Exergy Int. J.

1(2) (2001) 107–111

www.exergyonline.com

Optimal performance of an irreversible solar-assisted heat

pump

E. Torres-Reyes a∗ , J. Cervantes de Gortari b

a
Instituto de Investigaciones Científicas, Universidad de Guanajuato, L. de Retana No. 5, C.P. 36000, Guanajuato, Guanajuato, Mexico
b
Departamento de Termoenergía y Mejoramiento Ambiental, Facultad de Ingeniería, Universidad Nacional Autónoma de México, Mexico

Abstract — The thermodynamic optimization of a mechanically driven solar heat pump is presented. A new expression to describe
the optimal thermal performance under finite operating conditions considering the internal and external irreversibilities during
actual operation is derived. The optimum ratio between the condenser and collector–evaporator conductances (UA) determines the
coefficient of performance (COP) for the maximum heating load of the system. An experimental air-R22 heat pump was used to
determine the traditional performance parameters (COP and second law efficiency) which are compared with those obtained using
the expressions derived in this work. Results show that the new model very closely represents the performance of real systems.
 2001 Éditions scientifiques et médicales Elsevier SAS

Nomenclature τ temperature ratio between TLC and THC

τc temperature ratio between TL and TH
A heat transfer area . . . . . . . . . . . . . m2
Subscripts
COP coefficient of performance
G solar radiation . . . . . . . . . . . . . . . W·m−2 ce collector–evaporator
h enthalpy change . . . . . . . . . . . . . W·kg−1 co condenser
ṁa air mass flow rate . . . . . . . . . . . . . kg·s−1 f refrigerant
max maximum
ṁf refrigerant mass flow rate . . . . . . . . kg·s−1
rev reversible
Q̇i internal heat transfer leak . . . . . . . . W
real real
Q̇H heat transfer absorbed by air stream . . . W
Q̇L heat transfer absorbed by collector . . . W
Q̇LC evaporating heating load . . . . . . . . . W
Q̇HC condensing heating load . . . . . . . . . W 1. INTRODUCTION
Ri internal thermal resistance
TH thermodynamic temperature of air . . . K Thermodynamic finite time concepts are considered
TL ambient temperature . . . . . . . . . . . K a close way of representing the performance of real
THC refrigerant temperature in the condenser K processes. Most of the work done on the subject is based
TLC refrigerant temperature in the evaporator K on the efficiency of a Carnot heat engine for which the
U heat transfer coefficient . . . . . . . . . W·m−2 ·K−1 maximum power output is obtained when the efficiency is
Ẇk electric power . . . . . . . . . . . . . . . W 1 − (TL /TH )1/2 [1]. For the analysis of a reversed power
Greek letters plant there are two relevant forms, the exergy balance
and the Gouy–Stodola relation. By a thermodynamic
α absorptance
finite time analysis, some authors have found the optimal
ηII second law efficiency
operating temperature of solar collection devices used
in thermodynamic cycles [2–4]. Some of these results
∗ Correspondence and reprints. agree with those obtained using the minimization of
E-mail addresses: [email protected] (E. Torres-Reyes), entropy generation method [5] and by the maximization
[email protected]. (J. Cervantes de Gortari). of the exergetic efficiency during the energy collection

 2001 Éditions scientifiques et médicales Elsevier SAS. All rights reserved 107
S1164-0235(01)00016-4/FLA
 E. Torres-Reyes, J. Cervantes de Gortari / Exergy Int. J. 1(2) (2001) 107–111

[6]. There are various arguments for the application of

the different approaches. Since the task of maximizing
the exergy delivered by a collector of fixed cross section
is equivalent to minimizing the rate of entropy generation
from the environmental temperature to the apparent
temperature of the sun.
This work presents a description of the thermal per-
formance of a solar heat pump cycle under finite operat-
ing conditions. A thermodynamic optimization is based
on the definition of the second law efficiency and deter-
mines the optimal performance through the derivation of
the coefficient of performance for finite operating con-
ditions for the maximum heating load. Optimal perfor-
mance is related to the optimal condenser and collector–
evaporator conductances. This derivation considers the
internal and external irreversibilities of the heat pump cy- Figure 1. Heat pump model with two heat exchenger (con-
denser and collector–evaporator).
cle as described elsewhere [7, 8].

fluid plus the heat losses due to internal cycle irreversibil-

2. FINITE-SIZE AND FINITE-TIME MODEL ities, and this is expressed as,
DESCRIPTION
Q̇HC = Q̇H + Q̇i (3)

A heat pump is analyzed using a model that accounts where Q̇HC is the refrigerant heating load in the con-
for both, the internal and the external irreversibilities denser. The thermal energy (known as the useful heat)
that originate during the heat pump process, as shown given to the refrigerant by the collector–evaporator, Q̇LC
in figure 1. From the overall energy balance about the can be expressed as
system in figure 1, we have,
Q̇LC = Q̇L + Q̇i (4)
Ẇk = Q̇H − Q̇L (1)
Q̇LC is determined by the climatic conditions and by
where Q̇L is the energy absorbed by the collector– the physical characteristics of the solar collector where
evaporator. This term is the product between the solar evaporation takes place, [6]. It can be also represented by
irradiation and the collector surface area, GAce , and Q̇H
is the energy delivered by the refrigerant to an airflow on Q̇LC = ṁf (h)f,ce (5)
its pass through the condenser (ṁa h)co .
There are two sources of irreversibilities in this model. The thermal performance of the system can be rep-
These are: (a) the exergy losses due to internal irre- resented by the coefficient of performance COPreal in a
versibilities Q̇i , originated by the temperature difference general way as
between TH and TL , and (b) the finite temperature differ- Q̇H
ences (THC − TH ) and (TL − TLC ), associated with the COPreal = (6)
Ẇk
heat transfer between the working fluids, represented by
where Ẇk represents the electric power supplied to the
TH − TL cycle.
Q̇i = (2)
Ri

where Ri is the resistance to the internal heat transfer be-

tween TH and TL that correspond to the air thermody- 3. SECOND LAW EFFICIENCY
namic temperature and the ambient temperature.
The heat delivered by the refrigerant during conden- The second law efficiency for the heat pump can be
sation is the sum of the heating load given to the cooling expressed as the ratio between the reversible work and

108
 E. Torres-Reyes, J. Cervantes de Gortari / Exergy Int. J. 1(2) (2001) 107–111

the electric work supplied to the system per unit time Therefore, these conductances can be expressed as

Ẇrev (UA)co = xUA (15)

ηII = (7)
Ẇk
and
and as a function of the exergy associated with the heat (UA)ce = (1 − x)UA (16)
transfer, Q̇H , by
where x represents the fraction of the total conductance.


Q̇H TL Combining equation (1)–(4) and from (12)–(16), and
ηII = 1− (8) simplifying, the following expression results
Ẇk TH


Defining τc and τ as 1 1
Q̇H = x(1 − x)UA TL − − Q̇i (17)
τc τ
TLC
τc = (9)
THC This expression shows the heating load as a function
of the ratio of conductances. By differentiating Q̇H with
and respect to x and setting the result to zero, the optimum
TL conductance ratio for the maximum heating load is
τ= (10)
TH
xopt = 1/2 (18)
and combining equations (1)–(4) with (8), and using the
definitions given in (9) and (10), and introducing the
entropy balance, and rearranging gives This result agrees with the findings of other authors,
where similar derivations have been applied to refriger-


1−τ Q̇i −1 ation systems and other types of thermodynamic cycles
ηII = 1− (11) [9–11].
1 − τc Q̇H
Combining equation (14) and (17) and substituting
Equation (11) represents the second law efficiency of xopt , the maximum heating load can be expressed as
an irreversible solar heat pump under finite operating 1 
conditions, as a function of the heating load. Q̇H,max = (UA)co + (UA)ce TL
4


−1 
TLC −1 TL
× − − Q̇i (19)
THC TH
4. MAXIMIZATION OF THE HEATING
LOAD According to general definition given by equation (6),
the COP for maximum heating load is
The heating load supplied by the heat pump can be −1
maximized in order to obtain the optimal ratio of the 4 UA TL [τc
1
− τ −1 ] − Q̇i
COPmax = (20)
conductances of the two heat exchangers of the system. Ẇk
The heat flows Q̇HC and Q̇LC can be expressed as
This equation represents the coefficient of perfor-
Q̇LC = UAce (TL − TLC ) (12) mance of an irreversible heat pump at finite operating
conditions.
and
Q̇HC = UAco (THC − TH ) (13)
where the conductances are (UA)ce and (UA)co corre- 5. EXPERIMENTAL STUDIES
spond to the collector–evaporator and condenser of the
heat pump, respectively. Figure 2 shows a schematic of the experimental sys-
The total conductance of the system is the sum of the tem. The R-22/air condenser (CD-101), made of stainless
two conductances steal, has a heat transfer area Aco of 9.8 m2 . The com-
pression stage is carried out by a hermetic compressor
UA = (UA)co + (UA)ce (14) (CP-101) and expansion of the R-22 proceeds through a

109
 E. Torres-Reyes, J. Cervantes de Gortari / Exergy Int. J. 1(2) (2001) 107–111

Figure 2. Experimental solar-conventional assisted heat pump.

Figure 3. Average temperature of refrigerant (THC ) and the Figure 4. Average temperature of the refrigerant in the
air thermodynamic temperature (TH ) in the condenser as a condenser (THC ) and evaporator (TLC ) as a function of the solar
function of time. radiation.

The temperature variation of the refrigerant and the

thermal expansion valve (VP-107). The evaporation step
air stream during condensation for a typical experiment
takes place within a solar flat collector (ES-101) with a
as shown in figures 3 and 4.
solar energy collection area Ace of 4.5 m2 . The system
was built to operate with a conventional cross-flow ex-
changer as an evaporator (EC-101) as well.
Experimental tests were carried out within a solar 6. RESULTS
radiation range of 200 to 1 100 W·m−2 and ambient
temperature TL variations between 20 and 32 ◦ C. The real Figure 5 shows that the second law efficiency values
amount of work supplied to the compressor was taken as obtained using equation (7) fall within the same range
the amount of electricity consumed. as those obtained from equation (11). The latter was

110
 E. Torres-Reyes, J. Cervantes de Gortari / Exergy Int. J. 1(2) (2001) 107–111

calculated for an irreversible heat pump as an extension 7. CONCLUSIONS

of the exergy method [12]. It was applied to the model
described previously in [7]. The values in figure 5 are The results obtained form the expression derived for
shown against the heating load. the under finite operating conditions for the heat pump
The real and the maximum heating load coefficients of very closely agree with those obtained through the exergy
performance were calculated from equations (6) and (20), or the entropy generation method. This is so since these
respectively, and are shown in figure 6. On working out three methodologies are applied to real systems and are
the values of COPmax , the term UA was taken as the sum based on the same theoretical principles.
of the condenser and evaporator conductances. These Expressions for the maximum heating load and for
vary with the climate conditions as observed during the the under finite operating conditions were also derived.
experimental tests. These expressions allow establishing the optimum ther-
Second law efficiency values computed using equa- mal performance as a function of the optimal ratio of ex-
tion (7) (exergy efficiency) vary from 0.067–0.13 and the ternal conductances.
ones calculed with equation (11) are between 0.083 and The model proposed for the coefficient of performance
0.14. The coefficient of performance of the heat pump for the maximum heating load of the solar assisted system
computed using equation (6) and equation (20) show val- that describes closely the behavior of the experimental
ues in a range among 2.56–4.36. The range of variation system. The predicted values are close to the real COP’s
of the resulting values are due mainly to variations of cli- measured experimentally.
matological conditions. These results are directly given
through the variations (2.8–5.37 kW) of the maximum REFERENCES
heating load calculated with equation (19) and the real
heating load measured. Also through the variations regis- [1] Curzon F.L., Ahlborn B., Efficiency of a Carnot
engine at maximum power output, Amer. J. Phys. 43 (1975)
trated (approximatly 6.5 K) of the temperature difference 22–24.
between the condenser and evaporator, which are among [2] Gordon J.M., in: Sieniutycz S., Salamon P. (Eds.),
43.2–49.8 K. Finite-Time Thermodynamics and Thermoeconomics, Tay-
lor & Francis, New York, 1990, pp. 95–120.
[3] Lund K.O’F., in: Sieniutycz S., Salamon P. (Eds.),
Finite-Time Thermodynamics and Thermoeconomics, Tay-
lor & Francis, New York, 1990, pp. 95–120.
[4] Lin G., Yan Z., in: ECOS ’96 STOCKHOLM, Effi-
ciency, Costs, Optimization, Simulation and Environmental
Aspects of Energy Systems, June 25–27, 1996, pp. 141–
145.
[5] Bejan A., in: Entropy Generation Minimization. The
Method of Thermodynamic Optimization of Finite-Size
Systems and Finite-Time Processes, CRC Press, New York,
1996, pp. 21–31, 227–231, 286.
[6] Torres-Reyes E., Picón-Núñez M., Cervantes-de G.J.,
Energy—Internat. J. 23 (4) (1998) 337–344.
Figure 5. Second Law efficiency computed using equations (7) [7] Bejan A., Theory of heat transfer-irreversible refrig-
and (11), respectively, as a function of the heating load. eration plants, Internat. J. Heat Mass Transfer 32 (9) (1989)
1631–1639.
[8] Agrawal D.C., Menon V.J., Finite-time carnot refrig-
erators with wall gain and products loads, J. Appl. Phys. 74
(1993) 2153–2158.
[9] Bejan A., A relation between thermodynamic per-
formance and economics for heat engines, J. Energy Res.
Technology 115 (1993) 148–150.
[10] Bejan A., Vargas J.V.C., Sokolov M., Optimal alloca-
tion of a heat exchanger inventory in heat driven refriger-
ators, Internat. J. Heat Mass Transfer 38 (16) (1995) 2997–
3004.
[11] Ibrahim O.M., Klein S.A., J.W. Mitchell, A relation
between thermodynamic performance and economics for
heat engines, ASME AES 24 (1991) 15–20.
Figure 6. COPreal and COPmax as a function of the temperature [12] Kotas T.J., The Exergy Method of Thermal Plant
difference between the condenser and evaporator. Analysis, Butterworths, London, 1985.

111