1 s2.0 1359431196000038 Main
1 s2.0 1359431196000038 Main
1 s2.0 1359431196000038 Main
907-919, 1996
~ Pergamon P I h S1359-4311(96)00003-8
Copyright © 1996 Elsevier Science Ltd
Printed in Great Britain. All rights reserved
1359-4311/96 $15.00 + 0.00
E X E R G Y A N A L Y S I S F O R THE E V A L U A T I O N OF A
T H E R M A L S T O R A G E SYSTEM E M P L O Y I N G P C M S W I T H
DIFFERENT MELTING TEMPERATURES
R o m a n D o m a f i s k i a n d G i u m a Fellah*
Warsaw University of Technology, Institute of Heat Engineering, 00-665 Warsaw,
Nowowiejska 25, Poland
NOMENCLATURE
Greek symbols
fl operational factor
e operational factor
#. dynamic viscosity (N s m 2)
~P rate of the supplied exergy (W)
exergy (J)
0 dimensionless time (rhcTocpt/ML)
r dimensionless temperature ( T - To)/To
rhs, first-law efficiency
r/_,nd second-lawefficiency
2~ summation
Subscripts
0 ambient condition
1 unit (1)
2 unit (2)
c charging
d discharging
e exit
f final
i inlet
g gas
gen generation
m melting
P due to pressure drop
Q due to cooling the heat transfer fluid to ambient condition
T due to temperature difference
INTRODUCTION
A treatment of sensible heat storage units based on exergy analysis was documented by Bejan [1].
The analysis was performed for a storage process and indicated that the amount of exergy loss
for best second-law efficiency is about 50% of the supplied exergy. An analysis of a complete
charging-discharging cycle of sensible-storage systems was performed by Krane [2]. The analysis
indicated that about 70-90% of the supplied exergy is lost by irreversibility. The pioneer work of
using exergy analysis for the evaluation of phase-change thermal-storage systems was introduced
by Bjurstr6m and Carlson [3]. Their analysis indicated that for a complete charging~iischarging
cycle the second-law efficiency could be as low as 12%. Other articles published by Adebiyi and
Russel [4], Krane [5], De Lucia and Bejan [6, 7], Taylor et al. [8, 9], Charach and Zemel [10],
Charach [11] and Saborio et al. [12], among others, have indicated that the second-law efficiency
is low, and thus could be improved by utilizing multi-storage units in series. Lim et al. [13]
documented the advantages of employing more than one storage unit in series. In their analysis,
a charging process was performed by using a hot working fluid (gas), while a discharging process
was performed by transferring heat to reversible heat engines. The analysis showed the optimal
two-element's design promises to produce 28.2-29.8% more power than the optimum
single-element design. Their analysis was based on a steady-state power production. In other words,
the thermal energy is transferred immediately from the heat transfer fluid to heat engines through
the storage elements. Thus, the thermal energy could be transferred directly to the heat engines
without even employing the storage elements. In reality, the necessity of storing energy is raised
from the time gap between the generation and the use of energy. A discharging process in the
present analysis is accomplished by using a cold heat transfer fluid which is more realistic. The
analysis is limited to phase-change materials at their phase-change temperatures. Seeking an
improvement in the second-law efficiency, the thermal capacity of both units are sized such that
they complete their phase change at the same period of time. In other words, the process of charging
(discharging) is terminated when the two storage units complete their phase transition. The present
analysis is performed by using the iteration technique and extends the analysis of Lim et al. [13]
to include the effect of mass-flow rate, charging temperature, and the ratio of the number of
heat-transfer units (NTUc2/NTUcl) on the best second-law efficiency.
M O D E L L I N G OF A T H E R M A L E N E R G Y - S T O R A G E SYSTEM
The model assumed is depicted in Fig. 1, which shows a thermal energy-storage system during
a charging~lischarging cycle. The system consists of two storage units such that the upstream unit
Evaluation of a thermal storage system 909
has the higher of the two melting temperatures. The whole storage system is well insulated. The
charging process is performed by closing valves C and D then allowing a constant mass flow rate
(rhc) of hot air at Tot (higher than T~) and Pc~ to enter the system through valve A. The heat transfer
fluid leaves the storage system at Tec and Pec and then flows through valve B. The heat transfer
fluid is then cooled to the surrounding conditions. The surrounding conditions are To (298 K) and
P0 (0.1 MPa). During the discharging process both valves A and B are closed. A constant mass
flow rate (#td) of cooled air at Td2 (lower than Tm2) and P~2 is allowed to enter the system from
the opposite side of the storage system through valve C and leaves the storage system through valve
D at Ted and Pc~.
The recovered exergy will be used for some purposes that are out of the range of the present
study. Air is assumed to be the working fluid with constant thermal properties and behaving as
an ideal gas. It is also assumed the number of heat transfer units for each unit would be the same
during the charging-discharging cycle. Lumped system analysis is adopted and pure materials,
initially at their phase-change temperatures, i.e. either completely solid or completely liquid, are
assumed. To avoid an excessive waste of exergy to the atmosphere, the problem is formulated such
that both units complete their phase change at the same period of time.
The optimum performance is found to be at the minimum entropy generation number. The
influence of different designing and operational parameters on the exergy loss is sought.
Energy balance
The charging process. An energy balance for unit (1) results in
Tc~ - To2
(2)
Y c l - T o , - Tin, "
insuhtion
A Tml, M I
I I T~, P~
D C
B , - L ~ - Tm,
Tm, (8)
The p a r a m e t e r fll must always be greater than zero. We m a y define a dimensionless temperature
as
T-To
- To (9)
~'cl -- "/Tml
J~l -- "1 ~ G " (1 1)
As T~2 must always be greater than Tin2, we then introduce a n o t h e r dimensionless factor fl, such
that
"Cc2 - - ~'m2
(15)
i -G-
The effectiveness o f each unit can be written as
Tel - - ~'c2
(16)
Yc, - /~1(1 + Tml)
"~c2 - - Tec
Yc2 -- /32(1 + %2) "
(17)
A n energy balance for each unit gives the fraction o f energy to be stored in unit (1):
M1Lj
Q' - ML = O~yc,fl,(1 + "Cml) • (23)
M2L2
Q2 - ML - O~vc2fl2(1 + Tin2); (24)
O~ is the charging period for b o t h units (the thermal capacities o f both units are sized such that
b o t h units complete their phase change at the same period o f time). Since
Q, + Q2 = 1.0, (25)
we find
1
0c = f l l Y ¢ l ( l + .Cml) "1- fl2Yc2(l "1- ,,Cm2) . (26)
1
Q' = flzvc2(1 + Zm2) " (27)
1 + fl,y¢,(1 + Zm,)
T . I - T~
(28)
Yd, -- T d l - Tml "
T d-2- Tdl
Yd2- T d 2 - Tin2 (30)
Y02 = 1 -- e - NTU"2, (3 l)
Tin2 must always be greater than Td2 so a new dimensionless factor e2 is introduced such that
Td, must always be less than Tin,. We introduce a n o t h e r factor et such that
The effectiveness during the discharging process can be written for unit (1) as
or
0d-- 0~ = m.'~ 1
ma /3tyro(1 q- "Cml) + /32YR2(1 + Zm2) " (41)
The heat transfer rate during discharging is
Qo = rhdCp(Teo- To.,). (42)
Second-law analysis
The charging process. The rate of entropy generation may be written as
dS Q0 (45)
S'e"= -aT +'~XSec--S¢,)+ Too"
The first term on the right-hand side of equation (45) represents the rate of increase (decrease) of
entropy within the storage material, the second term represents the entropy rate change between
the inlet and the outlet of the heat exchanger, and the third term represents the entropy transfer
as a result of cooling the heat transfer fluid to ambient temperature (during charging only). Here
Qo =/~/ccp(Tec - - To), (46)
(~)¢.T
= T00
O~{fl~y~+ln[l+fl'(1-Y~') ]-~ ]}. (49)
Evaluation of a thermal storage system 913
insulation
r
/ T~a, M2
.... 00
MAR
oc{
¢,T = Too /~,Yc2+ In 1 + ~-2 "
(50)
Entropy generation due to cooling the heat transfer fluid to ambient conditions (CV2) may written
as
Control volume (2) is a hypothetical control volume simulating the process of cooling the heat
transfer fluid at Toc, P~c to atmosphere at To (298 K) and P0 (0.1 MPa).
Entropy generation (for both units) due to frictional effect (assume Pec = P0) may be written as
ML/c,p= ~ [ k---~l
k In( ~Co~)] ' (52)
z,¢ is given by
The discharging process. It can be shown that the entropy generation due to finite temperature
difference for unit (1) can be written as
where
Entropy generation (for both units) due to friction (assume P_, = P,,) may be written as
(57)
p=Cz[(h-ho)-T,(s-so)]. (58)
(59)
(60)
!P
Pdl + h) - 1nU + h)(l + PI)1 + (61)
ML= PIYcdll+ Lll)
h +
-=Yd
I[e
1
-ln[(e)+ yln($$)]. (62)
ML Y&*(1 + k)
Adding equations (61) and (62) gives the total exergy supplied during the cycle. The entropy
generation number can now be written as
To~Sgen
NC ~ (63)
Y, + Yy, ’
qlnd. = 1 - N . (66)
Evaluation of a thermal storage system 915
The o p t i m u m parameters are defined as those corresponding to the minimum exergy loss.
Irreversibility attributed to the later use of the recovered exergy is not considered in our analysis.
A computer program in F O R T R A N based on iteration technique is developed to search for those
parameters and their corresponding entropy generation numbers. To find the influence of mass flow
rate on the best second-law efficiency, we assign some values to the operational and designing
parameters, namely, Zcl = 0.25 (Tcx = 1.25T0), zd2 = 0.0 (Td2 = To), mass flow ratio rhd/rhc= 1.0,
and NTU¢I = NTU~2 = 2.5. Different values of dimensionless mass velocities are considered, they
range between t~ = 0.01 (G = 3.42 kg/m2.s) and (~ = 0.1 (G = 34.2 kg/m2.s). In the iteration
process, the parameter fl~ is varied between 0.1 (Tc~ = 1.1 Tin0 and 0.5 (Tc~ = 1.5Tm0, and f12 is varied
between 0.01 (Tc2 = 1.01 Tin0 and 0.5 (T¢: --- 1.5Tin2). For each value of (~, the program searches for
the minimum entropy generation number taking into consideration that T~ is always greater than
T~, and T~2 is greater than Tm2.
Figure 3 shows the influence of mass velocity on best (optimum) second-law efficiency. It drops
from a value of 48.2% at 0 = 0.01, to a value of 15.1% at 0 = 0.1. The influence of charging
temperature (T~,) on best second-law efficiency, optimum melting temperatures, and optimum
fraction of the total thermal energy stored in the first unit is shown in Figs 4 - 6. Values assigned
to the operational and designing parameters are N T U ~ = NTUc2 = 2.5, Zd2 = 0.0, (~ -~ 0.05, and
mass flow ratio = 1.0. Different values of dimensionless charging temperatures between 0.2
(Tel = 1.2T0) and 1.0 (T~ = 2.0T0) are considered. The best second-law efficiency reaches its
m a x i m u m value of about 66.1% at a dimensionless charging temperature of 0.7. The optimum
melting temperature of unit (1) varies linearly with the charging temperature. This behaviour is
needed to reduce the irreversibility due to the temperature difference between the heat transfer fluid
and the storage material. For a dimensionless charging temperature between 0.2 and 0.7, the
optimum melting temperature of unit (2) is found to be as small as possible (close to ambient
temperature), in other words both Tin2 and To~theoretically approach the ambient temperature, that
0.50 i I
\
"\..
0.40
\
\\
q 0.30
2ncl
\
0.20
0.10
I ' I l I ' I '
0.02 0.04 __ 0.06 0.08 0.10
G
Fig. 3. Effect of dimensionless mass velocity on best second-law efficiency
916 R. Domafiski and G. Fellah
1.00 I I I I t
0.~0 --
/
0.40 /
--
/
/'
/
/
/
020
I I ' l ' I '
0.20 0.40 0.60 O.BO 1.00
1;
c
1.00
First Unit
+ Second Unit
0.80 J)
1;
m
0.60
jf/
0.40
0.20
0.00
i I
0.20 0.40 0.60 0.80 1.00
,¢
c
0.60 I i I i I
'i \
0.50 -
\
\,
\
QI 0.40
'\
",\
0.30 - \
\
v x
0.20
' I t I f I '
0.20 0.40 0.60 0.80 1.00
"C
c
0.36 I I I , I I I I
/
/
0.32
rlz~
/
0.28
0.24
' I ' I ' I '
0.40 0.80 1.20 1.60 2.00
RNTU =, N T U 2 / N T U 1
Fig. 7. Dependence of best second-law efficiency on the ratio of number of heat transfer units.
918 R. Domafiski and G. Fellah
the finite temperature difference between heat transfer fluid and storage material. However, the
amounts of exergy and energy rejected by the heat transfer fluid to the surroundings still dominate
and a decrease in both efficiencies is observed. The foregoing discussion indicates the need for
employing multi-storage units (more than two) for large charging temperatures, such that the last
storage unit should have a melting temperature close to the ambient temperature.
An interesting observation of the effect of charging temperature on the fraction of total energy
stored in unit (1) corresponding to best second-law efficiency is shown in Fig. 6. Here we have zd2
= 0.0, (~ = 0.05, mass flow ratio = 1.0, and N T U d = NTUc2 = 2.5. The fraction decreases from
a value of 57.2% at Zcl = 0.2, to 22.9% at r~, = 0.7. The reason for this behaviour can be deduced
from equation (27). An increase in Z~l causes an increase in Zml (seeking best second-law efficiency).
In other words equation (27) tells us when high ~1 and Tm| are employed, the storage capacity of
unit (1) should be lowered (to ensure both units complete their phase-change at the same period
of time). This can be achieved either by reducing the mass of the storage material or by employing
materials with low latent heat or both. According to equation (25), the storage capacity of unit
(2) should be increased. When Q~ reaches its minimum value at Tc~ = 1.7T0, Q2 [or the storage
capacity of unit (2)] attains its m a x i m u m value. For a charging temperature higher than Tcj = 1.7 To,
more exergy will be lost to the surroundings.
The dependence of the best second-law efficiency on the ratio R N T U is shown in Fig. 7. Here
we let zd2 = 0.0, 6; = 0.05, and mass flow ratio (rhdrhc = 1.0. Different values of R N T U are
considered for NTUc~ = 2.5 and 5.0. For NTUc~ = 2.5, the best second-law efficiency increases from
a value of 0.26 at R N T U = 0.2 to a value of 0.36 at R N T U = 1.0. This result is in excellent
agreement with that given by Lim et al. [13]. We observed a decrease in the second-law efficiency
for larger values of R N T U . For NTUcl = 5.0, the best second-law efficiency increases from a value
of 0.28 at R N T U = 0.2 to a value of 0.32 at R N T U = 0.6, again we observe a reduction in best
second-law values for larger values of R N T U . Those values of R N T U depend on the NTUc~, such
that small values (of R N T U ) are required (to have best second-law efficiency) for large NTUc~ (to
reduce the frictional effect) and vice versa. In other words, when N T U d is large, the second unit
should have a small number of the heat transfer units. A system of practical interest is selected
to show the advantages of using a two-storage system over a single store system. The number of
heat transfer units identical to that considered by Lira et a/.[13] is considered, i.e.
N T U d = NTU~2 = 0.9. Other operational parameters are selected such that: rc~ = 1.0 (Tel = 2.0T0),
z~2 = 0.0 (Td2 ---- To), mass flow ratio = 1.0, G = 0.05 (G = 17.1 kg/m2.s). The present analysis
indicates such a system would have a best second-law efficiency of 31.4%. A single storage unit
with N T U = 1.8 and under the same flow condition (re = 1, and (~ = 0.05) according to Domafiski
and Fellah [14] would give a best second-law efficiency of 22.4%.
CONCLUSION
REFERENCES
1. A. Bejan, Two thermodynamic optima in the design of sensible heat units for energy storage. A S M E J. Heat Transfer
100, 708-712 (1978).
2. R. J. Krane, A second-law analysis of the optimum design and operation of thermal energy storage systems. Int. J.
Heat Transfer 30, 43-57 (1987).
3. H. Bjurstrom and B. Carlson, An exergy analysis of sensible and latent heat storage. Heat Recovery Systems & CHP
5(3), 233-250 (1985).
4. G. A. Adebiyi and L. D. Russell, A second-law analysis of phase-change thermal-energy storage systems. A S M E HDT
80, 9-20 (1987).
5. R. J. Krane, A second-law analysis of a thermal energy storage system with Joulean heating of storage element. ASME
paper No. 85-WA/HT-19 (1985).
6. M. De Lucia & A. Bejan, Thermodynamics of energy storage by melting due to conduction or natural convection.
A S M E J. Solar Energy Engng, 112, 110-116 (1990).
7. M. De Lucia and A. Bejan, Thermodynamics of phase-change energy storage: the effects of liquid superheating during
melting, and irreversibility during solidification. A S M E J. Solar Energy Engng 113, 2-10 (1991).
8. M. J. Taylor, R. J. Krane and J. R. Parsons, Second-law optimization of a sensible heat thermal energy storage system
with a distributed storage element--part 1: development of the analytical model. A S M E J. Energy Resources Technol.
113, 20-26 (1991).
9. M. J. Taylor, R. J. Krane and J. R. Parsons, Second-law optimization of a sensible heat thermal-energy storage system
with a distributed storage element--part 2: presentation and interpretation of results. A S M E J. Energy Resources
Technol. 113, 2%32 (1991).
10. C. Charach and A. Zemel, Thermodynamic analysis of latent heat storage in a shell-and-tube heat exchanger. A S M E
J. Solar Energy Engng 114, 93 99 (1992).
11. C. Charach, Second-law efficiency of an energy storage-removal cycle in a phase-change material shell-and-tube heat
exchanger. A S M E J. Solar Energy Engng 115, 240-243 (1993).
12, S. A. Saborio, H. Nakamura and G. M. Reistad, Optimum efficiencies and phase change temperatures in latent heat
storage systems. A S M E J. Energy Resources Technol. 116, 79-86 (1994).
13. J. S. Lim, A. Bejan and J. H. Kim, Thermodynamic optimization of phase-change energy storage using two or more
materials. A S M E J. Energy Resources Technol. 114, 84-90 (1992).
14. R. Domafiski and G. Fellah, Exergy analysis for the optimum performance of phase-change thermal-storage units.
Archives ~?f Thermodynamics, Poland (accepted).