Energy Conversion and Management 157 (2018) 631–643

Contents lists available at ScienceDirect

Energy Conversion and Management

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

Thermodynamic analysis of a combined power/refrigeration cycle: T

Combination of Kalina cycle and ejector refrigeration cycle
Candeniz Seckin
Department of Mechanical Engineering, Faculty of Engineering, Marmara University, Goztepe Campus, Istanbul 34722, Turkey

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

Keywords: In the present study, a new power and refrigeration cycle is investigated which is a combination of a Kalina cycle
Kalina cycle and an ejector refrigeration cycle (ERC). In the proposed configuration of the combined cycle, an ejector re-
Ejector refrigeration cycle frigeration cycle is inserted into the Kalina cycle to recover heat from ammonia poor solution which leaves the
Combined refrigeration and power cycle separator at high temperature/pressure and does not contribute to power generation in Kalina cycle Working
Cogeneration
fluid of the Kalina cycle and ERC are ammonia-water solution and R134a, respectively. The combined cycle is
Thermodynamic analysis
simulated by EES software and details of the applied mathematical model and developed simulation program are
extensively reported. The effect of five key operational parameters of the combined cycle (i.e. turbine inlet
pressure, turbine inlet temperature, concentration of ammonia-water basic solution, condenser outlet tem-
perature and pressure of refrigerant in heat exchanger) on the combined cycle performance parameters (re-
frigeration capacity, power production, thermal efficiency, exergy of produced power, exergy of refrigeration
and exergy efficiency) is analyzed and physical mechanisms behind the determined results are reported.
Additionally, variation of performance parameters with heat exchanger pressure is examined with different
refrigerants (R134a, R152a and R290) to determine the effect of refrigerants on system performance. The results
show that thermal efficiency of the combined cycle increases with increasing turbine inlet temperature and
concentration of ammonia-water solution but decreases with rising condenser outlet temperature and heat ex-
changer pressure. A maximum thermal efficiency point is determined in the analyzed range of the turbine inlet
pressure. Exergy efficiency increases with rising turbine inlet pressure, turbine inlet temperature and con-
centration of ammonia-water solution but decreases with increasing condenser outlet temperature and heat
exchanger pressure. Refrigeration capacity and thermal efficiency results of the combined cycle are the highest
for the operation of ERC with R290 and the lowest with R134a. Exergy efficiency is the lowest for ERC operation
with R290 and the highest with R134a.

1. Introduction
Energy is strongly necessary and directly related with economic and
social development of countries and well-being of human beings. Due to
the increasing depletion rates of high quality energy sources, the rapid
growth in population and growing rate of industrialization, efficient use
of energy and minimum consumption of natural energy sources are
regarded as critical issues of industrial development and environmental
protection. In the sense of more efficient energy-resource conversion,
the combined cycles of cooling and power display higher efficiency
with respect to conventional cycles and this makes the combined cycles
(cogeneration cycles) promising as a future trend of power and re-
frigeration production [1–3].
In a conventional steam power cycle (Rankine cycle), pure fluids are
used as working fluid (usually water) which leaves the boiler at high
temperature/pressure and expands in a turbine to produce power.
Instead, using a binary mixture in the cycle provides a better tem-
perature match than that of pure working fluids in heat transfer pro-
cesses due to the variable-temperature phase change capability of
binary mixtures. Due to better temperature match of the binary mixture
working fluid, less irreversibility occurs in heat transfer process and
total cycle efficiency gets higher. For power generation and refrigera-
tion, among possible candidates of binary mixtures, the most known
and used binary mixture is ammonia–water mixture in both of the in-
dustrial applications and academic studies. Ammonia–water mixture is
also the working fluid of the analyzed power production cycle (Kalina
cycle) of this present study [1,4,5].
Maloney and Robertson [6] performed one of the first studies of
power cycle with binary fluid: absorption power cycle with ammo-
nia–water mixture as the working fluid. Then, in the early 1980s, Kalina
[7,8] introduced his absorption power cycle (Kalina cycle) with a novel
configuration which uses ammonia–water mixture and offers higher

E-mail address: candeniz.seckin@marmara.edu.tr.

https://doi.org/10.1016/j.enconman.2017.12.047
Received 7 August 2017; Received in revised form 12 December 2017; Accepted 13 December 2017
Available online 23 December 2017
0196-8904/ © 2017 Published by Elsevier Ltd.
C. Seckin Energy Conversion and Management 157 (2018) 631–643

Nomenclature cond-II condenser II

d diffuser
A cross-sectional area (m2) dest destroyed
C speed of sound (m/s) eva evaporator
Eẋ rate of exergy (kW) ex exergy
h specific enthalpy (kJ/kg) H20 water
ṁ mass flow rate (kg/s) in-I input in the HE-I
M mach number, molecular weight in-II input in the HE-II
P pressure (kPa) is isentropic
q quality K Kalina cycle
Q̇ rate of heat transfer (kW) m primary nozzle exit
s specific entropy (kJ/kg K) mass mass transfer
T temperature (°C) net net
V velocity (m/s) NH3 ammonia
v specific volume (m3/kg) p-I pump I
w entrainment ratio p-II pump II
Ẇ power (kW) Q heat transfer
x concentration ref refrigeration
1,2,3 … ,21 number of state-points in Fig. 1 s secondary nozzle exit
tot total
Greek symbols tur turbine
W work interaction
ψ specific exergy (kJ/kg)
η efficiency Abbreviations
ρ density
ERC ejector refrigeration cycle
Subscripts HE I heat exchanger I
HE II heat exchanger II
0 dead state HTR high-temperature recuperator
a mixed flow/cross section a in Fig. 3 LTR low-temperature recuperator
b shocked flow/cross section b in Fig. 3 TTD terminal temperature difference
cc combined cycle
cond-I condenser I

efficiency. After that, numerous studies have investigated the perfor-
mance of Kalina cycle under different conditions and have reported the
higher performance of Kalina cycle relative to that of conventional
power cycles. [9–12]. The proposed configuration by Kalina can be
regarded as further development of Rankine cycle and absorption
power cycle which also utilize from the above mentioned characteristic
advantage of the binary mixture use. Hence, irreversibility of the cycle
is kept low but efficiency is higher [13].
Regarding the combined cycles which use ammonia–water mixture
as working fluid and generate power and refrigeration concurrently,
Goswami and Xu [14] proposed a new combined cycle of power/
cooling generation (Goswami cycle) in which the ammonia-rich vapor
separated by a rectifier unit is employed as the turbine working fluid to
generate power, and then the turbine exhaust provides cooling by
transferring sensible heat to the chilled water. Later, power generation
in the turbine of the cycle is further increased by slight modification of
Goswami cycle configuration such as installing a superheater before the
turbine to increase the turbine inlet temperature and pressure [15]. In
the literature, parametric analysis [16,17], exergy based analysis
[18–20], parametric optimization [21–24], thermoeconomic analysis
[25] and experimental studies [26,27] are carried out for Goswami
cycle. In some studies, several design modifications are applied to im-
prove the refrigeration and power generation capacity of the Goswami
cycle [28–31]. Besides Goswami cycle, other combined cooling and
power cycle configurations which use ammonia water mixture as the
working fluid are presented by Liu and Zhang [32], Zhang and Lior
[33,34], Wang et al. [35].
Studies of Kalina cycle based combined cycles which produce power
and refrigeration output, are also seen in the literature. Zheng et al.
[13] added a condenser, an evaporator and a rectifier to the Kalina
cycle to have higher concentration of ammonia in the working fluid and
to enhance the refrigeration capacity. Yu et al. [36] proposed a novel
combined cooling/power generation system which consists of a mod-
ified Kalina cycle and an ammonia absorption cooling cycle. The cycles
are interconnected by mixers, splitters, absorbers and heat exchangers.
The proposed combined cycle can produce different ratios of cooling
and power. Jing and Zheng [37] presented a new power/cooling co-
generation cycle which is a combination of Kalina cycle and the double-
effect ammonia–water absorption refrigeration cycle and analyzed
heat-to-power ratio and exergy efficiency improvement of the com-
bined cycle. Hua et al. [38] combined a modified Kalina cycle and an
ammonia-water absorption refrigeration cycle to generate power and
cooling output. Analysis of the impact of key parameters on the thermal
and exergy efficiencies is presented. Srinivas et al. [39] analyzed
cooling/power cogeneration cycle which is designed by coupling a
Kalina cycle with a absorption refrigeration cycle. They reported in-
creased energy utilization factor for the cogeneration cycle and de-
termined the operational conditions of maximum performance. Then
Shankar and Srinivas [40,41] proposed a new cooling/power cogen-
eration cycle (Srinivas cycle) in which the working fluid (ammonia–-
water mixture) is condensed at the turbine exit to the phase of the sa-
turated liquid and then is brought to a lower pressure before entering
into the evaporator to increase the cooling capacity of the cycle. Zare
[42] conducted a comparative thermodynamic analysis and optimiza-
tion for two different tri-generation cycles: Kalina cycle based and or-
ganic Rankine cycle (ORC) based tri-generation cycles driven by geo-
thermal heat. A LiBr/water absorption chiller and a water heater are
coupled to the tri-generation cycles to produce cooling and heating. The
Kalina cycle based tri-generation is more efficient than the other cycle
in terms of exergy efficiency. Another Kalina cycle based trigeneration

632
C. Seckin Energy Conversion and Management 157 (2018) 631–643

study is performed by Mohammadi and Mehrpooya [43] where a and pressure of the refrigerant in the heat exchanger) on selected per-
modified Kalina cycle is coupled to a reverse osmosis system to provide formance parameters: thermal efficiency, net power production, re-
heating, cooling and power and potable water. Wang et al. [4] proposed frigeration capacity and exergy efficiency of the Kalina based combined
a novel design of ammonia–water mixture using combined power and cycle. Additionally, exergy content of system products (power and re-
cooling cycle to utilize from low grade heat sources. Energy and exergy frigeration) with respect to variation in system key parameters is pre-
analyses of the proposed cycle are presented. Cao et al. [44] in- sented. Both of the thermal efficiency and exergy efficiency analyses
vestigated a Kalina-based combined cooling and power cycle which is results indicate useful information to evaluate the thermodynamic
driven by a low-grade heat source. The proposed cycle consists of a performance of the considered combined cycle. But, the exergy analysis
Kalina cycle and an absorption refrigeration cycle. Parametric analysis of the system provides deeper insight to determine the inefficient points
of the combined cycle is performed by means of a developed numerical of the system and provides opportunities for exergy loss minimization
simulation program. Rashidi et al. [45] analyzed two new power and of unit operations [18–20]. Discussion of the physical mechanism be-
cooling cogeneration cycles based on Kalina cycle and absorption re- hind the determined effect of the system key parameters is reported.
frigeration cycle from thermodynamic and economic aspects. As the Impact of different refrigerant use (R134a, R152a and R290) on given
working fluid, absorption refrigeration cycle uses ammonia-water performance parameters is comparatively reported at different heat
mixture in one cogeneration cycle and lithium bromide-water in the exchanger pressures.
other one.
Ejector refrigeration cycles (ERCs) are used in many trigeneration
2. Description of the combined cycle
and cogeneration systems due to several advantages of ejectors such as:
improved energy efficiency, simplicity of conception, low installation
In this study, analyzed combined cycle consists of a power cycle
and operational cost (due to the absence of moving parts) and ability to
(Kalina cycle) and a refrigeration cycle (ejector refrigeration cycle,
operate with a wide range of refrigerants [46,47]. As a result, several
ERC). Schematic representation of considered Kalina cycle and ERC are
attempts are reported in the literature to insert ejectors or ERC into the
presented in Fig. 1 and 2, respectively. State points of the cycles are
power/refrigeration cycles. Wang et al. [48] combined the Rankine
given in the figures below.
cycle and the ejector–absorption refrigeration cycle. They reported the
As seen in Fig. 1, Kalina cycle is made up: a heat exchanger (HE I)
improvement of combined refrigeration/power cycle performance by
which transfers heat from external heat source into the Kalina cycle
employing the ejector. Xu et al. [49] performed parametric analysis of
working fluid, a separator, a turbine, a high-temperature recuperator
combination of turbine added Brayton cycle and a transcritical ejector
(HTR), a low-temperature recuperator (LTR), a condenser (condenser
refrigeration cycle. Wang et al. [50] employed an ejector expander
I), a pump (pump I) and a valve (valve I). Ammonia- water solution is
instead of the expansion valve in a solar assisted trigeneration cycle
used as the working fluid to generate power in the Kalina cycle. The
which is a combination of Brayton cycle and transcritical CO2 re-
frigeration cycle with ejector-expansion device. Habibzadeh et al. [51] 2
presented the thermodynamic study of ORC and ERC coupled cogen-
eration cycle. The performance of the combined cycle is examined with Separator
different working fluids (R123, R141b, R245fa, R600a, R601a). Ghaebi Turbine Wtur
et al. [52] investigated the performance characteristics of four different
combined cooling and power cycles with ejector. They reported the
cooling capacity increase due to the ejector use in the combined cycles.
Barkhordarian et al. [53] proposed a cogeneration cycle which is a
combination of Kalina cycle and ERC. The peculiar novelty of the design 1 3
is: there are two evaporators that can produce refrigeration output in HE I
two different temperature levels and capacities. Ghaebi et al. [3] pre- ERC
sented a new design of combined cycle (Kalina cycle and ERC) in which
Qin-I 7
the exhaust of the turbine is fed to the ejector as the motive flow to
draw the secondary flow into the ejector. Energy, exergy, and ex-
ergoeconomic analyses of the combined cycle are carried out. Sun et al.
[54] performed an exergy analysis for ORC based combined cycle 13 4
driven by low temperature waste heat. The ORC is combined with two
HTR
different refrigeration cycles (absorption refrigeration cycle and ERC).
The impact of operational parameters on exergy efficiency is in-
vestigated.
In this study, a novel Kalina based combined power and refrigera-
tion cycle (combination of Kalina cycle and ejector refrigeration cycle, 5 6
ERC) is analyzed. Utilization potential of the ammonia-poor solution 12
(which leaves the separator at high temperature and pressure in Kalina Valve I
8
cycle) is used in refrigeration generation by extracting heat from am-
LTR
monia-poor solution and transferring this heat to the cooling cycle
(ERC). To achieve this goal, an ERC is inserted into Kalina cycle before
the high-temperature recuperator (details are presented in Section 2).
To the best of the author's knowledge and by surveying the above 11 9
mentioned literature review, the combination of the Kalina cycle and
ERC with proposed configuration in this present study, is not so far Wp-I Qcond-I
investigated. The working fluids of the Kalina cycle and ERC are am-
Pump I Condenser I
monia water solution and R134a, respectively. Parametric analysis of
the combined cycle is performed to determine the effect of system key 10
parameters (turbine inlet pressure, turbine inlet temperature, con-
Fig. 1. Schematic overview of Kalina sub-cycle of the combined cycle.
centration of ammonia water solution, condenser outlet temperature

633
C. Seckin
Wp-II HE II
3 4
Qin-II
14 15
Pump II
18 Condenser II
Ejector
17 16
19 21
Qcond-II
Valve II
Evaporator
20 21
Qref
Fig. 2. Schematic overview of the ejector refrigeration cycle (ERC).
ammonia-water basic solution absorbs heat in HE I and then, the
working fluid is delivered to the separator in two-phase region and at
high temperature/pressure (state-point 1). In the separator, ammonia-
water solution is split into saturated rich ammonia-water vapor (state-
point 2) and saturated weak ammonia-water liquid (state-point 3). The
saturated ammonia-rich vapor enters into the turbine and expands to
state-point 7 to generate power. The saturated weak ammonia-water
liquid (at state-point 3) is transferred to HE II which provides the ne-
cessary heat to operate the ERC (Figs. 1 and 2). After rejecting heat in
HE II, the ammonia-weak solution further rejects heat to the cold am-
monia-water basic solution coming from LTR (state-point 12) while
passing through the HTR (state-points 4–5). Then, the pressure of the
ammonia-weak solution is lowered down to turbine exit pressure in the
throttling valve I (state-point 6). Afterwards, the ammonia-weak solu-
tion (state-point 6) is mixed with ammonia-rich solution which exits
from the turbine (state-point 7) to reform the ammonia-water basic
solution (state-point 8). Heat transfer from the ammonia-water basic
solution occurs through the LTR (state-points 8–9) and condenser I
(state-points 9–10), to the working fluid exits from the Pump I and to
the environment, respectively. At the condenser exit (state-point 10),
ammonia-water basic solution is at saturated state. The pressure of the
basic solution is raised by pump I to the turbine inlet pressure. Tem-
perature of the ammonia-water basic solution rises while passing
through the LTR (state-points 11–12) and HTR (state-points 12–13) due
to the heat transfer processes. Finally, the cycle is closed when the
ammonia-water basic solution enters into the HE I (state-points 13)
[53,55].
Schematic diagram of the ejector refrigeration cycle (ERC) is seen in
Fig. 2. The ERC is composed of: a heat exchanger (HE II) which extracts
heat from the Kalina cycle working fluid and transfers it to the ERC to
operate the refrigeration cycle, an ejector, a condenser (condenser II), a
valve (valve II), an evaporator, a pump (pump II). At the exit of the HE
II (state-point 15), the refrigerant is the saturated vapor at high tem-
perature and pressure. The vapor (motive flow), enters into the ejector
and mixes with refrigerant stream which comes from the evaporator
(secondary flow). Details of the ejector working principles are pre-
sented further. At the exit of the ejector (state-point 16), the pressure of
the refrigerant is between the pressure of the HE II and the evaporator.
Hence, in an ERC, the function of integrating an ejector into the system
is twofold: expansion of the motive flow fed from the HE II, and com-
pression of the secondary flow drawn from the evaporator [46]. After
leaving the ejector, the refrigerant enters into the condenser II where
heat rejection process occurs. At the exit of the condenser II (state-point
17), the saturated liquid refrigerant is split into two streams: one stream
Energy Conversion and Management 157 (2018) 631–643
t a b
Primary nozzle
15 m
16
s
15
Secondary nozzle
Constant area Diffuser
mixing section
21
Fig. 3. Constant area ejector design.
pumped to HE II by pump II (state-points 18–14); and the other stream
expands through the valve II (state-points 19–20). At the exit of the
valve II (state-point 20), low pressure refrigerant enters into the eva-
porator where the refrigerant is evaporated by absorbing heat from the
refrigerated space to perform the cooling effect of the refrigeration
cycle. At the exit of the evaporator (state-point 21), the refrigerant is
saturated vapor which is sucked into the ejector and then, the re-
frigeration cycle is closed. Thus, pump II is the only component ERC
that has moving parts and requires mechanical energy. [56–58].
The ejector, as seen in Fig. 2, is integrated into the refrigeration
cycle (ERC) and receives flows drawn from the HE II (motive flow,
which acts as the activation source of the ejector) and from the eva-
porator (secondary flow). The refrigerant leaves the ejector at an in-
termediate pressure (P16) between the HE II and the evaporator pres-
sures which is equal to the compressor pressure [59]. Schematic
overview of the employed “constant area” type ejector with critical
cross-sections is presented in Fig. 3. The motive flow enters into the
ejector and passes through a convergent – divergent nozzle (primary
nozzle in Fig. 3). In the primary nozzle, the flow is chocked at the
nozzle throat (cross-section t in Fig. 3) and accelerates from subsonic to
supersonic velocity before leaving the nozzle at a pressure lower than
the evaporator pressure (cross-section m in Fig. 3). The pressure dif-
ference between the primary nozzle exit and the evaporator creates the
required suction for the secondary flow, hence, the secondary flow is
sucked into the ejector. The secondary flow enters into the secondary
nozzle (Fig. 3), accelerates and expands through the nozzle. Both of the
motive and secondary flows are mixed in the “constant area mixing
section” at constant pressure. Mixing process ends at cross-section a in
Fig. 3. Due to the supersonic flow in the mixing chamber, shock is
formed at a cross-section in the constant area mixing section (cross-
section b in Fig. 3) which causes pressure rise and velocity reduction of
the mixed flow from supersonic to subsonic. At the end of the mixing
chamber, the refrigerant enters into the diffuser part of the ejector in
which the velocity of the mixed flow is further stepped down and the
pressure is lifted to the condenser pressure (state-point 16 in Fig. 3)
[56–60].
3. Mathematical model and computational programing
In the analysis of both of the ERC and Kalina cycle, applied as-
sumptions are listed below [44,46,53,55,58,61]:
• Steady-state, one-dimensional, frictionless flow.
• The kinetic energy and potential energy changes in the devices are
neglected.
• The leakage of working fluid is neglected.
• Pressure loss in pipes and all components are neglected.
• Heat losses to the environment from all components and pipes
(except the condenser I and II) are neglected.
• Throttling processes in valve I and II are isenthalpic.
• Losses in the turbine, pump I, pump II and ejector are accounted for
634
C. Seckin Energy Conversion and Management 157 (2018) 631–643

by means of isentropic efficiencies. ṁ 1 = ṁ 8 = ṁ 9 = ṁ 10 = ṁ 11 = ṁ 12 = ṁ 13 (10)

ṁ 2 = ṁ 7 (11)
Special assumptions applied to analyze the Kalina cycle are pre-
sented as follows [44,53]: ṁ 3 = ṁ 4 = ṁ 5 = ṁ 6 (12)
In above equations, P, T, q, x and m are pressure, temperature,
– At the exit of the condenser I, the ammonia-water solution is at
quality, concentration and mass flow rate of the ammonia-water solu-
saturated liquid state.
tion at subscripted state-points, respectively (see Fig. 1).
– At the exit of the separator, the rich and weak ammonia-water
Other thermodynamic properties of the Ammonia-water solution
streams are saturated vapor and saturated liquid, respectively.
can be determined as a function of 3 properties of the solution by EES.

Below assumptions are made to perform the mathematical model of P10 = f (T10,x10,q10) (13)
ERC [46,56,62]:
x2 = f (T2,P2,q2) (14)
+ At the ejector inlet (the motive and secondary flows) and also at the x3 = f (T3,P3,q3) (15)
exit of the ejector, velocity of the refrigerant is zero.
+ The design pressure at the primary nozzle exit is uniform and the For the separator, the mass balance equation for ammonia-water
secondary flow reaches critical conditions at this level [58,63]. solution and ammonia is presented in Eqs. (16) and (17) below.
+ Critical-mode operation in the ejector, i.e. the primary and sec- ṁ 1 = ṁ 2 + ṁ 3 (16)
ondary flows are both chocked and the entrainment is constant.
+ At the exit of the condenser II, the refrigerant is at saturated liquid ṁ 1x1 = ṁ 2 x2 + ṁ 3x3 (17)
state. Specific enthalpy of the ammonia-water solution at the turbine exit
+ At the exit of the HE II and evaporator (at the ejector inlet, the (h7) is calculated as follows:
motive and secondary flows), the refrigerant is at saturated vapor
state. h2 = f (T2,P2,q2) (18)
+ Pressure and temperature are constant through the condenser II and s2 = f (T2,P2,q2) (19)
evaporator.
h7,is = f (P7,x7,s2) (20)
To simulate the considered combined cycle under different opera-
tional conditions, a computational program is developed using EES Ẇtur h −h
ηtur = = 2 7
(Engineering Equation Solver) software. Ammonia-water mixture Ẇtur,is h2−h7,is (21)
properties in Kalina cycle and refrigerant properties of ERC are avail-
T7 = f (P7,x7,h7) (22)
able in EES data bank and are incorporated into the simulation pro-
gram. In the mathematical model of the performed thermodynamic where h and s are the specific enthalpy and specific entropy of the
analysis, the conservation of mass, energy and momentum principles ammonia-water solution at subscripted state-points, his is the specific
are successively applied to each element of the analyzed combined enthalpy of the ammonia-water solution at the end of an isentropic
cycle. Mathematical formulas used in the thermodynamic modeling of process, Ẇtur is the power generation of the turbine, Ẇtur,is is the power
Kalina cycle and ERC are reported below. generation of the isentropic turbine.
Hence, applying the conservation of energy to the turbine, Ẇtur is
3.1. Kalina cycle modeling determined by the following equation:
Ẇtur = ṁ 2 (h2−h7) (23)
Temperature and pressure at the turbine inlet (T2 and P2), con-
denser exit temperature (T10), concentration and mass flow rate of the Conservation of energy equation is presented in Eq. (25) for heat
ammonia-water basic solution (x1 and m1), isentropic efficiency of the transfer process in HE II (Fig. 2), namely, heat rejection between state-
turbine (ηtur) and pump I (ηp-I), terminal temperature difference (TTD) points 3 and 4 in Fig. 2.
of recuperator are the input parameters. Definition of concentration h3 = f (T3,P3,q3) (24)
(ammonia mass fraction, x) is given in Eq. (1) where ṁ NH3 is the mass
flow rate of the ammonia in the ammonia-water mixture, ṁ K is the mass Q̇ in-II = ṁ 3 (h3−h 4) (25)
flow rate of the ammonia-water mixture in the Kalina cycle.
Based on determined h4, below equations can be obtained for state-
ṁ NH3 points 4 and 13.
x=
ṁ K (1)
T4 = f (P4,x 4,h 4) (26)
Based on the assumption of isobaric and isothermal process in the
h1 = f (T1,P1,x1) (27)
separator:
Pmax = P1 = P2 = P3 = P4 = P5 = P11 = P12 = P13 (2) Q̇ in-I = ṁ 1 (h1−h13) (28)

T1 = T2 = T3 (3) Applying the principle of energy conservation to the mixing process

of ammonia-rich solution and ammonia-weak, Eq. (30) is obtained:
Pmin = P10 = P6 = P7 = P8 = P9 (4)
h8 = f (T8,P8,x 8) (29)
q2 = 1 (5)
ṁ 8h8 = ṁ 6 h6 + ṁ 7 h7 (30)
q3 = q10 = 0 (6) The isenthalpic process in valve I can be expressed as follows:
x1 = x 8 = x 9 = x10 = x11 = x12 = x13 (7) h5 = h6 (31)
x2 = x7 (8) Hence,
x3 = x 4 = x5 = x 6 (9) T5 = f (P5,x5,h5) (32)

635
C. Seckin Energy Conversion and Management 157 (2018) 631–643

T6 = f (P6,x 6,h6) (33) Specific enthalpy of the motive flow at the primary nozzle exit (hm)
is determined by following equations:
As for the processes in LTR and HTR, equations below must be sa-
tisfied: h15 = f (P15,q15) (55)
T12 = T8−TTD (34) s15 = f (P15,q15) (56)
h12 = f (T12,P12,x12) (35) hm,is = f (Pm,s15) (57)
ṁ 4 (h 4−h5) = ṁ 12 (h13−h12) (36) h15−hm
ηm =
Specific enthalpy of the ammonia-water basic solution at the pump I h15−hm,is (58)
exit (h11) is calculated as follows:
where hm,is is the specific enthalpy of the motive flow at the end of the
h10 = f (T10,P10,x10) (37) isentropic expansion process in the primary nozzle, Pm is the motive
flow pressure at the primary nozzle exit, h and s are the specific en-
s10 = f (T10,P10,x10) (38)
thalpy and specific entropy of the refrigerant at subscripted state-points
h11,is = f (P11,x11,s10) (39) in Figs. 2 and 3.
Applying the conservation of energy principle, velocity of the mo-
Ẇp-I,is h11,is−h10 tive flow at the primary nozzle exit (Vm) is determined by the following
ηp-I = =
Ẇp-I h11−h10 (40) equation:

Hence, power consumption of the pump I (Ẇp-I ) is determined by the Vm = 2(h15−hm) (59)
following equation:
Based on the conservation of mass equation, below equation must
Ẇp-I = ṁ 10 (h11−h10) (41) be satisfied at the exit of the primary nozzle:
vm = f (Pm,hm) (60)
T11 = f (P11,x11,h11) (42)
Energy balance equation of LTR is given below to determine the 1 ⎞ V A
ṁ 15 = ṁ m = ⎛ mtot = m m
thermodynamic properties at condenser inlet. ⎝ 1+w ⎠ vm (61)

ṁ 12 (h12−h11) = ṁ 8 (h8−h9)
T9 = f (P9,x 9,h9)
The rate of heat rejection from condenser I (Q̇cond −I ) is determined
by applying energy balance equation for condenser I as the following:
(43) where ṁ tot is the total mass flow rate of the refrigerant (ṁ 15 + ṁ 21), Am
and vm are the cross-sectional area and the specific volume of the re-
(44) frigerant at the primary nozzle exit, respectively.
Specific enthalpy of the secondary flow at the secondary nozzle exit
(hs) is determined by following equations:

Q̇cond −I = ṁ 9 (h9−h10) (45) h21 = f (P21,q21) (62)

s21 = f (P21,q21) (63)

3.2. ERC modeling
h s,is = f (Ps,s21) (64)
Governing equations which are used in modeling and simulation of h21−h s
ERC are presented below. Inputs of the simulation program are con- ηs =
h21−h s,is (65)
denser II pressure (Pcond-II), evaporator pressure (Peva), heat exchanger
pressure (PHE-II), total mass flow rate of the refrigerant (ṁ tot ), isentropic hs,is is the specific enthalpy of the secondary flow at the end of the
efficiency of pump II (ηp-II), primary nozzle (ηm), secondary nozzle (ηs), isentropic expansion process in the secondary nozzle, Ps is the sec-
diffuser (ηd), cross-sectional area of the primary nozzle exit (Am). ondary flow pressure at the secondary nozzle exit, h and s are the
Mathematical definition of entrainment ratio (w) is given in Eq. (46) specific enthalpy and specific entropy of the refrigerant at subscripted
which is the ratio of the secondary flow mass flow rate (ṁ 21) to the state-points in Figs. 2 and 3.
motive flow mass flow rate (ṁ 15). Cs = f (Ps,h s) (66)
ṁ
w= 21 Vs = 2(h21−h s) (67)
ṁ 15 (46)
Based on above listed assumptions for ERC, below equations are vs = f (Ps,h s) (68)
obtained:
w ⎞ VA
ṁ 21 = ṁ s = ⎛ ṁ tot = s s
PHE-II = P14 = P15 (47) ⎝ 1+w ⎠ vs (69)
Pcond-II = P16 = P17 = P18 = P19 (48) where Cs is the speed of sound at cross-section s, As and vs are the cross-
sectional area and the specific volume of the refrigerant at the sec-
Pevap = P20 = P21 (49)
ondary nozzle exit, respectively.
q15 = q21 = 1 (50) If the phase of the refrigerant is saturated mixture at the exit of the
nozzle (which is not very often encountered in this study), Henry and
q17 = q18 = q19 = 0 (51) Fauske method is applied to determine the critical flow properties at the
ṁ tot = ṁ 16 = ṁ 17 (52) nozzle exit [64–67].
In the constant area mixing section, motive and secondary flows
ṁ 14 = ṁ 15 = ṁ 18 (53) preserve their identity over a short distance following the exit from
their respective nozzles, before mixing takes place at the cross-section
ṁ 19 = ṁ 20 = ṁ 21 (54)
“a” in Fig. 3. Mixing is assumed to occur at constant pressure over a
In above equations, P, q, and ṁ are pressure, quality and mass flow short distance in the constant area mixing section. Based on mass,
rate of the refrigerant at subscripted state-points, respectively. momentum and energy conservation equations, the enthalpy and

636
C. Seckin Energy Conversion and Management 157 (2018) 631–643

velocity of the mixed fluid can be calculated using the following through the evaporator (refrigeration capacity, Q̇ ref ) and power re-
equations: quirement of pump II (Ẇp-II ).
A a = Am + A s (70) h17 = h18 = h19 = f(P17,q17) (82)

ṁ a = ṁ m + ṁ s = ṁ tot (71) Q̇cond-II = ṁ tot (h16−h17) (83)

Conservation of momentum equation: v18 = f(P18,q18) (84)
A a V2a v (P −P )
Pa A a + ṁ tot Va = φ (Pm Am + ṁ m Vm + Ps A s + ṁ s V)
s = Pa A a + Ẇp-II = 18 14 18 ṁ 14
va ηp-II (85)
Am V2m A sV2s ⎞
= φ ⎛⎜Pm Am + + Ps A s +
Ẇp-II = ṁ 14 (h14−h18)
⎟

⎝ vm vs ⎠ (72) (86)

In above equations, Pa is the pressure of the mixed flow before Q̇ in-II = ṁ 14 (h15−h14) (87)
shock, Aa is the cross-sectional area at cross-section a, φ is the coeffi-
cient of frictional loss [68].
h19 = h20 (88)
Conservation of mass equation: Q̇ ref = ṁ 20 (h21−h20) (89)
VA
ṁ tot = a a Mathematical definition of the net power output (Ẇnet ) and com-
va (73) bined cycle thermal efficiency (ηcc) is presented in Eqs. (90) and (91).
where Va and va are velocity and specific volume of the refrigerant at Ẇnet = Ẇtur−Ẇp-I−Ẇp-II (90)
cross-section a in Fig. 3, respectively.
Specific enthalpy at cross-section a (ha) is determined by applying Ẇnet + Q̇ ref
conservation of energy principle which is presented below: ηcc =
Q̇ in-I (91)
V2a ⎤ V2m ⎤ V2s ⎤ The general exergy balance equation of a control volume under-
ṁ tot ⎡ha + = ṁ m ⎡hm + + ṁ s ⎡h s +
⎢ 2⎥ ⎢ 2 ⎥ ⎢ 2⎥ (74) going a steady state process is expressed as the following equation [72]:
⎣ ⎦ ⎣ ⎦ ⎣ ⎦
Eq. (75) must be satisfied. Eẋin−Eẋ out = Eẋ dest (92)

ha = f (Pa,va) (75)
Cross-section b is the cross-section where the shock occurs.
According to Lear et al. [69] the most common and realistic position is
occurrence of normal shock wave in the constant area mixing section
which results in subsonic flow after the shock. Thermodynamic mag-
nitudes after the shock wave are determined by Rankine–Hugoniot
equations [70,71].
where Eẋin and Eẋ out are the rate of exergy entering into the control
volume and the rate of exergy leaving the control volume, respectively.
Eẋ dest represents the rate of exergy destruction or the irreversibility
occurring in the process.
Term of exergy rate (Eẋ ) in the previous equation can be expressed
as:
Eẋ = Eẋ Q + EẋW + Eẋmass (93)

Pb 2k M2a −(k−1) in which Eẋ Q and EẋW are the rate of exergy transfer due to the heat and
= work interactions at the system boundary. Eẋmass refers to the exergy
Pa k+1 (76)
transfer via mass transfer across the system boundary. Explicit form of
ρb v ( k+ 1)M2a each term in Eq. (93) are given by:
= a =
ρa vb (k−1)M2a + 2 (77) T
Eẋ Q = ∑ ⎛1− T0 ⎞ Q̇K
⎜ ⎟

where k is the ratio of specific heats (cp/cv), cp and cv are specific heats ⎝ K⎠ (94)
at constant pressure and constant volume, respectively. ρa and ρb are
EẋW = Ẇ (95)
density at section a and b, respectively. Ma is the Mach number at cross-
section a. Eẋmass = ∑ ṁ ψ (96)
hb = f (P b,vb) (78)
where subscript “K” signifies the heat transfer surface, Q̇ K is the rate of
Specific enthalpy at the diffuser exit (h16) can be obtained by ap- heat transfer through the system boundary at heat transfer temperature
plying conservation of energy principle as presented below: of TK. Subscript “0” signifies the dead state (P0 = 1 atm., T0 = 25 °C)
and ψ is the specific exergy of a stream which can be expressed as sum
ṁ tot h16 = ṁ 15 h15 + ṁ 21h21 (79)
of physical exergy (ψph) and chemical exergy (ψph) as follows:
Using the isentropic efficiency formula of the diffuser part of the ψ = ψph + ψch (97)
ejector, the actual enthalpy at the diffuser outlet (h16) can also be de-
termined by following:
x ⎞ 0 1−x ⎞ 0
ψ = [(h−h 0)−T0 (s−s 0)] + ⎡ ⎛ e ⎜ +⎛⎟ e ⎜ ⎟

h16,is = f (P16,sb) (80) ⎢ ⎝ MNH3 ⎠ ch,NH3 ⎝ MH2O ⎠ ch,H2O

⎣

ηd =
h16,is−hb
h16−hb
where ηd is the diffuser isentropic efficiency, h16,is is the specific en-
thalpy at the end of the isentropic process in the diffuser, P16 is the
pressure at the exit of the diffuser and sb is the specific entropy at the
diffuser inlet.
Below presented computational procedure is applied to obtain the
rate of heat rejection from condenser II (Q̇cond-II ), rate of heat transfer
between the Kalina cycle and ERC (Q̇ in-II ), rate of heat absorption
+ R T0 ∑ x ln(x)⎤
⎥ (98)
(81) ⎦
where e0ch,NH3
and e0ch,H2O
are the standard chemical exergies of ammonia
and water, respectively. Standard chemical exergies are obtained from
Szargut et al. [72]. M is molecular weight of the substance and R is gas
constant. The term [R T0 ∑ x ln(x)] is the part of chemical exergy for-
mula related to the mixing effect of two fluids but is negligible relative
to the other terms in Eq. (98) and is disregarded.
Mathematical definition of combined cycle exergy efficiency (ηex) is

637
C. Seckin Energy Conversion and Management 157 (2018) 631–643

presented as follows. The rate of exergy supplied to the combined cycle Based on the above given equations, a simulation program is de-
(Eẋin-I ) is the difference of exergy content in the working fluid entering veloped and the flowchart of the program is presented in Fig. 4.
and exiting the HE I.

EẋW + Eẋref Ẇ + Eẋref 4. Results and discussion

ηex = = net
̇
Exin-I ṁ 1 (ψ1−ψ13) (99)
where Eẋin-I is the exergy transfer to the working fluid in HE I, Eẋref is Based on the developed mathematical model and numerical simu-
the refrigeration exergy (exergy of cooling) and EẋW is the exergy of the lation program, a parametric analysis is performed for the combined
produced power by the combined cycle. cycle of Kalina cycle and ERC. In the analysis, the effect of variation in
Given that cooling differs from power in exergy grade, the exergy of combined cycle parameters: turbine inlet pressure (P2), turbine inlet
cooling can be computed according to the following equation: temperature (T2), concentration of ammonia-water basic solution (x1),
condenser outlet temperature (T10) and pressure of refrigerant in HE-II
Eẋref = ṁ 20 [(h21−h20)−T0 (s21−s20)] (100) (PHE-II) on thermal efficiency (ηcc), exergy efficiency (ηex), net power

START

Pcond-II, Peva, PHE-II, mtot, p-II m s d, Am

Set w Det. m15, m21 Eqs. (61-69)

Set P16 Det. h16, h16,is, h16-new Eqs. (79-81)

h15, s15, hm,is, hm, Vm, vm, m15-new

Set Pm Eqs. (55-61) NO If
h16 = h16-new

NO If
If NO
m15,new = m15
P16 = Pcond-II

Det. Qcond-II, Wp-II, Qin-II, Qref

Set Ps Det. h21, s21, hs,is, hs, Cs, Vs
Eqs. (82-89)
Eqs. (62-67)

Qin-I, T2, P2, T10, x1, m1, tur, p-I, TTD

NO If
Cs = Vs

Det. m3, m2, Wtur, h7, h4, h13

Det. vs, As Eqs. (68-69) Eqs. (13-28)

Set Pa Det. Aa, Va, va, ha, ha-new Det. h8, h6, h5, h12, h12-new
Eqs. (70-75) Set T8
Eqs. (29-36)

NO If
ha = ha-new NO If
h12,new = h12

Det. Pb, vb, hb Eqs. (76-78)

Det. Wp-I, Qcond-I Eqs. (37-45)

Det. Wnet, cc Eqs. (90-91)

Fig. 4. Flowchart of the simulation program.

638
C. Seckin
output (Ẇnet ), refrigeration capacity (Q̇ ref ), exergy of produced power
(EẋW ) and exergy of refrigeration (Eẋref ) of the combined cycle is in-
vestigated. The operational conditions and simulation parameters for
Kalina cycle and ERC are reported in Table 1. In the analysis of the
effect of variation in one operational parameter, the other parameters
are kept constant as reported in Table 1.
In Fig. 5, Ẇnet , Q̇ ref and ηcc results of the considered combined cycle
with respect to variation in turbine inlet pressure (P2) is presented. It is
seen in Fig. 5 that, ηcc of the combined cycle is increasing with in-
creasing turbine inlet pressure but, after a certain point of P2, increase
in turbine inlet pressure causes a decreasing trend in ηcc and Ẇnet . It is
very predictable that, as the inlet pressure increases (for constant tur-
bine exit pressure, P7) enthalpy difference between the turbine inlet and
outlet fluxes (h2 − h7) rises. This leads Ẇtur and Ẇnet values to increase
based on Eqs. (23) and (90), respectively. Since Q̇ in-I is constant in the
analysis, ηcc is strongly correlated with Ẇnet (as seen in Fig. 5) based on
Eq. (91). It must be also stated that, Ẇtur is by far the largest constituent
of the net power production (Ẇnet ) of the combined cycle, i.e., magni-
tude of power consumption by pump I (Ẇp-I ) and pump II (Ẇp-II ) is
negligible relative to that of Ẇtur . On the other hand, it is determined
that, as P2 increases, mass flow rate of the ammonia-water solution
which is transferred from the separator to the turbine (ṁ 2 ) decreases.
Based on Eq. (23), Ẇtur is also proportional to ṁ 2 and, after a certain
point of P2, reduction in the mass flow rate of the ammonia rich vapor
flow (ṁ 2 ) dominates the Ẇtur results and Ẇtur (hence, Ẇnet ) decreases.
Presented direct correlation between the ηcc and Ẇnet (in Fig. 5) is also
supported by the constant Q̇ ref values of the combined cycle (based on
Eq. (91)). Constant parameters of ERC (Pcond-II, Peva, PHE-II, etc.) lead the
flow conditions of the refrigerant to remain constant at the HE II inlet
(state-point 14 in Fig. 2). Due to the constant PHE-II, conditions of the
refrigerant at the HE II outlet (state-point 15) and Q̇ in-II are constant
(based on Eqs. (55) and (87), respectively). Thus, variation of P2 does
not affect the ERC operational parameters including Q̇ ref . As a result,
Q̇ ref of the combined cycle is constant as P2 increases which is seen in
Fig. 5.
In Fig. 6, response of the combined cycle to the variation in turbine
inlet pressure (P2) in terms of EẋW , Eẋref and ηex is presented. It is seen
in Fig. 6 that exergy efficiency (ηex) increases but the rate of increase
gets lower with increasing P2. As stated above, since the operational
conditions of ERC are not affected by P2, refrigeration exergy (Eẋref ) is
constant based on Eq. (100). The magnitude of produced power exergy
(EẋW ) is identical with that of the produced power itself (Ẇnet ). Exergy
content of refrigeration (Eẋref ) is much lower than the power produc-
tion of the combined cycle, hence, influence of the EẋW is much greater
on exergy efficiency (ηex) than that of Eẋref (based on Eq. (99)). It must
be stated that, Eẋin-I (in Eq. (99)) reduces with increasing P2. As a result,
rate of increase in ηex gets lower with increasing P2 (as seen in Fig. 6),
due to the combined effect of reducing EẋW and Eẋin-I on ηex, based on
Eq. (99).
In Fig. 7, effect of increasing turbine inlet temperature (T2) on Ẇnet ,
Q̇ ref and ηcc is shown. It is determined that as the turbine inlet tem-
perature increases, the specific enthalpy drop of the ammonia rich so-
lution across the turbine (h2 − h7) increases. Additionally, higher T2
leads to increase in mass flow rate (ṁ 2 ) and concentration (x2) of the
ammonia-water solution at the turbine inlet (for constant concentra-
tion, increase of ammonia vapor with the rise of ammonia-water solu-
tion temperature is also seen in the phase diagram of ammonia-water
solution, available in [5]). Due to the combined effect of increase in
mass flow rate of ammonia-rich vapor (ṁ 2 ) and the growth of specific
enthalpy drop across the turbine (h2 – h7), net power output (Ẇnet )
increases as T2 rises. On the cooling capacity (Q̇ ref ) side of the issue, as a
result of constant operational conditions of ERC (which is explained
above), Q̇ in-II and Q̇ ref remain constant as T2 increases (Fig. 7). It must
be reported that, at the presence of constant Q̇ ref , ηcc values of the
combined cycle is strongly influenced by Ẇnet (based on Eq. (91)) and in
Fig. 7, ηcc results follow a similar pattern with Ẇnet .
In Fig. 8, EẋW , Eẋref and ηex results of the considered combined cycle
639
Energy Conversion and Management 157 (2018) 631–643
with respect to variation in turbine inlet temperature (T2) is presented.
As seen in Fig. 8, exergy efficiency of the combined cycle (ηex) rises with
increasing T2. It is determined that, exergy of the produced power (EẋW )
strongly prevails the ηex results mainly due to slightly decreasing Eẋin-I
in the investigated range of T2. Based on Eq. (99), predominant effect of
EẋW over ηex is also supported by the presence of constant Eẋref mag-
nitudes (as mentioned above). As a result, exergy efficiency of the cycle
(ηex) gets higher with increasing rate of rise as T2 increases.
Effect of ammonia-water basic solution concentration (x1) on the
combined cycle thermal efficiency (ηcc), net power output (Ẇnet ) and
cooling capacity (Q̇ ref ) are presented in Fig. 9. It is determined that, at
constant T2, mass flow rate of the ammonia-rich vapor flow which is
delivered to the turbine (ṁ 2 ) gets higher with increasing concentration
of ammonia-water basic solution (x1) since the amount of ammonia
vapor gets higher at HE I exit (state-point 1 in Fig. 1) as x1 increases
(this result is also supported by phase diagram of ammonia-water so-
lution, available in [5]). Since the power production by the turbine
(Ẇtur ) and hence, net power production of the combined cycle (Ẇnet ) are
dominated by the mass flow rate of ammonia-rich vapor transferred to
the turbine (ṁ 2 ), Ẇnet and ηcc gets higher as x1 increases. Because the
cooling capacity of the combined cycle is constant due to the above
given reasons, Ẇnet and ηcc are positively correlated based on Eq. (91) as
presented in Fig. 9.
As indicated in Fig. 10, exergy efficiency (ηex) curve exhibits a si-
milar trend to that of thermal efficiency (ηcc). Because the exergy of
refrigeration (Eẋref ) is much lower than EẋW , effect of EẋW is much more
prevailing on ηex than that of Eẋref , based on Eq. (99). Hence, the exergy
efficiency and thermal efficiency share the same variation trend with
increasing ammonia-water basic solution (x1) and the trend is primarily
determined by EẋW (Ẇnet ).
In Fig. 11, variation of Ẇnet , Q̇ ref and ηcc as a function of condenser
exit temperature (T10) of the considered combined cycle is presented. It
is obvious that, as T10 gets higher, pressure at the exit of the condenser
(P10) gets higher, as well. This brings the result of higher turbine exit
pressure (P7 in Fig. 1) based on Eq. (4). It is determined that, as the
turbine exit pressure increases, magnitude of the enthalpy difference of
the ammonia water solution through the turbine (h2-h7) drops. Based
on Eq. (23), entropy difference is proportional with Ẇtur and Ẇtur is by
far the largest constituent of Ẇnet (Eq. (90)). As a result, decrease in
entropy difference across the turbine causes Ẇnet to decrease with in-
creasing condenser exit temperature (T10). It must be stated that, in-
creasing T10 causes Q̇ in-I to decrease slightly and decreasing Q̇ in-I can be
viewed as a balancing factor of thermal efficiency results (ηcc), i.e.,
variation of ηcc is quite limited due to the concurrent decrease of the
Ẇnet and Q̇ in-I (based on Eq. (91)). A similar characteristic of Eẋin-I to
that of Q̇ in-I is determined: Eẋin-I slightly gets lower as T10 gets higher in
the investigated range of T10. Variation of the EẋW , Eẋref and ηex of the
combined cycle with increasing T10 is reported in Fig. 12. As seen in
Fig. 12, combined cycle exergy efficiency (ηex) declines as T10 increases
Table 1
Operational parameters and boundary conditions of combined cycle simulation program.
Rate of heat transfer to the combined cycle – Q̇ in-I (kW) 1000
Temperature at turbine inlet – T2 (°C) 120
Pressure at turbine inlet – P2 (kPa) 3500
Temperature at condenser I exit – T10 (°C) 25
Concentration of NH3-H2O basic solution – x1 (%) 70
Terminal temperature difference of LTR-TTD (°C) 10
Isentropic efficiency of pumps – ηp-I, ηp-II (%) 60
Isentropic efficiency of turbine and ejector parts-ηtur, ηm, ηs, ηd (%) 80
Mass flow rate of the ammonia-water basic solution – ṁ 1 (kg/s) 0.98
Condenser II pressure – Pcond-II (kPa) 850
Evaporator pressure – Peva (kPa) 415
Heat exchanger pressure (PHE-II) (kPa) 2635.3
Total mass flow rate of the refrigerant – ṁ tot (kg/s) 0.85
Dead state temperature – T0 (°C) [72] 25
Dead state pressure – P0 (kPa) [72] 101.3
C. Seckin Energy Conversion and Management 157 (2018) 631–643

. .
W net W net
. .
Q ref Q ref
cc cc

Fig. 5. Variation of Ẇnet , Q̇ ref and ηcc with turbine inlet pressure (P2). Fig. 9. Variation of Ẇnet , Q̇ ref and ηcc with concentration of ammonia-water basic solu-
tion (x1).

.
Ex W .
.
Ex W
Ex ref .

Ex ref
ex
ex

Fig. 6. Variation of EẋW , Eẋref and ηex with turbine inlet pressure (P2).
Fig. 10. Variation of EẋW , Eẋref and ηex with concentration of ammonia-water basic so-
lution (x1).

.
W net
.
Q ref .
W net
.
cc
Q ref
cc

Fig. 7. Variation of Ẇnet , Q̇ ref and ηcc with turbine inlet temperature (T2).

Fig. 11. Variation of Ẇnet , Q̇ ref and ηcc with condenser outlet temperature (T10).

.
Ex W
.

Ex ref .
Ex W
.
ex
Ex ref
ex

Fig. 8. Variation of EẋW , Eẋref and ηex with turbine inlet temperature (T2).

since decreasing EẋW (Ẇnet ) exerts a more dominating effect than de-
creasing Eẋin-I and ηex results get lower, based on Eq. (99). Determi-
native influence of EẋW on ηex is supported by constant Eẋref as pre-
sented in Fig. 12. It must be underlined that variation of ηex is quite
limited similar to that of ηcc.
Although the present study mainly focuses on the effect of the
Kalina cycle parameters on the considered performance parameters of
the combined cycle (thermal efficiency, net power output and cooling
Fig. 12. Variation of EẋW , Eẋref and ηex with condenser outlet temperature (T10).
capacity), the effect of PHE-II on the considered performance parameters
is also investigated. PHE-II is the pressure of the refrigerant flow in the
HE II which is also a part of the Kalina cycle. Variation of Ẇnet , Q̇ ref and
ηcc of the combined cycle with PHE-II is presented in Fig. 13. It is seen
that, since PHE-II is also an operational parameter of ERC, it has a direct
impact on Q̇ ref . It is seen in Fig. 13 that, ηcc of the combined cycle is

640
C. Seckin
.
W net
.
Q ref
cc
Fig. 13. Variation of Ẇnet , Q̇ ref and ηcc with pressure of refrigerant in HE II (PHE-II) – for
R134a.
decreasing with increasing PHE-II. It is determined that as PHE-II in-
creases, the corresponding primary nozzle exit pressure (nozzle back
pressure, Pm) increases with concurrent increase in pressure difference
of (PHE-II − Pm), i.e., the rate of increase in Pm is lower than that of PHE-
II. Increasing pressure difference between the HE II and primary nozzle
exit causes the motive flow mass flow rate (ṁ 15) of the ejector to in-
crease. As a result, the motive flow mass flow rate (ṁ 15) increases with
increasing HE II pressure (PHE-II). This is tantamount to saying, for
constant total mass flow rate of the refrigerant (ṁ tot ) in ERC, the sec-
ondary flow mass flow rate (ṁ 21) of the ejector decreases with in-
creasing PHE-II. Finally, increasing PHE-II causes reduction in refrigera-
tion capacity (Q̇ ref ) due to the decrease in secondary flow mass flow rate
(ṁ 21), based on Eq. (89). Meanwhile, decrease of the Ẇnet magnitudes
with increasing PHE-II is also presented in Fig. 13. Above mentioned rise
of motive flow mass flow rate (ṁ 15) as PHE-II increases results in greater
amount of heat transfer (Q̇ in-II ) between the Kalina cycle (state points
3–4 in Fig. 1) and ERC, based on Eq. (87). Hence, temperature of re-
turning ammonia-water solution to HE-I (state-point 13) gets lower and
this causes the amount of produced vapor to decrease at the exit of the
HE-I (state-point 1) which is equal to the mass flow rate of ammonia-
water solution transferred to the turbine (ṁ 2 ) in Kalina cycle. As
mentioned above, since the power production of the turbine (Ẇtur ) is
proportional to the ṁ 2 , Ẇtur and Ẇnet values decrease with increasing
PHE-II, based on Eq. (23) and Eq. (90), respectively. Combined effect of
decreasing Ẇnet and Q̇ ref leads ηcc results to decline with increasing PHE-
II (based on Eq. (91)) as presented in Fig. 13.
In Fig. 14, variation of Ẇnet , Q̇ ref and ηcc results is presented for
operation of ERC sub-cycle with two different refrigerants (R152a,
R290) to analyze also the effect of refrigerant on the system perfor-
mance. Selected refrigerants attained the highest ERC performance in
some of the previous studies [73–75]. Comparing the results in Figs. 13
and 14, it is determined that R290 yields the highest refrigeration ca-
pacity (Q̇ ref ) and thermal efficiency (ηcc). The refrigeration capacity and
thermal efficiency of R152a is slightly higher than those of R134a.
Better refrigeration performance of R290 and R152a is originated by
the higher heat absorption capacity of R290 and R152a in the eva-
poration process relative to that of R134a.
In Fig. 15, EẋW , Eẋref and ηex results of the considered combined
cycle with respect to variation in pressure of refrigerant in HE-II (PHE-II)
is seen. Eẋin-I scarcely gets lower as PHE-II increases with accompanying
decrease of EẋW (which is identical with Ẇnet and the reasons of the
decline in Ẇnet is explained above) as seen in Fig. 15. Owing to the
overwhelming influence of the EẋW on exergy efficiency (ηex), based on
Eq. (99), ηex of the considered combined cycle decreases as PHE-II in-
creases which is shown in Fig. 15.
In Fig. 16, variation of EẋW , Eẋref and ηex results is presented for
operation of ERC sub-cycle with two different refrigerants (R152a,
R290) to show also the effect of refrigerant on the system exergy effi-
ciency (ηex). Although the characteristics of the presented curves are
similar for R134a, R152a and R290 (in Figs. 15 and 16), numerical
results indicate that, high refrigeration capacity brings lower ηex since
641
Energy Conversion and Management 157 (2018) 631–643
the exergy content of the refrigeration capacity (Eẋref ) is much smaller
than the refrigeration energy itself (Q̇ ref ), based on Eq. (100). Hence,
declined work production potential and improved refrigeration capa-
city of R290 and R152a (relative to those of R134a) results in lower
total exergy content of system outputs (EẋW + Eẋref ) and lower exergy
efficiency (ηex) for R290 and R152a, based on Eq. (99).
5. Conclusion
In this present study, a novel combined power and refrigeration
cycle, namely, combination of Kalina cycle and ejector refrigeration
cycle (ERC), is analyzed. In the proposed configuration of the combined
cycle, to recover heat from ammonia poor solution which leaves the
separator at high temperature/pressure, an ejector refrigeration cycle is
inserted into the Kalina cycle which performs refrigeration. Based on
above presented mathematical model, a simulation program is devel-
oped to provide a deeper insight into the combined cycle physical
mechanism. A parametric analysis is performed which enlighten the
effect of combined cycle key parameters (i.e. turbine inlet pressure (P2),
turbine inlet temperature (T2), ammonia concentration of ammonia-
water basic solution (x1), condenser outlet temperature in Kalina cycle
(T10) and pressure of the refrigerant in the heat exchanger which op-
erates between Kalina cycle and ejector refrigeration cycle (PHE-II)) on
performance parameters of the combined cycle such as: combined cycle
thermal efficiency (ηcc), exergy efficiency (ηex) refrigeration capacity
(Q̇ ref ), net power production (Ẇnet ), exergy of produced power (EẋW )
and exergy of refrigeration (Eẋref ). The effect of the Kalina cycle para-
meters are mainly investigated in this study. The effect of PHE-II varia-
tion in HE II is analyzed since HE II is also a part of the Kalina cycle. In
the analysis, refrigerant is R134a in ERC but the effect of different re-
frigerant use (R152a and R290) on the combined cycle performance
parameters is also presented. An extensive discussion of the physical
mechanism behind the variation of the abovementioned operational
parameters is given in Section 4.
The main conclusions which are drawn from the findings of this
study are summarized as follows:
– As the turbine inlet pressure (P2) increases from 2500 kPa to
4500 kPa, Ẇnet initially increases but then drops. The optimum point
of P2 in terms of Ẇnet is determined around 3000 kPa under the
applied operational conditions. The net reduction of Ẇnet through
the investigated range of P2 is determined as 32%. ηcc of the cycle
follows the similar characteristics of the Ẇnet and the net reduction
of ηcc is determined as 22%. In the investigated range of P2, ηex of
the combined cycle increases by 11%.
– As the turbine inlet temperature (T2) increases from 100 °C to
135 °C, Ẇnet and ηcc increase. In the considered range of T2, increase
of Ẇnet and ηcc is 19% and 11%, respectively. In the investigated
range of T2, ηex of the combined cycle increases by 40%.
– In the range of ammonia-water basic solution concentration (x1)
from 0.6 to 0.9, Ẇnet and ηcc of the combined cycle increase by 50%
.
W net − R152a
.
W net − R290
.
Q ref − R152a
.
Q ref − R290
ηcc - R152a
ηcc – R290
Fig. 14. Variation of Ẇnet , Q̇ ref and ηcc with pressure of refrigerant in HE II (PHE-II) – for
R152a and R290.
C. Seckin
.
Ex W
. [10] Marston CH. Parametric analysis of the Kalina cycle. J Eng Gas Turb Power
Ex ref
ex
Fig. 15. Variation of EẋW , Eẋref and ηex with pressure of refrigerant in HE II (PHE-II) – for
R134a.
. [17] Kim KH, Kim G, Han CH. Performance assessment of ammonia–water based power
Ex W − R152a
.
Ex W − R290
.
Ex ref − R152a
. ASME 2003;125(1):55–60.
Ex ref − R290
ηex - R152a
ηex – R290
Fig. 16. Variation of EẋW , Eẋref and ηex with pressure of refrigerant in HE II (PHE-II) – for
R152a and R290.
and 30%, respectively. As x1 increases, increase of ηex of the com-
bined cycle is determined as 30%.
– The response of the system to increasing T10 (between 10 °C and
40 °C) in terms of Ẇnet and ηcc shows reverse pattern, i.e., Ẇnet de-
creases and ηcc increases with increasing T10. Decrease of Ẇnet is
obtained as 15% whereas increase of ηcc (5%) is not very con-
siderable relative to Ẇnet through the considered range of T10.
Variation of ηex is also not very high: 4% decrease as T10 increases
between 10 °C and 40 °C.
– Increase of PHE-II (in the range of 1700–3600 kPa) causes decrease in
Ẇnet , Q̇ ref , ηcc and ηex of the combined cycle. The effect of increasing
PHE-II is determined for three different refrigerants: R134a, R152a
and R290. Q̇ ref and Ẇnet decrease by around 50% and 32%, re-
spectively, for three of the analyzed refrigerants. However, reduc-
tion of ηcc is 39%, 43% and 46% for R134a, R152a and R290, re-
spectively. It is seen that, both of the Q̇ ref and ηcc performance of the
cycle are the highest for R290 and the lowest for R134a. Reduction
of ηex is 39%, 42% and 36% for R134a, R152a and R290, respec-
tively. It is determined that, the highest ηex is attained when R134a
is used as the working fluid in ERC and the lowest ηex is obtained
when R290 is used.
References
[1] Ayou DS, Bruno JC, Saravanan R, Coronas A. An overview of combined absorption
power and cooling cycles. Renew Sustain Energy Rev 2013;21:728–48.
[2] Cakir U, Comakli K, Yuksel F. The role of cogeneration systems in sustainability of
energy. Energy Convers Manage 2012;63:196–202.
[3] Ghaebi H, Parikhani T, Rostamzadeh H, Farhang B. Thermodynamic and thermo-
economic analysis and optimization of a novel combined cooling and power (CCP)
cycle by integrating of ejector refrigeration and Kalina cycles. Energy
2017;139:262–76.
[4] Wang J, Wang J, Zhao P, Dai Y. Thermodynamic analysis of a new combined
cooling and power system using ammonia–water mixture. Energy Convers Manage
2016;117:335–42.
[5] Wang E, Yu Z. A numerical analysis of a composition-adjustable Kalina cycle power
plant for power generation from low-temperature geothermal sources. Appl Energy
2016;180:834–48.
[6] Maloney JD, Robertson RC. Thermodynamic study of heat power cycles. Oak Ridge
National Laboratory report. CF-53-8-43; 1953.
642
Energy Conversion and Management 157 (2018) 631–643
[7] Kalina AI. Combined cycle and waste-heat recovery power systems based on a novel
thermodynamic energy cycle utilizing low-temperature heat for power generation.
ASME paper: 83-JPGC-GT-3; 1983.
[8] Kalina AI. Combined cycle system with novel bottoming cycle. J Eng Gas Turb
Power 1984;106:737–42.
[9] Ibrahim OM, Klein SA. Absorption power cycles. Energy 1996;21:21–7.
1990;112:107–16.
[11] Park Y, Sonntag R. A preliminary study of the Kalina power cycle in connection
with a combined cycle system. Int J Energy Res 1990;14:153–62.
[12] Zhang XX, He MG, Zhang Y. A review of research on the Kalina cycle. Renew
Sustain Energy Rev 2012;16:5309–18.
[13] Zheng DX, Chen B, Qi Y, Jin HG. Thermodynamic analysis of a novel absorption
power/cooling combined-cycle. Appl Energy 2006;83:311–23.
[14] Goswami DY, Xu F. Analysis of a new thermodynamic cycle for combined power
and cooling using low and mid temperature solar collectors. J Solar Energy Eng –
Trans ASME 1999;121(2):91–7.
[15] Xu F, Goswami DY, Bhagwat SS. A combined power/cooling cycle. Energy
2000;25(3):233–46.
[16] Padilla RV, Demirkaya G, Goswami DY, Stefanakos E, Rahman MM. Analysis of
power and cooling cogeneration using ammonia–water mixture. Energy
2010;35(12):4649–57.
and refrigeration cogeneration cycle. Int J Mater Mech Manuf 2013;1(1):36–40.
[18] Hasan AA, Goswami DY. Exergy analysis of a combined power and refrigeration
thermodynamic cycle driven by a solar heat source. J Solar Energy Eng – Trans
[19] Vidal A, Best R, Rivero R, Cervantes J. Analysis of a combined power and re-
frigeration cycle by the exergy method. Energy 2006;31(15):3401–14.
[20] Fontalvo A, Pinzon H, Duarte J, Bula A, Quiroga AG, Padilla RV. Exergy analysis of
a combined power and cooling cycle. Appl Therm Eng 2013;60(1–2):164–71.
[21] Lu S, Goswami DY. Optimization of a novel combined power/refrigeration ther-
modynamic cycle. J Solar Energy Eng – Trans ASME 2003;125(2):212–7.
[22] Lu S, Goswami DY. Theoretical analysis of ammonia-based combined power/re-
frigeration cycle at low refrigeration temperatures. In: Proceedings of solar sunrise
on the reliable energy economy, June 15–20, Reno, Nevada.
[23] Pouraghaie M, Atashkari K, Besarati SM, Nariman-zadeh N. Thermodynamic per-
formance optimization of a combined power/cooling cycle. Energy Convers Manage
2010;51(1):204–11.
[24] Demirkaya G, Besarati S, Padilla RV, Archibold AR, Goswami DY, Rahman MM,
et al. Multi-objective optimization of a combined power and cooling cycle for low-
grade and midgrade heat sources. J Energy Res Technol – Trans ASME
2012;134(3):032002.
[25] Zare V, Mahmoudi SMS, Yari M, Amidpour M. Thermoeconomic analysis and op-
timization of an ammonia–water power/cooling cogeneration cycle. Energy
2012;47(1):271–83.
[26] Tamm G, Goswami DY. Novel combined power and cooling thermodynamic cycle
for low temperature heat sources, part II: experimental investigation. J Solar Energy
Eng – Trans ASME 2003;125(2):223–9.
[27] Tamm G, Goswami DY, Lu S, Hasan AA. Theoretical and experimental investigation
of an ammonia–water power and refrigeration thermodynamic cycle. Sol Energy
2004;76(1–3):217–28.
[28] Martin C, Goswami DY. Effectiveness of cooling production with a combined power
and cooling thermodynamic cycle. Appl Therm Eng 2006;26(5–6):576–82.
[29] Sadrameli SM, Goswami DY. Optimum operating conditions for a combined power
and cooling thermodynamic cycle. Appl Energy 2007;84(3):254–65.
[30] Vijayaraghavan S, Goswami DY. On evaluating efficiency of a combined power and
cooling cycle. J Energy Res Technol – Trans ASME 2003;125(3):221–7.
[31] Goswami DY, Vijayaraghavan S. A combined power and cooling cycle modified to
improve resource utilization efficiency using a distillation stage. Energy
2006;31(8–9):1177–96.
[32] Liu M, Zhang N. Proposal and analysis of a novel ammonia-water cycle for power
and refrigeration cogeneration. Energy 2007;32(6):961–70.
[33] Zhang N, Lior N. Development of a novel combined absorption cycle for power
generation and refrigeration. J Energy Res Technol – Trans ASME
2007;129(3):254–65.
[34] Zhang N, Lior N. Methodology for thermal design of novel combined refrigeration/
power binary fluid systems. Int J Refrig 2007;30(6):1072–85.
[35] Wang J, Dai Y, Gao L. Parametric analysis and optimization for a combined power
and refrigeration cycle. Appl Energy 2008;85(11):1071–85.
[36] Yu Z, Han J, Liu H, Zhao H. Theoretical study on a novel ammonia–water cogen-
eration system with adjustable cooling to power ratios. Appl Energy
2014;122:53–61.
[37] Jing X, Zheng D. Effect of cycle coupling-configuration on energy cascade utiliza-
tion for a new power and cooling cogeneration cycle. Energy Convers Manage
2014;78:58–64.
[38] Hua J, Chen Y, Wang Y, Roskilly AP. Thermodynamic analysis of ammonia–water
power/chilling cogeneration cycle with low-grade waste heat. Appl Therm Eng
2014;64:483–90.
[39] Srinivas T, Reddy BV. Thermal optimization of a solar thermal cooling cogeneration
plant at low temperature heat recovery. J Energy Res Technol – Trans ASME
2014;136(2):021204.
[40] Shankar R, Srinivas T. Coupled cycle with Kalina cycle system and vapor absorption
refrigeration. Proc Inst Mech Eng A – J Power Energy 2014;228(8):953–64.
[41] Shankar R, Srinivas T. Investigation on operating processes for a new solar cooling
cogeneration plant. J Solar Energy Eng – Trans ASME 2014;136(3):031016.
C. Seckin
[42] Zare VA. Comparative thermodynamic analysis of two tri-generation systems uti-
lizing lowgrade geothermal energy. Energy Convers Manage 2016;118:264–74.
[43] Mohammadi A, Mehrpooya M. Energy and exergy analyses of a combined desali-
nation and CCHP system driven by geothermal energy. Appl Therm Eng
2017;116:685–94.
[44] Cao L, Wang JF, Wang HY, Zhao P, Dai Y. Thermodynamic analysis of a Kalina
based combined cooling and power cycle driven by low-grade heat source. Appl
Therm Eng 2017;111:8–19.
[45] Rashidi J, Ifaei P, Esfahani IJ, Ataei A, Yoo CK. Thermodynamic and economic
studies of two new high efficient power-cooling cogeneration systems based on
Kalina and absorption refrigeration cycles. Energy Convers Manage
2016;127:170–86.
[46] Seckin C. Parametric analysis and comparison of ejector expansion refrigeration
cycles with constant area and constant pressure ejectors. J Energy Res Technol –
Trans ASME 2017;139:042005-1–042005-10.
[47] Boumaraf L, Lallemand A. Modeling of an ejector refrigerating system operating in
dimensioning and off-dimensioning conditions with the working fluids R142b and
R600a. Appl Therm Eng 2009;29:265–74.
[48] Wang J, Dai Y, Zhang T, Ma S. Parametric analysis for a new combined power and
ejector absorption refrigeration cycle. Energy 2009;34(10):1587–93.
[49] Xu XX, Liu C, Fu X, Gao H, Li Y. Energy and exergy analyses of a modified combined
cooling, heating, and power system using supercritical CO2. Energy
2015;86:414–22.
[50] Wang JF, Zhao P, Niu X, Dai Y. Parametric analysis of a new combined cooling,
heating, and power system with transcritical CO2 driven by solar energy. Appl
Energy 2012;94:58–64.
[51] Habibzadeh A, Rashidi MM, Galanis N. Analysis of a combined power and ejector
refrigeration cycle using low temperature heat. Energy Convers Manage
2013;65:381–91.
[52] Ghaebi H, Rostamzadeh H, Matin PS. Performance evaluation of ejector expansion
combined cooling and power cycles. Heat Mass Transf 2017;4:1–17.
[53] Barkhordarian O, Behbahaninia A, Bahrampoury R. A novel ammonia-water com-
bined power and refrigeration cycle with two different cooling temperature levels.
Energy 2017;120:816–26.
[54] Sun W, Yue X, Wang Y. Exergy efficiency analysis of ORC (Organic Rankine Cycle)
and ORC-based combined cycles driven by low-temperature waste heat. Energy
Convers Manage 2017;135:63–73.
[55] Cao L, Wang J, Dai Y. Thermodynamic analysis of a biomass-fired Kalina cycle with
regenerative heater. Energy 2014;77:760–70.
[56] Yapıcı R, Ersoy HK. Performance characteristics of the ejector refrigeration system
based on the constant area ejector flow model. Energy Convers Manage
2005;46:3117–35.
[57] Ersoy HK, Yalcin S, Yapici R, Ozgoren M. Performance of a solar ejector cooling-
system in the southern region of Turkey. Appl Energy 2007;84:971–83.
643
Energy Conversion and Management 157 (2018) 631–643
[58] Khalil A, Fatouh M, Elgendy E. Ejector design and theoretical study of R134a ejector
refrigeration cycle. Int J Refrig 2011;34(7):1684–98.
[59] Chunnanond K, Aphornratana S. Ejectors: applications in refrigeration technology.
Renew Sustain Energy Rev 2004;8:129–55.
[60] Sadeghi M, Mahmoudi SMS, Saray RK. Exergoeconomic analysis and multi-objec-
tive optimization of an ejector refrigeration cycle powered by an internal com-
bustion (HCCI) engine. Energy Convers Manage 2015;96:403–17.
[61] Ahmad M, Karimi MN. Thermodynamic analysis of Kalina cycle. IJSR
2016;5:2244–9.
[62] Alexis GK, Karayiannis EK. A solar ejector cooling system using refrigerant R134a in
the Athens area. Renew Energy 2005;30:1457–69.
[63] Ouzzane M, Aidoun Z. Model development and numerical procedure for detailed
ejector analysis and design. Appl Therm Eng 2003;23:2337–51.
[64] Henry RE, Fauske HK. The two-phase critical flow of one component mixtures in
nozzles, orifices, and short tubes. J Heat Transf 1971;93:179–87.
[65] Seckin C. Effect of ejector internal efficiencies on cooling performance of an ejector
expansion refrigeration cycle with a two phase ejector. In: Proceedings of the in-
ternational conference on energy and thermal engineering: ICTE 2017. Istanbul,
Turkey; 2017.
[66] Seckin C. Effect of the primary nozzle throat diameter on the cooling performance
of an ejector expansion refrigeration cycle with a two phase ejector under critical
conditions. In: Proceedings of the istanbul international conference on progress in
applied sciences: ICPAS 2017. Istanbul, Turkey; 2017.
[67] Hassanain M, Elgendy E, Fatouh M. Ejector expansion refrigeration system: ejector
design and performance evaluation. Int J Refrig 2015;58:1–13.
[68] Huang BJ, Chang JM, Wang CP, Petrenko VA. A 1-D analysis of ejector perfor-
mance. Int J Refrig 1999;22:354–64.
[69] Lear W, Parker G, Sherif S. Analysis of two-phase ejectors with Fabri choking. Proc
Inst Mech Eng C J Mech Eng Sci 2002;216(5):607–21.
[70] Shu FH. The physics of astrophysics, vol. 2: gas dynamics. Mill Valley: University
Science Books; 1991.
[71] Carrillo JAE, Sanchez de La Flor FJ, Salmeron Lissen JM. Thermodynamic com-
parison of ejector cooling cycles. Ejector characterisation by means of entrainment
ratio and compression efficiency. Int J Refrig 2017;74:369–82.
[72] Szargut J, Morris DR, Steward FR. Exergy analysis of thermal, chemical, and me-
tallurgical processes. 1st ed. New York: Hemisphere Publishing Corporation; 1988.
[73] Tashtoush B, Alshare A, Al-Rifai S. Performance study of ejector cooling cycle at
critical mode under superheated primary flow. Energy Convers Manage
2015;94:300–10.
[74] Selvaraju A, Mani A. Analysis of an ejector with environment friendly refrigerants.
Appl Therm Eng 2004;24:827–38.
[75] Nehdi E, Kairouani L, Elakhdar M. A solar ejector air-conditioning system using
environment-friendly working fluids. Int J Energy Res 2008;32:1194–201.