Design and optimization of a small
compressed air energy storage
system for isolated applications
Hanif Sedigh_Nejad,
Dr. Tariq Iqbal, Dr. John Quaicoe, and Dr. Benjamin Jeyasurya
Memorial University of Newfoundland, St. John’s
May 2021
1
Outline
Motivations
Research methodology
Hybrid system design & modeling
System performance Assessment
Summary and conclusion2
Motivations
Reducing the fossil fuel consumption for isolated loads.
Environmental Impact
Difficulty of fuel delivery in winter
CAES system to support the wind based energy system
Random nature of the wind energy
Economical and technical challenges
Cost effective design of a wind based energy system.
Increase the harvested energy from the available wind energy
Enhance the reliability and economical feature of RES3
Research methodology
Develop a hybrid configuration
CAES system component modeling
Wind and load data generation
Evaluate the performance of different
energy storage systems
4
Energy Conversion in CAES
Charging or discharge cycles
Isothermal
Polytrophic ( 1<n<1.4)
Adiabatic (n=1.4)
2
1
21 1
1
( )
V
isothermal
V
PW P dV PV ln
P
1 ( 1)/1 1 1 1 1 2
1
( ) 1 ( ) 11 1
n n n
Polytropic
f
nPV V nPV PW
n V n P
5
Hybrid wind-diesel-CAES system
6
Wind turbine modeling
Wind turbine
Power Curve
Datasheet
Approximation
3
0.5 ( ) wind Turbine p w r wP C V A V
3 4
32
22 22
4
21
1
( )( ) ( ) ( )
1 32 4( )
ww w wVV V V
w
bb b b
cc c c
pC a e aV e a e a e
0 5 10 15 20 25 300
0.05
0.1
0.15
0.2
0.25
0.3
0.35
Wind Speed [m/s]
Cp
Valu
e f
or
Excel-
R 7
.5kW
W
ind
Tu
rbin
e
Calculated Cp Value
Approximated equation for Cp
0 2 4 6 8 10 12 14 16 18 200
1000
2000
3000
4000
5000
6000
7000
8000
9000
10000
Wind Speed [m/s]
Win
d T
urb
ine o
utp
ut
po
wer
[W]
Excel-R Manufacturer power curve
Calculated power curve
7
CAES System components
Compressor
1( )
. 11
n
nComp in Comp
nP P Q PR
n
1( )
( 1)
( 1)CS
Comp Nstage
Comp n
nN
CS in
n PQ
nN P PR
( ) ( )Comp
comp out in
T Comp T Comp
KP s P s
J s B
. .out Comp Copmm Q dt . nP V mRT
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 20
1
2
3
4
5
6
7
8
9
10
11
Time [s]
Ou
tPu
t p
res
su
re [
Bar]
8
Air Motor
Steady state model
2
2 1 0( , ) ( ) ( ) ( )mAM AM AM PW AM AM PW AM AM PW AMP n p C p n C p n C p
0 1000 2000 3000 4000 5000 6000 7000 80000
0.5
1
1.5
Speed [RPM]
Ou
tpu
t P
ow
er
[W]
1.4 bar
2.8 bar
4.2 bar
5.6 bar
7.0 bar
9
Air Motor Cont.
Dynamic model
Volume change
2 2 21 1) +( ( )( ) 2
2 4a AM sAM rAM AM AM AM AM sAMV L r r L e sin L e sir n c
2 2 21 1( ( )( ) 2
2 4)b AM sAM rAM AM AM AM A MM sAV L r r L e sin L e r sin c
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
1
1.5
2x 10
-5 (a): Va variation @ 300 rpm speed
time [s]
Ch
am
ber
A v
olu
me [
m3]
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
1
1.5
2x 10
-5 (a): Vb variation @ 300 rpm speed
time [s]
Ch
am
ber
B v
olu
me [
m3]
10
Air Motor Cont.
Dynamic Model
Pressure change
Developed Torque
Air motor Drive train
a s a a a
a a
dP RkT dm kP dV
dt V dt V dt
b s b b b
a b
dP RkT dm kP dV
dt V dt V dt
2 2( , , ) ( )( )2
AMAM a b a b a rAM
LT P P P P x r
AM AM AM d
dJ B T T
dt
11
Synchronous Generator
Block diagram
Experimental test results
0 0.2 0.4 0.6 0.8 1 1.20
50
100
150
200
250
Excitation current [A]
Op
en
cir
cu
it v
olt
ag
e [
V]
500 RPM
750 RPM
1000 RPM
1250 RPM
1500 RPM
1750 RPM
2000 RPM
0 0.2 0.4 0.6 0.8 1 1.20
0.2
0.4
0.6
0.8
1
1.2
Excitation current [A]
Op
en
cir
cu
it f
lux [
Wb
]
500 RPM
750 RPM
1000 RPM
1250 RPM
1500 RPM
1750 RPM
2000 RPM
( , ) ( )SG f re SG f reE I K I
3 20.4545( ) 1.911( ) 2.554 0.018( 24)f f f fI I II 12
Diesel Generator
Fuel consumption
Dynamic model
8 2 5(2.15 10 ) (6.29 10 ) 0.8782DG DG DGFC P P
1( ) [ ( ) ( )]rDG mDG d
DG DG
s T s T sJ s B
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 50
50
100
150
200
250
300
350
400
Time [s]
Sh
aft
Sp
eed
[ra
d/s
]
13
Supervisory Control Unit
14
Experimental Setup
15
Air motor model validation
16
Flow rate valve model
Flow coefficient
0
2fQ A
PC
P1 (Psi) 35 40 50 60
P2 (Psi) 28.200 31.560 35.860 42.275
Speed (rpm) 1150 1350 1475 1600
Output Flow rate (CFM) 30.448 35.301 42.879 49.195
Air consumption based on Datasheet (CFM) 29 34 42 50
Error (%) 4.992 3.825 2.094 1.610
Cf value based on ΔP=P1-P2 0.410 0.416 0.401 0.411
Cf value based on ΔP=P1-Patm 0.236 0.245 0.253 0.256
Cf value based on ΔP=P2-Patm 0.289 0.303 0.326 0.328
35 40 45 50 55 600
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
Pressure [Psi]
Ca
lcu
lete
d C
F V
alu
es
CF ( P)
CF (P1)
CF (P2)
35 40 45 50 55 60
0.23
0.24
0.25
0.26
Ca
lcu
late
d C
F v
alu
e b
as
ed
on
me
as
ure
d P
1 v
alu
es
Pressure (Psi)
35 40 45 50 55 600
2
4
6
Err
or
(%)
CF (P1)
Error (%)
17
Flow rate control
18
CAES system power control
Dynamic
Steady state
19
CAES system power control
Output voltage,
current and
power
20
Hybrid system design optimization
Impact of
Wind turbine selection
Energy storage system rating
Control strategy
Energy storage type
on total fuel consumption in an isolated application21
Wind speed databases
Available Wind data (1hr averaged)
Limited accuracy
Limited resolution.
Unreliable prediction of wind farm output power
Standard statistical methods to
regenerate wind speed data with desired time resolution.
Combining multiple databases
22
Wind speed distribution
Weibull probability distribution
( )1( ) ( )
kV
k Ck V
f V eC C
c is the scale factor
k is shape factor 23
Wind speed frequency distribution
0 10 20 30 40 50 600
0.05
0.1
0.15
0.2
0.25
Wind Speed [km/hr]
Win
d S
pe
ed
Pro
ba
bilit
y
h01
h02
h03
h04
h05
h06
h07
h08
h09
h10
h11
h12
h13
h14
h15
h16
h17
h18
h19
h20
h21
h22
h23
h24
0 10 20 30 40 50 600
0.05
0.1
0.15
0.2
Wind Speed [km/hr]
Win
d S
peed
Pro
bab
ilit
y in
(12:0
0am
– 1
:00am
)
0 10 20 30 40 50 600
0.05
0.1
0.15
0.2
0.25
Wind Speed [km/hr]
Win
d S
peed
Pro
bab
ilit
y in
(12:0
0p
m –
1:0
0p
m)
measured wind speed probability
approximated Weibull distribution equation
measured wind speed probability
approximated Weibull distribution equation
24
Monte Carlo simulation
Iteration process based on a specific probability
distribution function.
Monte Carlo simulation error = 1/√n (more than
1500 samples will result in less than 2.5% error)
Direct sampling method was applied to the
Monte Carlo simulation
25
Monte Carlo simulation, Cont.
A set of 1500 uniformly distributed numbers
between [0-1] was produced and applied to the
inverse of the Weibull Cumulative Distribution
Function of each hour
26
Monte Carlo simulation, Cont.
Proposed Method configuration
27
Wind Speed profile regeneration
0 10 20 30 40 500
5
10
15
20
25
30
35
40
45
50
Time [minute]
Win
d S
pe
ed
[k
m/h
r]
0 10 20 30 40 500
5
10
15
20
25
30
35
40
45
50
Time [minute]
Win
d S
pe
ed
[k
m/h
r]
0 10 20 30 40 500
5
10
15
20
25
30
35
40
45
Time [minute]
Win
d S
pe
ed
[k
m/h
r]
0 10 20 30 40 500
5
10
15
20
25
30
35
40
Time [minute]
Win
d S
pe
ed
[k
m/h
r]
0 10 20 30 40 500
5
10
15
20
25
30
35
40
Time [minute]
Win
d S
pe
ed
[k
m/h
r]
0 10 20 30 40 500
5
10
15
20
25
30
35
40
45
Time [minute]
Win
d S
pe
ed
[k
m/h
r]
0 10 20 30 40 500
5
10
15
20
25
30
35
Time [minute]
Win
d S
pe
ed
[k
m/h
r]
0 10 20 30 40 500
5
10
15
20
25
30
35
40
Time [minute]
Win
d S
pe
ed
[k
m/h
r]
0 10 20 30 40 500
5
10
15
20
25
30
35
Time [minute]W
ind
Sp
ee
d [
km
/hr]
28
Sample generated wind profile
first 3 hours (12:00 am – 3:00 am) in 3
days (1st, 7th and 14th) in January with 10
minute resolution
29
Mathematical model of Wind turbine
Obtain the Cp variation as a function of wind speed
Wind turbine model
Rating
Power
[KW]
Rotor
Diameter
[m]
Tower Height
[m]
Survival
Wind Speed
[Km/hr]
Sky Stream 2.4 3.72 13.7 (zone 3) 226.8
Wisper 500 3 4.5 13.7 198
Excel-5 5 6.2 24 216
Scirocco 6 5.6 24 216
Excel-R 7.5 7 24 201
Excel-S 10 7 24 201
𝐶𝑝 = 𝑎1𝑒−(
𝑊𝑆−𝑏1𝑐1
2
+⋯+ 𝑎4𝑒−(
𝑊𝑆−𝑏4𝑐4
2
Parameter Value Parameter Value Parameter Value
𝑎1 0.2 𝑏1 5.963 𝑐1 2.218
𝑎2 0.1555 𝑏2 3.825 𝑐2 1.228
𝑎3 0.2439 𝑏3 10.06 𝑐3 4.004
𝑎4 0.1056 𝑏4 15.59 𝑐4 5.769
Bergey Excel-S
30
Wind turbine performance assessment
Applying the wind speed profile to 6 different
wind turbines.
Different power rating
The wind speed profile should be corrected
based on the required wind turbine tower height.
22 1
1
( )h
V Vh
31
Wind turbine performance assessment, Cont.
Annual Average output power
SkyStream and Wisper 500 cannot provide the required power
Excel-S is considered an overdesign
Scirocco, Excel-5 and Excel-R able to meet the demand 32
Wind turbine performance assessment, Cont.
Excel-5 wind turbine has the highest value and it can deliver
the required power to the load.
SkyStream and Wisper 500.
Scirocco, Excel-R and Excel-S
Average Annual output PowerPerformance index
Rated output Power
33
Harvested Energy Index
New criterion for energy storage performance assessment
Limited capacity of energy storage systems
Large amount of energy in a short time
Significant wind speed fluctuation
Rejected energy
Wind turbine control
Dump load in isolated applications
0 0
( ( ) ) / ( ( ) )s st t
stored excessHEI E t dt E t dt 34
Case study for storage system sizing
Wind power and demand
HEI for different storage system ratings
0 200 400 600 800 1000 1200 14000
1000
2000
3000
4000
5000
6000
7000
8000
9000
10000
11000
Time [minute]p
ow
er
[W]
Wind Power
Load demand
0 200 400 600 800 1000 1200 14000
0.2
0.4
0.6
0.8
1
minute
Ha
rves
ted
En
erg
y In
dex
storage capacity = unlimited
storage capacity = 20kWhr
storage capacity = 15kWhr
storage capacity = 10kWhr
storage capacity = 5kWhr
storage capacity = 3kWhr
storage capacity = 1kWhr
35
HEI for CAES system
Wind power and demand
HEI for CAES system
0 200 400 600 800 1000 1200 14000
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Time [minute]
Harv
este
d E
nerg
y In
dex
10 bar
12 bar
15 bar
20 bar
30 bar
40 bar
50 bar
0 200 400 600 800 1000 1200 14000
1000
2000
3000
4000
5000
6000
7000
8000
9000
10000
11000
Time [minute]p
ow
er
[W]
Wind Power
Load demand
36
Max. HEI tracking control strategy
Maximum HEI tracking
Compression cycle configuration0 200 400 600 800 1000 1200 1400
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
minute
Ha
rves
ted
En
erg
y In
dex
10 bar
12 bar
15 bar
20 bar
30 bar
40 bar
50 bar
HEI maximum point track
Decrease in harvetsed Energywithout considering HEI in
control strategy
Increase in harvested energyconsidering HEI in control strategy
37
Daily Averaged HEI
Averaged HEI for fixed compression ratios
Comparison0 10 20 30 40 50 60 70 80 90 100
15
20
25
30
35
40
Working pressure [Bar]
Ha
rves
ted
En
erg
y In
dex
[%
]
1 2 3 4 5 6 7 8 90
5
10
15
20
25
30
35
40
45
50
Scenario Number
Ha
rves
ted
En
erg
y In
dex
[%
]
#1: 5 bar, #2: 10 bar, #3: 15 bar, #4: 20 bar, #5: 25 bar, #6: 35 bar, #7: 50 bar, #8: 80 bar, #9:max HEI track
38
Impact of control on total shortage
Tank pressure
Total shortage0 200 400 600 800 1000 1200 1400
0
5
10
15
20
25
30
35
40
Time [minute]
Tan
k P
ressu
re [
bar]
Fixed 10 bar
Fixed 20 bar
Fixed 26 bar
Fixed 30 bar
Fixed 40 bar
Fixed 50 bar
HEI 4Stages
HEI 25 stages1000 1050 1100
4
6
8
10
12
14
Control method Fixed 10 bar Fixed 26 bar HEI in 25 stages HEI in 4 stages
Shortage Duration 326 [min] 225 [min] 163 [min] 192 [min]
39
Diesel Generator Fuel consumption
On-off mode, only on shortage duration
Standby operation
No load fuel consumption
Control method Fixed 10 bar Fixed 26 bar HEI in 25 stages HEI in 4 stages
Shortage Duration 326 [min] 225 [min] 163 [min] 192 [min]
0 10 20 30 40 50 60 70 80 90 1001160
1180
1200
1220
1240
Pressure [Bar]
To
tal F
uel co
nsu
mp
tio
n [
L]
Standy by Operation of Diesel Generator
0 10 20 30 40 50 60 70 80 90 100200
300
400
500
Pressure [Bar]
To
tal F
uel co
nsu
mp
tio
n [
L]
ON OFF Operation of diesel generator
40
HEI for Battery storage system
Battery terminal voltage
SOC
0exchangedBQrated
rated exchanged
QE E K Ae
Q Q
exchangedQ idt
0 1000 2000 3000 4000 5000 6000 70000
2
4
6
8
10
12
Time [s] B
att
ery
Vo
lta
ge
[V
]
0.6 Ahr
1.2 Ahr
3 Ahr
4.8 Ahr
0 0.5 1 1.5 2 2.5 3
x 105
0.7
0.75
0.8
0.85
0.9
0.95
1
Time [s]
Sta
te O
f th
e C
harg
e
1kWhr
3kWhr
5kWhr
7kWhr
41
HEI for Battery storage system
Power balance
Stored, Delivered and rejected power
0 1 2 3 4 5 6 7 8
x 104
0
200
400
600
800
1000
Time [s]
Sto
red
Po
we
r[W
]
0 1 2 3 4 5 6 7 8
x 104
0
500
1000
Time [s]
Delivere
d P
ow
er[
W]
0 1 2 3 4 5 6 7 8
x 104
0
2000
4000
6000
8000
Time [s]
Re
jecte
d P
ow
er[
W]
0 1 2 3 4 5 6 7 8
x 104
-10000
-8000
-6000
-4000
-2000
0
2000
4000
6000
Time [s]
Po
wer
bala
nce [
W]
42
Annual HEI for BES system
Annual HEI
Fuel consumption0 50 100 150 200 250 300 350
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Day A
nn
ual C
HE
I
1 kW
2 kW
3 kW
5 kW
7 kW
10 kW
1 2 3 4 5 6 7 8 9 100
500
1000
1500
2000
2500
3000
Power Rating [kW]
To
tal F
uel co
nsu
mp
tio
n [
Lit
er]
43
Pumped Hydro Energy Storage
Annual HEI
Fuel consumption
1 2 3 4 5 6 7 8 9 100
500
1000
1500
2000
2500
3000
Power Rating [kW]
To
tal F
uel co
nsu
mp
tio
n [
Lit
er]
0 50 100 150 200 250 300 3500.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
Day A
nn
ual C
HE
I
1 kW
2 kW
3 kW
5 kW
7 kW
10 kW
44
CAES system
Annual HEI
Fuel consumption
0 50 100 150 200 250 300 3500.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Day
Co
nti
no
us H
EI
5bar& 0.54kW
6bar& 1.09kW
8bar& 2.18kW
10bar& 3.27kW
12bar& 4.35kW
14bar& 5.45kW
16bar& 6.53kW
18bar& 7.62kW
22bar& 9.8kW
1 2 3 4 5 6 7 8 9 10 110
500
1000
1500
2000
2500
3000
3500
Scenario Number
To
tal F
uel co
nsu
mp
tio
n [
Lit
er]
#1:5bar, #2:6bar, #3:8bar, #4:10bar, #5:12bar, #6:14bar, #7:16bar, #8:18bar, #9:20bar, #10:22bar, #11:24bar
45
Annual HEI comparison
1 kWh & 2 kWh
5 kWh & 10 kWh0 50 100 150 200 250 300 350
0.2
0.3
0.4
0.5
0.6
0.7
Day
An
nu
al C
HE
I
CAES 1kWhr
CAES 2kWhr
Battery 1kWhr
Battery 2kWhr
Pumped Hydro 1kWhr
Pumped Hydro 2kWhr
5 10 15 20 25 300.2
0.25
0.3
0.35
0.4
0 50 100 150 200 250 300 350
0.4
0.5
0.6
0.7
0.8
0.9
1
Day
An
nu
al C
HE
I
CAES 5kWhr
CAES 10kWhr
Battery 5kWhr
Battery 10kWhr
Pumped Hydro 5kWhr
Pumped Hydro 10kWhr
640 650 660 670 680 690 700
0.28
0.3
0.32
0.34
0.36
0.38
0.4
650 660 670 680 690 700 710
0.38
0.4
0.42
0.44
0.46
0.48
2 4 6 8 10 12 14 16 18 200.5
0.6
0.7
0.8
46
Annual HEI comparison Cont.
Annual Average
Annual Shortage
0 1000 2000 3000 4000 5000 6000 7000 8000 9000 1000010
20
30
40
50
60
70
80
90
100
Power Rating [W]
An
nu
al avera
ged
CH
EI [%
]
Battery Storage System
Pumped hydro Storage System
CAES System
0 2000 4000 6000 8000 10000 120000.5
1
1.5
2
2.5x 10
10
Power Rating [W]
To
tal S
ho
rtag
e E
nerg
y [
J]
Battery
Pumped hydro
CAES
0 2000 4000 6000 8000 10000 12000500
1000
1500
2000
2500
3000
3500
Power Rating [W]
To
tal F
uel co
nsu
mp
tio
n [
Lit
er]
Battery
Pumped hydro
CAES
1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 20001.6
1.8
2
2.2x 10
10
8000 8050 8100 8150 8200 8250890
900
910
920
930
47
Annual HEI comparison Cont.
Low efficient system with single stage, high pressure and no
heat exchanger
Annual shortage to
Excess power
5 10 15 20 25 30 35 40 45 5020
30
40
50
60
70
80
90
En
erg
y c
on
vers
ion
eff
icie
ncy [
%]
Compressor Working Pressure [bar]
0 2000 4000 6000 8000 10000 1200020
25
30
35
40
45
50
55
60
65
70
Power Rating [W]
An
nu
al S
ho
rtag
e t
o E
xcess p
ow
er
rati
o[%
]
Battery Storage System
Pumped hydro Storage System
CAES System
48
Summary of contributions
A CAES system was developed for isolated applications.
Mathematical modelling of CAES system
Simplified air motor and flow rate control valve validation
through experiment
Wind speed profile regeneration using Combined direct
sampling method and Monte Carlo simulation. 49
Summary of contributions
Hybrid energy system optimization and component sizing
Development of a new criterion based on HEI.
Performance evaluation of different energy storage systems
Impact of control strategy and storage rating on total fuel
consumption
50
Research outcomes
H.SedighNejad, T.Iqbal and J.Quaicoe,” Compressed Air
Energy Storage System Control and Performance Assessment
Using Energy Harvested Index”, Electronics Special Issue on
Renewable Energy Systems, 2014, 3, 1-21.
H.SedighNejad, T.Iqbal and J.Quaicoe, “Effect of the sizing of
compressed air storage system on overall performance of
Hybrid systems”, poster presentation at CanWEA’s 26th Annual
Conference and Exhibition, November 1-3, 2010, Montreal,
Quebec.
Hanif Sedighnejad, T. Iqbal, J. Quaicoe,” Design
Considerations for Compressed Air Energy Storage Systems”,
2010, PKP Open Conference Systems, presented by IEEE
Newfoundland and Labrador Section.
51
Research outcomes cont.
H.SedighNejad, T.Iqbal and J.Quaicoe, “Performance
evaluation of a hybrid wind-diesel-compressed air energy
storage system”, 24th Canadian Conference on Electrical and
Computer Engineering (CCECE), 8-11 May 2011, Niagara
Falls, ON Page(s): 000270 – 000273.
H.SedighNejad, T.Iqbal and J.Quaicoe, “Design and dynamic
modeling of a micro compressed air energy storage system”,
poster presentation at CanWEA’s 27th Annual Conference and
Exhibition, October 3-6, 2011, Vancouver, BC.
H.SedighNejad, T.Iqbal and J.Quaicoe, “A compressed air
storage system Design and Steady-State Performance Analysis
of CAES”, The Twentieth Annual Newfoundland Electrical and
Computer Engineering Conference (NECEC), Nov. 1st, 2011.
52
Research outcomes cont.
H.SedighNejad, and T.Iqbal, “Simplified dynamic model for
vane type air motor”, The 21th Annual Newfoundland
Electrical and Computer Engineering Conference (NECEC),
Nov. 8th, 2012.
53
Suggested Future Work
Application of heat exchanger to improve the efficiency
Dynamic control of the CAES system in conjunction with
another energy source,
Evaluation of the system with capability of working in
series/parallel configuration and its impact on round trip
efficiencies and system power ratings
54
Thanks for your attention
55
