Applied Thermal Engineering 31 (2011) 3904e3913

Contents lists available at ScienceDirect

Applied Thermal Engineering

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

Energy and exergy analysis of solar power tower plants

Chao Xu*, Zhifeng Wang, Xin Li, Feihu Sun
Key Laboratory of Solar Thermal Energy and Photovoltaic System, Institute of Electrical Engineering, Chinese Academy of Sciences, Beijing 100190, China

a r t i c l e i n f o a b s t r a c t

Article history: Establishing the renewable electricity contribution from solar thermal power systems based on energy
Received 25 May 2011 analysis alone cannot legitimately be complete unless the exergy concept becomes a part of that analysis.
Accepted 24 July 2011 This paper presents a theoretical framework for the energy analysis and exergy analysis of the solar
Available online 29 July 2011
power tower system using molten salt as the heat transfer fluid. Both the energy losses and exergy losses
in each component and in the overall system are evaluated to identify the causes and locations of the
Keywords:
thermodynamic imperfection. Several design parameters including the direct normal irradiation (DNI),
Exergy analysis
the concentration ratio, and the type of power cycle are also tested to evaluate their effects on the energy
Solar power tower plant
Central receiver
and exergy performance. The results show that the maximum exergy loss occurs in the receiver system,
Supercritical power cycle followed by the heliostat field system, although main energy loss occurs in the power cycle system. The
energy and exergy efficiencies of the receiver and the overall system can be increased by increasing the
DNI and the concentration ratio, but that increment in the efficiencies varies with the values of DNI and
the concentration ratio. It is also found that the overall energy and exergy efficiencies of the solar tower
system can be increased to some extent by integrating advanced power cycles including reheat Rankine
cycles and supercritical Rankine cycles.
Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction that is based on the first law analysis, the exergy analysis can give
a clearer assessment of various losses occurring in energy system
In the current world, electricity is still dominantly generated by both quantitatively and qualitatively. Exergy analysis can evaluate
burning conventional sources such as coal, oil and natural gas, quantitatively the causes and locations of the thermodynamic
which not only have a limited life but also release gaseous or liquid imperfection in the energy system, and thus indicate the possibil-
pollutants during operation. Because solar energy is an inex- ities of thermodynamic improvement. As a result, exergy analysis
haustible, clean and safe source of energy, it has received much has been widely used in the design, simulation and performance
attention as one of the most promising candidate to substitute for evaluation of energy systems. The conclusions from the exergy
the conventional fuels for electricity supply [1e3]. Recently, rapid analysis play a significant role on improving the existing processes,
development occurred worldwide in the basic technology and components or systems, or developing new processes or systems.
market strategy for the concentrating solar power (CSP) technolo- Energy analysis (or first law analysis) method has been
gies, including parabolic trough, power tower, and dish/engine. employed to conduct the energetic analysis and performance
However, the power generation efficiencies of the CSP systems are evaluation of the solar thermal power system or some components
found to be low, which indirectly increases the capital costs of [1,2,12e19]. Yao et al. [1] carried out modeling and simulation of the
electricity generation, and great efforts have to be concentrated on pioneer 1 MW solar thermal central receiver system (CRS) in China
the future research and development of CSP systems. (i.e., DAHAN). Based on the energy balance, they developed the
In the power generation system, exergy analysis (or second law mathematical models of the main basic components in the CRS and
analysis) has proven to be a powerful tool in thermodynamic integrated them to be a whole plant model. Prakash et al. [16]
analyses of the system [4e11]. Exergy is defined as the maximum carried out an experimental and numerical study of the steady
useful work that can be done by a system interacting with a refer- state convective heat losses occurring from a downward facing
ence environment. Different from the conventional energy analysis cylindrical cavity receiver. Various fluid inlet temperatures and
receiver inclinations were tested, and it was found that the
convective loss increases with mean receiver temperature and
* Corresponding author. Tel.: þ86 10 82547036; fax: þ86 10 62587946. decreases with the increase in receiver inclination. Montes et al.
E-mail address: mechxu@gmail.com (C. Xu). [17,18] investigated the influence of the solar multiple on the

1359-4311/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.applthermaleng.2011.07.038
 C. Xu et al. / Applied Thermal Engineering 31 (2011) 3904e3913 3905

annual performance of parabolic trough solar thermal power plants subsystem, the solar tower system can be considered as consisting
with direct steam generation. Both thermal analysis and economic of four subsystems: heliostat field subsystem, central receiver
analysis were performed in their work. Larbi et al. [19] presented subsystem, steam generation subsystem (SGSS), and the conven-
the performance analysis of a solar chimney power plant based on tional power cycle subsystem. The sun’s rays fall on a field of
the energy balance, and they also investigated the effects of various mirrors known as heliostats, and are reflected into the aperture
factors on the thermal performance including the solar irradiance, area of the central receiver, located at the top of a high tower, with
the ambient temperature, the height of the tower, and the surface the help of tracking system for each heliostat. The concentrated
of the collector. Recently, Li et al. [2] developed a thermal model of rays onto the central receiver result in high temperature of the
molten salt cavity receiver based on the energy balance. Various receiver which is thus used to heat the molten salt flowing
factors, such as receiver area, heat loss, tube diameter, were through the pipes embedded inside the receiver. Then the heated
analyzed to investigate the thermal performance of the cavity molten salt passes through the SGSS heat exchangers and transfers
receiver. the thermal energy to the water which flows through the heat
However, very limited papers have appeared on the exergy exchangers typically in a counter flow mode. After passing
analysis and performance assessment of the solar thermal power through the SGSS, water is heated from subcooled liquid to
plant [4,5,10,20e22], which can offer insights not available from the superheated steam, which is fed to the steam turbine for elec-
energy analysis alone. Bejan et al. [4] analyzed the potential for tricity generation.
exergy conservation in solar collector systems based on the second The energy and exergy analysis of the whole system can be
law of thermodynamics. It was shown that the exergy efficiency of carried out by combining the analysis for each subsystem, which is
the solar collector system is affected by heat transfer irreversibil- in general based on the principle of energy and exergy balance for
ities occurring between the sun and the collector, between the a control volume. The following assumptions are made in the
collector and the ambient air, and inside the collector. Kaushik et al. analysis:
[5] presented an exergy analysis for a solar trough power plant.
Basic energy and exergy analysis for the system components (a) The system runs at steady state with a constant solar isolation.
including parabolic trough collector/receiver and Rankine heat (b) Pressure drop and heat loss in pipe lines are all neglected.
engine were carried out to evaluate the energy and exergy losses as (c) Kinetic and potential energy and exergy are ignored.
well as exergy efficiency for typical solar thermal power system. It (d) Chemical exergy of materials are neglected.
was found that the main exergy loss takes place at the collector- (e) Parasitic efficiency of the whole system is assumed to be 88%,
receiver assembly. You et al. [21] investigated the optimal which is a typical value for systems of this kind.
thermal and exergy efficiencies for the combined system of the (f) The efficiency of the power cycle is independent of the power
reheat-regenerative Rankine power cycle and the parabolic trough output of the turbine.
collector. It was found that, to make most advantage of the
collector, the exiting fluid is supposed to be at the maximum For the steady-state process, the energy balance and exergy
temperature the collector can harvest. Lu et al. [10] established balance for a control volume can be expressed as Eqs. (1) and (2),
a basic physical model for solar absorber pipe with solar selective respectively [23].
coating, based on which the associated heat transfer and exergy
performances were analyzed and optimized. Most recently, Gupta
et al. [22] carried out the energy and exergy analysis for the
different components of a proposed conceptual direct steam
generation solar trough power plant. It was found that the
maximum energy loss takes place in the condenser followed by the
solar collector field (including the trough concentrators and
absorbers), while the maximum exergy loss occurs in the solar
collector field.
The literature review indicates that the energy and exergy
analysis of solar power tower plant, which uses heliostat field and
central receiver as the concentrator-receiver system, has not been
reported till now. In this regard, the objective of this article is to
construct a theoretical framework for the energy and exergy
analysis of the solar power tower plant using molten salt as the heat
transfer fluid, which can be used to evaluate the energy and exergy
losses in each component and in the overall power plant. The
energy and exergy efficiencies have also been computed and
compared for the individual components as well as for the overall
plant. Several influencing factors including the direct normal irra-
diation (DNI), the concentration ratio, and the power cycle were
also tested to evaluate their effects on the energy and exergy
performance.

2. Energy and exergy analysis

Fig. 1 depicts the schematic of a solar power tower plant using

molten salt as the heat transfer fluid. Since this paper only
considers the energy and exergy performance of the system at
steady state, the thermal storage subsystem will not be discussed Fig. 1. Schematic of a solar tower power plant (a), and the temperatureeentropy (Tes)
and not shown in Fig. 1. Without considering the thermal storage diagram of the corresponding power cycle (power cycle 1).
3906 C. Xu et al. / Applied Thermal Engineering 31 (2011) 3904e3913

X X X
_ c:v: ¼
W Q_ j þ _ in hin 
m _ ex hex
m (1) 2.2. Central receiver subsystem
j in ex
The central receiver, which is typically installed at the top of the
X  X X solar tower, has been developed with various shapes, including
_ c:v:
W ¼ 1  T0 =Tj Q_ j þ _ in jin 
m _ ex jex  T0 S_ gen
m (2) cavity receivers and cylindrical receivers. This analysis is only based
j in ex
on the cavity receiver. In operation, the receiver absorbs the
*
where the exergy destruction ðT0 S_ gen Þ, which is due to the entropy isolation Q_ rec and transports part of the energy to the heat transfer
generation, is identical to the so-called irreversibility ðI RÞ._ The fluid (e.g., molten salt) flowing through it. The rest of energy is lost
physical exergy of each state point can be considered as to the environment by convective, emissive, reflective and
conductive heat losses [2]. The energy balance and exergy balance
_ ¼ m
J _ j ¼ mððh
_  h0 Þ  T0 ðs  s0 ÞÞ (3) for the central receiver are:
Besides the above balance expressions, the energy efficiency and *  
exergy efficiency, which refer to comparison of the desired output Q_ rec ¼ Q_ rec;abs þ Q_ rec;totloss ¼ m
_ ms hms;b  hms;a þ Q_ rec;totloss
of a process with the input in terms of energy or exergy, are also (10)
used in the analysis. The analysis for each subsystem of the solar
tower power plant is given as follows.
_* ¼ J
_ _ _
Jrec rec;abs þ Jrec;loss þ I Rrec (11)
2.1. Heliostat field subsystem
where the exergy loss associated with the heat loss is expressed as

The heliostat field, which consists of dozens or hundreds of _  

_
J rec;loss ¼ Q rec;totloss 1  T0 =Trec;sur (12)
heliostats and has a total aperture area of Ah, reflects and
concentrates the sun’s rays to the central receiver. The total and the useful exergy absorbed by the flowing molten salt is
isolation is proportional to the total area and can be given
* _
byQ_ ¼ Ah q_ *. Here q_ * means the amount of solar radiation J _ ms ððhb  ha Þ  T0 ðsb  sa ÞÞ
rec;abs ¼ m
received per unit area, which in this paper is treated as the direct _ ms cpms ðTb  Ta  T0 lnðTb =Ta ÞÞ
¼ m (13)
normal irradiation (DNI). It should be noted that the DNI varies
with several factors such as the geographical position on the earth, The energy efficiency and exergy efficiency of the central receiver
the meteorological condition, and the time of day. In the present subsystem are defined as
analysis, it is assumed that q_ * is a constant and the system oper-
*
ates at steady state. hI;rec ¼ Q_ rec;abs =Q_ rec (14)
*
The incident solar radiation ðQ_ Þ is partly delivered to the
*
central receiver as solar isolation Q_ rec by the heliostat field, while _
hII;rec ¼ J _*
rec;abs =Jrec (15)
*
the remaining fraction ðQ_ 0 Þ
is lost to the environment due to
From the above analysis, in order to calculate the efficiency of
various loss mechanisms. These mechanisms include losses due to
the receiver, it is necessary to know the total heat loss from the
cosine efficiency, blocking and shading, mirror reflectivity, tracking
receiver as well as the average surface temperature of the absorber.
error, clean error, etc. [1]. In this analysis, it is assumed that the
These can be determined by modifying a validated thermal model
radiation energy loss is only accounted for by the overall field
for designing molten salt cavity receivers developed by Li et al. [2].
efficiency (hh), which is defined as the ratio of the power incident
Details about the modified model can be found in Appendix.
on the receiver absorber surface to the power incident on the
heliostats tracking the receiver. Thus, the energy balance and
exergy balance for the heliostat field are given by 2.3. Steam generator subsystem (SGSS)
* * *
Q_ ¼ Q_ rec þ Q_ 0 (4) The SGSS usually consists of a series of heat exchangers, through
* which the high-temperature molten salt from the receiver heats
_ ¼
J _*
J þ _*
J (5)
rec 0 water from the subcooled liquid to superheated steam. Assuming
the heat exchangers are well insulated and the heat loss to the
where _*
J
is the exergy delivered to the receiver and is the _*
J
rec 0 environment is negligible, the energy balance and exergy balance
exergy loss (irreversibility). for the SGSS are
*
The exergy J _ associated with the solar irradiation on the
*
heliostat mirror surface ðQ_ Þ can be expressed as
   
 
_ ms hms;b  hms;a ¼ Q_ st;abs ¼ m
Q_ rec;abs ¼ m _ st hst;5  hst;4 (16)
* *
_ ¼ Q_
J 1  T0 =T * (6)
_ _ _
J rec;abs ¼ Jst;abs þ I Rsgs (17)
where T* is the apparent sun temperature as an exergy source and
taken to be 4500 K [4]. Similarly, the exergy delivered to the where the useful exergy absorbed by the heated water is
receiver is written as
  _ _ st ððh5  h4 Þ  T0 ðs5  s4 ÞÞ
_ * ¼ Q_ * 1  T =T * J st;abs ¼ m (18)
J rec rec 0 (7)
The energy efficiency and exergy efficiency of the SGSS are:
Then, the energy efficiency and exergy efficiency of the heliostat
field subsystem are given by Eqs. (8) and (9), respectively.
* * hI;sgs ¼ Q_ st;abs =Q_ rec;abs ¼ 100% (19)
hI;h ¼ Q_ rec =Q_ ¼ hh (8)

_ * =J_* _
hII;sgs ¼ J _
hII;h ¼ J rec (9) st;abs =Jrec;abs (20)
 C. Xu et al. / Applied Thermal Engineering 31 (2011) 3904e3913 3907

2.4. Power cycle subsystem 3.2. Base case study

The power cycle used in the solar tower power plant is generally This section will give a general discussion about the energy and
a conventional Rankine cycle, which is depicted in Fig. 1. The exergy analysis of the solar tower power plant based on System 1.
Rankine cycle mainly consists of high and low-pressure turbine The properties of the base case system and properties of System 1
stages, feed water heaters, condensers and pumps. A regenerative are shown in Tables 1 and 2, respectively. The properties of molten
Rankine cycle, which uses feed water heaters, is employed here to salt (i.e., a mixture of 60 wt% NaNO3 and 40 wt% KNO3) are [25]:
avoid too low water temperature at the inlet of the SGSS (Point 4),  
which is necessary to prevent the solidification of molten salt in the Density; r kg=m3 ¼ 2090  0:636  Tð CÞ (28)
SGSS. Only one stage of extraction is used in the present analysis for
simplicity. Different power cycles will be discussed in the following
sections and the power cycle shown in Fig. 1 is termed as System 1 Specific heat cp ðJ=ðkg KÞÞ ¼ 1443 þ 0:172  Tð CÞ (29)
hereafter. The corresponding temperatureeentropy (Tes) diagram
of System 1 is shown in Fig. 1b. Thermal conductivity; lðW=ðm KÞÞ
For the power cycle subsystem, the energy balance and exergy 4
¼ 0:443 þ 1:9  10  Tð CÞ (30)
balance are
The results of energy and exergy analysis of the base case system
_ net þ Q_
Q_ st;abs ¼ W ps;totloss (21) are listed in Table 3. The percentages of energy and exergy losses at
each subsystem are shown in Fig. 2, which are the energy (exergy)
_ _ _ losses at each subsystem divided by the total energy (exergy) losses
J st;abs ¼ W net þ Jps;totloss (22)
of the whole system. From the energy analysis, it is found that the
where the net output work (electricity) from the power cycle isolation energy of 8.0 MW can generate net output electricity of
subsystem is 1.83 MW from the base case solar tower power plant. The total
energy efficiency of the whole system is 22.9%, while the subsystem
_ tW
_ net ¼ W _ _
W p1  W p2 (23) energy efficiencies are 75%, 90%, 100% and 37.85% for the heliostat
field, central receiver, SGSS and the power cycle, respectively. The
The energy efficiency and exergy efficiency of the power cycle power cycle subsystem has not only the lowest energy efficiency,
are but also the main energy loss. The percentage energy loss in the
power cycle is as large as 53.8%, followed by 32.4% in the heliostat
_ net =Q_
hI;ps ¼ W (24)
st;abs field and 9.7% in the central receiver.
However, the results of the exergy analysis show a distinct
hII;ps ¼ W _
_ net =J (25) behavior. The total exergy efficiency of the whole system is 24.5%,
st;abs
while the subsystem exergy efficiencies are 75%, 55.5%, 89.8% and
Finally, the overall energy efficiency and exergy efficiency of the 74.5% for the heliostat field, central receiver, SGSS and the power
whole system can be defined as the ratio of net electricity output cycle, respectively. The power cycle subsystem has a relatively large
ðW_ net h
para Þ from the whole system to the energy or exergy input exergy efficiency (74.5%), and the corresponding percentage exergy
associated with the solar irradiation on the heliostat surface, i.e., loss is as small as 12.6%, although main energy loss occurs there. On
the contrary, although the central receiver has a large energy effi-
_ net h
hI;overall ¼ W _*
para =Q (26) ciency and a small energy loss percentage (9.7%), it has the largest
percentage exergy loss (44.2%), followed by 33.1% in the heliostat
_ net h
hII;overall ¼ W _* field subsystem. This is because the fact that solar isolation is the
para =J (27)
energy of very high quality (T* ¼ 4500 K from Eq. (6)), and great
irreversibilities occur when the high-quality isolation is absorbed
to be thermal energy with the temperature of about 700e900 K. As
3. Results and discussion a result, the energy loss in the central receiver has high quality, i.e.,
containing a lot of exergy. On the contrary, although the power
3.1. Model validation

From the preceding section, it is known that the present analysis Table 1
is based on the energy balance and exergy balance of each Properties of the base case solar tower power plant (System 1).
subsystem. The analysis for the SGSS heat exchangers and the Subsystem Properties Values Unit
Rankine power cycle depends only on the thermodynamic prop- Heliostat field Beam radiation (DNI) 800 W m2
erties of molten salt and steam at each state as shown in Fig. 1, Overall field efficiency 75% e
which is well developed and self-evident. The analysis for the Total heliostat aperture area 10,000 m2
central receiver is based on a thermal model, which is modified Central receiver Aperture area 12.5 m2

Inlet temperature of molten salt 290 C
from a validated model developed by Li et al. [2]. To validate the 
Outlet temperature of molten salt 565 C
modification, the present model was used to calculate the thermal View factor 0.8 e
performance of the Sandia National Laboratories’ molten salt Tube diameter 0.019 m
electric experiment (MSEE) based on the parameters provided in Tube thickness 0.00165 m
Ref. [2]. The calculated energy efficiency of the receiver is 87.77%, Emissivity 0.8 e
Reflectivity 0.04 e
which agrees well with the predicted value of 87.73% from Li’s Wind velocity 5.0 m s1
work, and the experimental average efficiency of 87.5% from Ber- Passes 20 e

gan’s experiments [24]. Therefore, the calculated results of the SGSS Inlet temperature of water 239 C

paper are reasonable, which are useful for guiding the design and Outlet temperature of steam 552 C

e Ambient temperature 20 C
operation of solar tower power plants.
3908 C. Xu et al. / Applied Thermal Engineering 31 (2011) 3904e3913

Table 2
Properties of state points in different power cycles.

State point System 1 System 2 System 3

Temperature ( C) Pressure (kPa) Temperature ( C) Pressure (kPa) Temperature ( C) Pressure (kPa)

1 45.8 10 45.8 10 45.8 10
2s 45.9 3150 45.9 3150 45.9 3000
2 46.0 3150 46.0 3150 46.0 3000
3 236.6 3150 236.6 3150 233.8 3000
4s 238.7 12,600 238.7 12,600 238.4 24,000
4 239.0 12,600 239 12,600 239 24,000
5 552.0 12,600 552 12,600 552 24,000
6s 327.4 3150 327.4 3150 470.5 3000
6 353.5 3150 353.5 3150 481.6 3000
7s 45.8 10 45.8 10 45.8 10
7 45.8 10 45.8 10 45.8 10
8 e e 327.1 2520 317.3 4800
9 e e 552 2520 552 4800

cycle subsystem has the main energy loss, it has little exergy loss temperature is increased from 565 to 645  C. It is seen that, with
since the lost energy is of very low quality. the increase in the outlet temperature, the exergy efficiency can be
Hence, in the CSP system, great efforts should be concentrated slightly increased from 55.5 to 56.5%, while the energy efficiency is
to reduce the exergy loss in the central receiver subsystem, as well decreased from 90 to 88.5%. The decrease in the energy efficiency is
as the heliostat field subsystem. The factors affecting the exergy due to the larger heat loss associated with higher receiver
efficiency of the central receiver can be analyzed from the following temperature. Therefore, when trying to increase the outlet
expression for the exergy efficiency, which is developed based on temperature for improving the exergy efficiency, efforts must also
Eqs. (7), (10), (13), and (15): be concentrated on lowering the energy loss, which becomes much
trickier at higher working temperature.
1  T0 lnðTb =Ta Þ=ðTb  Ta Þ
hII;rec ¼ hI;rec (31)
1  T0 =T *
3.3. Effect of incident solar isolation
The above equation indicates that the exergy efficiency of the
receiver depends on the energy efficiency of the receiver (hI,rec), and The performance of the CSP plant depends highly on the inci-
the inlet (Ta) and outlet (Tb) temperatures of the molten salt. For the dent solar isolation, which varies with the geographical position,
base case system, hI,rec ¼ 90%, and from Table 1, the inlet and outlet the time of day, etc. To investigate the effect of incident solar
molten salt temperatures are 290 and 565  C, respectively. Hence, isolation on the energy and exergy performance of the CSP system,
the exergy efficiency is hII,rec ¼ 61.5%  hI,rec ¼ 55.4%. That means different DNIs ranging from 100 to 1000 W m2 were tested in this
that even there is no energy loss from the receiver (hI,rec ¼ 100%), work. Fig. 4 shows the variation in the energy and exergy effi-
the exergy efficiency cannot exceed 61.5% for the present temper- ciencies of both the receiver and the whole system as a function of
ature design of the molten salt. Since there is no great potential for the DNI. It is seen that both the energy and exergy efficiencies
the further improvement of the energy efficiency, the increase in increase with the increase in the DNI. And the trend for the effi-
the exergy efficiency depends mainly on the temperature design of ciency of the whole system is similar to that of the receiver, indi-
the working fluid. Specifically, the outlet temperature of the cating that the effect of the DNI on the efficiencies of the whole
working fluid should be greatly increased for further increasing the system mainly depends on its effect on the efficiencies of the
exergy efficiency of the receiver. For instance, when the outlet central receiver.
temperature is increased to 1000  C, the exergy efficiency is The increase in the energy and exergy efficiencies of the receiver
hII,rec ¼ 71%  hI,rec, meaning the ideal exergy efficiency is increased with the DNI can be explained by analyzing the heat loss mecha-
from 61.5 to 71% as a result of the increase in the outlet temperature nism from the receiver. From Eqs. (A-2) and (A-4) in Appendix, two
from 565 to 1000  C. of the main heat loss mechanisms, i.e., emissive heat loss and
However, the increase in the outlet temperature of working fluid convective heat loss, depend mainly on the surface temperature of
may be limited by material constraints of pipes and engine, which the receiver, instead of the DNI. The change of surface temperature
needs extensive future research work in this direction. Using the of the receiver with the DNI can be seen in Fig. 5, which shows
present working fluid, i.e., molten salt, the outlet temperature may clearly that when the DNI increases by ten times, from 100 to
be increased to some extent. Fig. 3 shows the variation in the 1000 W m2, the surface temperature only varies slightly from 510
energy and exergy efficiencies of the receiver when the outlet to 546  C. As a result, the heat loss from the receiver only increases

Table 3
Energy and exergy analysis of the base case solar tower system.

Subsystem Energy analysis Exergy analysis

Received (kW) Delivered (kW) Loss (kW) Energy efficiency (%) Received (kW) Delivered (kW) Loss (kW) Exergy efficiency
(%)
Heliostat field 8000.0 6000.0 2000.0 75.00 7478.8 5609.1 1869.7 75.00
Central receiver 6000.0 5401.3 598.7 90.02 5609.1 3111.7 2497.4 55.48
SGSS 5401.3 5401.3 0.0 100.00 3111.7 2793.5 318.2 89.77
Power cycle 5401.3 2080.6 3320.7 37.85 2793.5 2080.6 712.9 74.48
Overall 8000 1830.9 6169.1 22.89 7478.8 1830.9 5647.9 24.48
 C. Xu et al. / Applied Thermal Engineering 31 (2011) 3904e3913 3909

100

80 Overall energy efficiency

Overall exergy efficiency
Receiver energy efficiency

Efficiency, %
60 Receiver exergy efficiency

40

20
Fig. 2. Comparison of energy and exergy loss in the subsystems.
0 200 400 600 800 1000
DNI, W m-2
slightly when the input energy increases proportionally, which
results in the increase in the energy and exergy efficiencies with the Fig. 4. Effect of the direct normal irradiation on the energy and exergy efficiencies of
DNI. It should be pointed out that from Fig. 5 the surface temper- the receiver and the whole plant.
ature decreases first at very low solar isolation, e.g.,
DNI < 200 W m2. This is due to the assumption that the molten
effect of concentration ratio on the energy and exergy performance,
salt is always heated from 290 to 565  C through the receiver. When
a large range of concentration ratios ranging from 200 to 1400 were
the DNI is very low, to maintain a fixed hot molten salt temperature
tested. Fixing the total area of the heliostat field, with the increase
at the receiver outlet, the molten salt flow rate should be very low,
in the concentration ratio, the open aperture area of the receiver
bringing less heat away and leaving more heat which heats up the
decreases first sharply and then the decrease becomes slower, as
receiver. Actually, this assumption becomes invalid when the DNI is
shown in Fig. 6. Also, the concentration ratio affects the solar flux
decreased to a critical value, below which the molten salt cannot be
concentrated into the receiver, and thus the surface temperature of
heated to 565  C even with a very low flow rate.
the receiver, which can be seen in Fig. 6. With the increase in the
It is also seen that the effect of the DNI on the efficiencies varies
concentration ratio from 200 to 1400, the average surface
with the values of DNI. For instance, the energy efficiency of the
temperature of the receiver increases nearly linearly from 456 to
receiver increases greatly from about 45 to 85% when the DNI
618  C, which can be easily understood by referring to Eq. (A-6).
increases from 100 to 400 W m2, while it only increases slightly
The variation in the energy and exergy efficiencies of both the
from 85 to 91% when the DNI increases further from 400 to
receiver and the whole system with the concentration ratio is
1000 W m2. The variation in the net output electricity with the
shown in Fig. 7. Both the energy and exergy efficiencies increase
DNI is also shown in Fig. 5. It is seen that the net output electricity
with the increase in the concentration ratio. And similar to the
increases nearly linearly from 100 to 2300 kW with the DNI
effect of the DNI as discussed in the preceding section, the effect of
increase from 100 to 1000 W m2.
the concentration ratio on the efficiencies of the whole system
mainly depends on its effect on the efficiencies of the central
3.4. Effect of concentration ratio receiver. With the increase in the concentration ratio, although the
increased surface temperature of the receiver is disadvantageous
The concentration ratio of the receiver is a critical parameter for lowering the heat loss from the receiver, the simultaneous
when designing a high-efficient central receiver. To investigate the decrease in the aperture area plays a crucial role on lowering the

90.5 57.0 2500 560 Receiver surface temperature, C

Net output electricity

o

90.0 Receiver surface temperature 550

Net output electricity, kW

2000
56.5
Energy efficiency, %

Exergy efficiency, %

540
89.5 1500
56.0 530
89.0 1000
520
55.5
88.5 500
510

88.0 55.0 0 500

560 580 600 620 640 0 200 400 600 800 1000
o -2
Outlet temperature of molten salt, C DNI, W m

Fig. 3. Effect of the outlet temperature of the receiver on the energy and exergy Fig. 5. Effect of the direct normal irradiation on the surface temperature of the
efficiencies of the receiver. receiver and net output electricity of the whole plant.
3910 C. Xu et al. / Applied Thermal Engineering 31 (2011) 3904e3913

50 625
Sun 5
Receiver aperture area Turbine

Receiver surface temperature, C

600 Receiver

o
Receiver surface temperature b
2
Receiver aperture area, m

40
SGSS
575 a 8
6 xm 5 7
30 550 9 (1 – x)m 5
Condenser
Feedwater
20 525 heater
Pump Pump 1
500 Heliostat field
4
3
2

10
475 Fig. 8. Schematic of a solar tower power plant with a reheat process in the advanced
power cycle (System 2 or 3).
0 450
0 200 400 600 800 1000 1200 1400
Concentration ratio constraints at high temperatures. Based on the state-of-the-art
materials and technology, it is still effective to improve the
Fig. 6. Effect of the concentration on the aperture area and surface temperature of the energy and exergy efficiencies of the CSP system by improving the
receiver.
efficiency of the power cycle. In this section, we test two advanced
power cycles: System 2 which has a reheat process, and System 3
which has a reheat process and has supercritical working steam.
emissive, convective, and conductive heat loss as indicated by Eqs.
The schematic of the solar tower power plant with System 2 and
(A-2), (A-4), and (A-5) in Appendix, which finally leads to a lowered
System 3, and the corresponding Tes diagrams are shown in Figs. 8
total heat loss, and an increase in the energy efficiency and exergy
and 9, respectively. It should be noted that compared to System 1,
efficiency.
System 2 has a reheat process: the steam is withdrawn from the
It is also seen that the effect of the concentration ratio on the
exit of the high-pressure turbine and is reheated through the SGSS
efficiencies varies with the values of concentration ratio. For
heat exchangers, after which the heated steam is fed to the low-
instance, the exergy efficiency of the whole system increases
evidently from 24.0% to 27.4% when the concentration ratio
increases from 200 to 600, while it only increases slightly from
27.4% to 28.3% when the concentration ratio increases further from
600 to 1400. Therefore, the potential for further increasing the
energy and exergy efficiencies becomes much less when the
concentration ratio is beyond a critical value, e.g., 600 in this study.

3.5. Effect of advanced power cycle

As is known from the proceeding section, the receiver system

accounts for most of the exergy loss, while the percentage exergy
loss of the power cycle subsystem is much smaller, e.g., 12.6% for
the base power cycle. However, great improvement in the exergy
efficiency of the receiver system is impeded by the material

100

Overall energy efficiency

80 Overall exergy efficiency
Receiver energy efficiency
Efficiency, %

Receiver exergy efficiency

60

40

20

200 400 600 800 1000 1200 1400

Concentration ratio
Fig. 7. Effect of the concentration ratio on the energy and exergy efficiencies of the Fig. 9. Schematic of the temperatureeentropy (Tes) diagrams of the advanced power
receiver and the whole plant. cycles: System 2 (a) and System 3 (b).
 C. Xu et al. / Applied Thermal Engineering 31 (2011) 3904e3913 3911

a On the other hand, it is also clear from Fig. 10b that the exergy
efficiency of the SGSS can also be increased by using the advanced
power cycle. For instance, it can be increased from 89.9% to 95.0%
when System 1 is replaced by System 3. This is because that the
average temperature difference for the heat transfer process is
decreased in the advanced power cycle, which is obviously bene-
ficial for improving the overall exergy efficiency of the solar tower
plant.

4. Conclusions

In this paper, a theoretical framework for the energy and exergy

analysis of the solar tower power plant using molten salt as the heat
transfer fluid is developed. The energy and exergy losses in each
component and in the overall power plant are evaluated to identify
the causes and locations of the energy losses in the solar tower
power plant. Salient findings and conclusions, which are valuable in
b the improvement of the solar tower power plant, are summarized
as follows:

1) Due to the very low quality of energy losses from the power
cycle subsystem, the power cycle subsystem has a relatively
high exergy efficiency (74.5%), and the corresponding
percentage exergy loss is as small as 12.6%, although main
energy loss occurs there. On the other hand, although the
central receiver has a high energy efficiency (90%) and a small
energy loss percentage (9.7%), it has the largest percentage
exergy loss (44.2%), followed by the heliostat field subsystem
(33.1%). Therefore, in the solar power tower system, great
efforts should be focused on reducing the exergy loss in the
central receiver subsystem, which mainly depends on the
Fig. 10. Comparison of energy efficiency (a) and exergy efficiency (b) for the subsys-
operating temperature of the working fluid.
tems and the overall system using different power cycles.
2) The effect of DNI on the energy and exergy efficiencies of the
receiver, as well as the whole system, varies with the values of
DNI. The energy efficiency increases greatly from about 45% to
pressure turbine. Compared to System 2, System 3 has a higher inlet 85% when the DNI increases from 100 to 400 W m2, while it
steam pressure (i.e., 24 MPa) which makes the inlet steam to be only increases slightly from 85% to 91% when the DNI increases
supercritical steam. The reheat pressures in System 2 and System 3 further from 400 to 1000 W m2.
are all chosen to be 20% of the inlet pressure of the high-pressure 3) Fixing the total area of the heliostat field, with the increase in
turbine. the concentration ratio, the decrease in the aperture area plays
Fig. 10 shows the energy and exergy efficiencies of individual a crucial role on lowering the heat losses, which finally leads to
components and the whole system using different power cycles an increase in the energy efficiency and exergy efficiency.
(i.e., System 1, 2, and 3). For the energy efficiency, it is seen from Similarly, the effect of the concentration ratio on the efficien-
Fig. 10a that the efficiency of power cycle increases from 37.9% to cies varies with the values of concentration ratio.
39.7% after a reheat process is introduced (i.e., from System 1 to 4) The overall energy and exergy efficiencies of the solar tower
System 2). When the inlet steam becomes supercritical steam (in plant can be increased to some extent by introducing advanced
System 3), the energy efficiency of power cycle can be further power cycles. It is found that the overall exergy efficiency
improved to 42.1%. Accordingly, the overall efficiency of the solar increases from 24.5% to 25.6% after a reheat process is added,
tower plant increases evidently from 22.9% to 23.9% and then to and further to 27.4% after the inlet steam becomes supercritical.
25.7% after System 1 is replaced by System 2 and System 3. While in
terms of the exergy efficiency, it increases from 74.5% to 76.7% and Acknowledgements
to 78.9% for the power cycle when System 1 is updated to System 2
and System 3, respectively. Accordingly, the overall exergy effi- This work has been financed by China’s Ministry of Science and
ciency of the solar tower plant increases from 24.5% to 25.6% after Technology under the National High Technology Development
a reheat process is added, and further to 27.4% after the inlet steam “863” Plan with Project No. (2006AA050101) and the National Basic
becomes supercritical. Also, it should be pointed out that the tested Research Program of China (973 Program) with Project No.
power cycle with a reheat process (System 2) and the supercritical (2010CB227104), and Chinese Academy of Sciences.
power cycle (System 3) are all not well optimized which are used
here to illustrate the principle of the investigation of the advanced Nomenclature
power cycles, and there is still some potential to further improve
the efficiencies by optimizing the operating parameters. In general,
the overall energy and exergy efficiencies can be increased to some A area, m2
extent by introducing the reheat process in the power cycle, and C concentration ratio
especially, using supercritical power cycle in the solar thermal cp apecific heat, J (kg K)1
tower plant. d diameter, m
3912 C. Xu et al. / Applied Thermal Engineering 31 (2011) 3904e3913

Fr view factor The total heat losses from the central receiver can be contrib-
h enthalpy, J kg1; heat transfer coefficient, W (m2 K)1 uted by different mechanisms, including convective, emissive,
I R_ irreversibility, W reflective, and conductive heat losses, which gives:
m _ mass flow rate, kg s1
Q_ heat transferred, W Q_ rec;totloss ¼ Q_ rec;conv þ Q_ rec;em þ Q_ rec;ref þ Q_ rec;con (A-1)
q_ heat flux, W m2
The expressions for the different heat losses are listed as follows:
s entropy, J kg1
T temperature, K  
W work, W Q_ rec;em ¼ 3 avg
4
s Trec;sur  T04 Afield =C (A-2)

*
Greek Q_ rec;ref ¼ Q_ rec Fr r (A-3)
d thickness, m
l thermal conductivity, W (m K)1    
3 receiver surface emissivity Q_ rec;conv ¼ hair;fc;insi Trec;sur  T0 þ hair;nc;insi Trec;sur
s StefaneBoltzmann constant 5.67  108 W (m2 K4)1  
 T0 Fr Afield =C (A-4)
h efficiency
r density, kg m3
_  
J exergy, W Trec;sur  T0 Afield
j exergy, J kg1 Q_ rec;con ¼   (A-5)
dinsu =linsu þ 1=hair;o Fr C
where the surface temperature of the absorber can be determined
Superscripts
by the following equation:
* associated to solar rays
*  
Q_ rec Trec;sur  Tms
Subscripts ¼ (A-6)
Afield =Fr =C do =di =hms þ do lnðdo =di Þ=2=ltube
0 reference
abs absorbed When the solar isolation reflected by the heliostat field to the
*
avg average receiver ðQ_ rec Þ is known, the absorber’s surface temperature,
a at the inlet different heat losses, and the receiver thermal efficiency can be
b at the outlet determined in an iteration manner shown by Li et al. [2]. For more
con conduction information about the parameters and coefficients shown in Eqs.
conv convective (A-1)e(A-6), please also refer to Li et al. [2] for the simplicity of this
c.v. control volume paper.
em emissive
ex exit
fc force convection References
field heliostat field
gen generation [1] Z.H. Yao, Z.F. Wang, Z.W. Lu, X.D. Wei, Modeling and simulation of the pioneer
1 MW solar thermal central receiver system in China, Renewable Energy 34
h heliostat (2009) 2437e2446.
I the first law of thermodynamics [2] X. Li, W.Q. Kong, Z.F. Wang, C. Chang, F.W. Bai, Thermal model and thermo-
II the second law of thermodynamics dynamic performance of molten salt cavity receiver, Renewable Energy 35
(2010) 981e988.
in inlet
[3] W. Han, H.G. Jin, J.F. Su, R.M. Lin, Z.F. Wang, Design of the first Chinese 1 MW
i inner solar-power tower demonstration plant, International Journal of Green Energy
insi inner side of receiver 6 (2009) 414e425.
[4] A. Bejan, D.W. Kearney, F. Kreith, Second law analysis and synthesis of solar
ms molten salt
collector systems, Journal of Solar Energy Engineering 103 (1981) 23e28.
nc natural convection [5] S.C. Kaushik, R.D. Misra, N. Singh, Second law analysis of a solar thermal
o outer power system, International Journal of Solar Energy 20 (2000) 239e253.
p pump [6] A. Hepbasli, A key review on exergetic analysis and assessment of renewable
energy resources for a sustainable future, Renewable and Sustainable Energy
para parasitic Reviews 12 (2008) 593e661.
ps power cycle [7] T. Ganapathy, N. Alagumurthi, R.P. Gakkhar, K. Murugesan, Exergy analysis of
rec receiver operating lignite fired thermal power plant, Journal of Engineering Science
and Technology Review 2 (2009) 123e130.
ref reflective [8] Y.P. Dai, J.F. Wang, L. Gao, Exergy analysis, parametric analysis and optimi-
sgs steam generation subsystem zation for a novel combined power and ejector refrigeration cycle, Applied
st steam Thermal Engineering 29 (2009) 1983e1990.
[9] S. Farahat, F. Sarhaddi, H. Ajam, Exergetic optimization of flat plate solar
sur surface collectors, Renewable Energy 34 (2009) 1169e1174.
totloss total loss [10] J.F. Lu, J. Ding, J.P. Yang, Heat transfer performance and exergetic optimization
tube tube for molten salt flow in the receiver for solar receiver pipe, Renewable Energy 35 (2010) 1477e1483.
[11] R. Petela, Thermodynamic study of a simplified model of the solar chimney
power plant, Solar Energy 83 (2009) 94e107.
[12] J.E. Pacheco, M.E. Ralph, J.M. Chavez, Investigation of cold filling receiver
Appendix
panels and piping in molten-nitrate-salt central-receiver solar power plants,
Journal of Solar Energy Engineering 117 (1995) 282e289.
A modified thermal model for the molten salt cavity receiver. [13] S.A. Kalogirou, Solar thermal collectors and applications, Progress in Energy
A simple introduction for the thermal model of designing and Combustion Science 30 (2004) 231e295.
[14] J.M. Lata, M. Rodríguez, M.Á. Lara, High flux central receivers of molten salts
molten salt cavity receivers is provided as follows, which is based for the new generation of commercial stand-alone solar power plants, Journal
on a validated thermal model developed by Li et al. [2]. of Solar Energy Engineering 130 (2008) 021002 (1e5).
 C. Xu et al. / Applied Thermal Engineering 31 (2011) 3904e3913 3913

[15] T.Z. Ming, W. Liu, Y. Pan, G.L. Xu, Numerical analysis of flow and heat transfer [20] Y. Ying, E.J. Hu, Thermodynamic advantages of using solar energy in the
characteristics in solar chimney power plants with energy storage layer, regenerative Rankine power plant, Applied Thermal Engineering 19 (1999)
Energy Conversion and Management 49 (2008) 2872e2879. 1173e1180.
[16] M. Prakash, S.B. Kedare, J.K. Nayak, Investigations on heat losses from a solar [21] Y. You, E.J. Hu, A medium-temperature solar thermal power system and its
cavity receiver, Solar Energy 83 (2009) 157e170. efficiency optimisation, Applied Thermal Engineering 22 (2002) 357e364.
[17] M.J. Montes, A. Abánades, J.M. Martínez-Val, Performance of a direct steam [22] M.K. Gupta, S.C. Kaushik, Exergy analysis and investigation for various feed
generation solar thermal power plant for electricity production as a function water heaters of direct steam generation solar-thermal power plant,
of the solar multiple, Solar Energy 83 (2009) 679e689. Renewable Energy 35 (2010) 1228e1235.
[18] M.J. Montes, A. Abánades, J.M. Martínez-Val, M. Valdés, Solar multiple opti- [23] R.E. Sonntag, C. Borgnakke, G.J.V. Wylen, Fundamentals of Thermodynamics,
mization for a solar-only thermal power plant, using oil as heat transfer fluid sixth ed. John Wiley & Sons, Inc., 2003.
in the parabolic trough collectors, Solar Energy 83 (2009) 2165e2176. [24] N.E. Bergan, An external molten salt solar central receiver test, Solar Engi-
[19] S. Larbi, A. Bouhdjar, T. Chergui, Performance analysis of a solar chimney neering (1987).
power plant in the southwestern region of Algeria, Renewable and Sustain- [25] A.B. Zavoico, Solar Power Tower Design Basis Document, Sandia National
able Energy Reviews 14 (2010) 470e477. Laboratories, Report no. SAND2001e2100, 2001.