Energy 210 (2020) 118559

Contents lists available at ScienceDirect

Energy
journal homepage: www.elsevier.com/locate/energy

Thermoeconomic analysis and multi-objective optimization of a novel

hybrid absorption/recompression refrigeration system
Amir Reza Razmi a, **, Ahmad Arabkoohsar b, *, Hossein Nami c
a
Department of Mechanical Engineering, K.N. Toosi University of Technology, Tehran, Iran
b
Department of Energy Technology, Aalborg University, Denmark
c
Department of Mechanical Engineering, Faculty of Engineering, University of Maragheh, P.O.Box 83111-55181, Maragheh, Iran

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

Article history: In the present article, thermodynamic, exergoeconomic, economic, and sustainability investigations of a
Received 25 April 2020 recently developed environmentally friendly hybrid absorption/recompression refrigeration cycle is
Received in revised form proposed to evaluate its feasibility for decision making and marketing. The proposed system is a novel
6 July 2020
hybridization of the conventional vapor compression and absorption cycles, wherein a booster
Accepted 28 July 2020
Available online 15 August 2020
compressor has been used between the generator and condenser of the single-effect absorption cycle to
improve its performance. Also, two separate multi-objective optimization models are developed using a
combination of the nondominated-storing-genetic algorithm (NSGA-II) and artificial neural network
Keywords:
Thermodynamic and thermoeconomic
(ANN) to address the optimum performance values concerning the objective functions. The obtained
analysis results approve that the proposed cycle is a promising concept from both thermodynamic and economic
Absorption/recompression refrigeration viewpoints. The results indicate that the system presents a coefficient of performance and exergy effi-
cycle ciency of 4.88 and 37.43% under the optimum working conditions. The overall rate of exergy destruction
Sustainability index of the system is 20.23 kW, and a sustainability index of around 1.53 can be achieved at a cooling capacity
Artificial neural network of 150 kW. The economic results indicate that the reference system has a payback period of 4.17 years,
Multi-objective optimization which is reduced to less than 4 years after doing the optimizations.
ANN
© 2020 Elsevier Ltd. All rights reserved.
NSGA-II

1. Introduction applications [6]. Vapor compression cycles are widely implemented

because of their lower initial cost and simpler structure compared
Energy and environmental crises, due to prompt development to absorption cycles [7,8]. Despite operating with a higher coeffi-
of commercial divisions and the global economy, have become cient of performance (COP), vapor compression cycles generally use
progressively severe in the last decades [1,2]. In this regard, several chlorofluorocarbon (CFC) and hydro-fluorocarbon (HFC) as the
agreements such as Paris climate accord and Montreal and Kyoto working fluid which is considered as their major drawback [9].
protocols have been ordained to reduce environmental pollutions, These fluids cause some irreversible environmental impacts by
and Ozone depletion potential (ODP), and global warming potential emitting Carbone Dioxide, Sulphur, and Nitrogen, resulting in an
(GWP) [3]. According to statistics, 40% of total global power pro- augmentation in GWP and ODP [10,11]. On the contrary, absorption
duction is consumed in the buildings, 60% of which is consumed for cycles utilize environmentally friendly working fluids, as well as
heating, ventilation, and air conditioning (HVAC) purposes [4]. the possibility to be driven by renewable energy resources and
Hence, the development of efficient and green cooling and heating industrial waste heat [12]. However, their lower COP is the most
systems appears to be extremely crucial [5]. significant negative point for their vast application [13].
Vapor compression and absorption cycles are the most common As discussed, both of the aforementioned conventional cooling
HVAC systems, both of which can be applied for cooling and heating systems have their own advantages and disadvantages. To over-
come their drawbacks, hybridization of absorption and compres-
sion systems has attracted attention in recent years [14,15]. Some of
the benefits of the hybrid absorption/recompression (AR) systems
* Corresponding author.
** Corresponding author.
are [16,17]: (a) Notable improvement in heating/cooling capacities;
E-mail addresses: [email protected] (A.R. Razmi), [email protected] (b) Extension of the operational range: i.e. operating at lower
(A. Arabkoohsar). evaporator and heat source temperatures in comparison to

https://doi.org/10.1016/j.energy.2020.118559
0360-5442/© 2020 Elsevier Ltd. All rights reserved.
2 A.R. Razmi et al. / Energy 210 (2020) 118559

Nomenclature GWP Global warming potential

HVAC Heating; ventilation; and air
conditioning
A Area ½m2  HXCH Heat exchanger
AR Reference area ½m2  NSGA
II Nondominated
storing
genetic algorithm
c Cost per exergy unit ½$=GJ NPV Net present value
C_ Cost rate ½$=hr ODP Ozone depletion potential
Cr Compression ratio
DP Depletion numberE Greek symbols
ex Specific exergy ½kJ=kg t Annual operating hour
_
Ex Exergy rate ½kW h Isentropic efficiency
f Exergoeconomic factor ½% hex Exergy efficiency
h Specific enthalpy ½kJ:kg
1  r Density ½kg:m
3 
i Interest rate ½% n Volume fraction
m_ Mass flowrate ½kg=s f Maintenance factorEffectiveness ½%
N Service year of the cycle ½year
P Pressure ½kPa Subscripts
0 Reference condition
Q_ Heat transfer rate ½kW
Abs Absorber
r Relative cost difference ½%
Com Compressor
R2 Coefficient of determination
D Destruction
SC Saving cost ½$
e Exit
SI Sustainability index
Ev Expansion valve
T Temperature ½ C
ex Exergy
tp Payback period ½year
Evp Evaporator
W _ Power consumption ½kW f Fuel
X LiBr concentration ½% Gen Generator
Y Net cash flow ½$ Gen
Con Compacted generator
condenser
Z Investment cost ½$ i Inlet
ZR Reference cost ½$ k kth component
Z_ Cost rate ½$=hr lean Lean solution
p Product
Abbreviations P Pump
AR Absorption=recompression rich Rich solution
ANN Artificial neural network s Isentropic
COP Coefficient of performance tot Total
CRF Capital recovery factor

traditional absorption systems. (c) An improvement in COP, exergy advantage of this configuration, efficient operation at low heat
efficiency, and primary energy ratio. (d) A reduction in the initial source temperatures, indicating that the required heat source
investment cost in comparison to absorption cycles. temperature is reduced from 125 to 130  C in the conventional NH3/
The first configuration is when the compressor comes between H2O absorption cycles to 110e115  C in the proposed AR cycle [21].
the absorber and the evaporator. This design is the best option for In the second configuration, i.e. when a compressor is located
performance improvement and application of traditional absorp- between the absorber and generator, the cycle is driven by elec-
tion cycles in the lower heat source and ambient temperatures. Wu tricity rather than a heat source. This arrangement was initially
et al. [18] theoretically compared various working fluids for this presented by Osenbruck [22]; then, the cycle was developed by
arrangement, concluding that the maximum performance is ach- Altenkirch [23]. The generator of this configuration is capable of
ieved by using NH3/LiNO3 working fluid. Then, they substantiated extracting heat from low temperature fluids, while the absorber has
their findings with an experimental setup, representing that a COP the potential of releasing the heat of the absorption process to a
of 1.37 can be achieved at the evaporator and heat source tem- high-temperature fluid. Some of the major profits of this configu-
peratures of
25  C and 130  C by utilizing NH3/H2O as working ration are [24]: (a) Easy adjustment of heating and cooling capac-
fluid in the reference system [19]. As a result, this configuration can ities by solution composition; (b) A reduction in the unavoidable
be considered as an applicable heat pump for cold climates. heat losses in various processes; (c) A reduction in discharge tem-
Eventually, they investigated the annual performance of their perature and pressure of the compressor because of the lower
proposed system from technical and economic points of view un- saturation pressure of solution in comparison to the pure
der various heat source temperatures and compression ratios to refrigerant.
achieve an optimum operating condition [20]. It was concluded The third configuration uses a compressor between the low- and
that the obtained heat is markedly increased by employing a high-pressure stages. To develop this configuration, Ayala et al.
booster compressor between the absorber and the evaporator. [25,26] proposed theoretical and experimental analyses using NH3/
Accordingly, in spite the additional budget for purchasing the vapor LiNO3 working fluid, representing that the cooling and heating
compressor, the total capital cost of the system was remarkably capacities of this configuration can be simply and effectively
decreased due to a reduction in the size of the system. The second modulated. Their results disclosed that the provided heat at the
experimental analysis was done by We et al. to reveal another compressor exit is high enough to run an absorption subcycle.
 A.R. Razmi et al. / Energy 210 (2020) 118559 3

Recently, Leonzio [27] used a mathematical model to improve the

calculation accuracy of this arrangement. Besides, Wu et al. [28]
developed this configuration by offering a new design with some
adjustments and analyzed it from the thermodynamic point of
view, reaching greater efficiency and more flexible heating/cooling
capacities, making it preferable for colder climates.
The last configuration of AR systems is formed when a booster
compressor is located between the condenser and generator,
resulting in a reduction in the generation pressure of the conven-
tional absorption cycles. Thus, the cycle operates at lower heat
source temperature and an inordinate amount of heat is released in
the condenser, which is worth to be recovered. The basic concept of
this configuration was introduced by Riffat et al. [29], who analyzed
the reference AR system with a gas-driven compressor from the
thermodynamic perspective using various working fluids. Razmi
et al. [30] developed this configuration to exploit the entire of the
generated heat in the condenser to heat the solution in the
generator instead of using an external heat source. Providing a
compacted generator/compressor not only improved the cycle ef-
ficiency notably, but it also eliminated the use of bulky condenser at
the conventional absorption cycles. Their findings revealed that the
proposed cycle is superior to the traditional absorption cycle from
the viewpoints of first and second thermodynamics laws. Further-
more, owing to the use of a clean working fluid, it is superior to the
vapor compression systems from the environmental viewpoint. Fig. 1. The outline of the hybrid absorption/recompression refrigeration cycle [30].
Recently, Babaei et al. [31] scrutinized the effect of nanoparticles
addition on the performance of this AR configuration. They exam- 4th design in Table 1) is outlined in Fig. 1. Lithium bromide-water
ined three types of nanoparticles, concluding that Fe nanoparticle is (LiBreH2O) is considered as the solution pair of the proposed cy-
an ideal choice for improving the cycle performance. Despite the cle because of its non-flammable, non-toxic, and cost effective
mentioned researches, economic analysis and optimization of this characteristics [30].
system, which can remarkably justify its practical feasibility and Superheated vapor with high pressure and temperature (point
marketing, have remained as the gaps of the literature. To address 2) passes through the condenser coil to exchange heat to the so-
these gaps, this study presents a thorough thermoeconomic lution in the compacted generator-condenser, converting to liquid
assessment and optimization of the AR cycle with a compressor after cooling down. Water as the refrigerant fluid reaches the
between the generator and condenser. This will facilitate a precise evaporator (point 4) after being expanded in the first expansion
evaluation of system performance and provide a comprehensive valve. In the evaporator, chilled water is produced (point b) by
perspective for decision-making. Moreover, a sensitivity investi- transferring the heat of the inlet coil (point a) to the refrigerant,
gation is done to scrutinize the influence of critical parameters on which is converted to the vapor (point 5). Water vapor is taken by
the cycle performance; finally, a multi-objective optimization the rich LiBreH2O solution in the absorber to produce a lean so-
based on a combination of the nondominated-storing-genetic al- lution with a lower temperature (Point 6). Meanwhile, the extra
gorithm (NSGA-II) and the artificial neural network (ANN) is heat is captured by the coil (streams c and d). Subsequently, the
considered to address a trade-off between the thermodynamic lean solution enters the generator after pumping and taking some
performance and economic feasibility of the proposed cycle. booster heat in the heat exchanger. Solution temperature is
increased in the generator, where a portion of the solution is
separated as the water vapor because of its lower boiling temper-
2. System description
ature and then the solution enters the compressor (point 1). The
remaining fluid goes towards the absorber after coming out of the
As mentioned, an AR can come into four different configura-
heat exchanger and the expansion valve 2 to complete the cycle
tions. The hybridization can be designed in four general configu-
[31].
rations. Table 1 makes a comparison of these four cycles and their
For a better understanding of the reference AR cycle, a qualita-
main properties.
tive temperature-entropy diagram is illustrated in Fig. 2. As shown,
The schematic design of the understudy combined AR cycle (the

Table 1
Comparison of different configurations of AR cycles.

Configuration of AR system Properties

Compressor between absorber and evaporator Suitable for cold regions

Low generator temperature
Compressor between absorber and generator Running by electricity
Low discharge temperature and pressure of the compressor
Applicable for industrial waste heat recovery
Compressor between low- and high-pressure stages Easy modulating of cooling and heating capacities
More flexible cooling and heating capacity ratios
Compressor between generator and condenser Significant reduction or rejection of external heat supply
Low generator temperature
Elimination of bulky condenser
4 A.R. Razmi et al. / Energy 210 (2020) 118559

3.1. Thermodynamic analysis

The mass and energy conversion, as well as exergy balance

equations for different equipment are derived as follows [39]:
X X
m_ i ¼ m_ e (1)

X X
m_ i Xi ¼ m_ e Xe (2)

X X
m_ i hi þ Q_ ¼ m_ e he þ W_ (3)

X X
_
m_ i exi þ Ex _ W þ Ex
m_ e exe þ Ex _ D
Q ¼ (4)

here, subscripts i and e show the inlet and outlet streams of the
control volume and Ex_ D represents the exergy destruction of each
Fig. 2. Qualitative temperature-entropy diagram of the hybrid absorption/recom-
_ , and, Ex
control volume. Also, ex, Ex _ W are respectively specific,
pression system. Q
thermal, and power exergy rates, defining as follows [40]:

the refrigerant is saturated vapor and superheated vapor at the inlet ex ¼ ðh
h0 Þ
T0 ðs
s0 Þ (5)
and outlet of the compressor (streams 1 and 2). Besides, the
refrigerant at the condenser and evaporator outlets (streams 3 and
 
_ ¼ 1
T0 Q_
Ex (6)
5) is respectively saturated liquid and vapor. Q
Tk k
The following assumptions are supposed to simulate the pro-
posed AR cycle [32e36]:
_ W ¼W
Ex _ (7)
k
1. The understudy cycle operates at a steady state condition.
where subscript 0 relates to the reference condition. It is worthy to
2. Variations of kinetic and potential energies are negligible.
3. The isentropic efficiencies of the compressor and pump are mention that the kinematic and potential exergies have been
ignored in the formulation [41]. Since chemical reactions do not
constant.
4. The condenser coil outlet (point 3) is a saturated liquid. happen within the cycle components, absorber fluid’s chemical
state is constant. Furthermore, considering the insignificant rate of
5. The outlet stream of the evaporator (point 5) is supposed to be
saturated vapor. chemical exergy in comparison to physical exergy, it is discarded in
the present research like most of the publications in the literature
6. The pressure losses associated with frictional effects in pipes
[42e45]. To provide a more comprehensive perspective from the
and between components are ignored.
exergetic performance of the components, second law efficiency
are expressed as follows [46]:

3. System simulation _
Ex _
Ex
hex;k ¼ _ p ¼ 1
_ D (8)
Exf Exf
The governing equations for thermodynamic and exer-
goeconomic analyses of the reference cycle are proposed in this _ p andEx
_ denote the exergy rates of the product and fuel in
here Ex f
section. The primary assumptions and design conditions used for
the components. The product exergy is the desired exergy flow
the simulation of the current study are outlined in Table 2.
coming out from each component, whereas the fuel is the
consumed exergy within the component for providing that prod-
Table 2 uct. The governing energy balance and fuel-product equations for
The assumed design condition for the system simulation [30,37,38]. different equipment of the proposed cycle are tabulated in Tables 3
Foundation parameter Value Unit and 4.
 To assess the proposed system from the sustainable perspective,
Absorber temperature 35 C
Ambient pressure 1.01 bar a sustainability index (SI) is considered in this study. In fact, SI is a
Ambient temperature 25 
C determinative parameter that connects the exergy analysis to the
Annual operating hours 4320 Hr environmental aspect. It states that how decreasing the total exergy
Cost of electricity [37] 0.033 $/kWh destruction of the system in the energy conversion processes re-
Evaporator capacity 150 kW
Evaporator temperature 10 
C
sults in a drop in the environmental impact of system [47,48]:

Generator temperature 65 C
Heat exchanger effectiveness 0.9 eee 1

SI ¼ (9)
Inlet temperature of the absorber coil 30 C DP

Inlet temperature of the evaporator coil 18 C
Interest rate 15 %
here DP is an indicator of depletion number, defining as the pro-
Isentropic efficiency of compressor 88 %
Isentropic efficiency of pump 85 % portion of total exergy destruction to the inlet exergy. Hence, the
Maintenance factor 1.06 eee more the sustainability index, the less the environmental effect of

Outlet temperature of the absorber coil 40 C the reference system.

Outlet temperature of the evaporator coil 13 C The COP and exergy efficiency of the reference system are ob-
Service year of the system 20 eee
tained as follows [30,49,50].
 A.R. Razmi et al. / Energy 210 (2020) 118559 5

Table 3
Energy balance equations of the cycle components.

Component Equation

Absorber Q_ Abs ¼ m_ 5 h5 þ m_ 11 h11
m_ 6 h6

Compressor _ h
h1
W Com ¼ m _ 2 h2
m_ 1 h1 ; hCom ¼ 2;s
h2
h1
Evaporator Q_ Evp ¼ m_ 5 h5
m_ 4 h4
Expansion valve 1 h3 ¼ h4
Expansion valve 2 h10 ¼ h11
Generator-condenser Q_ Gen
Con ¼ m_ 1 h1 þ m_ 9 h9
m_ 8 h8 ¼ m_ 2 h2
m_ 3 h3
Heat exchanger T
T10
m_ 8 h8
m_ 7 h7 ¼ m_ 9 h9
m_ 10 h10 ; E ¼ 9
T9
T7
Pump _ P ¼ m_ 7 h7
m_ h ; h h7;s
h6
W 6 6 Com ¼
h7
h6

purchase cost of the kth component, and tis the annual working
Table 4
Fuel-product definition of the cycle components.
time of the cycle per hour. CRF is defined as [54]:

Component Fuel Product

ið1 þ iÞN
Absorber _ _ _ _
Ex
_ c CRF ¼ (15)
Ex 11 þ Ex5
Ex6 Exd ð1 þ iÞN
1
Compressor W_ _
Ex
Ex _
Com 2 1
Evaporator Ex _ 5
_
Ex _
Ex
Ex _ a
4 b where, N is the service years of the system and i indicates the in-
Expansion valve 1 _
Ex _
Ex
3 4 terest rate. To calculate the initial purchase cost of the equipment,
Expansion valve 2 _
Ex _
Ex
10 11
the capital investment cost functions of the equipment are listed in
Generator-condenser Ex _
_
Ex _ þ Ex
Ex _
Ex
_
2 3 9 1 8
_
_
Ex _
Ex
_ 7
Table 5.
Heat exchanger Ex 9 10 Ex8
Pump _ _
Ex
_ Finally, the updated purchase equipment cost is obtained using
W P Ex7 6
the following equation [55].

Cost at present year ¼ Original cost

Q_ Evp 
Cost index for the present year
(16)
COP ¼ (10) Cost index for the year of original cost
_ _
W Com þ W P
Along with the cost balance equations, some auxiliary equations
_ D;tot
Ex are needed to address the cost equilibrium equation, all of which
hex ¼ 1
_ _P
(11) are represented in Table 6.
W Com þ W
The intermediate cost per unit exergies of fuel and product, as
well as the cost destruction rate for the cycle components, as the
most well-known indicators of the exergoeconomic analysis, are
3.2. Exergoeconomic analysis calculated by using the following equations [56]:

The goal of this exergy based cost assessment is to specify the C_ f ;k

cf ;k ¼ (17)
cost development practices, as well as the calculation of the cost _
Ex f ;k
per exergy unit of the product, using for better economic perfor-
mance of the system [51]. To realize the unit cost of exergy streams,
cost-balancing equations along with the needed auxiliary equa- C_ p;k
cp;k ¼ (18)
tions are adopted to the cycle equipment. The cost balance equation _
Ex p;k
for the system components in exergy costing process is defined by
Ref. [52]:
C_ D;k ¼ cf ;k : Ex
_ (19)
X X D;k
C_ e;k þ C_ W;k ¼ C_ i;k þ C_ Q ;k þ Z_ k (12)
here, C_ f ;k and C_ p;k are respectively the cost rate of fuel and product at
here, kth component. Besides, Ex _ _
and Ex indicate the exergy rate of
f ;k p;k
the fuel and product at kth component.
C_ ¼ c : Ex
_ (13) The relative cost difference and exergoeconomic factor can be
defined as follows [57]:
where c presents the cost per unit of each exergy stream. Also,
C_ and C_ are respectively the cost rates related to the generated
W;k Q ;k cp;k
cf ;k
rk ¼ (20)
power and received thermal energy by the component. The lev- cf ;k
elized investment cost rate can be found by means of the capital
recovery factor (CRF) [53]:
Z_ k
fk ¼ (21)
Z : CRF: fr
Z_ k ¼ k (14) Z k þ C_ D;k
_
t
Since the exergoeconomic assessment provides a precise outline
where fr represents the maintenance factor,Zk is the equipment of the exergy cost rate, exergoeconomic performance parameters
6 A.R. Razmi et al. / Energy 210 (2020) 118559

Table 5
Cost functions and purchase cost of the cycle components [32,38].

Component Cost function Purchase cost ($)

Absorber  0:6 9037

A
ZAbs ¼ ZR AR ¼ 100 m2; ZR ¼ 16,500
AR
 
Compressor 1 653
ZCom ¼ 44:71 m_ 1 Cr : lnðCr Þ
0:95
hCom
Evaporator  0:6 8181
A
ZEvp ¼ ZR AR ¼ 100 m2; ZR ¼ 16,000
AR
Expansion valve 1 ZEv1 ¼ 114:5 m_ 3 7.7
Expansion valve 2 ZEv2 ¼ 114:5 m_ 10 99.3
Generator-condenser  0:6 3360
A
ZAbs ¼ ZR AR ¼ 100 m2; ZR ¼ 25,500
AR
Heat exchanger  0:6 3129
A
ZHXCH ¼ ZR AR ¼ 100 m2; ZR ¼ 12,000
AR
 _ 0:26  
Pump WP 1
hP 0:5 119
ZP ¼ 2100 
10 hP
Total capital cost 24,586

Table 6
Cost balance and auxiliary equations of the cycle components.

Component Cost balance equation Auxiliary equation

Absorber C_ d
C_ c ¼ C_ 5 þ C_ 11
C_ 6 þ Z_ Abs C_ 5 þ C_ 11 C_
cc ¼ 0 ; ¼ 6
E_ 5 þ E_ 11 E_ 6
Compressor C_ 2
C_ 1 ¼ C_ W;Com þ Z_ Com cw ¼ 0:033 $=kWh[37]
Evaporator C_ b
C_ a ¼ C_ 4
C_ 5 þ Z_ Evp c4 ¼ c5 ; ca ¼ 0
Expansion valve 1 C_ ¼ C_ þ Z_ Ev
4 3 1

Expansion valve 2 C_ 11 ¼ C_ 10 þ Z_ Ev2

Generator-condenser C_ 1 þ C_ 3 þ C_ 9 ¼ C_ 2 þ C_ 8 þ Z_ Gen
Con C_ 1 þ C_ 9 C_
c2 ¼ c3 ; ¼ 8
E_ 1 þ E_ 9 E_ 8
Heat exchanger C_ 8
C_ 7 ¼ C_ 9
C_ 10 þ Z_ HXCH c9 ¼ c10
Pump C_ 7
C_ 6 ¼ C_ W;P þ Z_ P

are considered for minimizing the total cost rate (C_ tot ). Total cost 4.1. Model validation
rate is achieved as the sum of the capital investment, operating, and
maintenance costs, as well as the cost related to the exergy To verify the simulation results for the hybrid AR cycle, a com-
destruction [58]: parison of the obtained data is considered with those of reported by
X X Riffat et al. [29]. This comparison has been outlined in Table 7. As it
C_ tot ¼ Z_ k þ C_ D;k (22) can be seen, there is a reasonable deviation between the current
study and the similar publication in the literature, meaning that the
model is reliable and authentic.

3.3. Economic analysis 4.2. Thermodynamic and exergoeconomic results

The net present value (NPV) method is applied to analyze the The thermodynamic properties of the streams in the reference
economic feasibility of the present cycle, which can address the cycle are listed in Table 8. A computer program was developed in
payback period of the invested capital budgeting and the cycle the EES software and reported data in Table 8 are extracted from
profit during its economic book life [59]. According to NPV method, the system thermodynamic simulation via simultaneous solving of
the net cash flow at the end of year is converted to its present value energy, exergy, and mass conversion equations.
based on the interest rate as follows [60]: The results associated with the exergy analysis, as well as the
specific exergy costs of product and fuel, cost of exergy destruction,
X
20 X
NPV ¼ Yð1 þ iÞ
N
CCapital (23)
N¼0 Table 7
Validation of the simulation results with the literature.
here, N and i express the year and interest rate. Y indicates the net
Parameters Present work Ref. [29] Deviation (%)
cash flow at the end of the year. According to Table 1, an interest
rate of 15% and 20 years of service time have been considered for COP 4.17 4.21 0.95
the proposed system. QGen-Con (kW) 129.7 128.5 0.93
QEvp (kW) 100 100 0.00
QAbs (kW) 124.6 123.7 0.73
4. Results and discussion TGen ( C) 60 60 0.00
Wcom (kW) 23.98 23.74 1.01
The simulation results of the hybrid absorption/recompression Xlean (%) 53.28 52.21 2.05
Xrich (%) 58.85 57.21 2.86
cycle are reported in this section.
 A.R. Razmi et al. / Energy 210 (2020) 118559 7

Table 8
The result of thermodynamic simulation for the system streams.

State Fluid T ( C) P (kPa) h (kJ/kg) s (kJ/kg.K) _ ðkg=sÞ

m _ ðkWÞ
Ex

1 H2O 65 4.1 2621 8.67 0.067 2.82

2 H2O 303 31 3082 8.77 0.067 31.90
3 H2O 70 31 292.5 0.95 0.067 0.87
4 H2O 10 1.2 292.5 1.03 0.067
0.78
5 H2O 10 1.2 2519 8.89 0.067
8.65
6 LiBre H2O 35 1.2 76.11 0.23 0.935 10.98
7 LiBre H2O 35 4.1 76.11 0.23 0.935 10.98
8 LiBre H2O 59 4.1 126.6 0.38 0.935 14.19
9 LiBre H2O 65 4.1 149.5 0.39 0.868 31.89
10 LiBre H2O 38 4.1 95.09 0.22 0.868 28.11
11 LiBre H2O 38 1.2 95.09 0.22 0.868 28.11
a H2O 18 101 75.47 0.26 7.169 1.68
b H2O 13 101 54.55 0.19 7.169 6.54
c H2O 30 101 125.7 0.43 4.328 0.38
d H2O 40 101 167.5 0.57 4.328 6.34

purchased equipment costs, exergoeconomic factor, and relative the exergy destruction streams of equipment.
cost difference for the system components are outlined in Table 9. The performance indicators and technical parameters of the
The most quantity of capital investment cost rate associates with system obtained from the thermodynamic and economic analyses
the absorber, followed by the evaporator, and the generator- are presented in Table 10. A vapor compressor with a compression
condenser components. Among the all, the compacted generator- ratio of 7.52 is needed to address the required heat power of the
condenser, owing to internal irreversibilities arisen at direct heat compacted generator-condenser (188 kW) for reaching the
exchange between the water vapor and LiBr-water solution, is designed generator temperature (65  C). As a result, 31.05 kW po-
responsible for about 52% of total exergy destruction in the system. wer is consumed in the compressor to deliver 150 kW refrigerating
As mentioned in Table 6, the specific exergy cost of the inlet water capacity in the evaporator with a COP of around 4.83. Moreover,
to the evaporator and the absorber (streams a and c) is supposed to 47.2 and 181 kW heat are respectively transferred in the heat
be zero. The specific exergy cost of cooling as the system product is exchanger and the absorber. The exergy analysis reveals that about
103.1 $/GJ. Also, the exergoeconomic factor of the second expansion 20.23 kW exergy is destructed in different processes and compo-
valve and the pump have the maximum values, meaning that they nents and the proposed cycle operates with an exergy efficiency of
have negligible cost of destruction. On the other hand, the lowest 34.84%. Also, the cycle has a sustainability index of around 1.53
values of exergoeconomic factor belong to the first expansion valve, which is a reasonable rate for refrigeration cycles. The economic
generator-condenser, compressor, and absorber. The calculated analysis denotes that an investment cost of around 24,586 $ is
values for the mentioned equipment indicate that the rate of cost needed to install a combined absorption/recompression refrigera-
destruction is higher than the capital cost rate. The total cost rates tion system with a refrigerating capacity of 150 kW. The reference
of the investment and the exergy destruction of the reference cycle cycle has a payback period of 4.17 years and 27,702 $ can be saved
are 0.96 and 3.35 $/hr, respectively, resulting in a total exer- during its service time (20 years). Finally, the exergoeconomic
goeconomic factor of 22.34%. It means that the cost of exergy analysis shows that the total cost rates of destruction and invest-
destruction is responsible for 77.66% of system cost rate, revealing ment are respectively 3.35 and 0.96 $/hr, leading to a total cost rate
the importance of applying the exergoeconomic analysis for the of 4.31 $/hr.
reference cycle. Thus, the use of expensive equipment that reduces
the cost destruction rate and enhances the investment cost rate,
4.3. Parametric analysis
improves the cycle performance from exergoeconomic perspective.
In addition, the second expansion valve has a low relative cost
In order to make a thorough assessment, the effect of the sig-
difference that indicates the negligible difference between the cost
nificant parameters on the system performance is examined in this
per exergy unit of its product and fuel, as well as the superior
section. Among all, the generator and evaporator temperatures are
benefit.
the most crucial factors that can substantially affect system oper-
A Sankey exergy flow diagram is also prepared and illustrated in
ation. Thus, the adjustment of the generator and evaporator tem-
Fig. 3 to give a better understanding of cycle operation for readers.
peratures are considered as the base of parametric analysis.
It is worthy to mention that the width of the vectors is comparative
The impact of the generator and evaporator temperatures on the
to the exergy values of the cycle streams and the dark arrows show
heat transfer rate of the compacted generator-condenser and the

Table 9
The result of exergy and exergoeconomic analyses for system components.

Component _ ½kW
Exf
_ p ½kW
Ex _ D ½kW
Ex hex ½% cf ½$=GJ cp ½$=GJ cD ½$=hr Z_ ½$=hr f ½% r ½%

Absorber 8.485 5.962 2.523 70.26 112.5 176.6 1.0220 0.3543 25.74 36.30
Compressor 31.05 29.08 1.974 93.64 9.167 10.03 0.0651 0.0256 28.21 8.639
Evaporator 7.867 4.859 3.008 61.76 74.69 102.9 0.8116 0.3207 65.33 37.45
Expansion valve 1 0.867
0.783 1.651 90.42 67.68 74.96 0.4022 0.0003 0.075 10.75
Expansion valve 2 28.11 28.11 0.000 100 54.78 54.81 0.0000 0.0039 100 0.071
Generator-condenser 31.04 20.53 10.51 66.15 67.68 15.99 2.5600 0.1317 4.894 76.37
Heat exchanger 3.789 3.220 0.569 84.97 54.78 75.05 0.1123 0.1227 52.21 27.01
Pump 0.002 0.000 0.002 2.86 9.167 eeee 0.0001 0.0046 98.44 99.95
8 A.R. Razmi et al. / Energy 210 (2020) 118559

Fig. 3. The Sankey exergy flow diagram for the hybrid absorption/recompression refrigeration system.

Table 10
The important result of the reference system.

Parameter Value Unit

Cr 7.52 e
C_D; tot
3.35 $/hr
C_ tot 4.31 $/hr
cp 76.49 $/GJ
COP 4.83 eee
_ D; tot
Ex 20.23 kW
hex 34.84 %
ftot 22.34 %
Q_Gen
con
188 kW
Q_ Evp 150 kW
Q_Abs
181 kW
Q_ HXCH 47.2 kW
SI 1.53 eee
NPV20 27,702 $
tp 4.17 year
W _ 31.05 kW
Com
W _ 0.002 kW
P
XLean 52.2 %
XRich 56.3 %
Ztot 24,586 $ Fig. 4. Variation in the required heat and power of the generator and compressor with
Z_ tot 0.96 $/hr the generator and evaporator temperatures.

compressor power consumption. As mentioned in Table 1, a con-

compressor power consumption is presented in Fig. 4. As it can be
stant refrigerating capacity of 150 kW was considered for the
observed, the more the evaporator temperature, the less the
reference cycle. Therefore, the COP is improved with decreasing the
needed heat and power of the generator and the compressor. There
compression ratio and power consumed by the compressor,
is an optimum generator temperature, wherein the heat required in
reaching its maximum value in a point around the generator tem-
the generator reaches its minimum quantity. Since the vapor
perature of 65  C where the power consumption becomes mini-
compressor supplies all the heat needed in the generator, this
mum. It is note mentioning that the optimum values for various
generator temperature value is responsible for the lowest power
evaporator temperatures will be presented in the following.
consumption of the compressor. In other words, the low heat ca-
The results of the second thermodynamic law including the
pacity of the generator-condenser means low power consumption
exergy efficiency of the system, total exergy destruction, and the
of the compressor.
ratio of exergy destruction in various equipment are shown in
The low values of power consumption by the compressor results
Figs. 6 and 7. Similar to the trend of power consumption, there is an
in a low compression ratio, shown in Fig. 5. The COP of the cycle is a
optimum generator temperature that conduces the minimum
ratio of the produced refrigerating capacity to the net power
exergy destruction and maximum exergy efficiency in the system.
consumed in the compressor and the pump. However, the power
Contrary to the results of the first thermodynamic law and COP, the
consumption of the pump is negligible and the COP highly depends
exergy efficiency of the system is improved in lower evaporator
on the values of the evaporator refrigerating capacity and the
temperature. Although total exergy destruction is increased in
 A.R. Razmi et al. / Energy 210 (2020) 118559 9

Fig. 5. Variation in the COP and compression ratio of the compressor with the
generator and evaporator temperatures.

Fig. 7. Variation in the exergy destruction ratio of the components with the generator
temperature.

considering a generator temperature, in which their exergy

destruction is moderate, can be more sensible for reducing total
exergy destruction. The results disclose that the exergy destruction
of the compressor is minimum at this point (65  C). It is note
mentioning that the exergy destructions of the first expansion
valve and the evaporator retain constant in various generator
temperatures. However, their exergy destruction ratios are altered
according to the overall exergy destruction of the cycle as repre-
sented in Fig. 6.
The capital cost of the system equals the sum of the equipment
purchased cost, which depends on many parameters as stated in
Table 5. The effect of the generator temperature on the cost of the
components is illustrated in Fig. 8. It is noted that the costs are
shown at the designed evaporator temperature of 10  C. Higher
mass flowrate of the LiBreH2O solution enters the solution heat
exchanger in lower generator temperatures, leading to a higher
Fig. 6. Variation in the exergy efficiency and total exergy destruction with the capital cost for this component. The economic results reveal that
generator and evaporator temperatures.
the capital investment cost is continuously decreased from 32300 $
in the generator temperature of 45  C to about 20500 $ in 85  C.
lower evaporator temperatures, its increment rate is lower than Therefore, in spite of the fact that the purchase cost of the evapo-
increasing compressor and pump power consumption, resulting in rator remains unchanged in different conditions due to its constant
higher exergy efficiency. In other words, the proportion of the cooling capacity, the capital cost ratio of the evaporator compared
overall exergy destruction to the exergy of the fuel (cycle power with the entire system capital cost is more at higher generator
consumption) is greater in higher evaporator temperatures, temperatures. As the cost equation of the compressor is a function
resulting in lower exergy efficiency based on Eq. (8). of its compression ratio, its capital cost follows the trend of the
Fig. 7 presents the ratio of exergy destruction in different compression ratio in Fig. 5, reaching the minimum value at a
components at an evaporator temperature of 10  C. The compacted generator temperature of around 65  C. Moreover, the capital cost
generator-condenser is responsible for a substantial amount of the of the compacted generator-condenser increases constantly until a
overall exergy destruction in the system. The rates of exergy specific generator temperature and after that begins to decrease.
destruction in the second expansion valve and the pump are so The sustainability index and exergoeconomic factor are the
negligible, not represented in Fig. 7. The alteration of the exergy other fundamental parameters in finding the environmental and
destruction in the absorber and the compacted generator- economic feasibility of the cycle in different operating settings,
condenser is more conspicuous compared to other components. shown in Fig. 9. The exergoeconomic factor reaches its minimum
Against to the generator-condenser, the exergy destruction in the value in a range of 55e60  C for the generator temperature based
evaporator is increased in higher generator temperatures. Hence, on the evaporator temperature, meaning that the maximum cost
10 A.R. Razmi et al. / Energy 210 (2020) 118559

index is higher at lower evaporator temperatures, meaning that

designing an operating condition with a lower evaporator tem-
peratures better meets the environmental aspect.
The influence of the generator and evaporator temperatures
variation on the cost rates of the system can be observed in Fig. 10.
As mentioned in Eq. (22), the total cost rate of the reference cycle
matches the sum of total cost rates of destruction and investment. A
higher price is required to purchase the cycle components in lower
evaporator temperatures, resulting in a higher investment cost rate.
Nevertheless, the total investment cost rate is reduced with a
decreasing slope in higher generator temperatures. The cost rate of
destruction becomes maximum in a generator temperature of
around 50  C, meaning that this point is the worst temperature
from the exergoeconomic perspective. Furthermore, the exer-
goeconomic results confirm the aforementioned claim about the
better feasibility of the cycle with greater evaporator temperature
at higher generator temperatures.
The ultimate cost of the chilled water production as the cycle
product, as well as the payback period of the cycle are represented
in Fig. 11 with a change in the generator temperature and for
different evaporator temperatures. The cost of the chilled water
directly depends on the total cost rate, presented in Fig. 10. Actually,
the more the total cost rate of the cycle, the more the cost of pro-
duction. Therefore, the maximum cost of production is achieved at
a generator temperature of around 55e60  C for various evaporator
temperatures, being responsible for the maximum cost rates. The
Fig. 8. Variation in the capital cost ratio of the components with the generator result of economic analysis implies that the proposed hybrid ab-
temperature.
sorption/recompression cycle has a reasonable payback period
compared with the available refrigeration cycles. Considering the
cycle operation during the hot months of the year, it has a payback
period of 4.17, 4.57, and 5.05 for the evaporator temperatures of 5,
7.5, and 10  C, respectively. Based on the obtained results, the
payback period is reduced with a rise in the generator temperature
and keeps almost constant at higher generator temperatures. It is
also obvious that lower profit is achievable in lower evaporator
temperatures during the service years of the cycle. According to
Fig. 12 and NPV method, a total profit of 21500, 24600, and 27700 $
can be achieved in the evaporator temperatures of 5, 7.5, and 10  C,
respectively.

Fig. 9. Variation in the sustainability index and total exergoeconomic factor with the
generator and evaporator temperatures.

destruction and the minimum investment cost happen in this

range. The more the evaporator temperature, the less the generator
temperature, in which the minimum exergoeconomic factor can be
achieved. Consequently, a greater generator temperature should be
designed for higher evaporator temperatures to improve the eco-
nomic feasibility of the reference cycle. Likewise, it is more
reasonable to consider a lower generator temperature for the
evaporator providing lower design temperature. Moreover, like the
exergy efficiency, the sustainability index is a function of exergy
destruction and is improved by increasing the generator up to an Fig. 10. Variation in the total rates of cost destruction, cycle cost, and investment cost
optimum value and then is decreased. Moreover, sustainability with the generator and evaporator temperatures.
 A.R. Razmi et al. / Energy 210 (2020) 118559 11

5. Multi-objective optimization

Providing a high efficiency and economic refrigeration concept

is the major objective of the proposed cycle. In this regard, two
separate multi-objective optimization models are performed to
find the optimal outputs. In the first scenario, maximum COP and
minimum payback period are the objective functions (model (1));
whereas, maximum exergy efficiency and minimum payback
period organize the objective functions of the second optimization
scenario (model (2)). Moreover, the generator, evaporator, and
absorber temperatures are applied as decision variables, search
spaces of which are expressed as follows.
 
Optimize COP or hex or tp TGen ; TEvp ; TAbs (24)

subject to:

50  TGen  85
5  TEvp  10 (25)
32  TAbs  38

Fig. 11. Variation in the payback period and cost of product with the generator and
evaporator temperatures.
5.1. Optimization methodology and accuracy check

Between numerous optimization methods, the genetic algo-

rithm (GA) is used for optimizing the proposed cycle. It has been
proved in many research works that GA has better performance in
comparison to the other traditional mathematical techniques [58].
The most conspicuous point about the GA method is that it does an
examine from a population of solutions rather than an individual
point [61]. In fact, the function of the GA method is based on a
population of potential solutions for producing superior approxi-
mations to reach the optimum solution. NSGA-II, operating based
on GA, is one of the most practical multi-criteria evolutionary al-
gorithms for optimizing energy systems [62]. It has many advan-
tages such as robustness and fast population distance estimation;
however, low operation speed is its main drawback [63].
The outline of the optimization methodology is shown in Fig. 13
to elaborate on the optimization process. To reduce run time and
prevent further computational cost, the developed simulation
model is repeated 750 times for 750 sets of input data to meet the
required data for producing the objective functions based on the
ANN. Estimating the objective functions in terms of ANN results in a
significant reduction in run time (from over 6 h to 10 min). The ANN
is a powerful mathematical structure which can be used for iden-
tifying complicated nonlinear relationships between input and
Fig. 12. Total profit (NPV) of the reference system in various evaporator temperatures output data sets. ANN models have been found useful and efficient,
during its service years.
particularly in problems for which the characteristics of the pro-
cesses are difficult to describe using physical equations [64]. It has

Fig. 13. The outline of the optimization methodology.

12 A.R. Razmi et al. / Energy 210 (2020) 118559

Fig. 15. Distribution of the Pareto frontier diagram for COP and payback period (model
(1)).

been proved in the literature that the ANN is capable of mapping

the implicit relationship between the decision variables and
objective functions, calculating the system performance with
excellent precision [65]. The 750 random generated data of ther-
modynamic and economic analyses from the developed model are
trained by the ANN to derive equations for the objective functions
(COP, hex , and tp) based on the decision variables (TGen , TEvp , and
TAbs ). The optimization is performed to address the optimum values
for two separate problems by simultaneous reflection of the eco-
nomic attractiveness of the reference cycle with the results of the
first and second thermodynamic laws. The model (1) is used for
simultaneous optimization of the COP and payback period, while
the model (2) is employed to present the optimum values by
considering the exergy efficiency and payback period at the same
time. Finally, the ideal optimum solution is addressed for two
optimization programs. The optimum solution has the minimum
geometric distance from the ideal point, or the maximum distance
from non-ideal point [66].
The accuracy verification of the predicted data for the objectives
by means of ANN and the actual data of the simulation is presented
in Fig. 14. The coefficient of determination (R2) is an indicator for
evaluating the performance of ANN. In fact, it measures the vari-
ance of the response data and the closer the R2 is to unity, the
higher the reliability of the ANN model. As presented in
Fig. 14(aec), the R2 values for COP, exergy efficiency, and payback
period are respectively 0.99945, 0.99994, and 0.99791, confirming
the high reliability of the ANN model.

5.2. Optimization results

The Pareto frontier solution of the model (1) for optimizing the
COP and payback period is shown in Fig. 15.
Although all of the points shown in the Pareto frontier diagram
reveal the optimal solutions, point B is the ideal solution of the
multi-criteria optimization for model (a), considering both the
optimized solutions of the payback period and COP, simultaneously.
At this ideal point, the COP and payback period are respectively
4.88 and 3.97 years. Moreover, point A with a COP of 4.44 and a
payback period of 3.84 years indicates the best optimal solution
Fig. 14. Accuracy verification of the ANN model for a) COP, b) exergy efficiency, and c) from the economic standpoint. Alternatively, point C shows the
payback period. best optimal solution from the perspective of the COP, where the
 A.R. Razmi et al. / Energy 210 (2020) 118559 13

Fig. 17. Distribution of the Pareto frontier diagram for the exergy efficiency and
payback period (model (2)).

However, Fig. 16 (b) represents that the optimal points for the
generator temperature are located between 68TGen72; most of
which are at the generator temperature of 71  C. Finally, as dis-
cussed in section 4.3, Fig. 16 (c) confirms the better performance of
the cycle in higher evaporator temperatures from the viewpoint of
the COP and economic. It is worthy to mention that designing a
refrigeration cycle with the evaporator temperature of more than
10  C is not so reasonable and not considered in the optimization.
Also, the Pareto frontier solution and scatter distribution of the
model (b) for optimizing the exergy efficiency and payback period
are presented in Figs. 17 and 18. Like the previous model, three
different points D, E, and F are considered in the Pareto frontier of
the model (2) to analyze the alterations of the exergy efficiency and
the payback period. Between all optimum points, the point E cor-
responds the best results for both exergy efficiency and the payback
period. Also, the point F can be preferred when the exergy effi-
ciency is the only objective function. In contrary, the point D rep-
resents the ideal solution for a single objective optimization based
on the payback period. Contrary to the scatter distribution of the
model (1), Fig. 18 (a) exhibits that there is a brilliant agreement for
the absorber temperature in the model (2), which should be
retained at the highest value (38  C). The more the generator
temperature, the less the payback period. Also, the exergy effi-
ciency is greater at the generator temperatures of higher than the
critical point (65  C); so, there is a good agreement between the
objective functions of the model (2) in higher generator tempera-
tures, as shown in Fig. 18 (b). In contrast to COP, the exergy effi-
ciency is greater at lower evaporator temperatures, whereas the
payback period increases. As a result, the scatter distribution of the
evaporator for the model (2) is distributed in a wide range and with
a more focus on the temperature of around 9.5  C.
Finally, the values of the optimum parameters and objectives for
Fig. 16. Scatter distribution for the important parameters of model (1); a) absorber
both models and points A, B, C, D, E, and F are indicated in Table 11.
temperature, b) generator temperature, and c) evaporator temperature.

6. Conclusions
payback period and COP are respectively 4.08 and 4.99. For a better
understanding of the key factors, the scatter distribution of model
Refrigeration systems play a crucial role in buildings’ energy
(a) is shown in Fig. 16.
saving and greenhouse gas emission reduction. The vapor
Fig. 16 (a) shows that the optimum points for the absorber
compression and absorption systems are the conventional kinds of
temperature are distributed at all ranges between 32  C and 38  C.
chillers, operating with opposite properties in terms of
14 A.R. Razmi et al. / Energy 210 (2020) 118559

Table 11
The optimum values for six Pareto frontier optimal solutions.

Model Point TGen ð CÞ TEvp ð CÞ TAbs ð CÞ COP hex ð%Þ tp ðyearÞ

(1): COP and tp

A 74.24 9.97 37.91 4.44 36.41 3.84
B 71.94 9.97 32.25 4.88 30.47 3.97
C 67.12 9.97 32.10 4.99 31.32 4.08
(2): hex and tp
D 84.31 9.87 35.91 3.76 31.19 3.35
E 71.83 9.33 37.93 4.45 37.43 4.11
F 70.52 5.13 37.83 4.07 39.32 5.51

feasibility and marketing. The substantial conclusions are drawn as

follows:

(1) The results of the first and second laws of thermodynamic

reveal that the proposed cycle has a coefficient of perfor-
mance and an exergy efficiency of 4.83 and 34.84%, respec-
tively. In a bid to address the dominant sources of
irreversibilities in the cycle, a Sankey diagram was designed,
indicating that for a refrigerating capacity of 150 kW, total
exergy of 20.23 kW destructed, in which the compacted
generator-condenser occupies 52% of the overall exergy
destruction.
(2) The exergoeconomic investigation represented that the
reference system has a total investment cost rate and the cost
destruction rate of 0.96 $/hr and 3.35 $/hr, and the chilled
water as the cycle product can be produced with a cost of
76.49 $/GJ. Moreover, a reasonable sustainability index of
1.53 was achieved for the reference cycle.
(3) Along with the feasibility of the proposed system from the
efficiency and environmental viewpoints, the economic
analysis approves its sustainability from the economic
perspective. A capital cost of 24,586 $ is needed to install the
reference cycle with the designed capacity. It was concluded
that the cycle has a payback period of around 4.17 years and a
total saving of 27,700 $ is achieved at the end of its service
years.
(4) Two separate multi-objective optimization models were
performed based on an integration of the payback period
with the COP and exergy efficiency as the objective functions,
concluding that the COP and exergy efficiency of 4.88 and
37.43% can be achieved at the optimum conditions. More-
over, the payback period reaches less than 4 years at the
optimum design values.

The present research identifies the relative merits of the pro-

posed hybrid absorption/recompression refrigeration cycle from
the perspective of performance, environmentally friendly charac-
teristics, and economic feasibility. Its experimental setup can be
developed as further research work to improve its real operation
and practical application in the future.

CRediT authorship contribution statement

Fig. 18. Scatter distribution for the important parameters of model (2); a) absorber
temperature, b) generator temperature, and c) evaporator temperature.
Amir Reza Razmi: Conceptualization, Methodology, Investiga-
tion, Software, Formal analysis, Validation, Writing - original draft,
performance, environmental condition, and capacity. Combining Visualization, Writing - review & editing. Ahmad Arabkoohsar:
the promising features of the available refrigeration cycles, a novel Conceptualization, Supervision, Writing - review & editing. Hos-
hybrid absorption/recompression cycle was developed to improve sein Nami: Methodology, Writing - review & editing.
the performance, economic feasibility, and environmental charac-
teristics of the refrigeration cycles. In this study, a precise and References
comprehensive assessment of the cycle performance and a multi-
objective optimization were performed to stimulate the system’s [1] Zeynalian M, Hajialirezaei AH, Razmi AR, Torabi M. Carbon Dioxide capture
from compressed air energy storage system. Appl Therm Eng 2020;178:
 A.R. Razmi et al. / Energy 210 (2020) 118559 15

115593. optimization of an ammonia e water power/cooling cogeneration cycle. En-

[2] Sohani A, Sayyaadi H. Providing an accurate method for obtaining the effi- ergy 2012;47:271e83.
ciency of a photovoltaic solar module. Renew Energy 2020;156:395e406. [33] Akrami E, Chitsaz A, Nami H, Mahmoudi SMS. Energetic and exergoeconomic
[3] Razmi A, Soltani M, Tayefeh M, Torabi M, Dusseault MB. Thermodynamic assessment of a multi-generation energy system based on indirect use of
analysis of compressed air energy storage (CAES) hybridized with a multi- geothermal energy. Energy 2017;124:625e39.
effect desalination (MED) system. Energy Convers Manag 2019;199:112047. [34] Razmi A, Soltani M, Torabi M. Investigation of an efficient and
[4] Razmi A, Soltani M, Aghanaja C, Torabi M. Thermodynamic and economic environmentally-friendly CCHP system based on CAES, ORC and compression-
investigation of a novel integration of the absorption-recompression refrig- absorption refrigeration cycle: energy and exergy analysis. Energy Convers
eration system with compressed air energy storage (CAES). Energy Convers Manag 2019;195:1199e211.
Manag 2019;187:262e73. [35] Behzadi A, Arabkoohsar A. Comparative performance assessment of a novel
[5] Sohani A, Sayyaadi H, Azimi M. Employing static and dynamic optimization cogeneration solar-driven building energy system integrating with various
approaches on a desiccant-enhanced indirect evaporative cooling system. district heating designs. Energy Convers Manag 2020;220:113101.
Energy Convers Manag 2019;199:112017. [36] Sohani A, Sayyaadi H. Design and retrofit optimization of the cellulose
[6] Akrami E, Nemati A, Nami H, Ranjbar F. Exergy and exergoeconomic assess- evaporative cooling pad systems at diverse climatic conditions. Appl Therm
ment of hydrogen and cooling production from concentrated PVT equipped Eng 2017;123:1396e418.
with PEM electrolyzer and LiBr-H2O absorption chiller. Int J Hydrogen Energy [37] Nami H, Anvari-moghaddam A. Small-scale CCHP systems for waste heat
2017;43:622e33. recovery from cement plants: thermodynamic, sustainability and economic
[7] Arabkoohsar A, Sadi M. Technical comparison of different solar-powered ab- implications. Energy 2020;192:116634.
sorption chiller designs for co-supply of heat and cold networks. Energy [38] Razmi AR, Janbaz M. Exergoeconomic assessment with reliability consider-
Convers Manag 2020;206:112343. ation of a green cogeneration system based on compressed air energy storage
[8] Sohani A, Sayyaadi H. Thermal comfort based resources consumption and (CAES). Energy Convers Manag 2020;204:112320.
economic analysis of a two-stage direct-indirect evaporative cooler with [39] Ghaebi H, Parikhani T, Rostamzadeh H, Farhang B. Thermodynamic and
diverse water to electricity tariff conditions. Energy Convers Manag 2018;172: thermoeconomic analysis and optimization of a novel combined cooling and
248e64. power (CCP) cycle by integrating of ejector refrigeration and Kalina cycles.
[9] Song JY, Lee JW, Kang YT. Comparisons of Nu correlations for H2O/LiBr solu- Energy 2017;139:262e76.
tion in plate heat exchanger for triple effect absorption chiller application. [40] Soltani M, Nabat MH, Razmi AR, Dusseault MB, Nathwani J. A comparative
Energy 2019;172:852e60. study between ORC and Kalina based waste heat recovery cycles applied to a
[10] Sun Z, Wang Q, Xie Z, Liu S, Su D. Energy and exergy analysis of low GWP green compressed air energy storage (CAES) system(exergy and economic
refrigerants in cascade refrigeration system. Energy 2019;170:1170e80. analysis). Energy Convers Manag 2020;222:113203. https://doi.org/10.1016/
[11] Ayou DS, Bruno JC, Coronas A. Integration of a mechanical and thermal j.enconman.2020.113203.
compressor booster in combined absorption power and refrigeration cycles. [41] Mojtaba S, Rahmani S, Ahmadi P, Wongwises S. Multi-objective design opti-
Energy 2017;135:327e41. mization of a multi-generation energy system based on geothermal and solar
[12] Arabkoohsar A, Andresen GB. Supporting district heating and cooling net- energy. Energy Convers Manag 2020;205:112426.
works with a bifunctional solar assisted absorption chiller. Energy Convers [42] Kaushik SC, Arora A. Energy and exergy analysis of single effect and series
Manag 2017;148:184e96. flow double effect waterelithium bromide absorption refrigeration systems.
[13] Nemati A, Nami H, Yari M, Ranjbar F. Effect of geometry and applied currents Int J Refrig 2009;32:1247e58.
on the exergy and exergoeconomic performance of a two-stage cascaded [43] Gomri R, Hakimi R. Second law analysis of double effect vapour absorption
thermoelectric cooler. Int J Refrig 2018;85:1e12. cooler system. Energy Convers Manag 2008;49:3343e8.
[14] Jain V, Sachdeva G, Singh S. Energy, exergy, economic and environmental (4E) [44] Mohtaram S, Omidi M, Lin J, Sun H, Chen W. Exergy analysis of a multi
analyses based comparative performance study and optimization of vapor mixture working fluid absorption refrigeration cycle. Case Stud Therm Eng
compression-absorption integrated refrigeration system. Energy 2015;91: 2019;15:100540.
816e32. [45] Avanessian T, Ameri M. Energy, exergy, and economic analysis of single and
[15] Kim J, Park S, Baik Y, Chang K, Ra H, Kim M, et al. Experimental study of double effect LiBreH2O absorption chillers. Energy Build 2014;73:26e36.
operating characteristics of compression/absorption high-temperature hybrid [46] Nami H, Arabkoohsar A. Improving the power share of waste-driven CHP
heat pump using waste heat. Renew Energy 2013;54:13e9. plants via parallelization with a small-scale Rankine cycle , a thermodynamic
[16] Gao P, Chang M, Zhang C, Shao L. System principles and applications of hybrid analysis. Energy 2019;171:27e36.
sorptionecompression heat pumpeA review. Int J Refrig 2019;108:14e25. [47] Ahmadi P, Rosen MA, Dincer I. Multi-objective exergy-based optimization of a
[17] Gazda W, Kozioł J. The estimation of energy efficiency for hybrid refrigeration polygeneration energy system using an evolutionary algorithm. Energy
system. Appl Energy 2013;101:49e57. 2012;46:21e31.
[18] Wu W, Wang BL, Shi WX, Li XT. Performance improvement of ammonia/ [48] Ahmadi P, Rosen MA, Dincer I. Greenhouse gas emission and exergo-
absorbent air source absorption heat pump in cold regions. Build Serv Eng environmental analyses of a trigeneration energy system. Int J Greenh Gas
Technol 2014;35:451e64. Control 2011;5:1540e9.
[19] Wu W, Wang B, Shang S, Shi W, Li X. Experimental investigation on NH3eH2O [49] Sohani A, Sayyaadi H. End-users’ and policymakers’ impacts on optimal
compression-assisted absorption heat pump (CAHP) for low temperature characteristics of a dew-point cooler. Appl Therm Eng 2020;165:114575.
heating in colder conditions. Int J Refrig 2016;67:109e24. [50] Rezayan O, Behbahaninia A. Thermoeconomic optimization and exergy anal-
[20] Wu W, Shi W, Wang B, Li X. Annual performance investigation and economic ysis of CO2/NH3 cascade refrigeration systems. Energy 2011;36:888e95.
analysis of heating systems with a compression-assisted air source absorption [51] Mojtaba S, Rostami M, Hamzeh A. Multi-criteria design optimization and
heat pump. Energy Convers Manag 2015;98:290e302. thermodynamic analysis of a novel multi- generation energy system for
[21] Wu W, Shi W, Wang J, Wang B, Li X. Experimental investigation on NH3eH2O hydrogen, cooling, heating, power, and freshwater. Int J Hydrogen Energy
compression-assisted absorption heat pump (CAHP) for low temperature 2020;45:15047e62.
heating under lower driving sources. Appl Energy 2016;176:258e71. [52] Mojtaba S, Rahmani S, Ahmadi P, Wongwises S. Multi-objective design opti-
[22] Osenbruck A. Verfahren zur Kalteerzeugung bei Absorptionsmaschinen. Dtsch mization of a multi-generation energy system based on geothermal and solar
Reichspatent, [DRP 84084] 1985:1e16. energy. Energy Convers Manag 2020;205:112426.
[23] Altenkirch E. Kompressionskaltemaschine mit Losungskreislauf Kaltetechnik [53] Alharbi S, Elsayed ML, Chow LC. Exergoeconomic analysis and optimization of
1950;2:310e5. 251-259,279-284. an integrated system of supercritical CO2 Brayton cycle and multi-effect
[24] Wu W, Wang B, Shi W, Li X. Absorption heating technologies: a review and desalination. Energy 2020;197:117225.
perspective. Appl Energy 2014;130:51e71. [54] Behzadi A, Habibollahzade A, Zare V, Ashjaee M. Multi-objective optimization
[25] Heard CL, Holland FA. AMMONIA/LITHIUM nitrate absorption/compression of a hybrid biomass-based SOFC/GT/double e ff ect absorption chiller/RO
refrigeration part I . simulation. Appl Therm Eng 1997;17:223e33. desalination system with CO2 recycle. Energy Convers Manag 2019;181:
[26] Ayala R, Heard CL. AMMONIA/LITHIUM nitrate absorption/compression 302e18.
refrigeration cycle. Part II. EXPERIMENTAL. Appl Therm Eng 1998;18:661e70. [55] Emadi MA, Chitgar N, Oyewunmi OA, Markides CN. Working-fluid selection
[27] Leonzio G. Mathematical model of absorption and hybrid heat pump. Chin J and thermoeconomic optimisation of a combined cycle cogeneration dual-
Chem Eng 2017;25:1492e504. loop organic Rankine cycle (ORC) system for solid oxide fuel cell (SOFC)
[28] Wu W, You T, Wang J, Wang B, Shi W, Li X. A novel internally hybrid waste-heat recovery. Appl Energy 2020;261:114384.
absorption-compression heat pump for performance improvement. Energy [56] Habibollahzade A, Gholamian E, Behzadi A. Multi-objective optimization and
Convers Manag 2018;168:237e51. comparative performance analysis of hybrid biomass-based solid oxide fuel
[29] Riffat SB, Wong CW. Gas-driven absorption/recompression system. Heat Re- cell/solid oxide electrolyzer cell/gas turbine using different gasification
covery Syst CHP 1994;14:165e71. agents. Appl Energy 2019;233e234:985e1002.
[30] Razmi A, Soltani M, Kashkooli F M, Garousi Farshi L. Energy and exergy [57] Nami H, Ranjbar F, Yari M. Methanol synthesis from renewable H2 and
analysis of an environmentally-friendly hybrid absorption/recompression captured CO2 from S-Graz cycle- Energy, exergy, exergoeconomic and exer-
refrigeration system. Energy Convers Manag 2018;164:59e69. goenvironmental (4E) analysis. Int J Hydrogen Energy 2019;44:26128e47.
[31] Babaei SM, Razmi AR, Soltani M, Nathwani J. Quantifying the effect of nano- [58] Baghernejad A, Yaghoubi M. Exergoeconomic analysis and optimization of an
particles addition to a hybrid absorption/recompression refrigeration cycle. integrated solar combined cycle system (ISCCS) using genetic algorithm. En-
J Clean Prod 2020;260:121084. ergy Convers Manag 2011;52:2193e203.
[32] Zare V, Mahmoudi SMS, Yari M, Amidpour M. Thermoeconomic analysis and [59] Arabkoohsar A, Nami H. Thermodynamic and economic analyses of a hybrid
16 A.R. Razmi et al. / Energy 210 (2020) 118559

waste-driven CHPeORC plant with exhaust heat recovery. Energy Convers geostatistical modeling using genetic algorithm and direct search methods.
Manag 2019;187:512e22. J Petrol Sci Eng 2008;63:34e42.
[60] Najafi A, Jafarian A, Darand J. Thermo-economic evaluation of a hybrid solar- [64] Sohani A, Sayyaadi H, Hoseinpoori S. Modeling and multi-objective optimi-
conventional energy supply in a zero liquid discharge wastewater treatment zation of an M-cycle cross-flow indirect evaporative cooler using the GMDH
plant. Energy Convers Manag 2019;188:276e95. type neural network. Int J Refrig 2016;69:186e204.
[61] Esfahani IJ, Yoo CK. Exergy analysis and parametric optimization of three [65] Jiang H, Xi Z, Rahman AA, Zhang X. Prediction of output power with artificial
power and fresh water cogeneration systems using refrigeration chillers. neural network using extended datasets for Stirling engines. Appl Energy
Energy 2013;59:340e55. 2020;271:115123.
[62] Sohani A, Sayyaadi H, Mohammadhosseini N. Comparative study of the con- [66] Behzadi A, Arabkoohsar A, Gholamian E. Multi-criteria optimization of a
ventional types of heat and mass exchangers to achieve the best design of biomass-fired proton exchange membrane fuel cell integrated with organic
dew point evaporative coolers at diverse climatic conditions. Energy Convers rankine cycle/thermoelectric generator using different gasification agents.
Manag 2018;158:327e45. Energy 2020;201:117640.
[63] Maschio C, Campane A, Jose D. A framework to integrate history matching and