Sustainable Energy Technologies and Assessments 42 (2020) 100857

Contents lists available at ScienceDirect

Sustainable Energy Technologies and Assessments

journal homepage: www.elsevier.com/locate/seta

Original article

Techno-economic analysis of a hybrid solar Thermal-PV power plant

Sadegh Khajepour , Mehran Ameri *
Mechanical Engineering Department, Shahid Bahonar University of Kerman, Kerman, Iran

A R T I C L E I N F O A B S T R A C T

Keywords: This paper introduces a new combination of solar fields to reduce LCOE and increase solar share. One Solar Field
Solar thermal power plant (PTC or LFR) with molten salt as heat transfer fluid is used to store energy during the day, and two independent
Photovoltaic LFRs are used for direct steam generation for high-pressure turbines and low-pressure turbines. Due to the low
Linear Fresnel Reflector
price of natural gas, using solar field is not economical, and using thermal energy storage is affordable only for a
Parabolic Trough Collector
Hybrid systems
higher price of natural gas. The PTC with a lower energy storage than LFR has a higher solar share, but using LFR
is more economical. In this paper, the effects of several parameters on the LCOE are investigated, and the key
factors are introduced. According to the results, three independent LFRs create the best combination to meet the
required energy. The LCOE of a photovoltaic power-plant is less than the LCOE of a solar thermal one, but the
photovoltaic power plant cannot supply the energy needed at night. Therefore, in the end, to further increase
system performance and reduce the LCOE, the solar thermal power-plant is combined with a photovoltaic system
and the effects of using different sizes of photovoltaic power-plant on the hybrid system are investigated.

increased by using thermal energy storage. Parabolic Trough Collector

(PTC), Linear Fresnel Reflector, and solar tower are three conventional
Introduction
solar collectors. There are numerous studies on PTC and LFR since they
can be used for small solar TPP.
The human need for energy is continually on the rise. From the in­
Reddy et al. [1] examined a PTC collector and showed that the
dustrial revolution onwards, most of the required electricity has been
thermal efficiency of the collector at less sunny and high-declination
produced by fossil fuels. But the increasing use of these fuels has resulted
areas is more sensitive to optical error. Khaksar et al. [2] examined
in global warming and environmental pollution. Therefore, many
the effect of wind speed and the addition of nanoparticles to the Heat
countries have established research projects for replacing fossil fuels
Transfer Fluid (HTF). They concluded that thermal efficiency decreases
with renewable energy.
7% in a wind speed of 10 m/s compared to still air, and 22% in a 5%
Renewable energy takes many forms, such as solar energy, wind
nanoparticle addition to pure HTF.
power, tidal power, wave power, hydropower, geothermal, and biomass.
Zineb et al. [3] studied the Noor Power Plant in Morocco and showed
Solar energy and wind power have attracted more attention because of
that DNI value is a crucial factor in determining annual electricity
their availability in more numerous parts of the world. These energies
generation.
are inexhaustible sources, but due to fluctuations in their energy pro­
Ma et al. [4] studied a PTC collector. They determined that ignoring
duction, they have little reliability.
the inflation of daily radiation leads to the non-optimal performance of
One-way to continuously produce energy or to produce the required
the system and causes problems in the operation of the solar system.
amount of electricity from renewable energy systems is using energy
Mansour et al. [5] used a 2D thermal modeling for analyzing a PTC
storage. Storage systems like battery, pumped storage, and thermal
collector. Their research showed that PTCs performance is highly sen­
storage are widely employed for storing energy. Batteries are used
sitive to solar radiation, the angle of incidence, aperture width, inlet
mostly for storing energy on small scales. Pumped-storage requires a
HTF temperature, and mass flow rate. The daily maximum energy effi­
large amount of water and a significant height difference, which are not
ciency is 67.91% and 31% for summer solstice and winter solstice,
available in every area. Moreover, developing countries tend to use
respectively.
traditional TPP, due to the low cost of fossil fuels.
Wang et al. [6] by studying a PTC collector in China showed that the
The Levelized Cost of Energy (LCOE) for solar power can be reduced
optical efficiency of this collector is about 70%, and the temperature
by connecting SF to traditional TPP. Also, the solar share can be

* Corresponding author at: P.O.B. 133-76175, Tehran, Iran.

E-mail address: [email protected] (M. Ameri).

https://doi.org/10.1016/j.seta.2020.100857
Received 24 January 2020; Received in revised form 19 July 2020; Accepted 4 October 2020
Available online 15 October 2020
2213-1388/© 2020 Elsevier Ltd. All rights reserved.
S. Khajepour and M. Ameri Sustainable Energy Technologies and Assessments 42 (2020) 100857

Nomenclature PL Laspeyres price index

PP Paasche price index
Qin the radiation emitted toward the SF n Year
Qabsorbed solar thermal radiation absorbed by the collector SFs Solar Fields
Afield area of the SF DS Direct Steam
DNI Direct Normal Irradiance HPT High-Pressure Turbines
ηopt optical efficiency LPT Low-Pressure Turbines
ηendloss end losses PNG Price of Natural Gas
θi incident angle TPP Thermal Power Plant
L receiver’s length LCOE Levelized Cost of Energy
Lf the focal length of the absorber tube to the mirrors PTC Parabolic Trough Collector
ηopt0 optical efficiency during zero angles of incidence of sun LFR Linear Fresnel Reflector
rays HTF Heat Transfer Fluid
IAML Longitudinal Incidence Angle Modifier TES Thermal Energy Storage
IAMt Transversal Incidence Angle Modifier GW Giga-watt
QhlHTF heat loss of a linear receiver in the collector’s focal line NREL National Renewable Energy Laboratory
Qhlpiping waste of heat from system pipes IMO Iranian Meteorological Organization
N system lifetime (per year) SM Solar Multiple
En the electricity generated in year n LF1 First Solar Field
r discount rate LF2 Second Solar Field
p1i the price of services or goods in the base year LF3 Third Solar Field (PTC kind)
q1i consumption of goods i, one year after the base year

difference between the outer surface of the absorber and HTF is incident angle is more influential than the transversal incident angle on
considerable. Ahelgren et al. [7] simulated a PTC collector for several the collector performance.
areas with different latitudes. They showed that the annual thermal Pulido-Iparraguirre et al. [10] studied an LFR and concluded that
performance could be estimated only by the total yearly DNI and the using a cylindrical mirror against the flat mirror with a suitable tilt can
latitude. This result helps to evaluate collector performance quickly and improve optical efficiency. It must be noted, however, that these solu­
saves time and money when the accuracy of 12.5% is enough. tions increase the initial cost of SF. These also showed that the receiver
Barbon et al. [8] analyzed a Linear Fresnel Reflector (LFR) and height does not need that much accuracy because the optical perfor­
showed that the sub-component of movement units and the number of mance remains constant for a wide range of these parameters. Bellos
mirrors have the most considerable influence on the total cost of the LFR et al. [11] Show that for high HTF temperatures, most of the heat loss in
collector. Bellos et al. [9] studied an LFR and showed that the maximum the LFR receiver occurs through radiation, and for temperatures higher
optical efficiency of the LFR collector is 61%, and the longitudinal than 900 K, thermal performance is close to zero. They also showed that
exergy efficiency is maximum for inlet temperature close to 650 K.
Lancereau et al. [12] showed that lift and drag forces for LFR col­
lectors are less than those of PTC collector. Therefore, a smaller force is

Table 1
IAM polynomial coefficients for SUPERNOVA (NOVA 1).
a0 0.9808 b0 1.0032

a1 0.047 b1 − 0.2987
a2 − 0.075 b2 0.5276
a3 − 0.242 b3 − 1.6328
a4 0.01 b4 0.7423

Table 2
Geometrics and economical parameters of LFR and PTC.
Parameter LFR PTC
2
ACollector 513.6m 656m2
Lf 7.4 m 2.15 m
L 44.8 m 115 m
ηopt0 0.643093 0.721319
Site improvements 20$/m2 30$/m2
SF 170$/m2 270$/m2
HTF system 40$/m2 80$/m2
Fig. 1. Definition of the transversal and longitudinal angles [28].

2
S. Khajepour and M. Ameri Sustainable Energy Technologies and Assessments 42 (2020) 100857

Table 3 heat transfer oil. However, economic and environmental parameters are
Thermal correction factors for PTR70 2008. more favorable for the salt-based configuration than the oil-based one.
PTR70 2008 Thermal correction factors The SF integrated with TES and fuel backup is techno-economically a
better configuration compared to the other ones.
4.05 A0
Mart et al. [18], by examining a hybrid CSP-PV power plant, showed
0.247 A1
that photovoltaic panels are more suitable for medium to high latitude
− 0.00146 A2
areas, especially in regions with semi-climatic conditions, and solar TPP
5.65e-06 A3
with thermal storage perform their best in low latitude, desert areas.
7.62e-8 A4
Research by Green et al. [19] showed that hybrid CSP-PV power
− 1.7 A5
plant’s performance could increase by considering a high angle for
− 0.025 A6
photovoltaic panels.
Research by Liu et al. [20] showed that due to the high cost of the
required to drive LFR systems. So, the cost of the structure of the LFR battery, using hybrid CSP-PV power plants is more economical than
collector is less than that of the structure of the PTC collector. using photovoltaic panels and batteries for the production of stationary
The reported researches show that by using hybrid solar system like electricity.
solar-biomass hybrid power generation system, the system’s thermo­ A study done by Roni et al. [21] showed that the cost of electricity
dynamic performance improves, while, LOCE of the solar system re­ generated by photovoltaic panels is less than solar TPP and occupies less
duces [13,14]. space. However, solar TPP is more efficient because of its ability to store
Baniasad and Ameri [15] used two separate LFR collectors. The first energy.
SF was used to provide the required energy and the second one to store A study by Bode et al. [22] showed that using photovoltaic panels as
energy. This research showed that using the second SF can improve an auxiliary system in solar TPP can reduce the time of capital return.
system performance. It must be considered that the SF area increased Petrollese and Cocco [23] showed that if photovoltaic capacity in­
without TES (Thermal Energy Storage), leading to a rise in investment creases in a hybrid CSP-PV power plant, it will increase solar defocus.
cost and a waste of energy [16]. Although using the battery with a hybrid CSP-PV power plant is not
Boukelia et al. [17] simulated eight configurations of PTC solar economically feasible, it can increase the capacity factor up to 90%.
thermal plants with/out TES and fuel backup systems, using heat Most of the battery effect is in winter [24,25].
transfer oil and heat transfer salt. The results showed that using a SF In previous studies, only the use of two independent SF was inves­
integrated with TES, fuel backup, and heat transfer salt leads to the tigated: the first one for providing the required energy and the second
highest overall energy efficiency compared to other configurations one for storing energy. Our previous research investigated the effects of
while the highest overall exergy efficiency is reached in this system by using three LFR fields [24]. It was concluded that the use of three LFR
fields could reduce LCOE and increase solar share.

Fig. 2. Solar resource map for Iran (“Kerman map”, “The World Bank, Solar resource data: Solargis”) [31,32].

3
S. Khajepour and M. Ameri Sustainable Energy Technologies and Assessments 42 (2020) 100857

tube covered by a transparent cover. The air existing in the gap between
the absorber tube and the transparent cover is vacuumed to minimize
the convection and conduction heat losses. This assembly is positioned
on a rigid structure tracking solar radiation.
In the 1960s, the first LFR was made at the University of Genoa. LFR
is made up of several flat mirrors placed on a flat structure. The mirrors
reflect the sunshine on the absorber, through which HTF flows, and
increase the temperature to a range of 150 to 520 degrees Celsius. LFR
has several advantages over PTC. Because of the small lift and drag
forces from the wind and the use of flat mirror plates, LFR investment
cost is lower than that of PTC. Unlike PTC, LFR has no challenge sealing
HTF from the absorber due to the constant absorber system.
In this research, the same absorber tube is used for both the PTC and
LFR. The following equation can calculate the useful heat of SF:
QLFR = Qin − QhlHTF − Qhlpiping (1)

whereQin is calculated by the characteristics of the SF and the efficiency

model [27].
Qin = ηopt .ηendloss .Qabsorbed (2)

where
Qabsorbed = Afield .DNI (3)
Fig. 3. Temperature changes.
Lf
ηendloss = 1 − tan(θi ) (4)
PTC has a higher efficiency than LFR. Previous research shows that L
combining a photovoltaic power plant with a solar TPP reduces the ηopt = ηopt0 .IAMt .IAML (5)
LCOE. This article has two innovations. In the first step, it investigates
the effect of using three SFs, LFR and PTC (two independent SF for DS For a North–South arrangement of LFR, the transversal and longi­
generation for HPT and LPT and one SF with TES using heat transfer salt tudinal angles that define in Fig. 1. are obtained via the following re­
for storing energy) for the first time, and it answers the question of lations [28]:
whether using PTC can further reduce the LCOE; in the second step, the
sinγs
effect of using a PV power plant along with the optimal solar system ∅T = arctan( ) (6)
tanαs
(which is reviewed in this study) on solar share and LCOE is
investigated. ∅L = arcsin(cosγs .cosαs ) (7)

Parabolic Trough collector and Linear Fresnel Reflector in which as expresses the sun elevation angle (sun altitude). trans­
verse and longitudinal incidence angle modifier expressions are based
PTC is a mature technology to the extent that its installations around on a polynomial which is modified to the following form:
the world produce 4.2 Gigawatt (GW) [26]. IAMt = a0 + a1 .∅t + a2 ∅2t + a3 ∅3t + a4 ∅4t (8)
PTC is made of a reflector plate of parabolic shape and an absorber

Fig. 4. DNI for Kerman province.

4
S. Khajepour and M. Ameri Sustainable Energy Technologies and Assessments 42 (2020) 100857

Fig. 5. Schematic diagram of the TPP for two combinations.

Table 4 Qhl4 =
A3 4
(T − T4i ) (15)
Thermodynamic properties of the state points of the Rankine power cycle that is 4 o
shown in Fig. 5. Constant factors A0, A1 … A6 are demonstrated in Table 3 for the
Point Pressure Temperature Mass flow rate Enthalpy (kJ/ surveyed recipient.
(bar) (◦ C) (kg/s) kg) Finally, Qhlpiping is considered 10 W per square meter of the SF.
1 0.34 73 96.45 303.5
2 10 73.2 96.45 305.8
Methodology
3 10 198 124.63 844.9
4 40 199 124.63 849.3
5 40 250 137.33 1087 Concentrated solar power (CSP)
6 100 253 137.33 1099
7 100 377 137.33 3023 Until now, the effects of using two types of solar collectors (PTC and
8 40 264 12.67 2850
9 10 198 28.18 2690
LFR) for three independent SFs at the same time have not been inves­
10 10 198 96.45 2690 tigated. This research studies several collector combinations. Here, two
11 10 377 96.45 3207 independent SFs are used to generate DS for HPT and LPT and another
12 0.34 73 96.45 2631 one to store energy during the day to supply the required energy for the
Pump efficiency and turbine efficiency are the same, and equal to 85%. TPP thermal plant during the night. LFR can be used to generate DS and
efficiency is 31%. increase melt salt temperature, while PTC can be used to increase melt
salt temperature only.

IAM l = b0 + b1 .∅l + b2 ∅2l + b3 ∅3l + b4 ∅4l (9) Air temperature and solar radiation
The values of coefficients an and bn for nova1 collector are provided Kerman (30.2839◦ N, 57.0834◦ E), the largest province of Iran with a
in table 1: total area of 181714km2 ,was chosen to simulate the new solar TPP. The
Geometrics and economical parameters of LFR and PTC are provided solar resource map for Kerman is shown in Fig. 2. The Iranian Meteo­
in table 2. The value of QhlHTF that provided by the National Renewable rological Organization (IMO) has announced that the annual solar ra­
Energy Laboratory (NREL) is determined by Eq. (10) [27]: diation in this province is 7625 MJ/m2 [29]. This city has the highest
{ } annual average of total sunshine hours per month, which is around 274
∫ To (A0 +A1 .(THTF − Tamb )+A2 .T2HTF +A3 .T3HTF + h [30].
√ ̅̅̅̅̅̅
̅ dTHTF
Ti
A4 .Ib .IAM.cosɵ.T2HTF + Vw .(A5 +A6 .(THTF − Tamb ))) The temperature graph and the DNI graph for Kerman province are
QhlHTF = given in Figs. 3 and 4, respectively.
(To − Ti )
Fig. 4 shows that solar radiation fluctuates during a year. Some
(10)
research considers simulated solar radiation data for sunny days only
This integrated result can be simplified to instead of the actual solar radiation data. Cloudy days, however, should
also be taken into account; otherwise, the solar TPP would be considered
HeatLossFactor Qhl1 + Qhl2 + Qhl3 + Qhl4
QhlHFT = . (11) in an ideal state, and simulation outcomes would show an ideal solar
Collectoraperture (To − Ti )
TPP, which, with lower PNG, introduces the use of solar thermal energy
where as affordable.

√̅̅̅̅̅̅̅ Solar fields

Qhl1 = (A0 + A5 . Vw )(To − Ti ) (12)
In the following, two different combinations for the solar power
√̅̅̅̅̅̅̅ T2o − T2i plant are analyzed. It should be noted that different scenarios for
Qhl2 = (A1 + A6 . Vw )( − Tamb .(To − Ti )) (13) Combination 1 have been studied in our previous articles [33].
2

(A2 + A4 .Ib .IAM.cosɵ) 3 Combination 1: Using two LFR fields to produce DS for HPT and LPT and an
Qhl3 = (To − T3i ) (14) LFR field for energy storage. In this combination, three LFR fields are
3
used at the same time to provide energy for HPT and LPT and to store
energy. LF1 and LF2 supply all or part of the energy required by the LPT

5
 S. Khajepour and M. Ameri
Table 5
The optimal solar multiple for each SF by changing the PNG.
Defocus Solar LCOE H SM for SM for SM for PNG (₵/m3 ) number of Defocus Solar LCOE H SM for SM for SM for PNG (₵/m3 ) number of
(%) Share (₵/kWh) (hours) #LF2 #LF1 #LF3 [$/MMBtu] combinations (%) Share (₵/kWh) (hours) #LF2 #LF1 #PTC [$/MMBtu] combinations
(%) (%)

0 0 5.32 0 0 0 0 8.85 Combination 0 0 5.32 0 0 0 0 8.85 Combination

[2.5] #1 [2.5] #2
0 0 5.99 0 0 0 0 10.61 0 0 5.99 0 0 0 0 10.61
[3] [3]
0.03 18.63 6.64 0 1.05 1.05 0 12.38 0.03 18.63 6.64 0 1.05 1.05 0 12.38
[3.5] [3.5]
0.65 20.47 7.18 0 1.14 1.16 0 14.15 0.65 20.47 7.18 0 1.14 1.16 0 14.15
[4] [4]
2.14 21.96 7.71 0 1.26 1.26 0 15.92 2.14 21.96 7.71 0 1.26 1.26 0 15.92
[4.5] [4.5]
4.19 23.42 8.22 0 1.35 1.38 0 17.69 4.19 23.42 8.22 0 1.35 1.38 0 17.69
[5] [5]
5.89 24.38 8.73 0 1.43 1.46 0 19.46 5.89 24.38 8.73 0 1.43 1.46 0 19.46
[5.5] [5.5]
6

7.84 25.33 9.23 0 1.54 1.55 0 21.23 7.84 25.33 9.23 0 1.54 1.55 0 21.23
[6] [6]
10.92 26.57 10.22 0 1.66 1.68 0 24.77 10.92 26.57 10.22 0 1.66 1.68 0 24.77
[7] [7]
13.92 27.60 11.20 0 1.8 1.81 0 28.30 13.92 27.60 11.20 0 1.80 1.81 0 28.30
[8] [8]
13.52 62.91 12.05 12 1.71 1.72 2.67 31.84 12.74 41.39 12.12 4 1.75 1.75 0.82 31.84
[9] [9]
16.53 66.38 12.53 12.5 1.79 1.79 3.02 35.38 13.66 66.89 12.62 12 1.79 1.79 2.3 35.38

Sustainable Energy Technologies and Assessments 42 (2020) 100857

[10] [10]
17.78 67.13 12.98 12.5 1.81 1.84 3.1 38.92 15.65 68.21 13.06 12 1.86 1.86 2.41 38.92
[11] [11]
19.76 69.18 13.41 13 1.89 1.89 3.33 42.46 17.25 69.18 13.49 12 1.93 1.93 2.49 42.46
[12] [12]
22.45 70.63 13.83 13 1.98 1.97 3.55 46.00 18.61 69.91 13.90 12 1.98 1.99 2.55 46.00
[13] [13]
S. Khajepour and M. Ameri Sustainable Energy Technologies and Assessments 42 (2020) 100857

Economic analysis
The Levelized Cost of Energy (LCOE) is calculated by the following
equation, which is usually used to survey the electricity prices of
different power generation systems [18]:
∑N
AnnualCostsn ∑ N
En
LCOE = (InitialCosts + n )/( ) (16)
n=1
(1 + r) n=1
(1 + r)n

The Fischer index shown in the following equation is used to calcu­

late the discount index. [34]:
√̅̅̅̅̅̅̅̅̅̅
PF = PL PP (17)
The following equations can determinePL andPP .
∑n 1 0
pi qi
PL = ∑i=1
n 0 0
(18)
i=1 pi qi

∑n 1 1
pi qi
PP = ∑i=1
n 0 1
(19)
i=1 pi qi

Finally, the discount rate in year n can be determined by the

following equation, after the discount index is calculated:

PnF − Pn−F 1
rn = (20)
Pn−F 1
Fig. 6. Solar share and defocus in terms of rising PNG.
Several parameters, such as collector’s cost, the component of the
traditional plant, fossil fuel cost, and cost of workers for design, con­
struction, and maintenance, affect the cost of the solar power plant.
The discount rate for each country is calculated based on the cur­
rency of the country; this amount for Iran is 7.2%. During the last
decade, the value of the rial has continuously decreased relative to the
dollar. Worker’s salary and PNG increase in rial terms every year based
on the discount rate. However, the rate of the falling value of the rial is
more than the amount of discount rate proclaimed by the Iranian Sta­
tistics Center. So, a worker’s salary and PNG decrease in dollar terms
every year. It seems that these decreases in cost and increases in the
price of other parts of plants neutralize each other in dollar terms. Since,
in this research, all values are calculated in the dollar, the amount of
discount rate is considering equal to zero.

Simulation of the system and results

Two combinations have been simulated, and optimizations for
several PNGs have been carried out. These results are demonstrated in
Table 5.
For Combinations 1 and 2, it is not profitable to use one SF with
molten salt as HTF for PNG up to 31.84 ₵/m3 , so LCOE is equal for
Combinations 1 and 2. At the PNG 31.84 ₵/m3 and the higher PNG,
LCOE for Combination 1 is lower than that for Combination 2.
Fig. 7. Effects of PNG increase on combinations. For Combinations 1 and 2 and PNG higher than 31.84 ₵/m3 , using
energy storage is economical; it is also more cost-effective due to the
increase in the area of the SF. According to Fig. 6, the amount of solar
share has sharply increased.
and the HPT. LF3 supplies the rest of the energy demanded by the LPT
In this study, the TPP should produce constant electricity in all hours
and the HPT. When the energy produced by LF3 is more than required,
during a year. During nights and cloudy days, the SF cannot supply the
the surplus energy is stored in thermal energy storage. At night and
required energy for the TPP. The effects of PNG increase on combina­
when LF1, LF2, and LF3 cannot supply the needed energy, thermal en­
tions are shown in Fig. 7. When the PNG is more than 28.30 ₵/m3 , by
ergy storage meets the required energy with the help of auxiliary fossil
increasing energy storage capacity, energy is stored in the thermal
foils.
storage during the day, and so some of the required energy can be
Information about solar TPP with the points aforesaid in Fig. 5 is
supplied by the thermal storage during the night; therefore, the con­
reported in Table 4.
sumption of natural gas and also LCOE decrease. According to Fig. 8,
LCOE is minimized in an optimum size of energy storage, and after that,
Combination 2: Using two LFR fields to produce DS for HPT and LPT and a
by increasing the size of energy storage, LCOE increases sharply.
PTC field for energy storage. According to Fig. 5b, combination 2 is the
The performance of PTC is higher than that of LFR, so, thermal
same as combination 1, except that instead of using an LFR field, a PTC
storage capacity in Combination 1 is higher than Combination 2. As a
field is used for energy storage.
result, according to Table 5, SM of the indirect steam generator SF for
Combination 1 is always higher than that of Combination 2. Still, SM of
other SFs for Combination 1 and 2 are almost the same.

7
S. Khajepour and M. Ameri Sustainable Energy Technologies and Assessments 42 (2020) 100857

Fig. 8. Changes in the LCOE by changing the storage capacity₵.

Fig. 9. Changes LCOE by changing the SM of the SF for each SFs, the PNG is 31.84 ₵/m3

Fig. 9 shows that when the PNG is 31.84 ₵/m3 , SM of the first (LF1), Sensitivity analysis of CSP
the second (LF2) and the third SF (LF3 or PTC) in Combination 1 are In Fig. 10, the effects of a 20 percent change in DNI and the cost of
1.72, 1.71 and 2.67, respectively; and in Combination 2, 1.75, 1.75 and several parameters of the SFs are depicted for Combinations 1 and 2 for
0.82, respectively. These SMs are equal to 782213m2 , 146890m2 , the PNG of 31.84 and 35.38 ₵/m3 . The cost of fuel and DNI are crucial
1,454,515m2 for Combination 1 and 796593m2 , 150,998m2 and factors in determining LCOE. For Combination 1 and 2, the cost of LFR
390,320m2 for Combination 2, respectively. It must be noted that and PTC are the most effective on LCOE, respectively. By increasing the
although SMs for the second and third SF in all combinations are the cost of fuel from 31.84 to 35.38 ₵/m3 , using a bigger TES and SF is
same because of different reference areas for these SFs, the area of these affordable, and natural gas consumption decreases. Therefore, the
SFs are different. For example, for Combination 2, when the PNG is impact of fuel cost decreases, and that of other parameters increases.
31.84 ₵/m3 , the area of the first SF is 5.28 times the area of the second When DNI is increasing, the amount of fuel consumption decreases, and
SF, and according to Fig. 9, SM of the first SF is more effective on LCOE because the cost of fuel is a crucial factor, LCOE decreases. Increasing
compared to SM of the second SF. the cost of each component of the solar system (cost of LFR, cost of PTC,
and cost of storage) makes the other components of the solar system
more significant and decreases the impact of the cost increase. As a

8
S. Khajepour and M. Ameri Sustainable Energy Technologies and Assessments 42 (2020) 100857

Fig. 10. The effect of 20 percent change in the amount of solar radiation, the cost of LFR, the cost of PTC, the cost of the TPP and the cost of fuel on the total cost.

result, the effect of increasing costs is less than that of decreasing costs. LCOE of the PV system. In this section, the effect of using a photovoltaic
system in several sizes and the impact of the angle of installation of solar
Hybrid CSP-PV panels on Combination 1 are investigated. The system is displayed
schematically in Fig. 11.
Combination 1 has the lowest LCOE among other combinations. As The highest output of the photovoltaic system is during peak hours of
mentioned at the beginning of the present work, studies show that sunlight. During these times, the photovoltaic system supplies part of the
combining photovoltaic panels with solar TPP can improve power plant energy consumed and causes the solar TPP to store more thermal energy,
performance and reduce LCOE. SAM, a performance and financial model leading to a decline in the use of fossil fuels. As shown in Fig. 12, when
developed by NREL [35], is used to estimate hourly energy output and the capacity of the photovoltaic system increases, the LCOE decreases.

9
S. Khajepour and M. Ameri Sustainable Energy Technologies and Assessments 42 (2020) 100857

Fig. 11. Hybrid Solar TPP with two SFs to produce DS for HPT and LPT and one LFR of SF for storing energy and a PV system.

Fig. 12. The effect of PV size system and PV inclination angle on LCOE for different PNG.

This decreases for higher PNG due to the use of heat storage, and hence photovoltaic capacity. This increase is significant in low PNG due to the
the reduction in the use of fossil fuels is more significant. Also, inadequate use of solar TPP. As the photovoltaic system capacity in­
increasing the angle of photovoltaic panels at high PNG (due to higher creases, the area of LFR fields decreases. For this reason, at PNG of 28.30
power generation in winter and more storage of thermal energy) reduces ₵/m3 , solar share for the hybrid CSP-PV power plant with 60 MW
the LCOE. The LCOE of solar PV for tilt angles is shown in Table 6. photovoltaic is higher than other systems (According to Fig. 14, in the
As shown in Fig. 13, solar share increases with the increase of larger sizes of the photovoltaic system, although the solar share

10
S. Khajepour and M. Ameri Sustainable Energy Technologies and Assessments 42 (2020) 100857

Table 6
The LCOE of solar PV for tilt angles.
The tilt angle of the PV array LCOE of solar PV
(Cent USD to kWh)

0 3.27
10 3.06
20 2.95
30 2.92
40 2.96
50 3.08
60 3.29
70 3.63

increases due to the greater use of the photovoltaic panel, the total solar
share reduces due to the reduced size of LFR of SF and the non-
optimization of the application of TES). Increasing the PNG to 31.84
₵/m3 increases the impact of solar thermal power on the solar share due
to the use of thermal storage. Also, in low PNG, the optimum angle of the
SF for increasing solar share is the latitude of the plant’s location. At
high PNG, increasing the angle of photovoltaic panels can increase solar
share. At the PNG of 28.30 ₵/m3 and zero-degree SF angles, the solar
share is above all other angles due to the increased energy production in
the summer and the optimized use of heat storage. In higher PNG and
the use of heat storage for different angles, the optimum angle of
photovoltaic panels for increasing solar share is shown to be more than
the latitude of the plant’s location. Fig. 14. The effect of PV size system and PV inclination angle on solar share for
The percentage of solar share of each component of the hybrid CSP- different PNG.
PV plant for two capacities of 10 and 100 MW photovoltaic systems is
shown in Fig. 15. For PNG below 10.61 ₵/m3 , the solar TPP is not
optimal since there is solar share only from the photovoltaic system.
Increasing the capacity of the photovoltaic system reduces the size of the the hybrid CSP-PV system decreases. LCOE of the PV system is cheaper
optimal solar TPP. It decreases the percentage of solar share of the solar than the LCOE of the CSP system, so PV cost is less important than other
thermal field in the total solar share. With an increase in PNG above parameters.
28.30 ₵/m3 , the total solar share will increase sharply due to the opti­
mization of the use of thermal storage. Conclusion

Sensitivity analysis of hybrid CSP-PV In this article, the effect of the concurrent use of two types of SFs
In Fig. 16, the effects of a 20 percent change in DNI, PV size and the integrated with a traditional TPP, and the impact of using a PV system
cost of several parameters of the SFs are depicted for Hybrid CSP-PV for with the solar TPP have been investigated. One SF (PTC or LFR) is used
the PNG of 31.84 and 35.38 ₵/m3 . For this combination, DNI is more to store energy during the day, and the other SFs are used to generate DS
important than the cost of fuel in determining LCOE. By increasing PV for HPT and LPT. During the night and when solar energy is not avail­
system size, the size of other parameters of CSP plant decreases, and also able, TES and auxiliary fossil fuel supply energy.
because of saving more energy in TES, natural gas consumption de­
creases. Therefore, by increase the size of the PV system, the net cost of • Using two independent SFs for the HPT and LPT can decrease LCOE.

Fig. 13. The effect of PV size system and PV inclination angle on solar share for different PNG.

11
S. Khajepour and M. Ameri Sustainable Energy Technologies and Assessments 42 (2020) 100857

Fig. 15. Hybrid CSP-PV Solar share VS PNG.

Fig. 16. The effect of 20 percent change in the amount of solar radiation, the cost of LFR, the cost of PTC, the cost of the TPP, the cost of fuel, the cost of PV and PV
size on the total cost.

• Using a SF with molten salt as HTF and TES are economical only Writing - original draft, Writing - review & editing, Supervision.
when the PNG is higher than 31.84 ₵/m3 .
• For the PNG higher than 31.84 ₵/m3 , using PTC instead of LFR leads Declaration of Competing Interest
to smaller TES. For the third SF, however, utilizing LFR is more
economical than using PTC for the third SF. The authors declare that they have no known competing financial
• for the PNG higher than 31.84 ₵/m3 , the concurrent use of three LFRs interests or personal relationships that could have appeared to influence
is more cost-effective the work reported in this paper.
• By combining the photovoltaic system with the solar TPP, the
optimal size of the solar thermal field is reduced. Acknowledgments
• Hybrid CSP-PV increases the possibility of more storage of solar heat
as photovoltaic panels supply a portion of the required energy during This research was supported by Kerman Regional Electric Company.
the hours with the highest solar radiation.
Appendix
CRediT authorship contribution statement

Sadegh Khajepour: Conceptualization, Methodology, Software,

Validation, Writing - original draft. Mehran Ameri: Conceptualization,

12
S. Khajepour and M. Ameri Sustainable Energy Technologies and Assessments 42 (2020) 100857

Fig. A1. Solar share and total defocus Changes by Changing the SM for each SF’s.

13
S. Khajepour and M. Ameri Sustainable Energy Technologies and Assessments 42 (2020) 100857

Fig. A2. LCOE Changes by Changing the SM for each SF’s.

14
S. Khajepour and M. Ameri Sustainable Energy Technologies and Assessments 42 (2020) 100857

Fig. A3. Solar share and total defocus changes by changing the SM for each SF’s.

References [6] Wang R, Sun J, Hong H, Liu Q. An on-site test method for thermal and optical
performances of parabolic-trough loop for utility-scale concentrating solar power
plant. Sol Energy 2017;153:142–52. https://doi.org/10.1016/j.
[1] Reddy KS, Ajay CS, Nitin Kumar B, Kumar BN. Sensitivity study of thermal
solener.2017.05.053.
performance characteristics based on optical parameters for direct steam
[7] Ahlgren B, Tian Z, Perers B, Dragsted J, Johansson E, Lundberg K, et al. A simpli fi
generation in parabolic trough collectors. Sol Energy 2018;169:577–93. https://
ed model for linear correlation between annual yield and DNI for parabolic trough
doi.org/10.1016/j.solener.2018.03.088.
collectors. Energy Convers Manag 2018;174:295–308. https://doi.org/10.1016/j.
[2] Khakrah H, Shamloo A, Kazemzadeh HS. Exergy analysis of parabolic trough solar
enconman.2018.08.008.
collectors using Al2O3/synthetic oil nanofluid. Sol Energy 2018;173:1236–47.
[8] Bay C, Barb A, Barbón A, Sánchez-Rodríguez JA, Bayón L, Bayón-Cueli C, et al. Cost
https://doi.org/10.1016/j.solener.2018.08.064.
estimation relationships of a small scale linear Fresnel re fl ector. Renew Energy
[3] Aqachmar Z, Allouhi A, Jamil A, Gagouch B, Kousksou T. Parabolic trough solar
2018. https://doi.org/10.1016/j.renene.2018.09.060.
thermal power plant Noor I in Morocco. Energy 2019. https://doi.org/10.1016/j.
[9] Bellos E, Tzivanidis C, Papadopoulos A. Secondary concentrator optimization of a
energy.2019.04.160.
linear Fresnel reflector using Bezier polynomial parametrization. Sol Energy 2018;
[4] Ma M, Zhao L, Deng S, Lin S, Zhang Y, Ni J, et al. Methodology for determining the
171:716–27. https://doi.org/10.1016/j.solener.2018.07.025.
design radiation for a PTC heating system based on non-guaranteed days. Sol
[10] Pulido-Iparraguirre D, Valenzuela L, Serrano-Aguilera J-J, Fernández-García A.
Energy 2018;174:97–107. https://doi.org/10.1016/j.solener.2018.08.079.
Optimized design of a Linear Fresnel Reflector for solar process heat applications.
[5] Mansour K, Boudries R, Dizene R. Optical, 2D thermal modeling and exergy
Renew Energy 2018. https://doi.org/10.1016/j.renene.2018.08.018.
analysis applied for performance prediction of a solar PTC. Sol Energy 2018;174:
[11] Bellos E, Tzivanidis C, Papadopoulos A. Optical and thermal analysis of a linear
1169–84. https://doi.org/10.1016/j.solener.2018.09.040.
Fresnel reflector operating with thermal oil, molten salt and liquid sodium. Appl

15
S. Khajepour and M. Ameri Sustainable Energy Technologies and Assessments 42 (2020) 100857

Therm Eng 2018;133:70–80. https://doi.org/10.1016/j. [23] Petrollese M, Cocco D. Optimal design of a hybrid CSP-PV plant for achieving the
applthermaleng.2018.01.038. full dispatchability of solar energy power plants. Sol Energy 2016;137:477–89.
[12] Lancereau Q, Rabut Q, Itskhokine D, Benmarraze M. Wind loads on Linear Fresnel https://doi.org/10.1016/j.solener.2016.08.027.
Reflectors ’ technology : a numerical study. Energy Procedia 2015;69:116–25. [24] Zurita A, Mata-torres C, Valenzuela C, Felbol C, Cardemil JM, Guzmán AM, et al.
https://doi.org/10.1016/j.egypro.2015.03.014. Techno-economic evaluation of a hybrid CSP + PV plant integrated with thermal
[13] Liu Q, Bai Z, Wang X, Lei J, Jin H. Investigation of thermodynamic performances energy storage and a large-scale battery energy storage system for base generation.
for two solar-biomass hybrid combined cycle power generation systems. Energy Sol Energy 2018;173:1262–77. https://doi.org/10.1016/j.solener.2018.08.061.
Convers Manag 2016;122:252–62. https://doi.org/10.1016/j. [25] Aguilar-jiménez JA, Velázquez N, Acuña A, Cota R, González E, González L, et al.
enconman.2016.05.080. Techno-economic analysis of a hybrid PV-CSP system with thermal energy storage
[14] Bai Z, Liu Q, Lei J, Wang X, Sun J, Jin H. Thermodynamic evaluation of a novel applied to isolated microgrids. Sol Energy 2018;174:55–65. https://doi.org/
solar-biomass hybrid power generation system. Energy Convers Manag 2017;142: 10.1016/j.solener.2018.08.078.
296–306. https://doi.org/10.1016/j.enconman.2017.03.028. [26] Prahl C, Röger M, Stanicki B, Hilgert C. Absorber tube displacement in parabolic
[15] Baniasad I, Ameri M. Techno economic feasibility analysis of Linear Fresnel solar fi trough collectors – A review and presentation of an airborne measurement
eld as thermal source of the MED / TVC desalination system. DES 2016:394. approach. Sol Energy 2017;157:692–706. https://doi.org/10.1016/j.
https://doi.org/10.1016/j.desal.2016.04.022. solener.2017.05.023.
[16] Martı JM, Valde M. Solar multiple optimization for a solar-only thermal power [27] Burkholder F, Kutscher C. Heat Loss Testing of Schott ’ s 2008 PTR70 Parabolic
plant , using oil as heat transfer fluid in the parabolic trough collectors 2009;83: Trough Receiver 2009. https://doi.org/https://www.nrel.gov/docs/fy09osti/
2165–76. https://doi.org/10.1016/j.solener.2009.08.010. 45633.pdf.
[17] Boukelia TE, Mecibah MS, Kumar BN, Reddy KS. Investigation of solar parabolic [28] Baniasad I, Ameri M. Techno economic feasibility analysis of Linear Fresnel solar fi
trough power plants with and without integrated TES (thermal energy storage) and eld as thermal source of the MED / TVC desalination system. DES 2016;394:1–17.
FBS (fuel backup system) using thermic oil and solar salt. Energy 2015. https://doi. https://doi.org/10.1016/j.desal.2016.04.022.
org/10.1016/j.energy.2015.05.038. [29] Askari IB, Ameri M. Techno-economic Feasibility Analysis of Stand-alone
[18] Martı JM. Analytical model for solar PV and CSP electricity costs : Present LCOE Renewable Energy Systems (PV / bat, Wind / bat and Hybrid PV / wind / bat) in
values and their future evolution 2013;20:119–32. https://doi.org/10.1016/j. Kerman. Iran 2011;7.. https://doi.org/10.1080/15567240903330384.
rser.2012.11.082. [30] Karimi M, Mijani N, Nadizadeh S. A Spatial and Mathematical Based Model for
[19] Green A, Diep C, Dunn R, Dent J. High capacity factor CSP-PV hybrid systems Solar Energy Potential Assessment and Optimal Lands for Solar Power Plant
2015;69:2049–59. https://doi.org/10.1016/j.egypro.2015.03.218. Construction in Iran 4 th International Conference on Solar Energy A Spatial and
[20] Liu H, Zhai R, Fu J, Wang Y, Yang Y. Optimization study of thermal-storage PV-CSP Mathematical Based Model for Solar Energy Potential Asses 2017.
integrated system based on GA-PSO algorithm. Sol Energy 2019;184:391–409. [31] kerman map n.d. http://www.kerman-info.ir/kerman-map.
https://doi.org/10.1016/j.solener.2019.04.017. [32] The World Bank, Solar resource data: Solargis n.d. https://solargis.com/maps-an
[21] Roni M, Hoque IU, Ahmed T. Comparative Study of Levelized Cost of Electricity d-gis-data/download/iran.
(LCOE) for Concentrating Solar Power (CSP) and Photovoltaic (PV) Plant in the [33] Khajepour S, Ameri M. Techno-economic analysis of using three Fresnel solar fields
Southeastern Region of Bangladesh. Int Conf Electr Comput Commun Eng 2019; coupled to a thermal power plant for different cost of natural gas. Renew Energy
2019:1–6. 2020;146:2243–54. https://doi.org/10.1016/j.renene.2019.08.075.
[22] Bode S, Cuellar A, Perez I. Retrofitting operating CSP plants with PV to power [34] De OJ, De OR, Eboli L, Mazzulla G. Index numbers for monitoring transit service
auxiliary loads – Technical consideration and case study Retrofitting Operating quality. Transp Res PART A 2015. https://doi.org/10.1016/j.tra.2015.05.018.
CSP Plants with PV to Power Auxiliary Loads – Technical Consideration and Case [35] The System Advisor Model (SAM) n.d.
Study 2019;090003.

16