Professional Documents
Culture Documents
Energy storage model with gridable vehicles for economic load dispatch
in the smart grid
Uttam Kumar Debnath a,, Iftekhar Ahmad a, Daryoush Habibi a, Ahmed Yousuf Saber b
a
b
Smart Energy Research Group, School of Engineering, Edith Cowan University, Joondalup, WA 6030, Australia
R&D Department, Operation Technology, Inc., ETAP, Lake Forest, CA 92630, USA
a r t i c l e
i n f o
Article history:
Received 8 June 2014
Received in revised form 24 August 2014
Accepted 4 September 2014
Available online 20 September 2014
Keywords:
Battery lifetime
Energy storage
Gridable vehicles
Participation rate
Renewable energy
Smart grid
a b s t r a c t
The intermittent nature of renewable energy sources (RESs) and unpredictable variable load demands
have necessitated the inclusion of energy storage devices in the smart grid environment. Electric vehicles
(EVs) and plug-in hybrid electric vehicles (PHEVs), with vehicle-to-grid capability, referred to as gridable
vehicles (GVs), have become an option as storage devices. However, unsupervised use of GVs as storage
devices presents new challenges due to concerns over battery lifetime and cost effectiveness of this twoway power transfer. These issues reduce the participation rate of GVs in the vehicle-to-grid discharge
program. In this study, we present a system model, for GVs to act as distributed storage devices, which
mitigates concerns over battery lifetime, and provides GV owners with a transparent cost-benet analysis
of their participation in the vehicle-to-grid discharge program. With this model in place, fuel and emissions cost has been reduced, as shown using particle swarm optimization. Simulation results show that
up to 52% of GVs might be discharging at a net loss. In contrast, our proposed system ensures that no GVs
are at loss when discharging to the grid. This factor alone is expected to increase the participation rate of
GVs by a signicant margin, and ensure economic load dispatch.
2014 Elsevier Ltd. All rights reserved.
Introduction
Sustainable energy systems have become a priority for almost
all countries in the world. A smart grid is meant to provide such
an energy system by integrating extra operational capabilities
and renewable energy sources (RESs) into the power system. RESs
such as wind and solar power are well accepted sustainable energy
sources. However, unreliability in continuous energy production
from RESs has been a major problem. A straightforward solution
to this problem is to use conventional energy storage systems, such
as electrochemical accumulators, which tend to be very costly. A
variety of energy storage provisions have been proposed to atten
the cost [1], although achieving an acceptable cost of storage is still
a very active area of research. In recent times, gridable vehicles
(GVs) have emerged as a signicant contender, for dealing with
the uncertainty of RESs, in order to keep the utility grid unaffected.
Research trends indicate that the use of GVs as energy storage
devices will bring about a breakthrough in the way that RESs are
managed.
Corresponding author. Tel.: +61 431530999, +61 863045318; fax: +61
863045811.
E-mail addresses: u.debnath@ecu.edu.au (U.K. Debnath), i.ahmad@ecu.edu.au
(I. Ahmad), d.habibi@ecu.edu.au (D. Habibi), aysaber@ieee.org (A.Y. Saber).
http://dx.doi.org/10.1016/j.ijepes.2014.09.004
0142-0615/ 2014 Elsevier Ltd. All rights reserved.
1018
U.K. Debnath et al. / Electrical Power and Energy Systems 64 (2015) 10171024
Nomenclature
Wdep/Wpre
Wmin/Wmax
f
FCi()/ECi()
wi
wc/we
D(t)
H
N
NGV(t)
NGV-Dsch(t)
NGVmax
Pi(t)
Pwind(t)
Ppv(t)
Pmax
/Pmin
i
i
Pvj
R(t)
Ncycle
CDremaining
CDstorage
CDcycling
Ccycle
Copp/Cdgdn
wopp/wdgdn
Vc(t)
Es(t)
ps(t)
pbest/gbest
Ite/MaxIte
min
N X
H
X
wc FC i Pi t we wi EC i Pi t
!
1
i1 t1
j1
Pwind t P Dt Losses Rt
NX
GV t
j1
From (2) and (3), it is evident that power transfer to/from the GVs is
the only determining factor for ensuring maximum utilization of
1019
U.K. Debnath et al. / Electrical Power and Energy Systems 64 (2015) 10171024
98
96
94
92
90
88
0
500
1000
1500
2000
2500
3000
3500
4000
Cycle Number
Fig. 1. A plot of the total capacity fading of a battery against the number of cycles
both from [7] and the tted Eq. (7). The quality of the t is given by R2 = 0.9985.
CDremaining f Ncycle
100
1020
U.K. Debnath et al. / Electrical Power and Energy Systems 64 (2015) 10171024
per cycle. From previous research [7], we can relate the DOD to
the cost per cycling of a battery. A DOD corresponding to discharge
cycles of either less/more than that specied, require an adjustment factor of more/less than unity, respectively. The capacity degradation cost then becomes a direct product of the cost and the
adjustment factor.
C. Proposed optimization model considering cost of vehicle energy
Wind and solar energy are emission free and their operating
costs are negligible. Fuel cost of a conventional thermal generator
is expressed as a quadratic function of the units generated power
as follows:
FC i Pi t ai bi Pi t ci P2i t
10
EC i Pi t ai bi Pi t ci P2i t
11
j1
Dt Losses
12
N
X
Pi t Ppv t Pwind t Dt Losses
i1
V c t
N GVDsch
X t
Es tps t
s1
N GVDsch
X t
13
j1
Pmin
6 Pi t 6 Pmax
i
i
Pimax
14
min
where
and Pi
are the maximum and minimum generation
limits of the i th unit.
Charging/discharging up to certain maximum/minimum levels
is ensured to prevent battery failure, and is given by:
Wmin Pv j 6 Pv j t 6 Wmax P v j
15
where Wmin and Wmax are the minimum and maximum levels of
charge respectively, of the individual vehicles. Vehicles that have
been registered for charging/discharging from/to the grid NGVmax,
can take part during a predened scheduling period; and are given
by:
H
X
NGV t Nmax
GV
16
t1
min
N X
H
X
wc FC i Pi t we wi EC i Pi t V c t
i1 t1
!
17
1021
U.K. Debnath et al. / Electrical Power and Energy Systems 64 (2015) 10171024
v pq k 1 v pq k c1 r1 pbestpq k xpq k
c2 r2 gbest q k
Range
xpq k 1
Ite 1
MaxIte
xpq k 1 xpq k v pq k 1
18
19
where velocity v, position x, accelerating parameters c1 and c2, random numbers r1 and r2, particle number p, problem dimension q,
and iteration index k, are standard terms of PSO [15]. The owchart
for the minimization of fuel and emissions cost with RESs and GVs
in a smart grid, using our proposed models is given in Fig. 3. If at
hour t, the schedule is; [P1(t), P2(t),. . ., PN(t), NGV(t), Ppv(t), Pwind(t)]T,
then power supplied to/from vehicles is fNGV(t)Pvi (Wpre Wdep).
The sign of this expression will indicate whether it is a load or
source; and the rest of the load demand, given by the expression;
[D(t) + fNGV(t)Pvi (Wpre Wdep) Ppv(t) Pwind(t)] will be met from
the conventional thermal units.
Table 1
Generating unit capacity and coefcients.
Fig. 3. Flowchart for fuel and emissions cost minimization with RESs and GVs in the
smart grid using proposed models.
Unit
Pmin (MW)
Pmax (MW)
a ($)
b ($/MW)
c ($/MW2)
1
2
3
4
5
6
100
50
80
50
50
50
500
200
300
150
200
120
240
200
220
200
220
190
7.0
10.0
8.5
11.0
10.5
12.0
0.0070
0.0095
0.0090
0.0090
0.0080
0.0075
1022
U.K. Debnath et al. / Electrical Power and Energy Systems 64 (2015) 10171024
Table 2
Generator emissions coefcients.
a (ton/h)
b (ton/MW h)
c (ton/MW2 h)
1
2
3
4
5
6
10.33908
32.00006
32.00006
30.03910
32.00006
30.03910
0.24444
0.38132
0.38132
0.40695
0.38132
0.40695
0.00312
0.00344
0.00344
0.00509
0.00344
0.00509
50
40
30
20
10
Unit
60
8000
Expected to discharge
For range of weights
(0.25-0.35, 0.8-0.9)
For range of weights
(0.25-0.35, 0.9-1.0)
For range of weights
(0.30-0.40, 0.8-0.9)
For range of weights
(0.25-0.40, 0.8-1.0)
For range of weights
(0.30-0.40, 0.9-1.0)
7000
6000
5000
4000
3000
2000
1000
0
0
10
15
20
25
8000
35
Expected to discharge
For weights (0.35, 0.95)
For weights (0.35, 1.00)
For weights (0.40, 0.90)
For weights (0.40, 1.00)
For weights (0.45, 0.90)
For weights (0.45, 0.95)
7000
6000
5000
40
4000
3000
2000
30
25
20
15
10
5
1000
0
0
10
15
20
25
1023
U.K. Debnath et al. / Electrical Power and Energy Systems 64 (2015) 10171024
Table 3
PSO results with vehicle energy and costs considered with no price-check.
Time
(H)
Unit 1
(MW)
Unit 2
(MW)
Unit 3
(MW)
Unit 4
(MW)
Unit 5
(MW)
Unit 6
(MW)
Solar
(MW)
Wind
(MW)
Vehicles
(MW)
Demand
(MW)
Loss
(MW)
Total cost
($)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
170.45
168.57
180.40
192.98
204.41
231.45
255.29
273.87
323.11
412.60
450.69
483.57
377.49
315.50
279.80
216.45
206.38
227.15
257.80
412.58
330.81
232.00
189.01
174.46
152.41
150.84
161.13
172.08
182.05
200.00
200.00
200.00
200.00
200.00
200.00
200.00
200.00
200.00
200.00
192.46
183.74
200.00
200.00
200.00
200.00
200.00
168.64
156.00
160.38
158.78
169.10
180.08
189.89
213.21
233.61
249.47
291.20
300.00
300.00
300.00
300.00
284.77
254.51
200.30
191.61
209.51
235.79
300.00
297.67
213.69
176.51
163.98
104.90
103.79
110.89
118.48
125.38
141.58
150.00
150.00
150.00
150.00
150.00
150.00
150.00
150.00
150.00
132.58
126.57
138.98
150.00
150.00
150.00
141.90
116.10
107.34
149.80
148.22
158.37
169.12
178.87
200.00
200.00
200.00
200.00
200.00
200.00
200.00
200.00
200.00
200.00
189.16
180.54
198.18
200.00
200.00
200.00
200.00
165.69
153.32
101.68
100.55
107.67
115.21
120.00
120.00
120.00
120.00
120.00
120.00
120.00
120.00
120.00
120.00
120.00
120.00
120.00
120.00
120.00
120.00
120.00
120.00
112.80
104.12
0.00
0.00
0.00
0.00
0.00
0.00
0.09
17.46
31.45
36.01
38.06
35.93
36.78
31.59
9.70
12.92
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
10.54
22.27
25.50
25.50
25.50
25.50
25.50
25.50
25.50
25.50
25.50
25.50
25.50
24.82
20.74
14.62
25.50
19.04
25.50
18.02
25.50
21.42
0.00
2.55
131.76
85.00
42.5
0
0
0
0
0
+0.1020
+2.1420
+13.8720
+34.7820
+34.7820
+13.8720
+2.1420
+0.2359
7.8157
+18.2070
+45.6514
+45.6514
+18.2070
+2.8114
6.2411
42.50
700.00
750.00
850.00
950.00
1000.00
1100.00
1150.00
1200.00
1300.00
1400.00
1450.00
1500.00
1400.00
1300.00
1200.00
1050.00
1000.00
1100.00
1200.00
1400.00
1300.00
1100.00
900.00
800.00
18.40
18.02
20.56
23.45
26.11
31.74
34.49
36.30
41.36
46.25
48.13
49.78
44.56
40.55
36.89
28.73
26.53
31.07
34.74
46.25
42.19
31.81
22.51
19.27
17967.96
17707.78
19420.21
21362.12
23158.92
27092.55
29286.39
30849.98
35524.14
42317.36
48783.07
57904.84
50047.09
39027.93
32034.26
25050.59
24321.10
32248.91
43644.59
55803.82
41926.79
28017.35
20772.68
18552.13
Table 4
PSO results with vehicle energy and costs considered with price-check.
Time
(H)
Unit 1
(MW)
Unit 2
(MW)
Unit 3
(MW)
Unit 4
(MW)
Unit 5
(MW)
Unit 6
(MW)
Solar
(MW)
Wind
(MW)
Vehicles
(MW)
Demand
(MW)
Loss
(MW)
Total cost
($)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
170.45
168.57
180.40
192.98
204.41
231.45
255.29
273.87
323.13
413.65
457.20
500.00
394.13
319.00
280.34
216.46
206.70
228.79
265.27
452.85
333.81
232.29
189.00
174.46
152.41
150.84
161.13
172.08
182.05
200.00
200.00
200.00
200.00
200.00
200.00
200.00
200.00
200.00
200.00
192.50
184.00
200.00
200.00
200.00
200.00
200.00
168.64
156.00
160.38
158.78
169.10
180.08
189.89
213.21
233.61
249.47
291.21
300.00
300.00
300.00
300.00
287.73
254.99
200.34
191.86
210.94
242.13
300.00
300.00
213.96
176.48
163.98
104.90
103.79
110.89
118.48
125.38
141.58
150.00
150.00
150.00
150.00
150.00
150.00
150.00
150.00
150.00
132.60
126.72
139.95
150.00
150.00
150.00
142.08
116.10
107.34
149.80
148.22
158.37
169.12
178.87
200.00
200.00
200.00
200.00
200.00
200.00
200.00
200.00
200.00
200.00
189.17
180.86
199.60
200.00
200.00
200.00
200.00
165.68
153.32
101.68
100.55
107.67
115.21
120.00
120.00
120.00
120.00
120.00
120.00
120.00
120.00
120.00
120.00
120.00
120.00
120.00
120.00
120.00
120.00
120.00
120.00
112.86
104.12
0.00
0.00
0.00
0.00
0.00
0.00
0.09
17.46
31.45
36.01
38.06
35.93
36.78
31.59
9.70
12.92
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
10.54
22.27
25.50
25.50
25.50
25.50
25.50
25.50
25.50
25.50
25.50
25.50
25.50
24.82
20.74
14.62
25.50
19.04
25.50
18.02
25.50
21.42
0.00
2.55
131.76
85.00
42.5
0
0
0
0
0
+0.0765
+1.1411
+7.6946
+18.4429
+18.9401
+7.7775
+1.1730
+0.1339
9.0461
+13.0560
+32.5571
+33.0289
+13.1899
+2.1038
6.2539
42.50
700.00
750.00
850.00
950.00
1000.00
1100.00
1150.00
1200.00
1300.00
1400.00
1450.00
1500.00
1400.00
1300.00
1200.00
1050.00
1000.00
1100.00
1200.00
1400.00
1300.00
1100.00
900.00
800.00
18.40
18.02
20.56
23.45
26.11
31.74
34.49
36.30
41.36
46.30
48.45
50.62
45.35
40.92
36.95
28.74
25.50
31.38
35.46
46.90
42.50
31.85
22.51
19.27
17967.96
17707.78
19420.21
21362.12
23158.92
27092.55
29286.39
30849.98
35517.73
42081.92
47375.61
54302.50
46242.31
37488.73
31783.01
25022.97
23985.69
30864.90
40205.66
52848.47
40673.34
27827.76
20769.16
18552.13
1024
U.K. Debnath et al. / Electrical Power and Energy Systems 64 (2015) 10171024
Table 5
Overall benets of using our proposed models.
Items
With conventional
model
GVs at loss
Expected GV
participation rate
Total cost
Potential for RESs
integration
Basis of discharging
decision
Up to 52%
As low as 48%
Close to none
Close to 100%
More
Less due to participation
rate
Non-transparent
Less
More
Transparent and
objective