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.
Journal of EEA, Vol. 27, 2010
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

1
23
4

Superheater

Hp Turbine 1

Hp Turbine 2
Reheater

Lp Turbine 3
Lp Turbine 1

Lp Turbine 2
24 16
Solar Field

2 5
6
Generator
Steam

3
7
25 15

Deaerator
27
HP Feed

LP Feed

Condenser
Preheater

IP Feed
Heater

Heater

Heater
22

13 12 11 10 9
26
8
HTF Pump 14 Feed Pump

17 18 21 Condensate
28 19 20 Pump

Figure 1 Steam power plant

The optimum design plant efficiency is determined reference values. In these weather conditions, the
using reference weather conditions of Addis power plant operates by producing less power than
Ababa. Reference ambient temperature and wind the rated capacity. TRNSYS software is selected
speed weather conditions can be found by taking for the simulation work due to its simplicity and
annual average data. Reference direct normal flexibility.
radiation value is set to the actual direct normal
radiation value that has a cumulative annual The TRNSYS modeling includes the TRNSYS
frequency value of about 95 % [3]. field model and power model. The solar field
model shown in Fig. 4 includes weather data
The size of solar field area which is required to processors, Type 16g and Type 9a. HTF mixer
produce the given power at the optimum design which is use to mix the HTF from the reheater and
weather conditions could be found by making heat the preheater is modeled by Type 11. Type 4 is also
balance between the HTF from the solar field and used to model the expansion vessel. All the above
steam from the Rankine system. components are part of the Standard TRNSYS
library [4].
MATLAB code is written to determine the
effective solar field area, after determining the Parabolic trough collector model (Type 396) is part
different state points and steam and HTF mass flow of the STEC library release 3.0, 2006. The HTF
rates. The flow chart of the code is shown in Fig. 2 splitter (Type 351), which is used to split the HTF
and the T-S diagram for the simulated power plant mass flow rate to the super heater, the steam
is indicated in Fig. 3. generator, the preheater train and reheater train is
used from the same library.
Properties pressure, temperature, specific enthalpy
and specific entropy at all state point are indicated TRNSYS power model is used to estimate the input
in Table 1. These values are results of the and output temperatures of the steam from the
MATLAB model. preheater, steam generator, superheater, turbine
stages, condenser, feed water heaters, and closed
SIMULATION OF THE POWER PLANT feed water heaters. The model also determines their
USING TRNSYS variations with the operating conditions. Similarly
the mass flow rate of steam, the heat transferred
The purpose of the TRNSYS modeling and between component stages and the net power
simulation is to study the operation of the power output from the plant are determined applying
plant at weather conditions different from the TRNSYS software.

82 Journal of EEA, Vol. 27, 2010

 Modeling, Simulation and Performance Evaluation

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

Start

Set Initial Conditions

Power = 10 MWe Tout,HTF = 390 O C Tin,HTF = 300 O C P(1) = Psat(T=Tsteam -5)
Tsteam = 375 OC Pinch Point = 15 OC Pcondenser = 8 kPa

Component Modeling After Making the Necessary Assumptions

State Point Determination

Determine Net Energy Transfer

Qnet = QSH + QSG + QRH + QPH
P(1) = P(1) – 0.05

Reference Weather conditions for Addis Ababa

DNI = 806 N/m2 Tamb = 16.5 OC WS = 4 m/s

Calculate the Parabolic Trough Efficiency

 ∆T + ∆Tin  ∆T + ∆Tin
∆Tout .∆Tin +
1
(∆Tout + ∆Tin )2
η = K .EL.RS . A + B. out + (C + C w .WS ). out + D. 3
 2.DNI  2.DNI DNI

Calculate Solar Field Area and Number of Collectors Using STEC Model
• • • • •
• Q net Q= Q abs − Q pipe Q abs = Aeff .DNI .η
m= net
C p (Tout − Tin )

Error = Pinch Point Temp. Difference

Error < Tol.

No
Yes
End

Figure 2 Flow chart for the developed MATLAB code

Journal of EEA, Vol. 27, 2010 83

Mekuannint Mesfin and Abebayehu Assefa

400
1 4

83.43 bar
15
300
16 2
17
37.57 bar
14
5
T [°C]

19
14.10 bar 3
200 13
18
12
405.3 kPa
11
20
21 77.98 kPa
100 10 6

9
22 8 kPa 7
8

0
0.0 1.1 2.2 3.3 4.4 5.5 6.6 7.7 8.8

s [kJ/kg-K]

Figure 3 T-s diagram of the power plant

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.

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.
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.

84 Journal of EEA, Vol. 27, 2010

 Modeling, Simulation and Performance Evaluation

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

State P(i) T(i) h(i) s(i)

Point (kPa) (OC) (kJ/kg) (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

Journal of EEA, Vol. 27, 2010 85

Mekuannint Mesfin and Abebayehu Assefa

Figure 5 TRNSYS representation the power plant model

86 Journal of EEA, Vol. 27, 2010

 Modeling, Simulation and Performance Evaluation

SIMULATION RESULTS
160

On a clear day such as March 16, 2001, the 140

capacity of the power plant is achieved for a period

HTF Mass Flow Rate (kg/s)

120
of about 9 hours (9:00 – 17:00 hours), Fig. 7. The
distribution of the direct normal insolation (DNI), 100
the mass flow rate of the solar field HTF, the solar
field inlet and outlet temperatures and the steam 80

mass flow rate through the Rankine cycle are 60

shown in Fig. 6 and Fig. 8-11, respectively.
40

Similar results are shown for a cloudy day (May 20

15, 2001) in Fig. 12–17. The distribution of the
direct normal insolation (DNI) for May 15, 2001 is 0
4 6 8 10 12 14 16 18 20
uneven, Fig. 12, so is the net power output, Fig. 13. Time (hr)
Accordingly, conventional backup system should Figure 8 Solar-field HTF mass flow rate for
be available to attain the required power (10 MWe) clear day
delivery.

1100
300
1000
Solar Field Inlet Temperature ( oC)

900
250
800

700 200
DNI (W/m2)

600

500 150

400
100
300

200
50
100

0
4 6 8 10 12 14 16 18 20 0
4 6 8 10 12 14 16 18 20 22
Time (hr)
Time (hr)
Figure 6 Direct Normal Insolation (DNI) for Figure 9 Solar-field inlet temperature for clear
clear day day

12
450

400
Solar Field Outlet Temperature ( oC)

10
350
8
300
Power (MW)

250
6
200
4
150

100
2

50
0
4 6 8 10 12 14 16 18 20 0
4 6 8 10 12 14 16 18 20 22
Time (hr)
Time (hr)
Figure 7 Net power for clear day Figure 10 Solar-field outlet temperature for
clear day

Journal of EEA, Vol. 27, 2010 87

Mekuannint Mesfin and Abebayehu Assefa

16
120

14
100
12

HTF Mass Flow Rate (kg/s)

Steam Flow Rate (kg/s)

80
10

8 60

6
40

4
20
2

0
0 4 6 8 10 12 14 16 18 20
4 6 8 10 12 14 16 18 20 Time (hr)
Time (hr)
Figure 14 Solar-field HTF mass flow for
Figure 11 Mass flow rate of steam for clear
cloudy day
day

300
800

Solar Field Inlet Temperature ( oC)

700 250

600
200

500
DNI (W/m 2)

150
400

100
300

200 50

100
0
4 6 8 10 12 14 16 18 20 22
0 Time (hr)
4 6 8 10 12 14 16 18 20
Time (hr)
Figure 15 Solar-field inlet temperature for
Figure 112 Direct Normal Insolation (DNI) for cloudy day
cloudy day
450
10
400
Solar Field Outlet Temperature ( oC)

9
350
8
300
7
250
Power (MW)

6
200
5
150
4
100
3
50
2
0
1 4 6 8 10 12 14 16 18 20 22
Time (hr)
0
4 6 8 10 12 14 16 18 20
Time (hr) Figure 16 Solar-field outlet temperature for
cloudy day
Figure 13 Net power for cloudy day

88 Journal of EEA, Vol. 27, 2010

 Modeling, Simulation and Performance Evaluation

11

10
The performance of the model developed in the
9
TRNSYS is evaluated by considering available
8
data for the year 2001 EC. Two days have been
Steam Flow Rate (kg/s)

7
selected for the simulation, one for solar clear day
6 and the other for cloudy day. The results of the
5 simulation have shown that the mass flow rate of
4 the HTF from the solar field and gross power
3 produced are heavily dependent on DNI.
2

1 REFERENCES
0
4 6 8 10 12
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.
14 16 18 20
[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] Schwarzbözl, 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.

Journal of EEA, Vol. 27, 2010 89