Desalination 425 (2018) 37–67

Contents lists available at ScienceDirect

Desalination
journal homepage: www.elsevier.com/locate/desal

Energy, exergy and exergo-economic analysis of different water desalination MARK

technologies powered by Linear Fresnel solar field
Ighball Baniasad Askaria,b, Mehran Ameria,⁎, Francesco Calisec
a
Department of Mechanical Engineering, Faculty of Engineering, Shahid Bahonar University of Kerman, Kerman, Iran
b
Department of Mechanical Engineering, Faculty of Engineering, University of Zabol, Sistan & Baluchestan, Iran
c
Department of Industrial Engineering, University Federico II of Naples, P.le Tecchio 80, 80125 Naples, Italy

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

Keywords: The integration of Multi Effect Desalination (MED) unit with and without Thermal Vapor Compression (TVC)
MED system into the Linear Fresnel Rankine Cycle (LFRC) was investigated for different seawater temperatures when
RO it is located at the regions with different solar radiation levels. A comparison was made between the fresh water
Linear Fresnel costs of the dual purpose LFRC/MED (or LFRC/MED/TVC) plants and the case when the MED and MED/TVC
Dual purpose plants
systems use direct steam of the LF solar field to produce fresh water (LF/MED & LF/MED/TVC). The water
Exergo-economic analysis
production costs of the described dual purpose plants were compared with that of the LFRC/Reverse Osmosis
Fuel price
(RO) plant (LFRC/RO). An exergo-economic analysis was performed to determine the water unit costs of the
plants. The results show that the fresh water costs of the LF/MED and LF/MED/TVC configurations are higher
than that of the dual purpose plants. Also, it was shown that the fresh water cost is more affected by solar
radiation level rather than seawater temperature of a region. It was also found that at fuel prices of 0.23$/m3,
the water production costs of fuel based dual purpose plants would be equal to that of the plants with solar
thermal source.

1. Introduction
Most of the power plants that are located next to the sea use Once-
Through (OT) cooling method by running a large amount of water
through the condenser in a single pass and discharging it back into the
sea. In these cases, the efficiency of the plants is depended on the sea
water temperature. The integrating of the Multi Effect Desalination
(MED) unit to the steam power plants are one of the alternatives to
produce water and electricity. The water production rate and efficiency
of dual purpose (electricity/water) plants are highly depended on the
seawater temperature. The higher seawater temperatures results in
higher required heat transfer areas of the MED desalination unit to
produce specific amount of fresh water [1]. More than half of Iran's
international border of 4430 km is coastline, including 740 km along
the Caspian Sea in the north and 1700 km along the Persian Gulf and
adjacent Gulf of Oman in the south (Fig. 1). Iran has started to decrease
the fuel price subsidy and therefor the application of renewable energy
technologies have been received great attention during the recent years
by Iran government. Due attention to high annual Direct Normal Irra-
diance (DNI), especially in south of Iran, solar energy appears as the
most appropriate technology to be used as thermal source of the power
plants to reduce the fossil fuel consumptions in Iran. MED and RO
technologies are two major desalination technologies that are used in
Iran. Please refer to [2] to find the water desalination technologies and
capacities that have been installed in Iran until 2014. Among all de-
salination plants in Iran, Qeshm water-power cogenerating is the only
dual purpose plant that utilizes the waste heat of a gas turbine to
produce the fresh water in a MED system with water production rate of
18,000 m3/day.
Several research works have been conducted on MED and MED/TVC
performances under different operational conditions [3–11]. The in-
tegration of MED and MED/TVC units with solar Rankine cycles (SRC)
plants was investigated by Fiorenza et al. [12]. In that research, MED
and RO desalination systems with different capacities were considered
to be powered by Photovoltaic panels and solar thermal power plants,
respectively. The water production costs of the PV/RO and SRC/MED
plants were determined as 2.05$/m3 and 2$/m3, respectively. The ap-
plication of SRC/MED and SRC/RO plants for the Persian Gulf,
Mediterranean Sea and Red sea were investigated by Fichtner and DLR
[13]. The water and electricity costs of two dual purpose plants were
found to range from 1.8$/m3 to 2.1$/m3 and 0.21$/kWh to 0.24$/
kWh, respectively. Sharaf et al. [14] investigated two techniques for

⁎
Corresponding author at: P.O. Box 76175-133, Iran.
E-mail address: ameri_mm@mail.uk.ac.ir (M. Ameri).

http://dx.doi.org/10.1016/j.desal.2017.10.008
Received 14 July 2017; Received in revised form 2 October 2017; Accepted 4 October 2017
0011-9164/ © 2017 Elsevier B.V. All rights reserved.
I.B. Askari et al.
Nomenclature
BPE Boiling Point Elevation
CCAPEX capital annualized direct costs, $
COE cost of electricity, $/kWh
COW cost of water, $/m3
CRF capital recovery factor
Cr compression ratio
CSP concentrating solar power plant
DNI Direct Normal Irradiation (W/m2)
Dr entrained vapor mass flow rate (kg/s)
Ds motive steam mass flow rate (kg/s)
DSG direct steam generation
GOR Gain Output Ratio
i interest rate (%)
LF Linear Fresnel solar field
MED Multi Effect Desalination
ṁ mass flow rate, kg/s
N number of project life time, year
desalination. In one technic the direct steam of the solar field was used
to feed a MED/TVC unit. In another technique the electricity generated
from the SRC plant was considered to be used as the power source of an
MED unit with Mechanical Vapor Compression (MVC) system. The
water production costs of the first and second techniques were found to
be as 1.5$/m3 and 2.1$/m3, respectively. The integration of MED and
RO units with SRC plants was investigated by Iaquaniello et al. [15].
The authors considered a case in which the low pressure output steam
of the plant is used as thermal source of the MED unit. Also, the RO unit
was assumed to be fed by the electricity generated from the SRC plant.
That research has shown that increasing the lifetime of the described
systems from 20 years to 30 years would result in decreasing the water
production cost by about 8.8%. Moser et al. [16] investigated two de-
salination systems of RO and MED/RO powered by the fuel based and
solar based steam plants. Different fuel price scenarios were used in the
economical calculations of that research. The results of that study
shown that the water production cost of the RO and RO/MED systems
with fuel based steam power plants would be 0.85$/m3 and 0.80$/
m3,respectively. However, for the solar based power plants the water
production cost were determined to be 1.22$/m3 and 1.1$/m3 for RO
and RO/MED systems, respectively. Different combinations of the RO,
MED and MED/TVC systems with SRCs were investigated by Palenzuela
et al. [17]. In that research, the SRCs were assumed to cool down by
three cooling methods of dry, wet and once through. The results of that
work shown that only for high salinities of seawater and high cooling
temperatures of the SRC condenser, the SRC/MED has lower water
production costs as compared to the SRC/RO plant. The integration of a
MED/TVC unit with SRC plant was studied by Delgado et al. [18]. The
output steam of high pressure and low pressure steam turbines
(4540 kPa and 362 kPa, respectively) of SRC plant were considered to
be used as the motive steam of the MED/TVC unit. The fresh water
production rate of 10,000 m3/day with sweater temperature of 26 °C
was considered in that research. The integration of the MED/TVC unit
with different water production rates into a fuel based steam Rankine
cycle was studied by Tamburini et al. [19]. The steam extracted from
the steam turbines of the plant at different suction pressures were
considered to feed the MED/TVC unit. The results of that study revealed
that the efficiency of the plant (38%) would be decreased to 21% when
the high pressure steam of the 4880 kPa is extracted from the high
pressure turbine of the plant to produce 36,000 m3/day of fresh water
in the MED/TVC unit. A comparison between the Parabolic Trough
Collector (PTC) base and LF base dual purpose SRC plants was con-
ducted by the authors of the present study [20]. The results of that work
shown that the water and electricity production costs of the LF/MED
38
Desalination 425 (2018) 37–67
NEA non-equilibrium allowance
NGB natural gas boiler
PB power block
Pd discharge vapor pressure (kPa)
Pr entrained vapor pressure (kPa)
Ps motive steam pressure (kPa)
PTC Parabolic Trough Collector
Q specific heat consumption, kJ/kg
Ra entertainment ratio
RC Rankine cycle
RO Reverse Osmosis
SA specific area of MED evaporators (m2/kg/s of D)
SMcw ratio of distillate to cooling seawater mass flow rate
TC condenser temperature (°C)
Tf temperature of feed seawater (°C)
TVC Thermal Vapor Compression
TSW seawater temperature (°C)
TMS motive steam temperature (°C)
plant is approximately 10% and 25% lower than that of the PTC/MED
plant. Also, it was shown that because of the higher thermal efficiency
of the PTC solar field as compared to the LF, the electricity and water
production costs of both plants would be equal, if the capital cost of the
PTC solar field would be decreased to 66% of its first cost assumption
(420$/m2). In another research, the authors of the present paper in-
vestigated the application of the LF solar field direct steam as the heat
source of the MED/TVC [21]. The results of that paper shown that the
water production costs of the LF/MED/TVC system varies between
1.63$/m3 to 3.03$/m3 for the cases with no Thermal Storage System
(TES) and 15 h of TES, respectively. The energy and exergy analysis of a
polygeneration system was performed by one of the authors of the
present paper (Calise et al. [22]). The described system includes Con-
centrated Photovoltaic Thermal (CPT), biogas heater, absorption chiller
and an MED unit. An exergo-economic analysis was used in that work to
determine the fresh water, electricity, heating and cooling costs of the
so-called system. In another research by Calise et al. [23], a poly-
generation system comprising geothermal well, PTC, Organic Rankine
Cycle (ORC), MED and absorption chiller was investigated to determine
the water, electricity, cooling and heating energy costs of the system.
An exergo economic model was used in that research work to determine
the production costs of different products of the system. In that work,
the cumulative exergy analysis of the plant for two periods during the
cold and hot seasons was applied instead of the instantaneous exergy
analysis.
As it can be seed from the literature review, there are valuable re-
search works that address the integration of MED and RO systems with
SRC plants. However, to the best of our knowledge, the water pro-
duction costs of dual purpose SRC/MED, SRC/MED/TVC and SRC/RO
plants have not been reported to be compared with that of the MED and
MED/TVC plants feed by direct steam of the solar field. Also, the de-
tailed investigation on the integration of MED/TVC unit into the SRC
plants has not been vastly investigated. Only in two references of
[18,19] the detailed analysis of SRC/MED/TVC plants were reported.
As it has been shown in [18], the described plant was considered at a
specific seawater temperature and for a specific water production rate.
Also in [19], a fuel based Rankine cycle was considered and the GOR of
the MED/TVC plant was assumed to be constant for a specific seawater
temperature. However, the seawater temperature has an important ef-
fect on the GOR of the MED/TVC plant and consequently on the water
production cost of the plant. In the present paper, three different sea-
water temperatures corresponding to the Mediterranean Sea, North
Australia coastlines and Persian Gulf seawater temperatures were
considered in the calculations of the MED and MED/TVC plants. No
I.B. Askari et al.
considerable exergo-economic analysis on the SRC/MED, SRC/MED/
TVC and SRC/RO plants has been reported in the literature. Only, in
reference [24] the authors performed a detailed exergo-economic
analysis on the SRC/MED and SRC/RO plants. However, the water
production costs of that research were obtained for the seawater
39
Desalination 425 (2018) 37–67
Fig. 1. Iran southern coastlines.
temperature of 26 °C and the results are useful for the regions located in
the Mediterranean Sea. In another research conducted by Catrini et al.
[25], an exergo-economic analysis was performed to determine the
electricity and water production costs of a fuel based Rankine cycle
integrated with MED/TVC unit. In that research also the Mediterranean
Fig. 2. LFRC/RO plant-Configuration A.
I.B. Askari et al.
seawater temperature was assumed as the operational conditions of the
MED/TVC plant. In the present paper, an exergo-economic analysis was
performed to determine the water production costs of the dual purpose
LFRC/MED, LFRC/MED/TVC, LFRC/RO and single purpose LF/MED
and LF/MED/TVC configurations. The unit costs of the electricity and
water for dual purpose plant and also the water production costs of one
purpose plants were compared with each other for the plants located at
regions having different solar Direct Normal Radiation (DNI) levels and
three different seawater temperatures of 26 °C, 31 °C and 36 °C. The
water production costs of the described plants were determined for
three different capital costs of the MED unit. Also, the calculations of
the exergo-economic analysis were conducted for the plants with nat-
ural gas boiler heat source, and the water production costs of the plants
were determined under different fuel price scenarios.
2. System description
A regenerative 50 MWe SRC plant was selected based on the current
commercial SRC plants like Andasol-1, in Spain [26]. The input and
output temperatures of the plant were considered as 373 °C and 256 °C,
respectively. Five main configurations of LFRC/RO (A), LFRC/MED
(B1), LFRC/MED/TVC (C1 & D1), LF/MED/TVC (C2 & D2) and LF/MED
(B2) were considered as it is shown in Figs. 2 to 6. In the case of LFRC/
MED/TVC, the heating steam was considered to be supplied by ex-
tracting the steam from the high and low pressure turbines at two
different pressures of 4540 kPa (C2) and 362 kPa (D2), respectively. As
it is shown in Figs. 2 to 4, the SRC plant comprises of six regenerative
heat feeders (five numbers in the case of LFRC/MED) to preheat the
water before entering the solar field. The Direct Steam (DS) generated
in the Linear Fresnel (LF) solar field was assumed to be used as the
working fluid of the plant. The solar field with no Thermal Storage
System (TES) was considered providing a solar share of nearly 20.50%
for all configurations. The other part of the plants heating requirements
were considered to be supplied on natural gas boiler (NGB) back up.
Since, the governing equations in thermodynamic modeling of dual
purpose configurations (A, B1, C1 and D1) are same with the exceptions
of RO and MED components, only for configuration A (LFRC/RO), the
40
Desalination 425 (2018) 37–67
energy and exergy balances as well as the exergy efficiencies for each
component are presented in Table A1 (Appendix A). Also, for the MED
unit, the governing equations of energy balance, exergy balance and
exergy efficiency are shown in Table A2.
2.1. Configuration A (LFRC/RO)
As it can be seen from Fig. 2, the resulting steam at temperature of
373 °C and pressure of 100 bar is introduced into the high pressure
turbine and after two stages expansion, the output steam is flowed
through the regenerator. The hot steam leaves the regenerator at high
temperature of 373.4 °C and it is flowed through the second turbine to
be expanded into lower pressures. The low pressure steam turbine of
the plant has four extractions that are required to be used in three low
pressure Feed Water Heaters (FWHs) and deaerator. The final extrac-
tion pressure from the low pressure steam turbine is determined de-
pended on the seawater temperature and assuming the temperature
difference of 8 °C between the seawater and once through cooling
condenser.
2.2. Configuration B1 (LFRC/MED)
A process flow diagram of configuration B1 is shown in Fig. 3. This
configuration is similar to configuration A but with the difference that
four extractions are taken from the low pressure turbine unlike the
typical five that are used in configurations A. The low temperature MED
unit is replaced by condenser of the plant to be fed by the steam from
the low pressure turbine outlet. In this case, the exhaust steam is con-
sidered to be extracted at a slightly higher pressure than the condenser
pressure of configurations A, as it must feed the MED unit at 70 °C
(31.2 kPa). It is necessary to mention that, the steam with 67.5 °C is
flowed into the first effects of the MED unit considering the temperature
drop of 2.80 °C at each effect.
2.3. Configurations C1 and D1 (LFRC/MED/TVC)
It can be observed from Fig. 4 (dashed lines to TVC) that two
Fig. 3. LFRC/MED plant-Configuration B1.
I.B. Askari et al.
possible steam extraction options (indicated as t8a and t4a) are as-
sumed as alternatives to supply the motive steam to the TVC system of
the MED unit. These steam extractions (t8a and t4a) are respectively
assumed to be positioned at pressures of 4540 kPa and 362 kPa. The
amount of motive steam extracted to drive the TVC system may change
depending on the required fresh water that is considered in the calcu-
lations. As it is clear from Fig. 4, similar to the case of configuration A,
41
Desalination 425 (2018) 37–67
Fig. 4. LFRC/MED/TVC plant-Configuration C1
and D1.
totally six closed type feed water heaters are included in the SRC; one of
which has drains pumped forward and the others comprise drains
cascaded backward flows. Also, the plant has a direct-contact feed
water heater, which serves as a deaerator. The steam with 67.5 °C is
flowed into the first effects of the MED unit and the output steam from
the MED/TVC, which is sub-cooled water at 67.5 °C, is pumped and
returned back into a mixing chamber. To have the same evaporative
Fig. 5. LF/MED plant-Configuration B2.
I.B. Askari et al.
effect areas for the MED/TVC and MED units, the temperature differ-
ences between the effects was considered to be constant and approxi-
mately equal as 2.80 °C for both cases of MED (B1) and MED/TVC
(C1 & D1) units.
2.4. Configuration B2 (LF/MED), C2 and D2 (LF/MED/TVC)
The MED and MED/TVC plants were considered to fed by the direct
steam of the LF solar field. These cases are single purpose plants with
the only product of fresh water. Unlike the dual purpose plants, these
Fig. 7. MED system-Parallel feed.
42
Desalination 425 (2018) 37–67
Fig. 6. LF/MED/TVC plant-Configurations C2
and D2.
configurations do not need the power block and their LF solar fields are
smaller than the dual purpose plants solar fields. For configuration B2,
the output steam of the LF field at the temperature of 70 °C and pressure
of 31.2 kPa is introduced into the MED first effect. Also, for the LF/
MED/TVC configurations (C2 & D2), the output steam from the LF solar
field at two high pressure/temperature scenarios of 362 kPa/140.5 °C
and 4540 kPa/259 °C are introduced into the TVC unit. The high
pressure motive steams (362 kPa and 4540 kPa) are flowed throw the
TVC systems and their temperatures and pressures are decreased into
the lower values of 70 °C and 31 bar.
I.B. Askari et al.
Fig. 8. MED/TVC system-Parallel feed.
3. Method and thermodynamic modeling
3.1. MED and MED/TVC systems
Both MED and MED/TVC systems were investigated for different
numbers of effects (5 to 12 effects). Each effect has a horizontal tube
falling film evaporator. Both desalination systems are the parallel feed
configuration in which the evaporating brine and heating steam flows
have a same direction. The preheated feed sea water is divided into the
set of parallel streams to feed into each evaporation effect. The thermal
energy, which is called as heating steam, is introduced into the first
effect of the desalination plants. In the case of MED/TVC, part of the
heating steam that is condensed in the first effect (Ds) is returned back
into the mixing chamber and the other part (Dr) joins the fresh water
product (Fig. 8). The heating steam leaving the first effect of the MED
system is flowed into the mixing chamber as it is shown in Figs. 3 and 4.
The vapor formed in the first effect is directed into the second effect to
be used as the heat source. For the MED unit, this process continues up
to the last effect and the output steam of the last effect is flowed
through the condenser where its latent heat is transferred to the cooling
seawater (MC) and its temperature is raised from TC to Tf (Fig. 7). For
the MED/TVC unit, as it is seen from Fig. 8, the output steam of each
effect is introduced into the next effect and this process is continued up
to the effect that the TVC is located after that (i − 1th effect). Part of
the steam formed in the i − 1th effect is suctioned into the TVC system
and the other part is introduced into the next effect to be used as the
heating steam of that effect. For both MED and MED/TVC systems, the
generated vapor at each effect is flowed into the preheaters to warm up
the feed water before entering into the next effect. The model essen-
tially includes mass and energy balances for each plant component;
please refer to Appendix A (Table A3) for the formulations that were
used in modeling of the MED/TVC desalination unit. The following
assumptions were used in thermodynamic modeling of the MED and
MED/TVC unit:
• Steady state operation.
43
Desalination 425 (2018) 37–67
• Equal temperature difference of feed heaters.
• Negligible heat losses to the surrounding.
• The distillate is salt free for all effects.
• Variations of specific heat as well as boiling.
• Temperature difference between all effects is assumed to be the
same.
• The compression, mixing and expansion processes in the TVC are
adiabatic.
• The motive steam D and the entrained vapor D are assumed to be
s r
saturated.
• The temperature difference between the input and output seawater,
respectively into and from the condenser was considered to vary
between 3.76 °C and 9.11 °C.
• The seawater temperature spraying on the first effect was set as
60 °C, and for the last effect the temperature of the sprayed seawater
(Fn) was considered to be 2 °C less than the last effect heating
temperature.
• The maximum yearly seawater temperature was considered in the
calculations.
• The temperature difference between the heating steam (70 °C) and
the first effect (67.5 °C) was considered as 2.5 °C.
• The seawater salinity was considered as a value between
35,000 ppm and 46,000 ppm with the resultant salt concentration of
respectively between 63,000 ppm and 70,000 ppm; depending on
the seawater salinity of the study locations.
For MED and MED/TVC units, two computer programs were de-
veloped in MATLAB software in order to determine the important
parameters such as Gain Output Ratio (GOR) and the specific total heat
transfer area (SA) of the plant. For MED unit, GOR is defined as the
amount of distillate water per each unit of heating steam that is entered
into the first effect of MED and similarly for the MED/TVC unit, GOR is
defined as the amount of fresh water that is produced per each unit of
motive steam that is flowed through the TVC system. The thermo-
dynamic formulations of MED unit are not described in detail here,
because the thermodynamic modeling of the MED unit has been vastly
I.B. Askari et al. Desalination 425 (2018) 37–67

investigated and reported in the literature [1,6,9]. Also, the MED/TVC Table 1
unit with the TVC location at the last effect has been reported by many The main input data that are used in thermodynamic modeling of SRC plant.
researchers [1,3,4,5,8,10,11]. Therefore, only the formulations of the
Component Reference [28] Model
MED/TVC changing the TVC locations on the effects are shown in this
paper (Appendix A, Table A3). For both MED and MED/TVC systems, Turbine
the governing equations of mass and energy balances for each compo- High pressure turbine inlet 370 373
temperature (C)
nent of the system have been applied in the METLAB computer code.
High pressure turbine inlet pressure 9000 10,000
Since such balances are trivial and may be easily formulated, describing (kPa)
the model in details is not necessary. The same temperature difference High pressure turbine efficiency 85.5 85.5
of approximately 2.8 °C was set in the calculations applying the con- (%)
straints of entrainment and compression ratios (Eqs. (a80) and (a82), Low pressure turbine efficiency (%) 89.5 89.5

Appendix A) of Ra ≤ 4 and 1.8 ≤ Cr ≤, respectively [1]. Also, for each Extraction point pressures
location, depending on the seawater salinity (Xbn, Eq. (a63), Appendix), Extraction t8 (kPa) 4540 4540
Extraction t6 (kPa) 2060 2060
the allowable resultant salt concentration was set as a constant value. In
Extraction t5 (kPa) 875 875
the case of MED/TVC system, for each number of effects, by applying Extraction t4 (kPa) 362 362
the mass and energy balance equations for each component of the Extraction t3 (kPa) 122.4 122.4
system (Table A3), the calculations were made to determine the amount Extraction t2 (kPa) 34.6 34.6
of distillate changing the location of TVC unit from the last effect to the Pressure drop
middle effects. Please refer to [1,3–11] to find the detailed thermo- Extraction t8 (%) 2.5 3
dynamic modeling of MED and MED/TVC systems. The calculations of Extraction t6 (%) 3 3
Extraction t5 (%) 4.5 3
this part of the study were conducted to determine the proper sizes of
Extraction t4 (%) 3 3
the MED and MED/TVC systems when they are located in the regions Extraction t3 (%) 3 3
with three different sea water temperatures of 26 °C, 31 °C and 36 °C. Extraction t2 (%) 3.5 3
The maximum numbers of effects for both desalination systems was Reheating line (%) 11.5 11.5
considered as 12 numbers and for the regions with higher seawater Condenser pump P1
temperatures, the calculations were done to find the lower number of Isentropic efficiency 75 73.5
effects in order to keep constant the temperature differences between Electro-mechanical efficiency 98 –

the effects. The temperature difference between effects has a direct Feed water pump P2
impact on the size of the evaporator areas. For both MED and MED/TVC Isentropic efficiency 78 76.4
Electro-mechanical efficiency 98 –
units, the ratio between the distillate water and seawater mas flow rate
were determined to calculate the pumping power that is required to Low/high pressure closed feed water heaters
Terminal temperature difference 1.5 1.5
transfer the seawater to the plants. The GOR of the plants were de-
(C)
termined for different numbers of effects and different seawater tem- Drain cooling approach (C) 5.5 5.5
peratures. As it was formerly mentioned in the assumptions of the de-
Condenser
salination units, the maximum yearly seawater temperature was used in Condensing pressure (bar) 0.08 0.08
the calculations. Steam generator thermal efficiency 98 Direct steam generated
(%) from LF
3.2. LF solar field and dual purpose plants Total pressure drop (water side) 450 Direct steam generated
(kPa) from LF
Thermal efficiency (%) 38.21 38.98
System Adviser Model (SAM), a software developed by NREL, was
used to calculate the output thermal power of the LF solar fields [27].
For each region of the study, the hourly solar radiation data, tem- The LF solar field comprises of different numbers of Loops with a
perature data, wind speed data, latitude and longitude of that region total area of 6676.8 m2 for each loop. The specifications of the LF solar
are input into the SAM. The other parameters such as input temperature field are shown in Table 2. The total land area in Table 2, is sum of the
of the Heat Transfer Fluid (HTF) at the inlet of the solar field, the solar field required land area and the non-solar field land area which is
aperture area and operating pressure of the solar field are input into the considered for the distances between each loops of the mirrors and also
software to calculate the output temperature of the steam generated by other facilities such as power block and so forth. The total solar field
solar field [20,21]. The thermodynamic modeling of the plants was required land area is determined by multiplying the LF aperture areas
developed in EES (Engineering Equation Solver). The net electricity by non-solar field multiplier, which was considered as 1.4 in the present
generation of all plants was considered as 50 MWe, and the thermo- work [27].
dynamic property of the plant model was set based on the SRC plant
that has been reported in [28] with the exception of the plant top 3.3. Energy and exergy model
pressure which was considered as 10,000 kPa (Table 1). Since the
thermodynamic modeling formulations of the plant is trivial and rather In order to investigate the thermodynamic behavior of the plants,
than standard, describing the model in details is not necessary. As it can and to determine the amount of thermal energy that is wasted in each
be seen from Table 1, there are six steam extraction from two steam
turbines of the plant two of which were chosen as the high and low Table 2
pressures (4540 kPa and 362 kPa) to be used as the motive steam of the Geometrical and optical parameters of LF solar fields.
MED/TVC system in the calculations of the present study. The outputs
Design parameter LF solar field
of SAM and EES, which are the required thermal loads of the plants,
mass flow rates of Rankine cycles and solar fields, gross and net elec- Alignment North–South
trical works of the plants as well as water production rates, were used in Reflective aperture area (m2) 513.6
MATLAB to calculate the required aperture areas of the LF solar field Length of collector module (m) 44.8
Focal length (m) 7
providing the solar share of 20.5% for the all plants and based on a
Solar field area (m2) 1× (total aperture area (m2))
simple optimization method that has been presented in the previous Total land area (m2) 1.4× solar field area (m2)
work of the authors [20].

44
I.B. Askari et al. Desalination 425 (2018) 37–67

component of the plants, the respective calculations were performed for
the inlet and outlet enthalpies, exergies, mass flow rates, pressures and
temperatures. Exergy destructions were calculated to determine the
plant irreversibilities. For each component of the plants, the mass, en-
ergy and exergy balances were applied based on the inlet and outlet
thermodynamic properties. The calculations were developed in EES
making the following assumptions:
Ẇnet = ẆGross − ẆPumps − Ẇdes (6)
WPumps is the required powers of the condenser, feedwater, on-
cethough cooling and sewater intake pumps. Wdes is the required power
that is consumed in MED, MED/TVC or RO desalination plants. Ė fresh is
the chemical exergy flow associated with the produced fresh water that
was calculated based on the minimum theoretical work of separation
required in any reversible desalination process using the following

• The system operates in a steady state condition. equation [31,32].

• Solar field supplies only 20.5% of the plants required annual Ė fresh = ṄFresh × ∅ × Ru × T0 × X s,feed (7)
thermal loads.
• Because the NGB is used as the backup system, the inlet and outlet where, ṄFresh is the molar flow rate of product freshwater that is defined
as follows:
temperatures of the plants are constant.
• The changes in the kinetic and gravitational terms in the energy and ṄFresh =
ṁ fresh
=
ṁ 15
exergy balances are negligible. MMwater MMwater (8)
• The turbines and pumps efficiencies and the pressure losses in heat ∅ in Eq. (7) represents the dissociation factor of salts (equal to 2 for
exchangers and the pipelines were applied according to the speci-
fications that are shown in Table 1. sodium chloride), Ru is the universal constant of gases, and Xs, feed is the
• The MED or MED/TVC units were considered as a component with molar fraction of the dissolved salts that was calculated using the fol-
lowing formulation:
specific inlet and outlet flow rates.
• The reference (dead) state temperature was considered to be 2 °C X s,feed = (TDSSeawater × MMwater )/(MMNACL ) (9)
greater than the seawater temperature.
• The once through cooling temperature was considered to be 8 °C where, TDSSeawater is the salinity of seawater in terms of milligram salt
per each liter of water that differs for different seawaters across the
greater than the seawater temperature.
• For SRC/RO and SRC/MED/TVC plants that include once through world (0.035 g/L to 0.048 g/L). MMwater and MMNACL are the molecular
mass of water and salt that are equal to 18 kg/kmol and 58.5 kg/kmol,
cooling condenser, the specific water flow rate of 87 m3/MWh was
considered to exhaust the vapor extracted from the steam turbine respectively. For the seawaters with the salinities of 0.038 g/L
[29]. (38,000 ppm) and 0.07 g/L (70,000 ppm), Xs, feed (Eq. (7)) is obtained
• For all cases, the desalination plants (MED, MED/TVC and RO) were as 0.01160 and 0.02154, respectively.
considered to be near the sea with the maximum distance of 2.5 km.
4. Exergo-economic model
For all components of the plants, the majority of components doesn't
involve the chemical exergy considering that the working fluid of the Exergo-economics is usually used for systems with steady or quasi-
plants dose not undergone any change in the chemical compositions. steady operating conditions. The following formulation of cost balances
However, in the desalination units some important chemical processes was used in the present study:
occur, and chemical exergy must be taken into account in exergy bal-
∑ c i ∙Eẋi + Ż unit = ∑ c j∙Eẋj
ances [23]. The physical exergy of each state point can be calculated i ∈ {inlet} j ∈ {outlet} (10)
using following equation:
where, Eẋi and Eẋj are the inlet and outlet exergy powers of the stream
Eẋ = ṁ ∙ [(h − h o) − T0 ∙ (s − s 0)] (1) at inlet and outlet of a component, respectively, which are expressed in
For each component of the plant, the equation of the exergy de- kWex or kJex . ci and cj are the exergetic unit cost ($/kWh) at the inlet and
h
struction rate, Ėd , can be expressed as: outlet of a component, respectively. This approach inherently assumes
that instantaneous or average hourly exergy flows maintains constant
Ėd = ∑ Exin − ∑ Eẋ out − Ẇ (2) throughout the operating period, and has been proven reliable for ap-
plication in large power plants. Ż unit is the cost rate of the capital in-
where, Exin and Exout refere to the input and output exergy of a com- vestment and the operation and maintenance (ŻO & M) of each compo-
ponent, respectively. nent which is calculated as follows:
The following equations were used to determine the energy and
j=N
exergy efficiency of the plants at the hours with maximum solar ra-
Ż unit = ŻCapital,unit + ∑ ŻO & M, unit
diation level for each location:
j=1 (11)
Ẇ
ηI = net where, N is the life time of the project. Ż unit can be calculated as the
Q̇s (3) levelized cost divided per the number of hours of annual operation by
applying the Capital Recovery Factor (CRF) as follows:
Ẇnet + Ė fresh
ηII =
Ėin,SF (4) ŻCapital,unit ∙CRF + ŻO & M,
Ż unit =
unit
⎛$⎞
8760 ⎝h⎠ (12)
where, Ėin,SF is the inlet exergy to the system that is the function of the
sun's outer surface temperature (Ts = 5800 K) and defined as follows i. (1 + i) N
[30]: CRF(i, N) =
(1 + i) N − 1 (13)
4
4 T T
Ein,SF = Afield ∙DNI × ⎛⎜1 − ⎛ 0 ⎞ + ⎛ 0 ⎞ ⎞⎟
⎜ ⎟ ⎜ ⎟
where, i and N are the real interest rate (6%) and life time of the project
⎝ 3 ⎝ Ts ⎠ ⎝ Ts ⎠ ⎠ (5) (25 years), respectively. Eq. (10) can be written in terms of cumulative
exergy flows over a yearly period as follows:
where, DNI is Direct Normal Irradiance, and Afield is the LF apperture
area. Ẇnet is the total net power of the plant that is obtained by sub- ∫period Ci ∙Ei + ∫period Ż unit = ∫period Cj∙Ej̇ (14)
tracting the parasitic powers from total gross power generated by the
plants using the following equation: Considering the application of NGB back up system, the inlet and

45
I.B. Askari et al. Desalination 425 (2018) 37–67

Table 3 Z͠ unit = 8760 × Ż unit = ŻCapital,unit ∙CRF + ŻO & M, unit (16)
Different cost parameters of the Linear Fresnel solar field, MED desalination unit
[13,27,33,34]. As mentioned formerly, the solar fields were considered to supply
20.5% of total annual thermal requirements of the plants during the
MED or MED/TVC
year hours. Therefore, the exergy costs of the inlet and outlet stream
Direct costs (DC)
Main investment ($/m3/day) 1240 to 1800 should be calculated based on solar and NGB thermal sources. It is
Post-treatment plant ($/m3) 120 evident that the exergy costs corresponding to the solar thermal source
Open sea water intakes ($/m3) 313 is high as compared to the fuel base thermal source. The following
Drinking water storage and pumping 100
equations were used to determine the solar field and NGB inlet and
($/m3)
Indirect costs (IC)
outlet exergy costs, respectively, for the hours with pick solar radiation
Freight & insurance rate during 5.00% DC and for night hours when the NGB supplies the thermal loads of the
construction plants:
Owner's cost rate 10.00% of direct material and
labor cost solarshare × 8760 × Csin ∙Ein + Z͠ LF = solarshare × 8760 × CSout ∙Eout (17)
Contingency rate 10.00% of DC
Construction overhead (interest during 12.24% of DC f
NGBshare × 8760 × Cin ∙Ein + Z͠ NGB = NGBshare × 8760 × Cout
f
∙Eout
construction)
Operation costs (OC) (18)
Electricity costs ($/m3) Depending on electricity cost
where, solarshare and NGBshare are the percentage of solar share and
(assuming: 1.55 kWh/m3)
Spare parts replacement 1.5% of total DC NGB share which were considered as 20.5% and 79.5%, respectively, in
Chemical cost of product water ($/m3) 0.04 the present work. Also, Couts and Coutf are the exergy unit costs of the
Insurance 0.5% of total DC solar field and NGB outlet streams, respectively. The main aim of the
Natural Gas auxiliary boiler costs ($/m3) 0.02 to 0.8 present work is to determine and compare the exergy costs of the
Labor cost of product water ($/m3) 0.025
electricity and fresh water streams for the plants with solar field
RO thermal source. However, at the final part of this research, the afore-
Direct costs (DC)
mentioned costs were calculated for the plants with fuel based thermal
Main investment ($/m3/day) 900
Pretreatment plant ($/m3) 250 source. The initial capital costs, operation & maintenance costs and
Post-treatment plant ($/m3) 120 other relevant costs, which were used n the calculations of the present
Open sea water intakes ($/m3) 313 paper, are shown in Table 3.
Drinking water storage and pumping 100
($/m3)
Wastewater collection & treatment 50 5. Results and discussions
($/m3)
Indirect costs (IC) At the first part, the thermodynamic modeling of MED and MED/
Freight & insurance rate during 5.00% DC
TVC systems was conducted. Then, the results were used in the ther-
construction
Owner's cost rate 10.00% of direct material and modynamic modeling of the whole plants; which were done using the
labor cost calculations of EES program. The output results of EES were input in
Contingency rate 10.00% of DC MATLAB to calculate the required solar field area supplying 20.5% of
Construction overhead (interest during 12.24% of DC total annual thermal load of each plant that were assumed to be located
construction)
at the specific locations of the study.
Operation costs (OC)
Electricity costs ($/m3) Depending on electricity cost
(assuming: 3.5 kWh/m3 to 5.1. MED and MED/TVC thermodynamic modeling results
5.5 kWh/m3)
Spare parts replacement 1.5% of total DC
Chemical cost of product water ($/m3) 0.04
In order to have a proper comparison between the plants with MED
Insurance 0.5% of total DC and MED/TVC units, the evaporator surface areas of these units should
Natural gas auxiliary boiler costs ($/m3) 0.02 to 0.8 be same in the technical and exergo-economical calculations of the
Labor cost of product water ($/m3) 0.05 plants. The surface areas of the MED and MED/TVC units play an im-
Solar field, NGB and power block portant role in the unit costs of the water as it was mentioned formerly.
Direct costs (DC) Two programs were developed in MATLAB to determine the GOR of
Site improvement ($/m2) 20
MED and MED/TVC units when it is located in the regions with three
Solar filed ($/m2) 108, 144 and 180
HTF system ($/m2) 35
different seawater temperatures (26 °C, 31 °C and 36 °C). In order to
Thermal Storage System ($/kWht) 77 insure the accuracy of the MED unit thermodynamic model, the results
Contingency rate 10.00% total DC were compared with that reported in [35]. The input parameters that
Power block (PB) ($/kWh) 1540 were used in the calculations of the model are shown in Table 4. The
NGB (million $/MWht) 0.12
main outputs of the model are GOR, evaporative specific areas of the
Indirect costs (ID)
Design and construction 15% of total DC effects, condenser surface area and the specific cooling water flow rate
Land cost ($/m2) 10 (SMcw). The results from the validation are shown in Table 5. It is ne-
Insurance 1% of total DC cessary to mention that in order to match the performance parameters
of the present model with that is presented in reference [35], the total
heat transfer areas of the effects, condenser and preheaters were set as
outlet exergy of the components are constant during the year hour
constant in the model. As it can be seen from Table 5, there is a good
(8760 h). Therefore, the above equation can be written as follows:
agreement between the results of the present model and Darwish et al.
model [35]. It is important to note that in Darwish et al. model [35], the
8760 × Ci ∙Ei + Z͠ unit = 8760 × Cj ∙Ej (15)
specific surface area of the MED unit (SA) was calculated as 184 m2/kg/
s of D excluding the condenser heat transfer area. The present model
where, Z͠ unit is the yearly cost rate of the capital investment and the
operation & maintenance. Z͠ unit is obtained from the following equation results shown that considering the condenser heat transfer area, the SA
by using Eq. (12): would be obtained as 233 m2/kg/s of D.
Table 6 shows the results of the thermodynamic modeling of MED

46
I.B. Askari et al. Desalination 425 (2018) 37–67

Table 4 temperatures. In Table 6, the specific cooling water flow rate (SMcw) is
Input data in the MED system thermodynamic model [35]. the ratio between the distillate and intake seawater mass flow rates. We
used this ratio in the calculations of the seawater pumping power. It is
Design parameters Values
necessary to mention that SMcw ratio is different from the ratio between
Effects specific heat (Cp) (kJ/kg) 4 the distillate and feed water mass flow rates; which is approximately
Distillate (kg/s) 94.39 equal to 0.35 for the MED desalination process.
BPE (°C) 1
Similar calculations were conducted to determine the performance
Latent heat of the vapor inside the effects (λ) (kJ/kg) 2333
Seawater cooling temperature (TSW) (°C) 26 parameters of MED/TVC unit based on the formulations that are shown
Condenser temperature (°C) 35 in Appendix A, Table A3. The thermodynamic modeling of the MED/
Seawater salinity (ppm) 45,978 TVC unit has been vastly investigated in the previous literature. How-
Resultant brine salinity (ppm) 72,000 ever, with the exception of the research works that have been reported
Heating steam (kg/s) 72.8
in [11,18], most of the previous works are related to the MED/TVC
Heating steam temperature (Tc) (°C) 70
First effect Brine temperature (Tb) (°C) 65 systems with the TVC unit located after the last effect of the MED
Total heat transfer coefficients of the effects, preheaters and condenser 3 system. The parametric study of MED/TVC units reported in [18] has
(U) (kW/m2 °C) been conducted for seawater temperature/salinity of 26 °C/38,000 ppm
Numbers of effects (n) 6
and for MED/TVC motive steam pressures of 362 kPa, 875 kPa,
2060 kPa and 4540 kPa. In the present work, similar modeling was
conducted for two different heating steam pressures of 362 kPa and
Table 5
Comparison between the results obtained from the present model and Darwish et al. 4540 kPa and for three seawater temperatures/salinity of 26 °C/
model [35]. 38,000 ppm, 31 °C/35,000 ppm and 36 °C/46,000 ppm. The low and
high motive steam pressures of 362 kPa and 4540 kPa can be extracted
Results Present work Darwish et al. [35] from the low pressure and high pressure steam turbines of the SRC plant
Feed water mass flow rate (kg/s) 1107 1090 as it was formerly shown in Table 1. For the first, the model results were
GOR 5.46 5.40 compared with the results reported in [18]. The validation results and
Specific cooling water flow rate (SMcw) 8.22 7.78 design parameters that were used in the modeling of the MED/TVC are
SA (m2/(kg/s of distillate)) 191 184 shown in Table 7. As it can be seen from Table 7, the results are in a
Condenser heat transfer area (m2) 9827 9448
good agreement with the results that have been presented in reference
[18].
Table 6 The same assumptions were made for thermodynamic modeling of
MED system specifications for different seawater temperatures. MED/TVC desalination plant as in the case of MED unit. Three different
seawater cooling temperatures were considered with different numbers
Location Persian Gulf Indian Mediterranean sea of effects for MED/TVC unit. Table 8 shows the results of thermo-
Ocean
dynamic modeling of MED/TVC unit with motive steam pressure of
Seawater salinity (ppm) 46,000 35,000 38,000 4540 kPa and for three different seawater temperatures and salinities.
Resultant brine salinity (ppm) 72,000 63,000 66,000 As it can be seen from Table 8, the specific heat transfer areas (SA) of
Numbers of effects (n) 10 11 12 the systems are approximately equal to that are obtained for the MED
GOR 8.50 9.41 9.78
units (Table 6). The comparison between the GORs of MED and MED/
Heating steam flow rate (kg/s) 8.01 7.22 6.95
Heating steam temperature (Ts) 70 70 70 TVC units shows that the GORs of MED/TVC units are nearly 55% more
(°C) than that of MED unit. Also, as it can be seen in Tables 6 and 8, the total
Effects temperature difference 2.86 2.83 2.83 heat transfer areas of the MED and MED/TVC units are obtained as
(°C)
nearly equal to 35,400 m2 by multiplying the specific areas of the ef-
Seawater cooling temperature 35.9 30.9 25.9
(TSW) (°C)
fects (SA) by distillate mass flow rates. Table 6 shows that the specific
Last effect temperature T(n) (°C) 41.65 39.15 36.36 cooling water flow rate (SMcw) of MED units are lower than that of
Condenser temperature 3.76 6.25 8.46 MED/TVC units, which means for a specific distillate mass flow rate,
difference for Mc (°C) the MED units required more seawater intake flow rates than the MED/
First effect temperature T(1) 67.5 67.5 67.5
TVC units. Also, as it is shown in Tables 6 and 8, the motive steam flow
(°C)
Distillate (kg/s) 59.46 65.91 70.54 rates of the MED/TVC units are approximately 55% lower than the
SA (m2/(kg/s of distillate)) 520.43 520.75 520.90 heating steam flow rates of the MED units. The results of this part of the
Specific cooling water flow rate 0.070 0.127 0.189
(SMcw)
Table 7
Mathematical model predictions against of Trapani desalination unit.
system that were derived from the MATLAB program. The seawater
Variable Reference [18] Model
temperature/salinity of Mediterranean sea, north Australia (Indian
Ocean) and Persian Gulf were considered as, respectively, 26 °C Number of effect 12 12
(38,000 ppm), 31 °C (35,000 ppm) and 36 °C (46,000 ppm) in the pre- Optimum effect for TVC location 11 11
sent study. As it can be seen from Table 6, for the effect temperature Seawater salinity (ppm) 40,000 40,000
differences of approximately 2.8 °C, the specific area (SA) of MED plant Seawater concentration at output (ppm) 65,900 65,900
Motive pressure Ps, kPa 4540 4540
is obtained as constant for two different seawater cooling temperatures. Seawater cooling temperature (TSW) (°C) 25.9 25.9
In the calculations of the LFRC/MED plant, the MED unit was assumed Motive steam temperature (TMS) (°C) 70 70
to have a same SA for different seawater temperatures. As it can be seen Top brine temperature (T1) (°C) – 67.5
from Table 6, the MED plants for three regions should have different Minimum brine temperature (Tn) (°C) 37 37
Feed sea water temperature to nth effect (Tf) (°C) 35 35
numbers of effects to keep constant their evaporative surface areas. For
Feed sea water temperature to 1st effect (Tf1) (°C) 60 60
the regions with lower seawater temperatures, the required heating Temperature drop per effect (ΔT) 2.7–2.8 2.77
steam flow rate is lower, the numbers of effects are higher and conse- GOR 14.66 15.00
quently the GORs are higher than the locations with higher seawater SA (m2/(kg/s of distillate)) 523.46 520.52

47
I.B. Askari et al. Desalination 425 (2018) 37–67

Table 8 calculations of the dual purpose SRC plants.

MED/TVC system specifications for different seawater temperatures for motive steam
pressure of 4540 kPa.
5.2. Energy analysis of dual purpose plants
Salinity (ppm) 46,000 35,000 38,000
Resultant salinity (ppm) 72,000 63,000 66,000
GOR 13.69 14.61 14.98 As it was mentioned formerly, the LF solar fields were designed for
Motive steam flow rate (kg/s) 4.96 4.66 4.53 annual solar share of approximately 20.50%. At first, the required
Motive steam temperature (Ts) (°C) 70 70 70 thermal load and mass flow rate of each plant were calculated by using
Number of effects 10 11 12
the thermodynamic modeling of the plant in EES. Then, the results were
TVC effect 10 10 11
Effects temperature difference (°C) 2.77 2.77 2.77 used in MATLAB to determine the required LF aperture areas corre-
Seawater cooling temperature (TSW) (°C) 35.9 30.9 25.9 sponding to the thermal loads and mass flow rates of the plants. Finally,
Last effect temperature T(n) (°C) 42.57 39.8 37.03 the determined LF solar field areas, which entail the annual solar share
Condenser temperature difference for Mc (°C) 4.67 6.9 9.13 of 20.5%, were input in the calculations of the EES to determine the
First effect temperature (T1) (°C) 67.5 67.5 67.5
input exergy of the LF solar fields as well as the exergy efficiency of
Distillate (kg/s) 67.98 67.92 67.91
SA (m2/(kg/s of distillate)) 522.11 519.64 522.06 each plant by using Eqs. (5) and (4), respectively. The thermodynamic
Specific cooling water flow rate (SMcw) 0.168 0.248 0.340 specifications of the reference LFRC plant were shown formerly in
Table 1. As shown in Table 1, two suction pressures of 4540 kPa and
362 kPa were considered to be used in the MED/TVC unit for the dual
study were used in the calculations of the next parts where the MED and purpose plants of C1 and D1 (LFRC/MED/TVC). Eq. (5) was used to
MED/TVC units are feed with the heating steam of the LFRC plants calculate the net efficiency of the plants considering the electricity
steam turbines or direct steams of the LF solar fields. The results of consumptions of desalination units. For all dual purpose plants, the net
Table 8 show that for three MED/TVC units with 10 effects (or 11 ef- electricity generation rate of 50 MWe was considered in the calcula-
fects) and 12 effects, the maximum GORs are obtained when TVC units tions. Since the electricity consumptions of the RO and MED units vary
are located on the 10th and 11th effects of the MED system, respec- depending on seawater salinity and pumping power of the plants, two
tively. These results were obtained by changing the TVC locations in the electricity consumptions of 3.5 kWh/m3 and 5.5 kWh/m3 were con-
computer modeling of the MED/TVC system as it is shown in Figs. 9 to sidered for RO unit. For the MED and MED/TVC units, the electricity
11, for seawater temperature of 26 °C (12 effects), 31 °C (11 effects) and consumptions of 1.5 kWh/m3 and 2.5 kWh/m3 were used in the cal-
36 °C (10 effects), respectively. As it is clear from Fig. 9, for each culations. Figs. 12 and 13 show the variations of the electrical efficiency
heating steam pressure, the maximum GOR is obtained for a specific of the dual purpose plants with net electricity generation rate of
TVC location. These results are in good agreement with that has been 50 MWe versus the water production rates of the plants for seawater
reported in [5,18]. In fact, the MED/TVC system is turned into a MED/ temperatures of 26 °C and 36 °C, respectively. As it can be seen from
TVC + MED by changing the TVC position from the last effect to the Figs. 12 and 13, the fresh water flow rate of the MED unit in config-
middle effects and it has been proven that MED/TVC + MED system is uration B1 varies from 34,660 m3/day to 40,970 m3/day for seawater
slightly more efficient than the MED/TVC system [5,11,18,36]. Base on temperatures of 36 °C to 26 °C, respectively, depending on the output
Fig. 9, the maximum GOR of the MED/TVC unit with 12 effects is ob- mass flow rate of the LFRC/MED steam turbine. For seawater tem-
tained when TVC is located after the effects of 11 and 9 for suction perature of 36 °C, the water production rate of configuration B1 is ob-
pressures of 4540 kPa and 362 kPa, respectively. The specific areas of tained as 38,100 m3/day that is not shown here. As it is clear from
the evaporators corresponding to the maximum GORs for suction Figs. 12 and 13, for the seawater temperature of 26 °C with higher
pressures of 4540 kPa and 362 kPa are obtained approximately as numbers of effects in the MED unit, the fresh water production of
523 m2/kg/s of distillate. Similarly, for seawater cooling temperature of configuration B1 is higher than that of the seawater temperature of
36 °C with 10 effects, the maximum GORs of the MED/TVC system is 36 °C. Therefore, the higher numbers of effects the higher water pro-
obtained when TVC is located on 9th and 10th effects for the suction duction rate and consequently the higher electricity consumptions in
pressures of 362 kPa and 4540 kPa, respectively, as it is shown in the MED unit of the plants. That is why the electrical efficiency of
Fig. 11. It is also clear from Fig. 11 that for maximum GORs, the eva- configuration B1 in the seawater temperature of 26 °C (30.47%) is
porator specific area is approximately equal to 520 m2/kg/s which is slightly lower than that of in the seawater temperature of 36 °C
nearly equal to the evaporator specific area of MED/TVC system with
12 effects and seawater cooling temperature of 26 °C (Fig. 9).
15.5 540

The results of MED and MED/TVC thermodynamic modeling are 15

shown in Table 9 in brief. As it is clear from Table 9, for different 530
seawater cooling temperatures, the evaporator specific area of the 14.5
MED/TVC evaporators are obtained as approximately 520 m2/kg/s.
SA(m2/kg/s of D)

The higher GOR of the MED/TVC plant results in lower required intake 14 520
seawater mass flow rate to produce each unit of fresh water as com-
13.5
GOR

pared to the MED unit. This is why the ratio of distillate to cooling
510
seawater mass flow rate (SMcw) for the MED/TVC units are higher than
13
that for the MED units. Table 9 also shows that the percentage differ-
ence between the GOR of MED/TVC and MED units are varied de- 12.5 500
pending on the seawater temperature and suction pressure. For in- Pm=4540kPa, SA
stance, the GOR of the MED/TVC system with suction pressure of 12 Pm=362kPa, SA
362 kPa is higher than that of the MED system by about 41.88%, 490
Pm=4540kPa, GOR
37.83% and 36.90% for sweater cooling temperatures of 36 °C, 31 °C 11.5 Pm=362kPa, GOR
and 26 °C, respectively. These percentages are higher than the men-
11 480
tioned values for the suction pressure of 4540 kPa; about 61.22%, 5 6 7 8 9 10 11 12
55.31% and 53.61% for sweater cooling temperatures of 36 °C, 31 °C Thermocompresor Location
and 26 °C, respectively. Parts of the results (Table 9) such as GOR, SMcw,
Fig. 9. GOR and SA for sea water cooling temperature of 26 °C.
seawater cooling temperature and suction pressures were used in the

48
I.B. Askari et al. Desalination 425 (2018) 37–67

15.5 540 40

15 38
530 36
14.5
34

Energy efficiency (%)

14 520

SA(m2/kg/s of D)
32
13.5
GOR

30
510
13 28
26
12.5 500
Conf#A(3.5 kWh/m )
3

Pm=4540kPa, SA 24 3
Conf#A(5.5 kWh/m )
12 Pm=362kPa, SA Conf#B1(1.5 kWh/m )
3
22
Pm=4540kPa, GOR 490 Conf#B1(2.5 kWh/m3)
11.5 Pm=362kPa, GOR 20 Conf#C1(362 kPa, 1.5kWh/m3)
3
Conf#C1(362 kPa, 2.5kWh/m )
3
11 480 18 Conf#D1(4540 kPa, 1.5kWh/m )
5 6 7 8 9 10 11 3
Conf#D1(4540 kPa, 2.5kWh/m )
16
Thermocompresor Location 10000 20000 30000 40000 50000
3
Fig. 10. GOR and SA for sea water cooling temperature of 31 °C. Fresh water rate (m /day)
Fig. 12. Once through cooling temperature 34 °C (TSW = 26 °C).
15.5 540
Pm=4540kPa, SA
15 40
Pm=362kPa, SA
Pm=4540kPa, GOR 530 38
14.5 Pm=362kPa, GOR
36
SA(m /kg/s of D)

14 520
34
Energy efficiency (%)

13.5
GOR

32
510
13 30
2

28
12.5 500
26
12 Conf#A(3.5 kWh/m )
3

490 24 3
Conf#A(5.5 kWh/m )
3
11.5 22 Conf#B1(1.5 kWh/m )
3
Conf#B1(2.5 kWh/m )
11 480 20 Conf#C1(362 kPa, 1.5kWh/m3)
5 6 7 8 9 10 3
Conf#C1(362 kPa, 2.5kWh/m )
3
Thermocompresor Location 18 Conf#D1(4540 kPa, 1.5kWh/m )
3
Conf#D1(4540 kPa, 2.5kWh/m )
Fig. 11. GOR and SA for sea water cooling temperature of 36 °C. 16
10000 20000 30000 40000 50000
3
Fresh water rate (m /day)
(30.78%). By increasing the electricity consumptions of MED unit from
1.5 kWh/m3 to 2.5 kWh/m3, the electricity efficiency of configuration Fig. 13. Once through cooling temperature 44 °C (TSW = 36 °C).
B1 would be decreased by about 2.2%. Configuration A with RO de-
salination unit has the highest electrical efficiency among four plants extracted steam of the plants at high pressure of 4540 kPa, has the
even with the electricity consumptions of 5.5 kWh/m3 in RO unit. Also, lowest electrical efficiency among four configurations. The results of
as it is clear from Figs. 12 and 13, configurations D1, which uses the Figs. 12 and 13 show that for the constant electricity generation rate of

Table 9
The specifications of the MED and MED/TVC systems for different seawater cooling temperatures.

Desalination Suction GOR TVC Number Effects Seawater Last effect Condenser First effect SA (m2/(kg/ Specific
system pressure location of effects temperature cooling temperature temperature temperature s of cooling
(kPa) difference °C temperature (°C) difference for (°C) distillate)) water flow
(°C) (°C) rate (SMcw)

MED – 8.50 – 10 2.86 36 41.65 3.76 67.50 520.43 0.070

MED/TVC 362 12.06 9 10 2.77 36 42.57 4.67 67.50 522.23 0.133
MED/TVC 4540 13.69 10 10 2.77 36 42.57 4.67 67.50 521.22 0.168
MED – 9.41 – 11 2.83 31 39.15 6.25 67.50 520.75 0.1270
MED/TVC 362 12.97 9 11 2.77 31 39.80 6.90 67.50 520.24 0.216
MED/TVC 4540 14.61 10 11 2.77 31 39.80 6.90 67.50 519.64 0.248
MED – 9.78 12 2.83 26 36.36 8.46 67.50 520.90 0.189
MED/TVC 362 13.39 9 12 2.77 26 37.03 9.13 67.50 519.03 0.309
MED/TVC 4540 14.98 11 12 2.77 26 37.03 9.13 67.50 522.06 0.340

49
I.B. Askari et al.
50 MWe, configurations C1 and D1 with water production rates of
27,000 m3/day and 21,000 m3/day, respectively, have the same elec-
trical efficiency of approximately 30.5% as configuration B1 with water
production rate of 40,970 m3/day for sweater temperature of 26 °C.
Also, for seawater temperature of 36 °C, the electrical efficiency of
configurations C1 and D1 with water production rates of 22,000 m3/
day and 16,000 m3/day, respectively, are equal to that of configuration
B1 with fresh water production rate of 34,660 m3/day.
The results of Figs. 12 and 13 show that the electrical efficiency of
configuration A is higher than that of configuration B1 with MED unit.
In some cases where the once through cooling temperature of LFRC/RO
plant is high, the electrical efficiency of configuration A would be de-
creased to a value close to that of configuration B1. The variations in
electrical efficiency of configurations A and B1 versus the fresh water
production rate of these two dual purpose plants are shown in Figs. 14
and 15 for the once through cooling temperatures of 52 °C and 57 °C,
respectively, in the condenser of configuration A. The cooling tem-
peratures of 52 °C to 57 °C are commonly used in the conventional
steam plants that are operating within the Persian Gulf. As it is shown
in Figs. 14 and 15, for higher cooling temperatures of LFRC/RO plant
and for electricity consumptions of 5.5 kWh/m3 in RO unit, the elec-
trical efficiency of configuration A is equal to that of configuration B1
with the MED unit. It can be concluded that only in the specific con-
ditions when the once through cooling temperatures are high and the
electricity consumptions of RO unit is high as the result of high salinity
of seawater, the electrical efficiency of the LFRC/RO and LFRC/MED
plants approach to a same value. For the other cases, configuration A
with RO unit has higher electrical efficiency than configuration B1 with
MED unit.
5.3. Exergy analysis of dual purpose plants
Based on the obtained results of the previous parts of the present
study, configuration A has always higher electrical efficiencies as
compared to the other plants. However, the highest electrical effi-
ciencies of a dual purpose plant cannot guarantee its lowest fresh water
production costs. In this part of the study, an exergy analysis was made
to obtain the inlet and outlet exergy flow rates as well as the exergy
destruction rates of each component of the study plants. The exergy
flow rates were used in the exergo-economic analysis to determine the
electricity and fresh water unit of costs that are produced by each plant.
Firstly, the thermal loads and mass flow rates of all plants were cal-
culated using the EES calculations. Table 10 shows the mass flow rates,
temperatures and pressures of Configurations A, B1, C1 and D1 (Figs. 2
to 4) for seawater temperatures of 26 °C and 36 °C and fresh water flow
rates of 40,970 m3/day and 34,660 m3/day, respectively. As can be
seen from Table 10, because of the higher electricity consumption in the
RO unit, the total electricity generation of configuration A is higher
than that of the other plants. However, the higher electrical efficiency
of configuration A implies the lower mass flow rates for the plant and its
LF solar field. Configuration B1 with electrical efficiency of higher than
C1 and D1 has lower required thermal energy and lower mass flow rates
for its Rankine cycle and LF solar field as compared to C1 and D1. The
highest required thermal energy and mass flow rate are related to
configuration D1 with lowest electrical efficiency among the other
configurations.
Three locations with three different seawater temperatures and
different total annual solar radiation levels were chosen to obtain the
electricity and water production costs of the plants operating at dif-
ferent weather conditions around the world. The specifications of the
study locations are shown in Table 11. As it is shown in Table 11, the
seawater temperatures of the study areas are as equal as 26 °C, 31 °C
and 36 °C, respectively, for de Almeria in Mediterranean Sea, Port-
Hedland in north Australia (Indian Ocean) and Kish Island in Persian
Gulf. The solar radiation levels of three locations are different so that
Port-Hedland and de Almeria has the highest and lowest total annual
50
Desalination 425 (2018) 37–67
Direct Normal Irradiance (DNI) of respectively 2700 kWh/m2 and
1990 kWh/m2. For configurations A, C1 and D1 the condenser cooling
temperature were considered to be 8 °C more than the seawater tem-
peratures for all locations. Please refer to reference [37] to obtain the
seawater temperature of any location around the world.
The required aperture areas of the LF solar field can be calculated by
using the solar radiation data of a specific location, mass flow rate and
required thermal energy of the plants. The calculations were done for
solar share of 20.5% as it was mentioned formerly. Fig. 16 shows the
percentages of solar share and defocused fraction for the LF solar fields
of configurations A, B1, C1 and D1 located in Kish Island with seawater
temperature of 36 °C and total annual DNI of 2300 kWh/m2. The total
annual defocused thermal energy is defined as the total annual solar
thermal energy that is wasted by defocusing the solar field mirrors to
prevent the overheating of the plant. The defocused fraction was de-
fined as the ratio of total annual defocused solar thermal energy to the
total annual required thermal energy of the plant [21]. The LF solar
field comprises of different numbers of loops with aperture area of
6676.8 m2 for each loop. As it can be seen from Fig. 16, depending on
the thermal energy needs of each plant, the solar field with specific
number of loops satisfies the total annual solar share of 20.5%. For
instance, for configuration A with lowest required thermal energy
(148.60 MWht, Table 10), the LF solar field with 41 numbers of loops
supplies 20.5% of total annual required thermal energy of the plant.
Configuration D1 with highest thermal energy needs requires larger
aperture areas of LF solar field (63 loops) to supply 20.5% of total
annual thermal load of this plant. Similar calculations were performed
to determine the LF solar field numbers of loops supplying 20.5% of the
total annual thermal needs of the plants when they are located in Port-
Hedland and de Almeria with different seawater temperatures and solar
DNI levels. The results of the calculations are shown in Table 12. The
reference plant in Table 12 is the LFRC plant without desalination units.
As it is clear from Table 12, the required aperture areas of the LF solar
fields in the reference plants are lower than that of the dual purpose
plants due to lower required thermal energy needs and mass flow rate
of the reference plants. It is evident that the locations with higher solar
radiation levels require lower aperture areas of the LF solar field as it is
shown in Table 12. For Port-Hedland, the required LF numbers of loops
is approximately 22% and 32% lower than that for the Kish Island and
de Almeria, respectively. The increasing in the electricity consumptions
of RO unit from 3.5 kWh/m2 to 5.5 kWh/m2 increases the required
numbers of LF solar field by about 5.8% and 7.4% for the Port-Hedland
40
38
36
34
Energy efficiency (%)
32
30
28
26
3
Conf#A(3.5 kWh/m )
24 3
Conf#A(5.5 kWh/m )
3
22 Conf#B1(1.5 kWh/m )
Conf#B1(2.5 kWh/m3)
20
18
16
10000 20000 30000 40000 50000
3
Fresh water rate (m /day)
Fig. 14. Once through cooling temperature of 52 °C for configuration A-TSW = 36 °C.
I.B. Askari et al. Desalination 425 (2018) 37–67

40 Table 11
The specifications of three locations around the world that were used in the calculations
38 of present work.
36 Location DNI (kWh/m2/ TSW (°C) Condenser cooling temperature
34 day) (°C)
Energy efficiency (%)

32 Kish Island 2300 36 42

Port-Hedland 2700 31 39
30 de Almeria 1990 26 34

28
26 30 60

Solar thermal power that is defocused(%)

Def, Conf#A
24 Ssh, Conf#A
Conf#A(3.5 kWh/m3) Def, Conf#B1 50
22 3
Conf#A(5.5 kWh/m ) Ssh, Conf#B1
3
Conf#B1(1.5 kWh/m ) 25 Def, Conf#C1
20 Ssh, Conf#C1
Conf#B1(2.5 kWh/m3)

Solar Share(%)
Def, Conf#D1 40
18 Ssh, Conf#D1
16
10000 20000 30000 40000 50000 20 30
3
Fresh water rate (m /day)
Fig. 15. Once through cooling temperature of 57 °C for configuration A-TSW = 36 °C. 20
15
and Kish Island (or de Almeria), respectively. As it can be seen from 10
Table 12, the required solar field aperture area for the Port-Hedland
reference plant is approximately 23% and 63% lower than that for the
Kish Island and de Almeria, respectively. In the other words, the in- 10 0
creasing of the total annual solar radiation levels by about 17.40% 30 35 40 45 50 55 60 65 70
(from 2300 kWh/m2 to 2700 kWh/m2) and 35.13% (from 1990 kWh/ LF solar field number of loops
m2 to 2700 kWh/m2) results in decreasing the required LF solar field Fig. 16. The variations of solar share and defocused fraction versus the solar field re-
area of the reference plant by about 23% and 63%, respectively. quired number of loops-, TSW = 36 °C, DNI = 2300 kWh/m2/year.
The calculations of exergy analysis were performed by using the
computer code that was developed in EES. Table 13 shows the ther- destructions in each component to the total exergy destruction of the
modynamic specifications of the inlet and outlet streams for each
plant, are shown in Fig. 18a to d. As can be seen from Figs. 17 and 18,
component in configuration A and for seawater temperature of 26 °C. the highest percentage of exergy destruction is caused by the LF solar
Based on Table 13, the mass flow rate of the plant is equal to 62.2 kg/s
field. Based on Fig. 17, the highest exergy destructions are occurred
which is different from the solar field mass flow rate of this plant
respectively in MED/TVC units of configurations D1 and D2, high
(77.13 kg/s, Table 10). Part of the solar field mass flow rate is not pressure turbine of configuration D1, Reheat turbine and MED unit of
flowed into the power plant and that is used to increase the temperature
configuration B1. Fig. 18a shows that for configuration A, the reheat
of the output stream from the high pressure turbine (point 2, Fig. 2) in turbine has the highest percentage of exergy destruction after solar field
the re-heater of the plant.
and for high pressure turbine and condenser of the plant, the percen-
The main exergy destructions corresponding to the components of tage of exergy destruction is low; nearly 1%. The MED and MED/TVC
configurations A, B1, C1 and D1 are shown in Fig. 17 for seawater
units in configurations B1, C1 and D1 have the percentages of 5%, 7%
temperature of 26 °C. Also, the percentages of exergy destructions in and 8%, respectively, which are high as compared to the percentage of
main components of the plants, which are defined as the ratio of exergy
exergy destruction in the condenser of configuration A (1%). The

Table 10
The specifications of dual purpose plants for two cooling seawaters of 26 °C and 36 °C.

Conf. Desalination electricity Seawater cooling Condenser temperature LF mass flow LFRC thermal load Gross electricity Fresh water flow
consumption (kWh/m3) temperature (°C) (°C) rate (kg/s) (MWhth) (MWh) rate (m3/day)

A 3.5 26 34 77.13 145.26 57.22 40,975

A 3.5 36 42 78.91 148.60 56.27 34,660
A 5.5 26 34 82.84 154.13 60.71 40,975
A 5.5 36 42 83.05 156.40 59.22 35,580
B1 1.5 26 34 93.42 175.94 52.44 40,975
B1 1.5 36 42 92.14 173.51 52.07 34,660
B1 2.5 26 34 99.31 187.03 54.06 40,975
B1 2.5 36 42 97.11 182.88 53.46 35,580
C1 1.5 26 34 106.50 200.57 53.55 40,975
C1 1.5 36 42 104.80 197.43 53.19 34,660
C1 2.5 26 34 111.77 210.46 55.07 40,975
C1 2.5 36 42 110.40 207.81 54.58 34,460
D1 1.5 26 34 116.92 220.19 53.55 40,975
D1 1.5 36 42 114.90 216.49 53.19 34,660
D1 2.5 26 34 122.47 230.65 55.07 40,975
D1 2.5 36 42 121.00 227.99 54.58 34,460

51
I.B. Askari et al. Desalination 425 (2018) 37–67

Table 12 MED/TVC unit (8%). The lowest percentage of exergy destruction of

Solar field required numbers of loops supplying 20.50% of total annual thermal loads of solar field in Fig. 18a to d is obtained for configuration D1, which
the plants.
means that the other components of this configuration have higher
Plant A (3.5 kWh/ A (5.5 kWh/ B1 C1 D1 Referencea exergy destructions as compared to the other three plants.
m3 ) m3)

Port Hedland, TSW = 31 °C, DNI = 2700 kWh/m2/year 5.4. Exergo-economic analysis results
Numbers of 34 36 41 47 51 30
loops
In order to determine the electricity generation and water produc-
Kish, TSW = 36 °C, DNI = 2300 kWh/m2/year tion costs of the dual purpose plants and also the fresh water production
Numbers of 42 44 50 57 63 37
loops
costs of the single purpose plants, an exergo-economic analysis was
performed in this part of the study applying Eqs. (1) to (18). The
de Almeria, TSW = 26 °C, DNI = 1990 kWh/m2/year
electricity generation costs of the plants were calculated based on both
Numbers of 54 58 66 75 83 49
loops gross electricity (excluding electricity consumption in desalination
units) and net electricity rates. The electricity generation costs of the
a
LFRC plant without desalination unit. plants based on the gross electricity rate are slightly lower than that
based on the net electricity rates. The gross electricity unit of cost was
Table 13 used in the calculations of fresh water unit of cost considering different
The thermodynamic specifications of each state point of configuration A shown in Fig. 2. electricity consumption rates in the desalination units. The capital,
operating and maintenance costs of the plants were set based on the
Stream number T (°C) P (kPa) h (kJ/kg) S (kJ/kg ∙ K) ṁ (kg/s)
cost data that are shown in Table 3. The total exergy of the plants were
1 373.0 10,000 3008.0 6.07 62.20 calculated using Eq. (4) and based on solar thermal energy of the LF
t8 273.9 4540 2856.0 6.12 6.79 solar fields at the hour of maximum solar radiation level in a sunny day
d8 219.4 4540 941.4 2.50 6.79
of the year for three regions of the study. The LF solar thermal energy is
t6 213.9 2060 2703.0 6.13 3.67
d6 190.4 2060 809.8 2.24 10.46 capable of supplying 100% of required thermal load of the plants at the
2 213.9 2060 2703.0 6.13 51.57 hour of peak solar radiation level. The diagram blocks of exergy flows
3 373.4 1818 3192.0 7.08 51.57 between different components of configurations A, B1, C1, D1, D2, C2
t5 282.8 875 3017.0 7.12 3.49 and B2 are shown in Figs. 19 to 22. As can be seen from Figs. 19 to 21,
t4 188.9 362.7 2839.0 7.12 3.01
d4 110.9 362.7 465.2 1.42 3.01
LF solar field, power block, RO, MED and MED/TVC units were con-
t3 105.4 122.4 2659.0 7.22 2.69 sidered as a system with inlet and outlet exergy streams in kWhex. For
d3 77.94 122.4 326.3 1.05 5.70 configuration A, WRO is the only exergy stream that is flowed into the
t2 72.44 34.6 2481.0 7.28 2.96 RO unit from the LFRC. However, for configurations B1 and C1 (or D1),
d2 39.19 34.6 164.1 0.56 8.66
two exergy streams of WMED and E4 (Et4c and Et8c for configurations C1
4 33.90 5.3 2275.0 7.43 39.60
5 33.90 5.3 142.0 0.47 39.60 and D1, respectively) are flowed into the MED and MED/TVC units of
6 34.10 5.3 146.8 0.49 48.26 configurations B1 and C1 (or D1). Each exergy stream has a specific cost
7 34.83 1000 146.8 0.48 48.26 in $/kWhex that is determined by solving a set of equations. The exergy
8 140.10 1000 589.9 1.74 48.26 stream unit of costs is mainly related to the capital costs of components
9 179.90 1000 763.0 2.14 62.20
of the plants. Table 14 shows the exergy balance equations of the plants
10 149.90 10,460 790.0 2.17 62.20
11 258.10 10,460 1125.0 2.86 62.20 that were used to calculate the unit cost of each inlet or outlet stream
into or from the components of the study plants. The exergo-economic
model consists of a series of cost balances, one for each component, and
extraction of the steam at the temperature of 70 °C in configuration B1 a number of auxiliary equations for each component associated with
causes to increase the exergy destructions of the Reheat steam turbine multiple exiting exergy flows. As can be seen from Table 14, the exergy
of this plant. As can be seen from Fig. 18b, the Reheat turbine of con- balance equations are written based on Eqs. (10) to (18) so that each
figuration B1 has the percentage destruction of 8% which is twofold of parameter represents the total annualized cost of a component (Z͠LF ,
that for configurations C1 and D1 (4%). In configuration D1, the ex- Z͠RO , Z͠MED , Z͠PB ) or total annualized cost of the inlet or outlet exergy
traction of the steam from the high pressure turbine of the plant results streams into or from a component. For instance, the total annual cost of
in high exergy destruction in this component; so that the percentage of the outlet exergy stream from the solar field is calculated by multi-
exergy destruction in this component (7%) is nearly same as that in plying the total annual number of hours with solar thermal energy by
hourly cost of the outlet exergy stream from the solar field as

Fig. 17. Exergy destructions of main components

of dual purpose plants.

52
I.B. Askari et al.
18-a-Configuration A 18-b-Configuration B1
18-c-Configuration C1 18-d-Configuration D1
(solarshare × 8760 × C1s ∙ E1). C1s is the cost of outlet exergy stream
($/kWhex) from the solar field that is determined based on the LF solar
field capital cost. E1 is the hourly exergy of solar field outlet stream in
kWhex. As can be seen from Table 14, the seawater intake costs were set
as zero due to the free source of energy and water. Also, the exergy cost
of the wasted streams such as wasted brine or outlet cooling seawater
from the condenser of MED or MED/TVC unit (point 14 in configura-
tions B1, C1 and D1 also point 5 in configurations D2, C2 and B2) were
assumed to be zero. For configurations A, C1 and D1, the unit cost of
input and output exergy streams of the condenser unit were assumed to
be wasted with the exergy flow cost of zero.
Fig. 23 shows the exergy efficiency, the gross and net electricity
generation unit costs of dual purpose plants assuming net electricity
generation rate of 50 MWe and water production rate of 34,660 m3/day
for Persian Gulf with solar radiation level of 2300 kWh/m2/year and
seawater temperature of 36 °C. As can be seen in Fig. 23, the electricity
unit cost is defined in terms of cost of electricity (COE) ($/kWh). The
exergy efficiencies of the plants were calculated based on the peak solar
radiation level in a sunny day of summer in the Persian Gulf (ap-
proximately 960 W/m2). However, the unit cost of electricity was de-
termined based on the total annual thermal energy of the plants that is
supplied by the solar field; assuming the solar share of 20.5% and using
the equations that are shown in Table 14. Fig. 23 shows that config-
urations A and D1 have the highest and lowest exergy efficiencies, re-
spectively, among four dual purpose plants. The higher electrical effi-
ciency implies the lower aperture area and the lower mass flow rates of
the solar field as it was explained formerly in the explanations of
Table 10. Therefore, considering the same electricity generation and
water production rates, the plants with smaller solar fields has the
53
Desalination 425 (2018) 37–67
Fig. 18. Percentage of total exergy destructions
caused by the main component of dual purpose
plants.
lower input exergy in the solar field and consequently higher exergy
efficiencies. Extracting the high pressure steam of 4540 kPa from the
high pressure steam turbine of configuration D1 results in an exergy
efficiency of 14% for this configuration which is nearly 50% lower than
that of 21% for configuration A with the RO unit. The electricity unit
cost of configuration A is the lowest value among four plants. As it was
expected, the electricity unit costs calculated based on the gross elec-
tricity generation rate are approximately 7.8% lower than that are
calculated based on the net electricity generation rate. The electricity
unit costs that were obtained based on gross electricity generation rate
were used in the calculations of the fresh water unit costs of the plants.
The capital cost of MED and MED/TVC units plays an important role
in the calculations of fresh water unit of costs. According to the results
of former sections, assuming constant evaporative areas for the MED or
MED/TVC effects, the lower seawater temperatures entails higher fresh
water flow rates. Therefore, the capital cost of MED or MED/TVC units
can be different for the locations with different seawater temperatures.
Assuming the capital costs of 1240$/m3/day for the MED (or MED/
TVC) system located in a region with seawater temperature of 26 °C and
distillate flow rate of 40,970 m3/day (50.8M$/day), for a location with
seawater temperatures of 31 °C and 36 °C with distillate flow rates of
respectively equal to 38,100 m3/day and 34,660 m3/day, the capital
cost of MED unit would be obtained as equal to 1333$/m3/day and
1465$/m3/day, respectively. In the other words, with the specific in-
vestment of 50.8M$/day, the capital cost of an MED unit located in the
places with different seawater temperatures and consequently different
distillate flow rates is obtained as different values. Besides the seawater
temperature, the total annual DNI of a location has an important effect
on the electricity or fresh water unit cost of the plants. Based on
I.B. Askari et al.
Table 11, the locations with higher solar radiation level are not ne-
cessarily that with the higher seawater temperatures due to the factors
such as humidity of a region, shore wind speeds and so forth. That is
why the seawater temperature of a region like Port-Hedland with
higher DNI level is lower than that of the Kish Island in Persian Gulf
with DNI level of lower than Port Hedland. The calculation of the
electricity and fresh water unit of costs were done for three locations
(Table 11) considering different distillate flow rates and capital costs of
the MED unit for the plants comprising the MED system.
The low seawater temperature in the Mediterranean coastal regions
is an advantage for water productions using the MED or MED/TVC
plants. However, the low DNI level on one hand and the high fuel prices
on the other hand are two disadvantages of MED technologies in that
region. For the Persian Gulf, the high seawater temperature and high
solar radiation level are respectively the advantage and disadvantage of
solar based MED technologies within this region. However, with the
low fuel price of 0.02$/m3 (natural gas), the application of solar
thermal energy as the thermal source of the desalination plants is not
still economically justified for the Persian Gulf region. Port-Hedland
located in north Australia, which has high solar radiation level and
moderate seawater temperature, can be a suitable place for seawater
distillation using solar thermal source. In the present work, the exergo-
economic analysis of totally 8 configurations were done in order to
determine the electricity or fresh water unit of costs assuming that the
plants are located at three different regions around the world
(Table 11):
• The reference plant of configuration A without desalination unit.
• Configuration A with RO electricity consumption of 3.5 kWh/m 3
54
Desalination 425 (2018) 37–67
Fig. 19. Block diagram of exergy flows between
different components of Conf#A.
and 5.5 kWh/m3
• Configuration B1 with MED electricity consumption of 1.5 kWh/m3
and 2.5 kWh/m3
• Configuration C1 with MED/TVC having the motive steam pressure
of 362 kPa and electricity consumption of 1.5 kWh/m3 and
2.5 kWh/m3
• Configuration D1 with MED/TVC having the motive steam pressure
of 4540 kPa and electricity consumption of 1.5 kWh/m3 and
2.5 kWh/m3
• Configuration D2 with MED/TVC having the motive steam pressure
of 4540 kPa and electricity consumption of 1.5 kWh/m3 and
2.5 kWh/m3
• Configuration C2 with MED/TVC having the motive steam pressure
of 362 kPa and electricity consumption of 1.5 kWh/m3 and
2.5 kWh/m3
• Configuration B2 with MED electricity consumption of 1.5 kWh/m3
and 2.5 kWh/m3.
The inlet and outlet exergy streams of the main components of the
systems were calculated in terms of kWhex based on Figs. 19 to 22. The
unit cost of electricity (COE) and fresh water (COW) were determined
for dual purpose configurations. Also, the COW of LF/MED (B2) and
LF/MED/TVC (C2 and D2) systems were compared to that of dual
purpose configurations (B1, C1 and D1). For all configurations COW
were determined based on $/kWhex and $/m3 of produced fresh water.
For two configurations of D1 and D2, the results of exergo-economic
analysis are shown in Tables 15 and 16, respectively, for the case
when the plants are located in Port-Hedland with DNI of 2700 kWh/m2
and seawater temperature of 31 °C. It is necessary to mention that all
I.B. Askari et al.
the annual exergy streams (kWh/year) were determined by applying
the solar share of 20.50%. For instance, considering the net electricity
generation rate of 50 MWe, the total annual electricity that is
generated by solar thermal source is equal to 89.78 GWh
(50 MWe × 8760 × 0.205). As can be seen from Table 15, assuming
same electricity generation rate of 53.55 MWe for both reference and
dual purpose plants (Conf#D1 in Table 10), because the reference plant
of D1 (without desalination unit) requires lower thermal energy than
dual purpose plant of D1, the aperture area of the LF solar field and
consequently its capital cost for the reference plant is lower than those
for the dual purpose plant. That is way the solar field annualized costs
in the case of reference plant is approximately 58% lower than that in
the case of dual purpose plant as it is shown in Table 15. It can be also
seen from Table 15 that the COE of reference plant (0.1405$/kWh) is
approximately 13% lower than that of the dual purpose plant (0.1589$/
kWh). The difference between the COE of dual purpose and reference
plants were used to determine the thermal energy cost that is used in
the MED/TVC desalination unit as follows:
(LCOED1 − LCOERef,D1) × WGross = Cst4c ∙Et4c − Cst4e ∙Et4e (19)
where, WGross is the total electrical work (53.55 MWe) including the
required electricity that is consumed in the desalination unit and the
condenser pump of dual purpose plant (3.7 MWe). Also, Ct4cs and Ct4es
are respectively the costs of inlet and outlet exergy streams into or from
the MED/TVC desalination unit as it is shown in Fig. 21. Based on Eq.
(19), the total annual electricity cost that is induced into the plant due
to integration of the desalination unit is obtained as 9.24 × 106$/year.
Therefore, the cost of inlet and outlet exergy streams into or from the
55
Desalination 425 (2018) 37–67
Fig. 20. Block diagram of exergy flows between
different components of Conf#B1.
MED/TVC desalination unit is obtained as 0.034$/kWhex as it is shown
in Table 15. As it is clear from Table 15, the exergy efficiency of re-
ference and dual purpose plants (D1), which were obtained for Port-
Hedland at a typical hour with maximum solar radiation level, are
equal to 29.07% and 17.12%, respectively. The exergy efficiency dif-
ference is occurred due to decreasing of the dual purpose plant (D1)
electricity generation rate which is caused by integration of MED/TVC
unit into the plant. In fact, part of the steam in the high pressure steam
turbine of the dual purpose plant is not used to produce the electricity
and it is extracted to be used as the motive steam of the MED/TVC unit.
Table 15 also shows that the fresh water exergy unit of cost is equal to
1.84$/kWhex for configuration D1. Considering the total annual che-
mical exergy of the fresh water that is produced by solar field
(2.66 × 106 kWhex/year), the total annual exergy cost of the fresh
water is obtained as equal to 4.89 × 106$/year (1.84$/
kWhex × 2.66 × 106 kWhex/year). Finally, the unit cost of fresh water
(1.71$/m3 in Table 15) is obtained by dividing the total annual exergy
cost of the fresh water by total annual fresh water that is produced by
LF solar field thermal source which is equal to 2.85 × 106 m3/year
(38,100 m3/day × 365 × 0.205). Table 16 shows that the unit cost of
exergy stream that is entered (or exit) into (or from) the MED/TVC unit
of configuration D2 (0.053$/kWhex) is approximately 57% higher than
that of configuration D1 (0.034$/kWhex, in Table 15). That is why the
fresh water unit of cost for configuration D1 (dual purpose plant) is
lower than that for configuration D2 (2.06$/m3). In contrast to con-
figuration D1, which has two products of electricity and fresh water
with total hourly exergy value of 55.03 MWex (53.55 MWe
+ 1.48 MWex), the only product of configuration D2 is fresh water with
I.B. Askari et al.
a low hourly exergy value of 1.48 MWex. Therefore, applying Eq. (4),
the exergy efficiency of dual purpose plant (D1) is obtained con-
siderably higher than that of configuration D2 with the exergy effi-
ciency of 1.37% as it is shown in Table 16.
The exergy efficiency of all configurations and the reference plants,
which were calculated using Eq. (4), are shown in Fig. 24 for three
locations of the study at the hour of maximum radiation level. As it is
clear from Fig. 24, the reference plants have the higher exergy effi-
ciencies as compared to dual purpose plants for three locations of the
study. The locations with higher solar radiation level have higher ex-
ergy efficiencies; so that the maximum and minimum exergy effi-
ciencies are related to the Port-Hedland and De-Almeria, respectively.
The lower DNI level the higher required aperture area in the solar field
and consequently the lower exergy efficiency based on Eq. (4). For
three locations of the study, configuration D1 with moving steam
pressure of 4540 kPa in the MED/TVC unit has the minimum exergy
efficiency among the dual purpose plants. Fig. 24 also shows that be-
cause the low exergy of produced fresh water, the exergy efficiency of
configurations B2, C2 and D2 are lower than that of the dual purpose
plants. In configuration B1, part of the thermal energy that is used in
the desalination process is cost free because the first low of thermo-
dynamic that implies part of the thermal energy is wasted due to
cooling up the plant. Therefore, the thermal energy costs of the desa-
lination process in configuration B1 is lower than that of in config-
urations C1 and C2. Therefore, configuration B1 with MED unit has
higher exergy efficiency as compared to configurations C1 and D1 with
56
Desalination 425 (2018) 37–67
Fig. 21. Block diagram of exergy flows between
different components of Confs#C1 or D1.
MED/TVC units. However, for configuration B2 (LF/MED), the solar
field required aperture area is higher than that of configurations C2 and
D2 because the lower GOR of MED unit as compared to the GOR of
MED/TVC units in configurations C2 and D2 (LF/MED/TVC). There-
fore, the exergy efficiency of configuration B2 would be obtained lower
than that of configurations C2 and D2.
The fresh water unit costs that are obtained assuming the applica-
tion of different configurations within three locations of the study are
shown in Fig. 25; considering the electricity consumptions of 1.5 kWh/
m3 for MED and MED/TVC units as well as 3.5 kWh/m3 and 5.5 kWh/
m3 for RO desalination unit. Generally, Port-Hedland and de-Almeria
with the maximum and minimum DNI levels, respectively, have the
minimum and maximum fresh water units of costs for all configura-
tions. The advantage of low seawater temperature in Mediterranean Sea
(de-Almeria), which causes to have higher GORs in the MED and MED/
TVC units, cannot compensate the disadvantage of lower solar radiation
level at this region. Therefore, the high solar radiation levels of Port-
Hedland and Persian Gulf results in lower fresh water productions costs
of all configurations as compared to the case when the systems are
located at Mediterranean Sea with lower DNI level. As it is shown in
Fig. 25, with the exception of Port-Hedland region, the COW of con-
figurations with MED or MED/TVC units are lower than that of the dual
purpose configuration of A with RO desalination unit. In fact, the higher
electricity efficiency of configuration A implies the lower required
thermal energy and consequently lower solar field aperture area for this
configuration as compared to other dual purpose plants. Therefore, the
I.B. Askari et al. Desalination 425 (2018) 37–67

Fig. 22. Block diagram of exergy flows between

different components of Confs#B2, C2 or D2.

water production cost of configuration A is low as compared to that of
configurationsB1, C1 and D1. Fig. 25 shows that unlike two other re-
gions of the study, for Port-Hedland the COW of configuration A is
higher than that of configuration B1. In order to illustrate the reason,
the unit cost of exergy streams that are exit from the LF solar field and
entered into the MED unit of configuration B1 (C1s and C4s, Table 14
and Fig. 20) are tabulated in Table 17 for three locations of the study.
As it can be seen from Table 17, the highest and lowest unit costs of
exergy streams are related to De-Almeria with lowest DNI level and
Port-Hedland with highest DNI, respectively. The lower unit cost of
exergy stream that is entered into the MED unit results in the lower
water production cost. It can be concluded that even if the electricity

Table 14
The exergy balance equations that were used in the calculations of exergy stream unit of costs.

System Exergy balance equation Auxiliary equations

LF solar field of Confs. A, B1 and B2 s

solarshare × 8760 × C11 ∙E11 + Z͠ LF = solarshare × 8760 × C1s∙E1 C1s = C11s
Power block, Conf#A solarshare × 8760 × C1s∙E1 + C5s ∙E5 + Cst4e ∙Et4e + Z͠ PB = Cs4 ∙E4 + Cst4c ∙Et4c + Csel ∙WPB C4s = 0
C4s = C5s
Desalination unit, Conf#A s
C14 s
∙E14 + Csel ∙WRO + Z͠ RO = C15 s
∙E15 + C16 ∙E16 C12s = C13s = C14s = 0
E16 = Efresh (Eq. (7))
Power block, Conf#B1 solarshare × 8760 × C1s∙E1 + C5s ∙E5 + Z͠ PB = Cs4 ∙E4 + Csel ∙WPB C4s = C5s
Desalination unit, Conf#B1 Cs4 ∙E4 + C12
s s
∙E12 + Csel ∙Wdes + Z͠ MED = C14 s
∙E14 + C13 ∙E13 + C5s ∙E5 + C15
s
∙E15 C12s = C13s = C14s = 0
E15 = Efresh (Eq. (7))
Power block, Conf#C1 solarshare × 8760 × C1s∙E1 + C5s ∙E5 + Cst4e ∙Et4e + Z͠ PB = Cs4 ∙E4 + Cst4c ∙Et4c + Csel ∙WPB Ct4es = 0
Ct4es = Ct5es
Desalination unit Conf#C1 Cst4c ∙Et4c + C12
s s
∙E12 + Csel ∙Wdes + Z͠ MED = C14 s
∙E14 + C13 ∙E13 + Cst4e ∙Et4e + C15
s
∙E15 C12s = C13s = C14s = 0
E15 = Efresh (Eq. (7))
LF solar field of Confs. B2, C2 and D2 solarshare × 8760 × Cs2 ∙E2 + Z͠ LF = solarshare × 8760 × C1s∙E1 C1s = C2s
Desalination units of Confs. B2, C2 and D2s C1s∙E1 + C3s ∙E3 + Csel ∙Wdes + Z͠ MED = Cs2 ∙E2 + Cs4 ∙E4 + C5s ∙E5 + C6s ∙E6 C3s = C4s = C5s = 0
E6 = Efresh (Eq. (7))

57
I.B. Askari et al. Desalination 425 (2018) 37–67

0.4 26 4
2 o 3
DNI=2700kWh/m /yr, TSW=31 C, D=38100m /day
Exergy Efficiency 24 3.5
0.35 2 o
DNI=2300kWh/m /yr, TSW=36 C, D=34660m /day
3
LCOE (WNet)
2 o 3
LCOE (WGross) DNI=1990kWh/m /yr, TSW=26 C, D=40970m /day
22 3
0.3

Exergy efficiency (%)

3
5.5kWh/m
20 2.5

COW($/m3)

3.5kWh/m3
COE ($/kWh)

0.25
18 2
0.2
16 1.5
0.15
14 1

0.1 0.5
12

0.05 10 0
0 A A B1 C1 D1 D2 C2 B2
Plant
0 8
A B1 C1 D1
Fig. 25. Water unit cost of the plants for three locations of the study.
Plant
Fig. 23. Exergy efficiency and electricity unit costs of the plants for Persian Gulf. efficiency of configuration A with RO unit would be higher than that of
configuration B1 with MED unit, the high radiation level at a region can
be reduces the thermal energy costs and consequently water production
cost of configuration B1 to the values lower than that of configuration
A. Fig. 25 also shows that the COW of configurations B2, C2 and D2 are
40 higher than that of dual purpose configurations. In fact, the application
2 o
DNI=2700kWh/m /yr, TSW=31 C, D=38100m /day
3
of solar field as the thermal source of MED or MED/TVC units to pro-
35 2 o
DNI=2300kWh/m /yr, TSW=36 C, D=34660m /day
3
duce only fresh water results in high COW values because of the high
2 o 3
30
DNI=1990kWh/m /yr, TSW=26 C, D=40970m /day exergy value of solar thermal source and low exergy value of produced
Exergy efficiency(%)

fresh water.
25 Fig. 26 shows the unit cost of electricity generation for dual purpose
plants as well as the unit electricity costs of the reference plants that are
20 used in the calculations of COW of configurations B2, C2 and D2 when
they are located at three regions of the study. For configurations B2, C2
15
and D2, it was assumed that the electricity, which is generated by the
10 reference plant (configuration A without RO desalination unit) at the
condenser cooling temperature of 8 °C higher than the seawater tem-
5 perature, is used in the MED and MED/TVC units. As it is clear from
Fig. 26, the electricity generation cost of reference plants that are used
0 in the COW of configurations B2, C2 and D2 are lower than that of dual
Ref. A B1 C1 D1 D2 C2 B2

Plant purpose configurations of A, B1, C1 and D1. Generally, configuration A

with low electricity consumption of 3.5 kWh/m3 in RO unit has the
Fig. 24. Exergy efficiency of the plants for three locations at the hour with maximum DNI
lowest electricity unit cost among dual purpose configurations. The
level.
increasing of electricity consumption in RO unit of configuration A from
3.5 kWh/m3 to 5.5 kWh/m3 results in increasing the COE of config-
uration A to the values higher than that of configuration B1 for three

Table 15
The results of exergo-economic analysis of Conf#D1, DNI = 2700 kWh/m2/year, TSW = 31 °C (port-Hedland).

Number Stream Conf#D1 Reference plant without desalination unit

E (kWh/year) c ($/kWh) E (kWh/year) cs ($/kWh)

1 Solar field output 256.96 × 106 0.044 715.10 × 106 0.034

11 Solar field input 1167.29 × 106 0.044 161.70 × 106 0.034
12 Total electricity generation (by solar source) 95.52 × 106 0.158 502.50 × 106 0.1374
12 Net electricity generation (by solar source) 89.80 × 106 0.180 – 0.1405
t4c MED/TVC input 260.49 × 106 0.034 – –
t4e MED/TVC output 1.91 × 106 0.034 – –
Wdes MED/TVC electricity consumption 4.27 × 106 0.1405 – –
15 Fresh water 2.66 × 106 1.84 (1.71$/m3) – –

Costs Conf#D1 Reference plant without desalination unit

Solar field annualized costs ($/year) 8.28 × 10 6

4.87 × 106
MED/TVC annualized costs ($/year) 11.37 × 106 –
Power Block annualized costs ($/year) 7.97 × 106 7.97 × 106
Exergy efficiency (%) 17.12 29.07

58
I.B. Askari et al. Desalination 425 (2018) 37–67

Table 16
The results of exergo-economic analysis of Conf#D2, DNI = 2700 kWh/m2/year, TSW = 31 °C (port-Hedland).

Number Stream Conf#D2

E (kW/ cs ($/kWh)
year)

1 Solar field output 261.9- 0.053

0 × 106
2 Solar field input 2.0- 0.053
2 × 106
Wdes MED/TVC 4.2- 0.1405
electricity 7 × 106
consumption
6 Fresh water 2.6- 2.23 (2.06$/m3)
6 × 106

Costs Conf#D2

Solar field annualized costs ($/year) 2.92 × 106 –

MED/TVC annualized costs ($/year) 11.37 × 106 –
Exergy efficiency (%) 1.37 –

Table 17 0.75
The unit cost of exergy streams that are entered into the MED or exited from solar field of 0.7 2 o
DNI=2700kWh/m /yr, TSW=31 C, D=38100m /day
3

Conf#B1.
Solar field apperture area(km ) 0.65 2 o 3
DNI=2300kWh/m /yr, TSW=36 C, D=34660m /day
2

0.6 2 o 3
DNI=1990kWh/m /yr, TSW=26 C, D=40970m /day
Location The unit cost of exergy stream The unit cost of exergy stream 0.55
that is entered into the MED that is exited from the solar 0.5
unit (C4s) ($/kWhex) field (C1s) ($/kWhex)

5.5kWh/m3
3.5kWh/m3
0.45
Port-Hedland 0.027 0.043 0.4
Kish Island 0.044 0.054 0.35
De_Almeria 0.056 0.069 0.3
0.25
0.2
0.15
0.35 0.1
2 o
DNI=2700kWh/m /yr, TSW=31 C, D=38100m /day
3 0.05
0.3 2 o 3 0
DNI=2300kWh/m /yr, TSW=36 C, D=34660m /day 0 A A B1 C1 D1 D2 C2 B2
DNI=1990kWh/m2/yr, TSW=26oC, D=40970m /day
3
Plant
0.25
3

5.5kWh/m3
3.5kWh/m

Ref. cycle LCOEs unsed in Fig. 27. Required solar field aperture areas for different configurations located at three
COE($/kWh)

LCOW calculations regions of the study.

0.2

0.15
configuration A should be higher than that of configuration B1 as it is
0.1 shown in Table 10. The higher electricity generations rate of config-
uration A (5.5 kWh/m3) causes to increase the capital cost of Rankine
0.05 cycle (power block) of that plant to the higher values than that of
configuration B1. In consequence, in despite of lower capital cost of
0
0 A A B1 C1 D1 D2 C2 B2 solar field for configuration A as compared to configuration B1, the sum
Plant of solar field and Rankine cycle capital costs for the configuration A
(5.5 kWh/m3) would be obtained higher than that for configuration B1.
Fig. 26. Water unit cost of the plants for three locations of the study.
That is why the COE of configuration B1 is obtained as lower than that
of configuration A with electricity consumption of 5.5 kWh/m3 in RO
unit. It is worth to mention that the decreasing of Rankine cycle capital
cost may changes the results. Also, these results can be different for the
locations of the study. Since the electrical efficiency of configuration A configurations with PTC solar field due to the higher capital cost of PTC
(5.5 kWh/m3) is higher than that of configuration B1, the required solar field as compared to the LF solar field. In such a case, the sum of
aperture area and consequently capital cost of solar field for config- capital costs of the solar field and Rankine cycle of the plant with lower
uration A (5.5 kWh/m3) is lower than that for configuration B1. How- solar field aperture area can be lower than that of the plant with higher
ever, because of higher electricity consumption of RO unit in config- solar field aperture area.
uration A (5.5 kWh/m3), in order to obtain a constant net electricity For three locations of the study with different DNI level, the re-
generation rate of 50 MWe, the electricity generation rate of quired aperture areas of solar fields were calculated for all

59
I.B. Askari et al. Desalination 425 (2018) 37–67

0.25 4.5

0.225 LF Capital cost=180$/m2 MED Capital cost=1300$/m2

2
4
LF Capital cost=144$/m MED Capital cost=1593$/m2
0.2 LF Capital cost=108$/m2 2
3.5 MED Capital cost=1800$/m
0.175
3
COE($/kWh)

COW($/m3)
0.15
2.5
0.125
3.5kWh/m3

2
0.1

0.075 1.5

0.05 1

0.025 0.5
0
Ref A B1 C1 D1 0
B1 C1 D1 D2 C2 B2
Plant
Plant
Fig. 28. The effect of solar field capital cost on the COE of dual purpose plants for Persian
Gulf. Fig. 30. The effect of solar field capital cost on the COW of dual purpose plants for
Persian Gulf-MED capital cost = 1465$/m3/day.

4.5 (Mediterranean Sea) with DNI levels of respectively 17% and 35%
2 lower than that of Port-Hedland in Australia, the required solar field
LF Capital cost=180$/m aperture areas of dual purpose configurations (B1, C1 and D1) are 21%
4
LF Capital cost=144$/m2 and 62% higher than that of dual purpose configurations located in
2
3.5 LF Capital cost=108$/m Port-Hedland region, respectively.

3 5.4.1. The effect of solar field and MED capital costs on COW and COE of
COW($/m )

the plants located in Persian Gulf region

3

2.5 It has been shown by the authors of the present study [20] that the
COW of the systems with solar field thermal source is highly sensitive to
2 the capital cost of solar field. In this part, the COWs of all configurations
of the study were calculated for three different scenarios of solar field
1.5 capital cost when they are located in Persian Gulf region with seawater
temperature of 31 °C and DNI level of 2300 kWh/m2/year. Fig. 28
shows the effect of decreasing capital cost of LF solar field on the COE of
3

1
3.5kWh/m

0.5
0 COE of the study configurations by about 8.16% and 16.17%, respec-
A B1 C1 D1 D2
Plant
Fig. 29. The effect of solar field capital cost on the COW of dual purpose plants for
Persian Gulf-MED capital cost = 1465$/m3/day.
configurations (Fig. 27). As it was mentioned formerly, it was assumed
that the solar fields supply 20.50% of total annual required thermal
energy of the plants. It is evident that the locations with high solar
radiation level are those with low required aperture areas in the solar
field of the plants. The results of Fig. 27 show that for all three locations
of the study the solar field required aperture area of configurations B2,
C2 and D2 are approximately equal to 65%, 46% and 38% of the solar
field aperture areas of configurations B1, C1 and D1, respectively. In
fact, the low GOR of MED unit in configuration B2 causes to increase
the required solar field aperture area of this configuration as compared
to configurations C2 and D2 with higher GORs in MED/TVC units.
Fig. 27 also shows that for the Kish Island (Persian Gulf) and de-Almeria
reference and dual purpose plants for Persian Gulf region. The results of
Fig. 28 reveal that the decreasing of solar field capital cost to 80% and
60% of its first cost assumption (180$/m2) results in decreasing the
C2 B2 tively. The percentage decrease in COE of dual purpose configurations
is low as compared to the percentage decrease in capital cost of the LF
solar field. The reason is that the solar field capital cost is one of the
parameters that affect the COE of the plants, and COE is depended on
the other cost parameters such as capital cost of Rankine cycle and so
forth. Generally, it can be concluded that each 10% decrease in solar
field capital cost results in 4% decrease in the COE of dual purpose
plants when they are located in Persian Gulf.
Similar calculations were done to determine the effect of solar field
capital cost on COW of dual purpose and single purpose configurations
when they are located in Persian Gulf region. As can be seen from
Fig. 29, by decreasing of the solar field capital cost to 60% of its first
cost assumption, the COW of configurations A, B1, C1, D1, D2, C2 and
B2 would be decreased by about 5.06%, 9.42%, 14.67%, 18.93%,
21.47%, 22.10% and 24.55%, respectively. As it is clear, configurations
A (LFRC/RO) and B2 (LF/MED) with, respectively, maximum and
minimum exergy efficiencies have the minimum and maximum per-
centages decrease among the other configurations, respectively.

60
I.B. Askari et al. Desalination 425 (2018) 37–67

7 7
D1-NGB heat source B2-NGB heat source
6 D1-Solar heat source 6 B2-Solar heat source
C1-NGB heat source C2-NGB heat source
C1-Solar heat source C2-Solar heat source
5 B1-NGB heat source 5 D2-NGB heat source
B1-Solar heat source D2-Solar heat source

COW($/m )
COW($/m )

3
3

4 4

3 3

2 2

1 1

0 0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
3 3
Natural gas price ($/m ) Natural gas price ($/m )
Fig. 31. The effect of fuel price on COW of dual purpose plants for Persian Gulf-MED Fig. 32. The effect of fuel price on COW of single purpose plants for Persian Gulf-MED
capital cost = 1465$/m3/day. capital cost = 1465$/m3/day.

Because of the low GOR and exergy efficiency of configuration B2, the
sensitivity of COW of this configuration to the capital cost of solar field
is higher than that of the other configurations. For dual purpose con-
figurations with high electrical efficiency and consequently low re-
quired thermal energy, the COW is less sensitive to the solar field ca-
pital cost.
The other effective cost parameter that can affect the COWs of the
study configurations is capital cost of the MED and MED/TVC units.
Fig. 30 shows the effect of decreasing the capital cost of MED unit on
COW of configurations with MED or MED/TVC units for Persian Gulf
region. As can be seen from Fig. 30, the capital cost of MED unit was
considered in three scenarios of 1800$/m3/day, 1465$/m3/day (first
cost assumptions for Persian Gulf in the calculations of present study)
and 1333$/m3/day. As it is shown in Fig. 30, by decreasing the capital
cost of MED from 1800$/m3/day to 1465$/m3/day and 1333$/m3/day
(23% and 35% decrease, respectively), the COW of configuration B1
would be decreased by about 4.51% and 7.78%, respectively. Also, for
configuration B2, 2.98% and 5.08% of decrease would be obtained by
decreasing the capital cost of MED unit from 1800$/m3/day to 1465$/
m3/day and 1333$/m3/day, respectively. In fact, due to high cost of
source and solar field thermal source, the COWs of the plants with solar
field thermal sources are also shown in Figs. 31 and 32. As can be seen
from Fig. 31, because of the low natural gas fuel price in Persian Gulf
(0.02$/m3), the COW of fuel based plants are considerably lower than
that of the solar based plants. Fig. 31 also shows that assuming the
constant cost parameters at the fuel price of approximately equal to
0.23$/m3, the COW of the fuel based and solar based plants would be
equal to a same value. In the other words, for fuel price of 0.23$/m3,
the total electricity and thermal costs of the fuel based plants would be
equal to that of the solar based plants with LF solar field capital cost of
180$/m2. Also, as can be seen from Fig. 31, the slop of COW chart for
4
MED electrcicty need (2.5kWh/m3)
3.5 MED electrcicty need (1.5kWh/m3)
3
2.5
COW($/m3)

exergy stream that is flowed into the MED unit of configuration B2,
which causes this configuration has the minimum exergy efficiency and 2
maximum COW among all configurations, the COW of configuration B2
is more affected by the thermal energy cost rather than the capital cost
1.5
of MED unit. That is why the COW of configuration B1 has the low
decrease of 5.08% by decreasing the MED capital cost by about 35%.
Since nearly 79.5% of total annual required thermal loads of the 1
study plants were assumed to be supplied by natural gas boiler (NGB),
the fuel cost has an important role on the COWs of the study config- 0.5
urations with fuel thermal source. The COWs of the plants, which were
presented in former sections, were calculated based on the solar field 0
thermal source capital and other costs. In this part of the study, the B1 C1 D1 D2 C2 B2
effect of fuel cost on the COWs of configurations with MED or MED/
Plant
TVC unit were determined for dual purpose and single purpose con-
Fig. 33. The effect of MED electricity consumption on COW of plants for Persian Gulf-
figurations, respectively, as shown in Figs. 31 and 32. Also, in order to
MED capital cost = 1465$/m3/day.
make a comparison between the COWs of the plants with NGB thermal

61
I.B. Askari et al.
the plants with low exergy efficiencies is steeper than that with high
exergy efficiency. The low exergy efficiency means that a plant requires
high amount of thermal energy and fuel to produce a specific amount of
electricity and water. Therefore, the sensitivity of COWs of plants with
low exergy efficiencies to the fuel price is more than that with high
exergy efficiency. It is evident that at lower COWs for the solar based
plants, the COW of fuel based plants is equal to that of the solar based
plants at lower fuel prices and vice versa. That is why for single purpose
configurations of B2, C2 and D2 with higher COWs than dual purpose
plants, the COW of the systems with solar and fuel thermal sources are
equal at higher fuel prices as it is clear in Fig. 32. Based on Fig. 32 for
configurations of B2, C2 and D2, the COWs of configurations with solar
based and fuel based thermal sources would be equal at the fuel price of
approximately 0.31$/m3; which is approximately 35% higher than that
value that is obtained for dual purpose plants as it is shown in Fig. 31
(0.23$/m3). A comparison between Figs. 31 and 32 shows that the
slope of COW chart for dual purpose configurations in Fig. 31 is lower
than that for single purpose configurations in Fig. 32. This shows that
for configurations B2, C2 and D2, the sensitivity of COW to the fuel
price is higher than that for dual purpose plants of B1, C1 and D1. In
fact, the low exergy efficiency of single purpose configurations causes
to increase the sensitivity of their COWs to the fuel cost.
The effect of increasing the electricity consumptions of MED or
MED/TVC units on the COWs of dual purpose and single purpose
configurations are shown in Fig. 33. As it is shown in Fig. 33, the in-
creasing of electricity consumption from 1.5 kWh/m3 to 2.5 kWh/m3
would result in increasing the COWs of configurations B1, C1 and D1 by
about 11.72%, 11.04% and 9.53%, respectively. Also, the results of
Fig. 33 show that for configurations B2, C2 and D2, respectively,
3.86%, 4.75% and 5.07% increase in the COWs of these plants would be
obtained if the electricity consumption of desalination units is increased
from 1.5 kWh/m3 to 2.5 kWh/m3. The lower percentages increase in
COWs of single purpose plants as compared to dual purpose plants is
because of the higher thermal energy costs that are used in the desa-
lination units of single purpose plants. In fact, the thermal energy costs
and electricity costs are two factors that affect the COWs of desalination
plants. Therefore, the higher thermal energy costs of desalination pro-
cess decrease the impact of electricity costs in the COW of plant. That is
why in single purpose configurations with high thermal energy costs of
desalination process, the COWs of the systems are less sensitive to the
electricity consumption of the MED units as compared to dual purpose
plants.
6. Conclusion
In the present paper, the energy and exergy analysis of different
desalination technologies were performed. The effect of integration of
RO, MED and MED/TVC desalination units into the solar Rankine cycles
on the unit costs of water (COW) and electricity (COE) of dual purpose
plants were considered. The COWs for dual purpose plants (LFRC/RO,
LFRC/MED and LFRC/MED/TVC) were compared with that for the
single purpose configurations (LF/MED, LF/MED/TVC). The GORs of
MED and MED/TVC desalination units were determined for the loca-
tions with different seawater temperatures. Also, the effect of solar DNI
level on the COWs of all configurations was determined considering
three different DNI levels for three locations in the calculations of the
study. Finally, the effect of fuel price on the COWs of different con-
figurations was determined when they are located in Persian Gulf re-
gion. All dual purpose plants were considered to have a constant net
electricity rate of 50 MWe and for different water production rates
depending on the seawater temperatures of the study regions. The
62
Desalination 425 (2018) 37–67
following results were derived from the present work:
• At a same surface area for MED and MED/TVC evaporators and
considering the net electricity generation rate of 50 MWe, the water
production rate of dual purpose plants for the locations with sea-
water temperatures of 26 °C, 31 °C and 36 °C are obtained as
40,970 m3/day, 38,100 m3/day and 34,460 m3/day, respectively.
• The integration of RO unit into the solar Rankine cycle (LFRC/RO)
entails the higher electrical efficiency as compared to the dual
purpose plants with MED or MED/TVC units. Only in the specific
conditions when the once through cooling temperatures of LFRC/RO
plant are high and the electricity consumptions of RO unit is high as
the result of high salinity of seawater, the electrical efficiency of the
LFRC/RO and LFRC/MED plants approach to a same value.
• The results of exergy analysis show that for dual purpose plants, the
LF solar field causes the highest percentage of exergy destruction
among the other components of the plants. The portion of exergy
destruction for condenser of LFRC/RO plant was obtained as < 1%.
However, for LFRC/MED and LFRC/MED/TVC the percentages of
exergy destructions of MED and MED/TVC units were determined as
5% and 7%, respectively. Because of the steam extraction at high
temperature of 70 °C in configuration LFRC/MED, the percentage of
exergy destruction for the reheat steam turbine of this plant was
obtained as 8% which is twofold of that of the LFRC/MED/TVC
plants.
• Because the low exergy value of fresh water as compared to the
electricity, the exergy efficiency of single purpose configuration is
considerably low as compared to that of dual purpose plants with
two products of electricity and fresh water. Consequently, the unit
costs of fresh water for single purpose systems are higher than that
for the dual purpose plants.
• At the regions with high solar radiation levels (2700 kWh/m2/year),
the sum of solar field and Rankine cycle capital costs for the LFRC/
MED plant is obtained as lower than that for the LFRC/RO plant
with electricity consumption of 5.5 kWh/m3 of fresh water in RO
unit. Therefore, the unit cost of fresh water for LFRC/MED plant is
lower than that for LFRC/RO plant. Also, the high solar radiation
level results in low thermal energy costs that is required in the MED
unit of LFRC/MED plant, and in consequence, the electricity unit
cost of this configuration is obtained as lower than that of LFRC/RO
plant for the regions with high solar radiation levels.
• The effect of seawater temperature on the unit cost of fresh water is
lower than that of the solar DNI level. Therefore, the locations with
higher solar radiation levels are those with lower unit costs of fresh
water for the considered configurations.
• For dual purpose and single purpose configurations that are used in
Persian Gulf, when the fuel cost is increased from 0.02$/m3 to
0.23$/m3 and 0.31$/m3, respectively, the COW of plants with fuel
based or solar based thermal sources is equal. Otherwise, the COWs
of the fuel based plants are considerably lower than that of the solar
based plants.
• For single purpose configurations, because of the high thermal en-
ergy costs, the COW of these configurations are less sensitive to the
capital cost of MED unit and electricity consumption rate of desa-
lination process as compared to the dual purpose plants. The in-
creasing of MED electricity consumption from 1.5 kWh/m3 to
2.5 kWh/m3 results in increasing the COW of dual purpose plant of
LFRC/MED by about 11.72%. However for single purpose config-
uration of LF/MED, only 5.07% increase in the COW would be ob-
tained by increasing the MED electricity consumption from
1.5 kWh/m3 to 2.5 kWh/m3.
 Appendix A

Table A1
Energy conservation, exergy balance and exergy efficiency equations used in modeling of configuration A (LFRC/RO) shown in Fig. 2 [39,40].
I.B. Askari et al.

Parameter Description

Solar field
Q USF = ṁ SF × (h1 − h11) = ηsf × Qs (a1) Solar field useful thermal energy ηsf (solar field thermal
efficiency)
Qs = Afield ∙ DNI (a2) Solar field incident thermal energy, Afield: solar field
aperture area
̇
0 = (E11 − E1) + Ein,SF − Idestroyed (a3) Solar field exergy balance equation
4 (a4) The exergy, i.e. maximum available power output in W/m2
4 T T
⎜ ⎟
Ein,SF = Qs × ⎜⎛1 − ⎛ 0 ⎞ + ⎛ 0 ⎞ ⎟⎞ ⎜ ⎟
of the collector's total mirror area, for an LF at a certain
⎝ 3 T T
⎝ sun ⎠ ⎝ sun ⎠ ⎠
hour of the day
̇
Idestroyed (E1 − E11) (a5) Solar field exergy efficiency
ηII = 1 − =
Ein,SF Ein,SF

Regenerator heat exchanger

ZReheater = ((1 − y8 − y6)∗ (h3 − h2))/(ηhex ∗ (h1 − h11)) (a6) Re-heater energy conservation, ZReheater (re-heater mass
flow rate)
̇
0 = ZReheater × (e1 − e11) + (1 − (y8 + y6)) × (e2 − e3) − Idestroyed (a7) Re-heater exergy balance
̇
Idestroyed (1 − (y8 + y6)) × (e3 − e2 ) (a8) Re-heater exergy efficiency
ηII = 1 − =

63
ZReheater × (e1 − e11) ZReheater × (e1 − e11)
High pressure turbine
WH. P. T. = (h1 − ht8) + (1 − y8) × (ht8 − h2) (a9) High pressure turbine energy conservation
̇
WH.P.T. = (e1 − et8) + (1 − y8) × (et8 − e2) − Idestroyed (a10) High pressure turbine exergy balance
̇
Idestroyed WH.P.T. (a11) High pressure turbine exergy efficiency
ηII = 1 − =
(e1 − et8) + (1 − y8) × (et8 − e2 ) (e1 − et8) + (1 − y8) × (et8 − e2 )

Reheat turbine
WR. H. T. = (1 − y8 − y6) × (h3 − ht5) + (1 − y8 − y6 − y5) × (ht5 − ht4) + (1 − y8 − y6 − y5 − y4) × (ht4 − ht3) (a12) Reheat turbine energy conservation
+ (1 − y8 − y6 − y5 − y4 − y3) × (ht3 − ht2) + (1 − y8 − y6 − y5 − y4 − y3 − y2) × (ht2 − h4)
WR. H. T. = (1 − y8 − y6) × (e3 − et5) + (1 − y8 − y6 − y5) × (et5 − et4) + (1 − y8 − y6 − y5 − y4) × (et4 − et3) (a13) Reheat turbine exergy balance
+ (1 − y8 − y6 − y5 − y4 − y3) × (et3 − et2) + (1 − y8 − y6 − y5 − y4 − y3 − y2) × (et2 − e4) − T0 ∙ Sgen
̇
Idestroyed = T0 ∙Sgen = T0 × [(1 − y8 − y6) × (st5 − s3) + (1 − y8 − y6 − y5) × (st4 − st5) + (1 − y8 − y6 − y5 − y4) × (st3 − st4) (a14) Exergy destruction
+ (1 − y8 − y6 − y5 − y4 − y3) × (st2 − st3) + (1 − y8 − y6 − y5 − y4 − y3 − y2) × (s4 − st2)]
WR.H.T. (a15) Exergy efficiency
ηII =
̇
WR.H.T. + Idestroyed

Pump#1
WPump # 1 = zf × v6 × (P7 − P6) (a16) Energy balance
zf = 1 − (y5 + y6 + y8)
− WPump # 1 = zf × (e6 − e7) − T0 × Sgen (a17) Exergy balance
̇
Idestroyed z × (e7 − e6 ) (a18) Exergy efficiency
ηII = 1 − = f
WPump#1 WPump#1
Desalination 425 (2018) 37–67
 Pump#2
WPump # 2 = (h10 − h9) (a19) Energy balance
− WPump # 2 = (e9 − e10) − T0 ∙ Sgen (a20) Exergy balance
̇
Idestroyed (e − e9 ) (a21) Exergy efficiency
I.B. Askari et al.

ηII = 1 − = 10
WPump#2 WPump#2

Mixing chamber
z1 × h5 + (y2 + y3 + y4) × hd2 = zf × h6 (a22) Energy balance
z1 = 1 − (y2 + y3 + y4 + y5 + y6 + y8)
0 = z1 × e5 + (y2 + y3 + y4) × ed2 − z × e6 − T0 ∙ Sgen (a23) Exergy balance
z × e6 (a24) Exergy efficiency
ηII = z1 × e5 + (y2 + y3 + y4) × ed2

Feed water heater (FWH#2)

y2 = (zf × (hf2 − h7) − (y3 + y4) × (hd3 − hd2))/(ht2 − hd2) (a25) Energy balance
0 = y2 × (et2 − ed2) + (y3 + y4) × (ed3 − ed2) − zf × (ef2 − e7) − T0 ∙ Sgen (a26) Exergy balance
z f × e f2 + (y2 + y3 + y4) × ed2 (a27) Exergy efficiency
ηII =
z f × e7 + y2 × et2 + (y3 + y4) × ed3

Feed water heater (FWH#3)

y3 = (zf × (hf3 − hf2) − y4 × (hd4 − hd3))/(ht3 − hd3) (a28) Energy balance
0 = y3 × (et3 − ed3) + y4 × (ed4 − ed3) − zf × (ef3 − ef2) − T0 ∙ Sgen (a29) Exergy balance
z f × e f3 + (y3 + y4) × ed3 (a30) Exergy efficiency
ηII =
z f × e f2 + y3 × et3 + y4 × ed4

Feed water heater (FWH#4)

64
y4 = (zf × (h8 − hf3))/(ht4 − hd4) (a31) Energy balance
0 = y4 × (et4 − ed4) − zf × (e8 − ef3) − T0 ∙ Sgen (a32) Exergy balance
z f × e8 + y4 × ed4 (a33) Exergy efficiency
ηII =
z f × e f3 + y4 × et4

Feed water heater (FWH#6)

y6 = ((hf6 − h10) − (y8) × (hd8 − hd6))/(ht6 − hd6) (a34) Energy balance
0 = y6 × (et6 − ed6) + y8 × (ed8 − ed6) − (ef6 − e10) − T0 ∙ Sgen (a35) Exergy balance
e f6 + (y6 + y8) × ed6 (a36) Exergy efficiency
ηII =
e10 + y6 × et6 + y8 × ed8

Feed water heater (FWH#8)

y8 = (h11 − hf6)/(ht8 − hd8) (a37) Energy balance
0 = y8 × (et8 − ed8) − (e11 − ef6) − T0 ∙ Sgen (a38) Exergy balance
e11 + y8 × ed8 (a39) Exergy efficiency
ηII =
e f6 + y8 × et8

De-aerator
y5 = ((h9 − h8) − (y6 + y8) × (hd6 − h8))/(ht5 − h8) (a40) Energy balance
0 = y5 × (et5 − e8) + (y6 + y8) × (ed6 − e8) − (e9 − e8) − T0 ∙ Sgen (a41) Exergy balance
e9 (a42) Exergy efficiency
ηII =
z f × e8 + y5 × et5 + (y6 + y8) × ed6

Condenser
Qout = z1 × (h5 − h4) (a43) Energy balance
Desalination 425 (2018) 37–67
 T (a44) Exergy balance
⎜ ⎟
= z1 × (e4 − e5) + Qout × ⎛1 − 0 ⎞ − T0 ∙Sgen
⎝ T12 ⎠

T (a45) Exergy destruction

̇
Idestroyed ⎜
= T0 ∙Sgen = T0 × [z1 × (s5 − s4)] − Qout × ⎛1 − 0 ⎞ ⎟
I.B. Askari et al.

⎝ T12 ⎠
̇
Idestroyed (a46) Exergy efficiency
ηII = 1 −
z1 × (e4 − e5 )
RO
0 = ṁ 14 × h14 + WRO − (1 − Wrr) × ṁ 15 × h15 − Wrr × ṁ 14 × h16 (a47) Energy balance, Wrr (Water Recovery Ratio = 45%, [33]),
WRO (RO electricity consumption)
0 = ṁ 14 × e14 + WRO − (1 − Wrr) × ṁ 14 × e15 − Wrr × ṁ 14 × e16 − T0 ∙Sgen (a48) Exergy balance
ṁ 16 × e fresh (a49) Exergy efficiency, ṁ 16 × e fresh (Eq. (7))
ηII =
ṁ 14 × e14 + WRO

Table A2
Energy conservation, exergy balance and exergy efficiency equations used in modeling of MED unit in Configuration B1 (LFRC/MED) (Fig. 3).

MED

0 = z1 × (h5 − h 4) + ṁ 12 × h12 + WMED − ṁ 14 × h14 − ṁ 13 × h13 − ṁ 15 × h15 (a50) Energy balance, WMED (MED electricity consumption)
0 = z1 × (e4 − e5) + ṁ 12 × e12 − ṁ 14 × e14 − ṁ 13 × e13 − ṁ 15 × e15 + WMED − T0 ∙Sgen (a51) Exergy balance
ṁ 15 × e fresh (a52) Exergy efficiency, ṁ 15 × e fresh (Eq. (7))
ηII =

65
z1 × (e4 − e5) + Wdes + M12 × e12

Table A3
Some of the parameters of the MED/TVC system [3,9,11,36].

Parameter Description

Tj + 1 = Tj − ΔT (a53) Temperature at each effect (j = 1 to n − 1)

T − Tn (a54) The temperature drop in each effect
∆T = 1
n−1
Tvn = Tn − BPE − NEA (a55) Saturation temperature corresponding to the pressure in the effect
BPE ≅ 0.5 (a56) The Boiling Point Elevation (BPE) is the increase in the boiling temperature at a given
pressure due to the dissolved salts in the water
NEA ≅ 0.3 (a57) The non-equilibrium allowance (NEA) for effects 2 − n, which is a measure for the efficiency
of the flashing process.
F (a58) Feed seawater flow rate F is distributed equally to all effects
Fn = n
B1 = F1 − D1 (a59) The brine leaving the first effect
j=n (a60) The brine leaving the second effect to the last effect
Bj = ∑j = 1 Fj − Dj
F1 (a61) Salinity of brine leaving the first effect
Xb1 = F1 − D1
∙Xf
Xf ∙ Fj + 1 (a62) Salinity of brine leaving the jth effect
Xbj + 1 = j+1
∑j = 1 (Fj − Dj)
Desalination 425 (2018) 37–67
 Xf ∙Fn (a63) Brine salinity for the last effect
Xbn = n
∑ j = 1 (Fj − Dj )
C ∙ ΔT (a64) A small amount of vapor will be released by flashing when the condensed vapor (Dr) is
yi = λi
I.B. Askari et al.

introduced to the flashing box, which is equal to Dr. y

[(Ds + Dr ) ∙ (h d − hfd )] T − Tf1 ⎞ (a65) Vapor generated in the first effect (for the MED, Dr = 0)
D1 = − F1∙C∙ ⎛ 1 ⎜ ⎟

λ1 ⎝ λ1 ⎠
k−2 (a66) The vapor generated in the second effect to i − 1th effect (TVC location, Fig. 8):
⎡⎛ ⎛ ⎞ ⎞⎤ λ T − Tfk ⎞ C∙ΔT
Dk = ⎢ Dk − 1 + ⎜ ∑ Dk + Dr ⎟ ∙y − (k − 1) ∙Fk ∙y ⎥ ∙ k − 1 − Fk ∙C∙ ⎛ k + Bk − 1∙ ⎜ ⎟ (k = 2 to i − 1)
⎜ ⎟ λk λk
⎢
⎣⎝ ⎝ k = 1 ⎠ ⎠⎦⎥ ⎝ λk ⎠

λ i−1 T − Tfi ⎞ C∙ΔT (a67) The vapor generated in the first effect after TVC location (i)
Di = Df ∙ − Fi ∙C∙ ⎛ i ⎜ + Bi − 1∙⎟

λi ⎝ λi ⎠ λi
j−2 (a68) The vapor generated in the other effects after TVC location (i + 1 to n)
⎡⎛ ⎞ ⎤ λ j−1 Tj − Tfj ⎞ C∙ΔT
Dj = ⎢ Dj − 1 + − Fj∙C∙ ⎜⎛ ⎟ + Bj ∙
⎜ ∑ Dj∙y − (j − 1) ∙Fj∙y⎟ ⎥ ∙ λj λ j λj
⎢
⎣⎝ j=1 ⎠⎦⎥ ⎝ ⎠
[(Ds + Dr ) ∙ (h d − hfd )] (a69) The heat transfer area of first effect
A e1 = Ue1 ∙ (Td − T1)
3 (a70) The heat transfer area of the second effect to i − 1th effect (TVC location, Fig. 8)
[ (Dk − 2 + (∑kk −
= 1 Dk + Dr ) ∙ y − (k − 2) ∙ F k ∙ y) ] ∙ λk − 2
A ek − 1 = Uek − 1 ∙ (Tvk − 2 − T k − 1)
Df ∙λ i − 1 (a71) The heat transfer area of the first effect after TVC location (i)
A ei =
Uei ∙ (Tvi − 1 − Ti )
j−2 (a72) The heat transfer area for the other effects after TVC location (i + 1 to n)
⎡ Dj − 1 + ∑j = 1 Dj ∙y − (j − 1) ∙Fj∙y ⎤ ∙λ j − 1
( )
A ej = ⎣ ⎦
Uej∙ (Tvj − 1 − Tj)
1939.4 + 1.40562 ∙ Ti − 0.0207525 ∙ (Ti)2 + 0.0023186 ∙ (Ti)3 (a73) overall heat transfer coefficient of the effect

66
Uen = 1000
C ∙ ΔTfi ∙ ∑ik = 1 F k Tvi − Tfi + 1 (a74) Feed water preheater heat transfer areas i = 1,2, …, n − 1 (with the exception of the i − 1th
Afi = Uf ∙ (Tfi − Tfi + 1)
∙ln ⎡ ⎤
⎣ Tvi − Tfi ⎦ preheater)
Dn ∙ λ n (a75) Condenser heat transfer area
Ac = Uc ∙ (LMTD)c
(Tvn − Tf ) − (Tvn − Tc) (a76) Condenser logarithmic mean temperature difference
(LMTD)c = T −T
vn f⎤
ln ⎡
⎣ Tvn − Tc ⎦
Uc = 1.7194 + 3.2063 ∙ 10− 2 ∙ Tvn − 1.5971 ∙ 10− 5 ∙ (Tvn)2 + 1.9918 ∙ 10− 7 ∙ (Tvn)3 (a77) Overall heat transfer coefficient of the condenser
Dn ∙ λ n (a78) The cooling seawater flow rate
Mc = C ∙ (Tn − Tc)
λ = 2501.897149 − 2.407064037 × T + 1.192217 × 10− 3 × T2 − 1.5863 × 10− 5 × T3 (a79) Latent heat of evaporation (kJ/kg)
Pd (a80) The ratio of discharged pressure to entrained pressure (TVC unit)
Cr = Pr
Ps (a81) The ratio of the motive pressure to entrained pressure (TVC unit)
Er = Pr
Ds a3 a5 C a8 Cr C2 (a82) Entrained ratio (TVC unit). ai values are shown in Table A4
Ra = Dr
= a1 + a2∙Cr + Er
+ a 4∙Cr 2 + + a 6∙ Er + a7∙Cr 3 + + a 9∙ + a10∙ Er
Er 2 r Er 3 Er 2 r
D (a83) Gain output ratio of the desalination unit
GOR = Ds
i=n i=n−1 (a84) The specific heat transfer area
(∑i = 1 A e + A c + ∑i = 1 Afi)
SA = D
Desalination 425 (2018) 37–67
I.B. Askari et al.
Table A4
Constants of the polynomial model adopted for TVC performance [38].
Constant 10 ≤ Er ≤ 100
a1 − 3.20842211
a2 3.933353
a3 27.236
a4 − 1.19206949
a5 − 141.42328825
a6 − 22.54551842
a7 0.125813
a8 348.5066
a9 41.7961
a10 4.439929
References
[1] H.T. El-Dessouky, I. Alatiqi, S. Bingulac, H. Ettouney, Steady-state analysis of the
multiple effect evaporation desalination process, Chem. Eng. Technol. 21 (1988)
437–451.
[2] Sh. Gorjian, B. Ghobadian, T. Tavakkoli Hashjin, A. Banakar, Experimental per-
formance evaluation of a stand-alone point-focus parabolic solar still, Desalination
352 (2014) 1–17.
[3] A.O.B. Amer, Development and optimization of ME-TVC desalination system,
Desalination 249 (2009) 1315–1331.
[4] R. Kouhikamali, Thermodynamic analysis of feed water pre-heaters in multiple
effect distillation systems, Appl. Therm. Eng. 50 (2013) 1157–1163.
[5] R. Kouhikamali, M. Sanaei, M. Mehdizadeh, Process investigation of different lo-
cations of thermo-compressor suction in MED–TVC plants, Desalination 280 (2011)
134–138.
[6] P. Druetta, P. Aguirre, S. Mussati, Optimization of multi-effect evaporation desali-
nation plants, Desalination 311 (2013) 1–15.
[7] P. Palenzuela, A.S. Hassan, G. Zaragoza, D.C. Alarcón-Padilla, Steady state model
for multi-effect distillation case study: Plataforma Solar de Almería MED pilot plant,
Desalination 337 (2014) 31–42.
[8] M.T. Mazini, A. Yazdizadeh, M.H. Ramezani, Dynamic modeling of multi-effect
desalination with thermal vapor compressor plant, Desalination 353 (2014)
98–108.
[9] I.S. Al-Mutaz, I. Wazeer, Comparative performance evaluation of conventional
multi-effect evaporation desalination processes, Appl. Therm. Eng. 73 (2014)
1192–1201.
[10] I.S. Al-Mutaz, I. Wazeer, Development of a steady-state mathematical model for
MEE-TVC desalination plants, Desalination 351 (2014) 9–18.
[11] L. Shuai, Y. Zhou, M. Bi, J. Ren, Z. Cai, An integrated optimization model and
application of MEE-TVC desalination system, Desalination 371 (2015) 1–8.
[12] G. Fiorenza, V.K. Sharma, G. Braccio, Techno-economic evaluation of a solar
powered water desalination plant, Energy Convers. Manag. 44 (2003) 2217–2240.
[13] Fichtner, DLR, MENA regional water outlook, part ii, desalination using renewable
energy, task1: desalination potential; task2: energy requirements; task3: con-
centrate management, Available at: http://www.dlr.de/tt/Portaldata/41/
Resources/dokumente/institut/system/projects/MENA_regional_water_outlook.
pdf, .
[14] M.A. Sharaf, A.S. Nafey, L. García-Rodríguez, Exergy and thermo-economic ana-
lyses of a combined solar organic cycle with multi effect distillation (MED) desa-
lination process, Desalination 272 (2011) 135–147.
[15] G. Iaquaniello, A. Salladini, A. Mari, A.A. Mabrouk, H.E.S. Fath, Concentrating solar
power (CSP) system integrated with MED–RO hybrid desalination, Desalination 336
(2014) 121–128.
[16] M. Moser, F. Trieb, T. Fichter, J. Kern, D. Hess, A flexible techno-economic model
for the assessment of desalination plants driven by renewable energies, Desalin.
Water Treat. 55 (2014) 3091–3105.
[17] P. Palenzuela, G. Zaragoza, D.C.A. Padilla, J. Blanco, Large-scale solar desalination
by combination with CSP: techno-economic analysis of different options for the
Mediterranean Sea and the Gulf, Desalination 366 (2015) 130–138.
[18] B.O. Delgado, P. Palenzuela, D.C.A. Padilla, Parametric study of a multi-effect
distillation plant with thermal vapor compression for its integration into a Rankine
cycle power block, Desalination 394 (2016) 18–29.
[19] A. Tamburini, A. Cipollina, G. Micale, A. Piacentino, CHP (combined heat and
power) retrofit for a large MED-TVC (multiple effect distillation along with thermal
vapour compression) desalination plant: high efficiency assessment for different
design options under the current legislative EU framework, Energy 115 (2016)
1548–1559.
67
Desalination 425 (2018) 37–67
Er ≥ 100
−1.93422581
2.152524
113.4909
−0.52222106
−14,735.96533618
−31.85197010
0.047507
900,786
−495.58154134
10.02513
[20] I. Baniasad Askari, M. Ameri, The application of Linear Fresnel and Parabolic
Trough solar fields as thermal source to produce electricity and fresh water,
Desalination 415 (2017) 90–103.
[21] I. Baniasad Askari, M. Ameri, Techno economic feasibility analysis of Linear Fresnel
solar field as thermal source of the MED/TVC desalination system, Desalination 394
(2016) 1–17.
[22] F. Calise, M. Dentice d'Accadia, A. Macaluso, A. Piacentino, L. Vanoli, Exergetic and
exergo-economic analysis of a novel hybrid solar–geothermal polygeneration
system producing energy and water, Energy Convers. Manag. 115 (2016) 200–220.
[23] F. Calise, M. Dentice d'Accadia, A. Piacentino, Exergetic and exergo-economic
analysis of a renewable polygeneration system and viability study for small isolated
communities, Energy 92 (2015) 290–307.
[24] B.O. Delgado, L.G. Rodríguez, D.C.A. Padilla, Thermo-economic comparison of in-
tegrating seawater desalination processes in a concentrating solar power plant of
5 MWe, Desalination 392 (2016) 102–117.
[25] P. Catrini, A. Cipollina, G. Micale, A. Piacentino, A. Tamburini, Exergy analysis and
thermoeconomic cost accounting of a Combined Heat and Power steam cycle in-
tegrated with a Multi Effect Distillation-Thermal Vapour Compression desalination
plant, Energy Convers. Manag. 149 (2017) 950–965.
[26] S. Relloso, Y. Gutiérrez, Real application of molten salt thermal storage to obtain
high capacity factors in parabolic trough plants, Proceedings of 14th International
SolarPACES Symposium on Solar, Thermal Concentrating Technologies, Las Vegas,
USA, 2008.
[27] System Adviser Model (SAM), version 2015.6.30, PTR70 Parabolic Trough
Receiver, National Renewable Energy Laboratory, technical report no. NREL/TP-
550–45633, https://www.nrel.gov/analysis/sam/help/html-php/index.html?
linear_fresnel_system_costs.htm, (2009).
[28] M.J. Montes, A. Aba'nades, J.M. Martı'nez-Val, M. Valde's, Solar multiple optimi-
zation for a solar-only thermal power plant, using oil as heat transfer fluid in the
parabolic trough collectors, Sol. Energy 83 (2009) 2165–2176.
[29] US DOE, Concentrating Solar Power Commercial Application Study: Reducing
Water Consumption of Concentrating Solar Power Electricity Generation, (2009).
[30] R. Petela, Exergy analysis of the solar cylindrical-parabolic cooker, Sol. Energy 79
(2005) 221–233.
[31] K.S. Spiegler, Y.M. El-Sayed, The energetics of desalination process, Desalination
134 (2001) 109–128.
[32] K.H. Mistry, H.A. Hunter, J.H. Lienhard, Effect of composition and non-ideal so-
lution behavior on desalination calculation for mixed electrolyte solutions with
comparison to seawater, Desalination 318 (2013) 34–47.
[33] S. Loutatidou, H.A. Arafat, Techno-economic analysis of MED and RO desalination
powered by low-enthalpy geothermal energy, Desalination 365 (2015) 277–292.
[34] Technology data for energy plants, https://ens.dk/sites/ens.dk/files/Analyser/old_
technology_data_for_individual_heating_plants_and_energy_transport_aug2016.pdf,
(2016).
[35] M.A. Darwish, F. Al-Juwayhel, H.K. Abdulraheim, Multi-effect boiling systems from
an energy viewpoint, Desalination 194 (2006) 22–39.
[36] F.N. Alasfour, M.A. Darwish, A.O.B. Amer, Thermal analysis of ME-TVC + MEE
desalination systems, Desalination 174 (2005) 39–61.
[37] Seawater temperature, https://www.seatemperature.org/, (2017).
[38] A.S. Hassan, M.A. Darwish, Performance of thermal vapor compression,
Desalination 35 (2014) 41–46.
[39] V. Siva Reddy, S.C. Kaushik, S.K. Tyagi, Exergetic analysis of solar concentrator
aided natural gas fired combined cycle power plant, Renew. Energy 39 (2012)
114–125.
[40] S. Safarian, F. Aramoun, Energy and exergy assessments of modified Organic
Rankine Cycles (ORCs), Energy Rep. 1 (2015) 1–7.