ICSET 2008

Abstract This paper presents a hybrid technology for a

renewable energy power generation, which is developed as a
small-scale power plant. The proposed system is a new concept
of small-compressed air energy storage (S-CAES) integrated
with induction generator. The system consists of 3 main
components: air compressor, energy storage system and power
generation. The air compressor uses a general reciprocating
engine that produces a compressed air. It is stored in
pressurized vessels. The energy storage system uses S-CAES,
which is developed as a small volume and installed above
ground to avoid the site limitation as the conventional CAES
does. Therefore, it can be located in any location. The S-CAES
system is operated in low pressure that not more than 10 bar,
so, it easy to available component in country and inexpensive.
The induction generator (IG) is used in power generation,
which does not have combustion units as the conventional
CAES that used gas turbine. Thus, the system does not burn
fuel and exhaust pollution. This paper expresses the modeling
and simulation to obtain the characteristic of a new concept of
S-CAES power plant, which can be helpful in system designing
of renewable energy conversion.
I. INTRODUCTION
owadays, the energy consumption from fossil fuel
combustion is one cause of global warming phenomena.
Therefore, the efficient energy using by avoid combustion is
considered. The alternative choice is renewable energy, such
as, biomass, wave or tidal, water, wind and solar energy.
However, they are unreliable energy source because the
power, when is needed, cannot produce all time. Thus, the
more economical and efficient energy storage system is
needed. The compressed air energy storage system (CAES)
is the one choice, which are long service period, low cost of
energy, low cost of maintenance and operation and high
power efficiency [1]. The CAES power plant produces
power by storing energy, during off-peak periods, in the
form of compressed air and use on demand during the peak
periods to generate power with a turbo-generator/gas turbine
system. However, this system seems to be disadvantage as it
is quite large power facility and is needed large underground
carven, while having a limitation in terms of site installation.
This system have successfully implemented in Hantorf in
Geramany, McIntosh in Alabama, Norton in Ohio ,a municipality
in Iowa, in Japan and under construction Israel.
Varin Vongmanee and Veerapol Monyakul are with The Joint Graduate
School of Energy and Environment, King Mongkuts University of
Technology Thonburi, 91 Prachauthit Rd. Bangmod, Tungkru, Bangkok,
Thailand, 10140, Tel: (662)4708309-10 Ext. 4115-6 Fax: (662)8726736,
email: [email protected]* and [email protected]**, respectively.
Fig. 1 Small Compressed Air Energy Storage System Power Plant
The CAES is applied on the hybrid energy storage system
based on CAES and Super-capacitors energy storage with
maximum efficiency point tracking (MEPT) algorithm. It is
implemented for the real time optimization of energy
conversion by using power electronics converter, which is
applied to the thermodynamic converter to obtain a variable
power. This system improves flexibility and dynamic
performances of the storage system and suitable for a wide
range of low power. The advantage over classical CAES
plants include site independence, fuel free operation and
environmental harmlessness [2].
The transportable compressed air energy storage (T-
CAES) system is applies from classical CAES. It discussed
herein consists of 2.5 MW of wind turbine, the logic circuit
for input power acceptance, a 500 kW electric motor,
compressor, and turbo-expander/generator, storage tank
and power conditioner. T-CAES system uses ambient
temperature compressed air and does not use combustion.
The output of this system is electrical power and chilled air
that obtain in expansion process. The draw back of this
system is work at high wind speed because the system needs
the electrical power from wind turbine for driven compressor
to produce compressed air. The thick tanks about 3-inch are
used for store compressed air in high pressure, it is special
order and expensive [3]. Therefore, it is considered not
appropriate for using as a small renewable power generation
in Thailand.
This paper, the small-compressed air energy storage (S-
CAES) integrated with induction generator is proposed that
is shown in figure 1. This system is a composite technology
of energy storage and electric power generation. The energy
will transfer to the S-CAES system by using the air
compressor, which produces high-pressure compressed air at
ambient temperature. It stores in above ground pressure vessel,
which is a temporary storage. When it full stored or on peak
condition, it will be supplied to drive the air expansion unit
A New Concept of Small-Compressed Air Energy Storage System
Integrated with Induction Generator
Varin Vongmanee, Veerapol Monyakul
N
866
978-1-4244-1888-6/08/$25.00 c 2008 IEEE
that the shaft coupling with induction generator to electric
generation, while traditional CAES use gas turbine that
complicates mechanic and very expensive. The induction
generator operates in variable speed to available more
power. In theory, the variable speed drives can produce 15%
to 30% more energy output compared with constant speed
[4]. The AC/DC Converter works as variable speed control
of induction generator and converts the alternating current
(AC) to the direct current (DC). The Grid Connected Converter
is used to invert from DC to AC for connecting with the power
line as well as control the constant voltage and the frequency. In
addition, the power quality is controlled in this section.
This paper will determine the modeling and simulation to
obtain the characteristic of a new concept of S-CAES power
plant, which can be helpful in system designing of
renewable energy conversion. The paper is consisted of 5
sections as follow: I) Introduction, II) Theory, III)
Simulations and Results, IV) Conclusion and V) Reference.
II. THEORY
The energy conversion process of the proposed system
consists of three main components: air compressor, energy
storage vessel and compressed air generator. The simplified
diagram is shown in figure 2. The energy conversion starting
with the compressor; the electrical power from renewable
energy is converted into mechanical torque for suck and
compress air from the atmosphere and store in the tank. Then
the compressed air transmits the air power to drive the prime
mover coupling with generator, which the mechanical power
converts back to electric power. The compressed air is
released back into atmosphere again. The energy flow and
air state changing of the proposed system is described in
figure 3.
Fig. 2 Simplified Diagram of Proposed System
Fig. 3 Energy Flow and Air State Changing.
The corresponding of two parameters, torque (T) and shaft
speed (N) of electric power, are converted to two parameters
of air power, pressure (P) and flow (Q) by using compressor.
Then, the pressure and flow are converted back to torque and
speed for drive generator. Thus, two theories, which are the
thermodynamic of air energy conversion and variable speed
induction generator, are considered.
A. Thermodynamic Model of Air Energy Conversion
The thermodynamic of air energy conversion both
compression and expansion are considered an isothermal
process which the temperature is constant. The ideal gas law,
which relates temperature (T), pressure (P) and volume (V),
is used to derive the power and work. The relationship is
PV nRT = , where n is the number of mole and R is the
gas constant. In addition, the first law of thermodynamics
(Q-W=U) is used as U=0 because of the isothermal process,
T = 0 that result to Q=W where W is work done on the
system, Q is work in or out of the system and U is internal
work of the system.
1) Air Compression Process: work for producing the
compressed air as isothermal process is shown with the
integration that
2 2
1 1
1 2
2 1 1 1 2 1
(ln ln )
(ln ln ) (ln ln )
V V
V V
nRT
W PdV dV nRT V V
V
nRT P P PV P P
= = =
= =
) )
(1)
where P
1
,P
2
and V
1
,V
2
are the pressure and volume of the
initial state and the final state that are atmospheric state and
compressed air state, respectively. However, power is work
per unit time and air flow (Q) is volume per unit time
(Q=V/t), so the compression power,(P
c
) is

1 2 1
(ln ln )
c
P PQ P P = (2)
Applying the electric power, which is P
e
=VI , and the
compressed air power equation (2), the compressed air flow
driven by electric power can be expressed by
1 2 1
*
(ln ln )
V I
Q
P P P
=

(3)
where, V,I are voltage and current, respectively. We can see
that power is directly depend on pressure and air flow, which
vary with the available power from renewable energy.
2) Compressed Air Expansion Process: The work
potential of compressed air can calculate as the same
equation (2) but the air state change is opposite that is
compressed air state change to atmospheric state (show in
figure 3). Therefore, in during compressed air discharge to
drive the reciprocating machine that the shaft couple with
shaft of generator, the air pressure in vessel will decrease. As
the result, the speed of generator will change following as
condition leading to electric power generation is change.
867
The energy efficiency of system may be determined by the
ratio of isothermal expansion power to electric input power
as follow:

1 2 1
(ln ln )
*
*
s
PQ P P
V I

=
(4)
where
s
is the efficiency of compressor, compressed air
engine and electric machine, respectively. Moreover, the
both powers must multiply with time of compression until
full store and time of expansion until empty tank to be same unit.
B. DQ Induction Machine Model
(a) d-axis circuit
(b) q-axis circuit
Fig. 4 dq induction machine model in stationary reference frame
The induction machine can be modeled in dq stationary
reference frame circuits as shown in figure 4. The dq model
is powerful for analyzing the transient and steady state
conditions, which give complete dynamic solution. The
matrix equations are given as following[4-6]:
0 0
0 0
qs qs s s m
ds ds s s m
qr qr m r m r r r r
r m m r r r r dr dr
v i R pL pL
v i R pL pL
v i pL L R pL L
L pL L R pL v i


( ( (
+
( ( (
( ( ( +
= ( ( (
+ ( ( (
( ( (
+
( (
(5)
where R
s
,R
r
are stator and rotor resistance, L
ls
,L
lr
are stator
and rotor inductance, L
m
is mutual or magnetizing
inductance,
qs ds
, are dq magnetic flux of stator,
qr dr
, are dq magnetic flux of rotor,
qs ds
v v , are dq satator
voltage,
qr dr
v v , are dq rotor voltage,
qs ds
i i , are dq satator
current,
qr dr
i i , are dq rotor current,
r
is rotor angular velocity.
C. DQ Induction Generator Model
Generally, the induction generator requires the
reactive power, which will supply the excitation current
to the core and generate a rotating magnetic field, for
build up the output voltage. However, the excitation
current is provided from an external source such as
battery, charged capacitor and grid utility. One
application connects the charged capacitor across terminal
stator and drives rotor shaft by external prime mover, the
voltage at stator terminal will be build up. This process
call self excited induction generator. The matrix
equations for dq model in stationary reference frame
without load are shown as following,[4].
1
0 0
0
1
0 0 0
0
0
s s m
qs
ds s s m
qr
m r m r r r r
dr
r m m r r r r
R pL pL
i
pC
i R pL pL
pC
i
pL L R pL L
i
L pL L R pL


(
+ +
(
( (
(
( (
(
+ +
( (
= (
( (
(
+
( (
(
(
(
+

(6)
When it connects with load, ) (
L
R ,(considered resistive load
only), the output voltage (
qs ds
v v , ) are following:
qs qs qs
L
dv i v
dt C R C
=
(7)
ds ds ds
L
dv i v
dt C R C
=
(8)
The general mechanical equation in generating mode is given as

m
m e
d
T T J
dt

+ =
(9)
where
e
T
is electromagnetic torque and J is inertia. Finally,
the mechanical power required to drive the induction generator
is given by

m m m
P T = (10)
where,
m
T
is mechanical torque in shaft,
m
is mechanical
shaft speed. The whole equations are used to determine
the algorithm of field oriented control, FOC, for induction
generator.
D. The Model of CAES Integrated with IG
In field oriented control, the relationship of flux,
) (
* e
ds
i
and torque,
) (
* e
qs
i
, producing current that are function of
electromagnetic torque are determined as command to
control. The
) (
* e
ds
i
sets the machine flux level, which is
defined constant so it able to produce full torque. Set
) (
* e
ds
i
equals to magnetizing current,
) (
m
i
, which is determined
by machine open circuit test. R
s
is neglected
because of
s m
R L >> . Therefore,
868

) ( *
*
m ls e
e
ds
L L
V
i
+
=

(11)
At steady state, the rotor flux linkage is constant that
results to
* e
r m ds
L i = , therefore, the relation of the
electromagnetic torque and control current are shown as

* *
'
2
2 2
3
e
qs
e
ds
r
m
e
i i
L
L P
T =
(12)
Set the electromagnetic torque equal to mechanical torque
and the mechanical power equals to the air expansion power,
which is the prime mover of generator. Apply equation (2),
(10) and (12). Therefore, the torque current demand
) (
* e
qs
i
is

e
ds
r
m
m
e
qs
i
L
L P
P P Q P
i
2
1 2 1 *
2 2
3
*
) ln (ln *

=
(13)
The current
) (
* e
qs
i
and ) (
* e
ds
i relate with every equations,
therefore, the both currents are command in control. The
current ) (
* e
ds
i will control the machine flux linkage and the
current
) (
* e
qs
i
will control the machine torque and generated
the reference slip frequency, ) (
sl
that correspond to the
efficiency of drive system is shown as:

*
*
*
e
ds
e
qs
r
r
sl
i
i
L
R
=
(14)
The slip frequency is added to the instantaneous rotor
speed,
r
, in order to produce the angular frequency, ( )
e
.
Then,
e
is integrated to give the flux angle, ( )
e
which is
expressed as:

)
+ = dt
r sl e
) ( (15)
Since, the control current
) (
* e
qs
i
and
) (
* e
ds
i
are expressed in
synchronous rotating reference frame. They are
considered same as dc quantities, they will be changed
to ac quantities in stationary reference frame, which are
expressed as:

+ =
+ =
+ =
) sin cos (
2
1
) cos sin (
2
3
-
) sin cos (
2
1
) cos sin (
2
3
) cos sin (
* * * * *
* * * * *
* * *



i i i i i
i i i i i
i i i
e
qs
e
ds
e
qs
e
ds
s
cs
e
qs
e
ds
e
qs
e
ds
s
bs
e
qs
e
ds
s
as
(16)
Fig. 5 Compression Work when Vary Compression (or Expansion) Pressure
Fig. 6 Compression Power when Vary Compression (or Expansion)
Pressure and Air flow
III. SIMULATIONS AND RESULTS
The simulation consists of two sections. First is to
find the thermodynamic relation of air energy
conversion process both compression and expansion.
Second is to find characteristic of induction generator is
driven by compressed air engine. Thermodynamic
simulates under conditions, 2 cylinder of compressor,
temperature T=293 K. and Atmospheric pressure=1 atm.
Figure 5 shows the compression (or Expansion) work
from compressor when change the compression (or
Expansion) pressure. We can see that the compression
(or Expansion) work will increase following the
pressure. This compression (or Expansion) work is the
energy that must use for producing the compressed air.
Figure 6 expresses the power from compression (or
Expansion) process under conditions of compression (or
Expansion) pressure and air flow change. The
compression power, which is power consumption,
directly depends on air flow rate and compression (or
Expansion) pressure.
Figure 7 express that the energy efficiency of the
system under change of expansion pressure. Assuming
200 minute for time of compression until full store of
20,000 liter and 50 minute for time of expansion until
empty tank at 0.4 m
3
/min of air flow rate to drive
generator, and 0.85 of
s
. As the result, we can see that the
energy efficiency of the system is increase follow
869
increase of pressure ratio. Therefore, we can improve
the system efficiency by increase the stage of expansion.
Although more stage can increase efficiency, but the
system is complexity and high initial and maintenance
cost.
Fig. 7 Energy Efficiency under Change of Expansion Pressure
The induction generator simulation is implemented by
using MATLAB/SIMULINK to determine the behavior
of the drives system. It consists of 2 simulations:
constant supply voltage and frequency condition and
open loop variable speed field oriented control. The
simulation starts at compressed air engine, the power
from expansion process under various air pressures at
constant air flow, 1000 l/min and atmospheric pressure 1
atm, will determine by Eq. (2). After that mechanical
torque will be calculated by Eq.(10). Both values are
shown in Table I.
TABLE I.
THE EXPANSION POWER AND TORQUE CALCULATION
Expansion
Pressure (MPa)
0.2 0.4 0.6 0.8 1.0
Power (kW) 1.16 2.31 2.99 3.45 3.84
Torque(Nm) 0.77 1.54 1.99 2.30 2.56
The first simulation, the supply terminal voltage of
induction machine is set constant frequency at 50 Hz.
The mechanical torque, 0.77, 1.99 and 2.56 N.m, from
Table I are applied to the induction machine model at
time 1.5, 2.5 and 3.5 second, respectively. Figure 8
shows the dynamic response of electromagnetic
torque,
( )
e
T
, rotor speed, ( )
r
, and line currents,
( )
abc
i
, will
change as mechanical torque command. It reaches to
steady state at 1 second in which the induction machine
works as motor mode. After that the mechanical torque
are applied starting at 1.5 sec., the electromagnetic
torque is negative in which the induction machine works
as generator mode.
Finally, the open loop variable speed field oriented
control is simulated by change the speed command and
Fig. 8 Dynamic Respond of Induction Generator to Vary the Mechanical
Torque from Air Expansion
Fig. 9 MATLAB/SIMULINK Diagram of the open loop variable speed field
oriented control
assume flux constant. This condition is simulated
closely to the characteristic of the air pressure supplying
from the tank changed. The SIMULINK diagram shows
in figure 9. The simulation results show the response of
voltage
( )
dq
v
, current
( )
dq
i
, line current
( )
abc
i
, phase voltage
( )
a
v
, rotor speed ( )
r
and the electromagnetic torque
( )
e
T
that change follow speed command.
In figure 10, we can see that the rotor speed,
r
, line
current,
a
I
,and phase voltage,
a
v , change as
m

change.
These results mean that induction generator extracts the
power from air expansion engine.
* e
ds
i and
* e
ds
v are constant
because we assume flux constant.
* e
qs
i ,
* e
qs
v and
e
T are
negative and change follow
m
change. It means that the
induction machine is working in generator mode.
870
IV. CONCLUSION
This study is to determine the modeling and simulation
to obtain the characteristic of a new concept of S-CAES
power plant.. The simulation results show that, the
compression and expansion pressure directly depends on
air flow rate and system efficiency. Therefore, to improve
the system efficiency of thermodynamic conversion of S-
CAES, the system should be able to operate at the high
pressure by increase the stage of expansion, which the
system is complexity, high initial and maintenance cost.
For Induction generator simulation result, it will show the
dynamic response, which run on motoring and generating
mode. It can extract the compressed air power follow as
input speed conditions. Therefore, it can produce the
power in wide speed range.
The simulation results can be helpful in system designing
for the renewable energy conversion system, which can
use the local available contents resulting to low initial
cost.
The proposed system can apply to uninterruptible
power supply, peak shaving for the energy building
management ,pneumatic application and air power vehicle
(APV)
REFERENCES
[1] R.C. Bansal et al. On some of the design aspects of wind energy
conversion systems, Energy conversion and management,2002, 43,
2175-2187.
[2] http://leiwww.epfl.ch/publications/lemofouet_rufer_cyphelly_barrade
_grasser_store_03.pdf, Access at January 10,2008.
[3] E Ben et al, Transportable compressed air energy storage (T-CASE)
system driven by a 2500 kW wind turbine, In: Proceeding of
international conference on efficiency, cost, optimization, simulation
and environment impact of energy system, 2006.
[4] B.K. Bose, "Power Electronics and Variable Frequency Drives",
IEEE Press.,1997.
[5] D.W. Novotny, T.A. Lipo, "Vector Control and Dynamics of AC
Drives", Clarendon Press Oxford,1997.
[6] Chee-Mun-ong, "Dynamic Simulation of Electric Machinery",
Prentice Hall PTR, 1997.
Fig. 10 Dynamic Respond of Induction Generator when change the speed of
prime-mover
871