J ournal of EEA, Vol.

27, 2010
MODELING, SIMULATION AND PERFORMANCE EVALUATION OF
PARABOLIC TROUGH
SOLAR COLLECTOR POWER GENERATION SYSTEM

Mekuannint Mesfin and Abebayehu Assefa
Department of Mechanical Engineering
Addis Ababa University

ABSTRACT

Model of a parabolic trough power plant, taking
into consideration the different losses associated
with collection of the solar irradiance and thermal
losses is presented. MATLAB software is employed
to model the power plant at reference state points.
The code is then used to find the different reference
values which are used as inputs for the TRNSYS
programme implemented in the modeling and
simulation of the power plant. TRNSYS software is
used to simulate the performance of the model at
off design weather conditions as well.

Keywords: Parabolic Trough Collector (PTC);
Heat Transfer Fluid (HTF); TRNSYS power plant
model; STEC library; Solar Advisor Model (SAM);
TRNSYS solar field model; Solar Electric
Generation System (SEGS).

INTRODUCTION

Parabolic troughs are currently most used means of
power generation option of solar sources. Solar
electric generation systems (SEGs) employ solar
collectors to track the sun and use its energy to
produce steam. These plants replace the boiler part
of a conventional Rankine cycle power plant with
solar fields that are used to increase the temperature
of heat transfer fluids. The solar field area must
however be wide enough to satisfy the power
demand. Heat exchangers are used to transfer heat
energy from the heat transfer fluid (HTF) to water
coming from feedwater heaters.

In this paper a proposed SEGs power plant with a
capacity of 10 MWe is discussed, with Addis
Ababa selected as the site of operation. For the
mentioned capacity of the plant, MATLAB
software is used to model the power plant at
reference weather conditions. The outputs of the
code are used as reference values in TRNSYS
modeling and simulation of the plant at off
optimum design weather conditions. In the
TRNSYS simulation, a library of Solar Thermal
Electric Component (STEC) models are used for
both solar and conventional power cycle elements,
in addition to the built-in TRNSYS components.
SOLAR-FIELD POWER PLANT MODEL

Solar Thermal Electric Components (STEC)
developed for a parabolic trough is applied to
model the solar field. The solar field model bases
itself on the model of Lippke [1] that uses and
integrates efficiency equations to account for the
different fluid temperatures at the field inlet and
outlet of the collector field. It calculates the
demanded mass flow rate of the HTF to achieve a
pre-defined outlet temperature [2].

MODELING OF THE POWER PLANT

In the modeling part, each component of the power
plant is analyzed by properly identifying the inputs
and outputs as well as defining and evaluating its
efficiency. The power plant to be investigated is
assumed to operate using the Rankine power cycle.
It has two high pressure and three low pressure
turbines, arranged for optimized power generation.
Accordingly there are two high and one low
pressure closed feedwater heaters. Between the
high pressure and low pressure feedwater heaters a
deaerator exists. Steam leaving the LP Turbine 3
and cascaded wet-vapor from the feed water
heaters is condensed in the condenser. The
condensate is pumped to the deaerator pressure,
first passing through the LP feed water heater. The
feed pump discharges the circulating water to the
pressure of the HP Turbine 1, first passing through
the train of heat exchangers. The heat exchanger
train consists of a feedwater preheater, a steam
generator and a superheater, in series and a
reheater, in parallel with the other three heat
exchangers. The power cycle begins by collecting
the HTF returning from the solar field in an
expansion vessel. The expansion vessel serves to
compensate for variation in the volume of the heat
transfer fluid throughout the day, since the specific
volume of the HTF is dependent on temperature.
The heat transfer fluid is pumped from the
expansion vessel and delivered to the heat
exchanger train as the energy source for the power
cycle. The complete schematic diagram of the
power plant is shown in Fig. 1.


Mekuannint Mesfin and Abebayehu Assefa

J ournal of EEA, Vol. 27, 2010

82
H
p

T
u
r
b
i
n
e

1
L
p

T
u
r
b
i
n
e

3
R
e
h
e
a
t
e
r
HTF Pump
S
o
l
a
r

F
i
e
l
d
Feed Pump
D
e
a
e
r
a
t
o
r
L
P

F
e
e
d
C
o
n
d
e
n
s
e
r
Condensate
1
2
3
4
5
6
7
8
22
21
20 19
18 17
14
9 10 11 12 13
15
16
23
25
27
28
26
Pump
H
e
a
t
e
r
H
P

F
e
e
d
H
e
a
t
e
r
I
P

F
e
e
d
H
e
a
t
e
r
24
P
r
e
h
e
a
t
e
r
S
t
e
a
m
G
e
n
e
r
a
t
o
r
S
u
p
e
r
h
e
a
t
e
r
H
p

T
u
r
b
i
n
e

2
L
p

T
u
r
b
i
n
e

1
L
p

T
u
r
b
i
n
e

2
Figure 1 Steam power plant

The optimum design plant efficiency is determined
using reference weather conditions of Addis
Ababa. Reference ambient temperature and wind
speed weather conditions can be found by taking
annual average data. Reference direct normal
radiation value is set to the actual direct normal
radiation value that has a cumulative annual
frequency value of about 95 % [3].

The size of solar field area which is required to
produce the given power at the optimum design
weather conditions could be found by making heat
balance between the HTF from the solar field and
steam from the Rankine system.

MATLAB code is written to determine the
effective solar field area, after determining the
different state points and steam and HTF mass flow
rates. The flow chart of the code is shown in Fig. 2
and the T-S diagram for the simulated power plant
is indicated in Fig. 3.

Properties pressure, temperature, specific enthalpy
and specific entropy at all state point are indicated
in Table 1. These values are results of the
MATLAB model.

SIMULATION OF THE POWER PLANT
USING TRNSYS

The purpose of the TRNSYS modeling and
simulation is to study the operation of the power
plant at weather conditions different from the
reference values. In these weather conditions, the
power plant operates by producing less power than
the rated capacity. TRNSYS software is selected
for the simulation work due to its simplicity and
flexibility.

The TRNSYS modeling includes the TRNSYS
field model and power model. The solar field
model shown in Fig. 4 includes weather data
processors, Type 16g and Type 9a. HTF mixer
which is use to mix the HTF from the reheater and
the preheater is modeled by Type 11. Type 4 is also
used to model the expansion vessel. All the above
components are part of the Standard TRNSYS
library [4].

Parabolic trough collector model (Type 396) is part
of the STEC library release 3.0, 2006. The HTF
splitter (Type 351), which is used to split the HTF
mass flow rate to the super heater, the steam
generator, the preheater train and reheater train is
used from the same library.

TRNSYS power model is used to estimate the input
and output temperatures of the steam from the
preheater, steam generator, superheater, turbine
stages, condenser, feed water heaters, and closed
feed water heaters. The model also determines their
variations with the operating conditions. Similarly
the mass flow rate of steam, the heat transferred
between component stages and the net power
output from the plant are determined applying
TRNSYS software.
Modeling, Simulation and Performance Evaluation

J ournal of EEA, Vol. 27, 2010 83
In the power modeling of the plant using TRNSYS
most of the components are part of STEC model
library release 3.0, 2006. Fig. 5 shows the
TRNSYS power plant model.






















Figure 2 Flow chart for the developed MATLAB code
Set Initial Conditions
Power = 10 MWe T
out,HTF
= 390
O
C T
in,HTF
= 300
O
C P(1) = Psat(T=T
steam
-5)
T
steam
= 375
O
C Pinch Point = 15
O
C P
condenser
= 8 kPa
T

Component Modeling After Making the Necessary Assumptions
State Point Determination
Determine Net Energy Transfer
Q
net
= Q
SH
+ Q
SG
+ Q
RH
+ Q
PH

Reference Weather conditions for Addis Ababa
DNI = 806 N/m
2
T
amb
= 16.5 OC WS = 4 m/s
Calculate the Parabolic Trough Efficiency
( )
( )
DNI
T T T T
D
DNI
T T
WS C C
DNI
T T
B A RS EL K
in out in out
in out
w
in out
2
3
1
.
.
. 2
. .
. 2
. . . .
A + A + A A
+
A + A
+ +
(

A + A
+ = q

Calculate Solar Field Area and Number of Collectors Using STEC Model
( )
-
-
=

net
p out in
Q
m
C T T

- - -
=
net abs pipe
Q Q Q
-
=
abs eff
Q A .DNI.q

Error = Pinch Point Temp. Difference
Error < Tol.
End
Yes
No
Start
P(1) = P(1) 0.05
Mekuannint Mesfin and Abebayehu Assefa

J ournal of EEA, Vol. 27, 2010

84
0.0 1.1 2.2 3.3 4.4 5.5 6.6 7.7 8.8
0
100
200
300
400
s [kJ/kg-K]
T

[

C
]
83.43 bar
37.57 bar
14.10 bar
405.3 kPa
77.98 kPa
8 kPa
1
2
3
4
5
6
7
8
9
11
13
14
17
18
20
22
10
21
12
19
15
16


Figure 3 T-s diagram of the power plant




Figure 4 TRNSYS solar field model

In the power modeling of the plant using TRNSYS
most of the components are part of STEC model
library release 3.0, 2006. Fig. 5 shows the
TRNSYS power plant model.

RESULTS OF THE TRNSYS SIMULATION

The reference values for the TRNSYS simulation
are found from the full load operation of the power
plant which is modeled using MATLAB. The
reference values include, area of the solar field,
over all heat transfer coefficients of the heat
exchangers, design mass flow rate of water/ steam,
design inlet and outlet pressures of each turbine
stage and, outlet pressures of each turbine stage and
mass flow rates of the steam/water and HTF in the
heat exchangers.

The simulation is done for year 2001 EC. The
weather conditions for Addis Ababa are found from
SWERA [5] in a text and TMY formats. The result
of the simulation for two days of operation, one
clear day (March 16) and one cloudy day (May 15)
are given below.

For the clear day simulation, all the mass flow rates
and inlet and exit temperatures as well as the power
generated are established to be equal to the design
value. The cloudy day simulation, however, shows
that all the results are highly dependent on DNI
values and reinforces the necessity of thermal
storage for improved output from this kind of
power plants.
Modeling, Simulation and Performance Evaluation

J ournal of EEA, Vol. 27, 2010 85


Table 1: State points of the power plant (MATLAB output)

State
Point
P(i)
(kPa)
T(i)
(
O
C)
h(i)
(kJ/kg)
s(i)
(kJ/kg K)
1 375.00 83.43 3058.55 6.2265
2 275.48 37.57 2899.30 6.2782
3 195.38 14.10 2731.23 6.3415
4 375.00 14.10 3203.89 7.2193
5 235.24 4.05 2933.99 7.3152
6 93.24 0.78 2664.96 7.4449
7 41.51 0.08 2380.27 7.6046
8 41.51 0.08 173.85 0.5925
9 41.57 4.05 174.39 0.5931
10 92.74 4.05 388.75 1.2240
11 144.09 4.05 606.77 1.7815
12 145.60 83.43 618.25 1.7885
13 194.70 83.43 831.62 2.2702
14 246.61 83.43 1069.54 2.7518
15 297.96 83.43 1333.38 3.2354
16 297.96 83.43 2753.40 5.7219
17 246.67 37.57 1069.54 2.7629
18 195.38 14.10 1069.54 2.7948
19 195.38 14.10 831.62 2.2870
20 144.09 4.05 831.62 2.3204
21 92.80 0.78 388.75 1.2250
22 41.51 0.08 388.75 1.2755



Mekuannint Mesfin and Abebayehu Assefa

J ournal of EEA, Vol. 27, 2010

86



Figure 5 TRNSYS representation the power plant model
Modeling, Simulation and Performance Evaluation

J ournal of EEA, Vol. 27, 2010 87
SIMULATION RESULTS

On a clear day such as March 16, 2001, the
capacity of the power plant is achieved for a period
of about 9 hours (9:00 17:00 hours), Fig. 7. The
distribution of the direct normal insolation (DNI),
the mass flow rate of the solar field HTF, the solar
field inlet and outlet temperatures and the steam
mass flow rate through the Rankine cycle are
shown in Fig. 6 and Fig. 8-11, respectively.

Similar results are shown for a cloudy day (May
15, 2001) in Fig. 1217. The distribution of the
direct normal insolation (DNI) for May 15, 2001 is
uneven, Fig. 12, so is the net power output, Fig. 13.
Accordingly, conventional backup system should
be available to attain the required power (10 MWe)
delivery.

4 6 8 10 12 14 16 18 20
0
100
200
300
400
500
600
700
800
900
1000
1100
D
N
I

(
W
/
m
2
)
Time (hr)

Figure 6 Direct Normal Insolation (DNI) for
clear day

4 6 8 10 12 14 16 18 20
0
2
4
6
8
10
12
P
o
w
e
r

(
M
W
)
Time (hr)

Figure 7 Net power for clear day
4 6 8 10 12 14 16 18 20
0
20
40
60
80
100
120
140
160

H
T
F

M
a
s
s

F
l
o
w

R
a
t
e

(
k
g
/
s
)
Time (hr)

Figure 8 Solar-field HTF mass flow rate for
clear day

4 6 8 10 12 14 16 18 20 22
0
50
100
150
200
250
300

S
o
l
a
r

F
i
e
l
d

I
n
l
e
t

T
e
m
p
e
r
a
t
u
r
e

(
o
C
)
Time (hr)

Figure 9 Solar-field inlet temperature for clear
day

4 6 8 10 12 14 16 18 20 22
0
50
100
150
200
250
300
350
400
450

S
o
l
a
r

F
i
e
l
d

O
u
t
l
e
t

T
e
m
p
e
r
a
t
u
r
e

(
o
C
)
Time (hr)

Figure 10 Solar-field outlet temperature for
clear day

Mekuannint Mesfin and Abebayehu Assefa

J ournal of EEA, Vol. 27, 2010

88
4 6 8 10 12 14 16 18 20
0
2
4
6
8
10
12
14
16

S
t
e
a
m

F
l
o
w

R
a
t
e

(
k
g
/
s
)
Time (hr)

Figure 11 Mass flow rate of steam for clear
day


4 6 8 10 12 14 16 18 20
0
100
200
300
400
500
600
700
800
D
N
I

(
W
/
m
2
)
Time (hr)

Figure 112 Direct Normal Insolation (DNI) for
cloudy day

4 6 8 10 12 14 16 18 20
0
1
2
3
4
5
6
7
8
9
10
P
o
w
e
r

(
M
W
)
Time (hr)

Figure 13 Net power for cloudy day
4 6 8 10 12 14 16 18 20
0
20
40
60
80
100
120

H
T
F

M
a
s
s

F
l
o
w

R
a
t
e

(
k
g
/
s
)
Time (hr)

Figure 14 Solar-field HTF mass flow for
cloudy day

4 6 8 10 12 14 16 18 20 22
0
50
100
150
200
250
300

S
o
l
a
r

F
i
e
l
d

I
n
l
e
t

T
e
m
p
e
r
a
t
u
r
e

(
o
C
)
Time (hr)

Figure 15 Solar-field inlet temperature for
cloudy day

4 6 8 10 12 14 16 18 20 22
0
50
100
150
200
250
300
350
400
450

S
o
l
a
r

F
i
e
l
d

O
u
t
l
e
t

T
e
m
p
e
r
a
t
u
r
e

(
o
C
)
Time (hr)

Figure 16 Solar-field outlet temperature for
cloudy day
Modeling, Simulation and Performance Evaluation

J ournal of EEA, Vol. 27, 2010 89
4 6 8 10 12 14 16 18 20
0
1
2
3
4
5
6
7
8
9
10
11

S
t
e
a
m

F
l
o
w

R
a
t
e

(
k
g
/
s
)
Time (hr)

Figure 17 Mass flow rate of steam for cloudy
day

CONCLUSION

In this paper a power plant is designed using
appropriate considerations and assumptions. The
designed power plant is modeled in two simulation
programs, MATLAB and TRNSYS. MATLAB
software is used to only to model the power plant at
full load operation and reference weather
conditions. The outputs of the code are used as
reference values in TRNSYS modeling and
simulation of the plant at off optimum design
weather conditions. The TRNSYS model of the
power plant is used to model both the solar field
and the power field. This model is developed by
using component from STEC library, the TRNSYS
built in library and a developed component, Type
850. The main purpose of the model developed by
the TRNSYS is to investigate the part load
operation of the power plant.

The performance of the model developed in the
TRNSYS is evaluated by considering available
data for the year 2001 EC. Two days have been
selected for the simulation, one for solar clear day
and the other for cloudy day. The results of the
simulation have shown that the mass flow rate of
the HTF from the solar field and gross power
produced are heavily dependent on DNI.

REFERENCES

[1] Lippke, Frank. Simulation of the Part Load
Behavior of a 30MWe SEGS Plant. Prepared
for Sandia National Laboratories,
Albuquerque, NM, SAND95-1293, June 1995.

[2] Schwarzbzl, Peter and Zentrum, Deutsches. A
TRNSYS Model Library for Solar Thermal
Electric Components (STEC). A Reference
Manual, Release 3.0, 2006.

[3] Sam User Guide, SAM Software,
https://www.nrel.gov/analysis/sam.

[4] TRNSYS - A Transient System Simulation
Program. Madison, WI: University of
Wisconsin-Madison Solar Energy Laboratory,
http://sel.me.wisc.edu/trnsys/

[5] Weather Data, http://swera.unep.net, Last
Accessed: November 15, 2008.