HRSG. The
model was first applied to a single pressure level HRSG. More complex configurations have been
modeled in similar way, as the only difference is the number of components of the HRSG. Figure 3 shows
the diagram of the methodology.
Rapún [12] proposed a method which consists of an iterative process that modifies the geometric
parameters of the HRSG, until the desired UA value is reached. Although this methodology works well, it
might lead to large values of the total heat exchange area and thus to high HRSG costs. Using the work of
Rapún as a reference, a new method has been developed in order to determine the geometric design of the
HRSG surfaces with the following objectives:
a) Calculate, through equation (3), the overall heat transfer coefficient and then the exchange
surface (A), in order to obtain the product of them called UAgeom whose value must be almost the same
that the UAtherm calculated through equations (2).

b) Achieve a low value of the total HRSG exchange surface area that is useful in the
thermoeconomic optimization process of the CCGT power plant. This thermoeconomic optimization
process was developed in previous works and is described in [15 and 16].

c) Keep the same width and tube length for all the sections of the HRSG to have an homogeneous
transverse section.

d) Keep the gas side and water side pressure losses within an acceptable range.

Taking all the objectives already mentioned into consideration, the model proposed in this work
minimizes the error defined as follows:

Error=Abs
( UA geom −UA therm
UA therm ) (4)

It is important to say that, even though this is not an optimization problem, the solution of the
equation (4) may be achieved by applying an optimization methodology like genetic algorithms and in
this case the characteristics of the problem are:

a) Design Variables

The independent variables chosen are defined in table 2. The tube length is considered common to all
the HRSG sections. The problem has 16 independent variables for a single pressure CCGT power plant 4.

b) Fitness function

Since the goal is to find the minimum value of the error determined by the equation (4), the problem
may be considered as an optimization one. As it was mentioned, the methodology applied for its solution

4
For the two pressure level (without and with reheat) CCGT power plant the number of independent variables is 35
and 40 respectively.
was a genetic algorithm. This method had already been employed in [5 and 15] –which is preceding of
this one–.It is important to say that according to the nature of this method, it is helpful for finding
maximum values so, in order to apply it, the objective function must be redefined in the following way:

1 1
f e= =

( )
1+ Abs (Error ) UA geom −UA therm
1+ Abs
UA therm
(5)

The above function defines the error for each HRSG section so it must be applied to all of them. For
the one pressure level CCGT power plant, which has three different exchange sections, the objective
function is:

(6)

As the error decreases, the value of increases until it reaches the maximum value of three5.

Nevertheless, when genetic algorithms are applied a fitness function must be defined. This function
includes the objective function and a number of penalization functions which depend on the physical
constraints of some dependant variables [3, 8]. Without the penalizations, it would be nearly impossible
to find a reliable optimum. In this case, the penalizations depend on constrains described in section 2.2.
Including them in the fitness function yields:

f f =f obj−K Δ pg Pen Δp g −K Δ ps Pen Δp s −K vs Penv s −K a ( PenA ec +PenAev + PenA sh )

(7)

Pen Δp s
where is the penalization for the gas side pressure drop, is the one for the steam pressure

Pen v s
drop at the superheater, is the penalization for steam velocity and finally , ,

are penalizations for big values of the heat exchange surface corresponding to the economizer,
evaporator and superheater, respectively.

Each penalization in equation (7) is defined so that this is only applied when the value of the variable is
Pen Δp s
greater than the restriction. For instance, for the penalization of the gas side pressure drop, could
be defined as:

PenΔp g=¿ { Δpg−Δp Lim if Δpg>ΔpLim ¿ ¿¿¿

(8)

5
This value matches with the number of HRSG sections.
 Δp Lim
were is the limit of gas side pressure drop, it means that is the maximum value that this variable
could reach in order to avoid performance troubles. In the same way are defined the others penalization
functions.

The reasons for including these penalizations are explained in section 2.2. The last three penalizations in
eq. (7) were included because an additional objective of this work is to find the geometric parameters that,
not only give an accurate value for UA, but also give reasonable small heat exchange area. The value of
the constants in the eq. (7) was established by the authors, this value should be selected so that the
penalties have the same order of magnitude as the objective function, in order to discriminate those
designs that do not satisfy constrains established. For instance, we have already mentioned that the gas
side pressure drop should be between 0,005 and 0,015 bar, if the algorithm reach a configuration that has

almost the maximum value of the objective function ( ) but with a gas side pressure droop of 0.060
bar, according with eq. (8) the value of the penalization will not be large enough to rule out this
K Δp
g
configuration, so should be at least 100 in order to have the same order of magnitude as the
objective function. The same consideration was taken for establishing the value of all the constants of eq.
(7).

4.2. Description of the program for the HRSG surface calculation.

The tool proposed here works as follows (A diagram that describes the methodology is shown in fig.
3:
1. A population composed of a certain number of individuals is randomly generated. The
individuals are identified by the design variables as described in section 2.1.
2. Before evaluating the fitness function for all the individuals, the value of the HRSG width (b),
which is determined by constrains described in section 2.2, is calculated. The calculation of this
parameter is an iterative process described as follows:
- An initial value of the HRSG width b is calculated using the following equation:
mg
b=
L⋅ρg v g
(9)

- Starting from the obtained value of b, the separation between the tubes (lperp) is calculated with
the equation (1) and the gas and water/steam side convective heat transfer coefficients (he and hi)
are attained using the methodology proposed by Weir [19].

- If the values of lperp and he do not fulfill the restrictions described in section 2.2, the value of b
is increased by a given amount and the iterative process goes on until a reliable value of b is
reached. The process continues in the same way with following HRSG sections, trying to keep
the same value of b for all of them.
 3. Once a good value of the HRSG width is achieved, the fitness function, including its
penalizations, (eq. 7) is evaluated.
4. The genetic algorithm continues until the difference between the UA geom.and the UAtherm is minor
that the minimum error previously established. If the maximum value of the fitness function is
found during the process, the geometric parameters provide the actual UA that have a
conveniently low value of the heat exchange area and pressure losses.

5. RESULTS

The proposed methodology was applied for the calculation of the exchange surface of a HRSG
whose heat exchange area and thermodynamic design parameters are already known. The HRSG chosen
belongs to a Spanish two pressure level CCGT power plant. This plant began its operations some years
ago and its main design data are shown in table 3. The objective is to calculate the geometric design
parameters of the HRSG with the methodology proposed and then to compare the resultant area with the
one in the existing plant.
In Table 3 are also shown, the steam cycle design parameters as well as the results of its simulation.
In this table the UA’s of each section calculated with the known thermodynamic design parameters are
shown.
Using the thermodynamic design parameters of the steam cycle and the UA product obtained with the
eq. 2 the program calculates the HRSG geometry and the heat exchange surface. Table 4 shows the UA
obtained with the proposed program. A minimum difference between these values and those of table 3
may be observed. Besides, in the same table 4 there is a comparison between the real area of the HRSG
and the one obtained with the program. As it can be observed the latter is smaller by 8.4%. This is
because the proposal methodology finds, not only a geometric configuration that matches with the UA therm
value calculated, but also the smaller one.
Also table 5 shows the optimum geometric design parameters found by the genetic algorithm, in this
table can be observed the configuration for all the HRSG sections. Particularly is observed that for the LP
superheater the geometric design do not include fins, that is because the low value of the UA product
corresponding to this section. This aggresses with the suggestion of Ganapathy for the low pressure
superheater [7].
On the other hand, it is interesting to analyze that even though the HRSG total area is smaller than
the real one, the exchange surface calculated for the low pressure superheater is greater. This is because,
the HRSG length and width were restricted to be the same for all the sections, and this is the value that
the algorithm finds for this UA product with the conditions established.
Finally, table 4 also shows the gas and steam pressure losses of the calculated HRSG. As it was
mentioned, their values are located within the suggested interval in section 4.

6. CONCLUSIONS
 In the design of a HRSG it is important to take into account not only the thermodynamic design
parameters, but also the economic ones. Because of this, many authors develop thermoeconomic
optimization systems that find the best thermodynamic design parameters. However, the first conclusion
of the work is that it is possible to continue these methodologies by means of optimizing the geometric
design of the HRSG.
Through the results, it was observed that the proposed method achieves the desired UA for each
section and is able to find a HRSG geometric design with small heat exchange area and low pressure loss.
Firstly, the method proposed in this work is useful for obtaining a suitable geometric design of the
HRSG that gives accurate UA values and, as a consequence, correct thermodynamic design parameters.
Moreover, it is possible to find an HRSG design with a minimum heat exchange area while at the same
time controlling the pressure losses.
It is also important to point out that the results found by the algorithm applied match with the
suggestions for the design of HRSGs established by Dechamps [4] and Ganapathy [7], in particular
because Ganapathy suggests that a greater area does not necessarily mean a better heat transfer: the most
important parameter is the product of the overall heat transfer coefficient by the area.
This method, with the thermoeconomic optimization tool proposed previously in Valdés et. al. [16],
and used in Duran et. al. [5] allows finding an appropriate design of a CCGT power plant (see. fig. 3),
with minimum cost or maximum cash flow and, at the same time, with a small HRSG area. Hence, this
method is useful for both the design and the optimization of CCGT power plants. Furthermore, for future
works this methodology may be applied for the design of supercritical HRSG.

7. REFERENCES
1. Attala L., Facchini B. and Ferrara G. (2001), Thermoeconomic optimization method as design tool in
gas-steam combined plant realization. Energy Conversion and Management. 42, 2163-2172.
2. Bejan A and Kraus A. (2003), Heat Transfer Handbook, Ed. Wiley, 2nd. Ed. USA.
3. Bentley P. (1999), An Introduction to Evolutionary Design by Computers. Evolutionary Design by
Computers, 1-71.
4. Dechamps P.J. (1995), Incremental cost optimization of heat recovery steam generators. In ASME
COGEN-TURBO, Viena, Austria.
5. Durán M. and Galindo S. (2007), Thermoeconomic Study of CCGT Plants, Towards cleaner planet,
Environmental Science and Engineering, Ed. Springer, 1st. Ed. New York, USA.
6. Franco A. and Giannini N. (2006), A general method for the optimum design of heat recovery steam
generators, Energy. 31, Issue 15, 3342–3361.
7. Ganapathy V. (2002), Industrial Boilers and heat steam generators. Design applications and
calculations, Ed. Marcel Deker Inc. 1st. Ed. New York, USA.
8. Goldberg D. (1989). Genetic Algorithms in Search, Optimization, and Machine Learning, 1ª Ed.,
Addison-Wesley, Michigan.
9. Gunn and Horton, (1989), Industrial Boilers, Ed. Longman Scientific and Technical, 1st. Ed.
10. Manassaldi J., Mussati S. and Scenna N. (2011), Optimal synthesis and design of Heat Recovery
Steam Generation (HRSG) via mathematical programming, Energy 36, Issue 1, 475-485.
11. Rovira A., Sanchez C., M. Muñóz, M. Valdés y M.D. Durán, (2011), Thermoeconomic optimisation
of heat recovery steam generators of combined cycle gas turbine power plants considering off-
design operation, Energy Conversion and Management. 52, Issue 4, 1840-1849.
12. Rapún J.L. (1999), Modelo matemático del comportamiento de ciclos combinados de turbinas de gas
y vapor, PHD Thesis, ETSII-UPM.
13. Sadik Kakaç, “Boilers, evaporators and condensers” Ed. Wiley Interscience. 1st. Ed. USA.
14. Toffolo A. and Lazzareto A. (2002), Evolutionary algorithms for multi-objective energetic and
economic optimization in thermal system design. Energy. 27, 549-567.
15. Valdés M. and Rapún J. (2001), Optimization of a heat recovery steam generators for combined cycle
gas turbine power plants. Applied Thermal Engineering. 21, 1149-1159.
16. Valdés M., Durán M. and Rovira A. (2003), Thermoeconomic optimization of combined cycle gas
turbine using genetic algorithms, Applied Thermal Engineering, 23, 2169-2182.
17. Various, (1994), Steam: Its Generation and Use, 40th Ed., Babcok and Wilcox Co.
18. Wagner W. and Kretzschmar H. (1988), International Steam tables, properties of water and steam
based on the industrial formulation IAPWS-IF97. Springer.
19. Weir C. D., (1988). Estimating the performance of gas turbine heat-recovery boilers off-design, Proc.
Instn. Mech. Engrs. V202 A4.
 b

4
e

E C O N O M IZ E R

3
d
E VAP O R ATO R

c
2
b
S U P E R H E ATE R

a
STE AM TU R B IN E
1

G AS TU R B IN E

Fig. 1. Schema of the one pressure level heat recovery steam generator
xb/2 x
Fig. 2. Arrangement
b
of finned tubes in a row of the HRSG.

d0 lperp et d2
b
System Thermoeconomic optimization
Optimization
GT HRSG+ST
Simulation Opt.

Geometric Design HRSG Thermodynamic

Geometric Design Design
Procedure GA Parameters
UAtherm

Error
Calculation of the
UA geom (UA geom, UAtherm)

Calculation of Decrease of the

HRSG sections Area of the HRSG
Area GA
Health function

Calculation of
HRSG sections Same width of the
width HRSG

Calculation of Minimum pressure

Pressure losses losses.

Fig. 3. Flow diagram of the HRSG geometric design procedure.

Table 1. Lower and upper limits for the HRSG geometric parameters.
Geometric parameter Section Lower limit Upper limit
Fins per length unit (Z) (fins/m) Economizer 50 230
Fin diameter (d2) (m) 0.050 0.078
Number of tubes (Nt) 9 24
Fins per length unit (Z) (fins/m) Evaporator 50 190
Fin diameter (d2) (m) 0.050 0.079
Number of tubes (Nt) 10 24
Fins per length unit (Z) (fins/m) Superheater 0 100
Fin diameter (d2) (m) 0.050 0.078
Number of tubes (Nt) 10 24
Number of tubes per column (Nprof) All 5 42
Tube Diameter (d0) (m) 0.050 0.065
Tube Length (L) (m) * 10 44
Table 2. Independent parameters for the HRSG geometry calculation.
PARAMETERS SECCTION
Nt
Nprof Economizer
z Evaporator
d0 Superheater
d2
L Common to all sections
Table 3. Design and simulation data of CCGT power plant analyzed.
Gas Cycle design data
Gas mass flow (kg/s) 178,5
Compression ratio 15.7
Gas turbine inlet temperature 1373
HRSG and steam cycle design data
Drum low pressure (bar) 7
Drum high pressure (bar) 72
Condensation pressure (mbar) 111
T -lp(K) 57
T-hp(K) 29
pp-lp (K) 9
pp-hp (K) 8
ap-lp (K) 5
ap-hp (K) 13
Steam cycle simulation results Calculated with the thermodynamic parameters
(kW) 79533.9
HRSG inlet gas temperature (SHHP inlet) 784 K
SHHP outlet gas temperature 726 K
EVHP inlet gas temperature 726 K
EVHP outlet gas temperature 576 K
SHLP inlet gas temperature 576 K
SHLP outlet gas temperature 565 K
EC2HP inlet gas temperature 565 K
EC2HP outlet gas temperature 496 K
EVLP inlet gas temperature 496 K
EC1HP inlet gas temperature and EVLP outlet gas temperature 426 K
ECLP inlet gas temperature 460 K
HRSG outlet gas temperature (EC1HP and ECLP outlet temp.) 406 K
UAtherm calculated during the simulation
UA-ECLP (W/K) 66.63
UA-EC1HP (W/K) 127.60
UA-EVLP (W/K) 439.48
UA-EC2HP (W/K) 375.49
UA-SHLP (W/K) 10.07
UA-EVHP (W/K) 742.20
UA-SHHP (W/K) 191.037
Table 4. Results obtained with the program.
Section UA’s calculated with the geometric parameters (W/K)
UA-ECLP 67.58
UA-EC1HP 127.79
UA-EVLP 439.20
UA-EC2HP 376.06
UA-SHLP 10.12
UA-EVHP 741.08
UA-SHHP 191.15
Pressure losses
Gas pressure loss (mbar) 7.27
Steam pressure loss (bar) 1.85
Element area Exchange real area(m2) Exchange calculated area (m2)
A-EC1HP 4725 3308.07
A-EC2HP 6300 6871.92
A-EVHP 12265 10842.25
A-SHHP 3985 5281.33
A-ECLP 4725 1925.62
A-EVLP 7610 7669.09
A-SHLP 176 544.11
Total area 39786 36442.42
Table 5. Geometric design parameters found by the algorithm.
Section Geometric design parameter Value
LP Economizer Nt 13
Nprof 10
z 167
d0 (m) 0.056
d2 (m) 0.073
1sr. HP Economizer Nt 14
Nprof 15
z 157
d0 (m) 0.054
d2 (m) 0.074
LP Evaporator Nt 22
Nprof 22
z 132
d0 (m) 0.052
d2 (m) 0.076
2nd. HP Economizer Nt 18
Nprof 21
z 191
d0 (m) 0.056
d2 (m) 0.075
LP Superheater Nt 11
Nprof 12
z 0
d0 (m) 0.068
d2 (m) 0.068
HP Evaporator Nt 23
Nprof 26
z 172
d0 (m) 0.0576
d2 (m) 0.078
HP Superheater Nt 21
Nprof 40
z 72
d0 (m) 0.067
d2 (m) 0.0735
Common b (m) 4.37
Common L (m) 19.45