Enhancement of self-healing property of smart grid in islanding mode using electric vehicles and direct load control

Abstract

Power system reliability enhancement is one of the main goals pursued by making the systems smart. Self-healing, which includes fault detection, fault isolation, reconfiguration and service restoration, is the most notable feature of the smart grids in this regards. Increased penetration of distributed generation sources in power grids has even more facilitated the service restoration process. Onn the other hand, the significant progress in the technologies used in battery manufacturing has economically justified the utilization of electric vehicles in the improvement of system reliability. Nevertheless, in order to optimally make use of plug-in hybrid electric vehicles, their owner behavior should be properly modelled in the first place. In this paper stochastic nature of electric vehicle's behavior is modelled and the parking lots needed for self-healing in the islanding situation are optimally placed in the power grid. The proposed method is experimented on a 35-bus system. The results of simulation reveal that optimal usage of electric vehicles has notable effect on the improvement of system self-healing in islanding mode which consecutively results in the enhancement of power system reliability.

Full text

Enhancement of Se
l
Islanding Mode
U
Seyed Mohsen
Mohammadi Hoseini Nezhad
Department of Electrical and
Computer Engineering
University of Tehran
Tehran, Iran
Sm_mohammadi@ut.ac.ir
Alireza
F
Department
o
Compute
r
K. N. Toos
i
Tec
h
Teh
r
Fereidunia
n
Abstract—
Power s
y
stem reliabilit
y
enhance
m
main
g
oals
p
ursued b
y
makin
g
the s
y
stems s
m
which includes fault detection, fault isolatio
and service restoration, is the most notable f
e
g
rids in this re
g
ards. Increased
p
enetrati
o
generation sources in power grids has even
m
service restoration process. On the other ha
n
p
ro
g
ress in the technolo
g
ies used in batter
y
m
economicall
y
j
ustified the utilization of elect
r
im
p
rovement of s
y
stem reliabilit
y
. Neverth
e
optimally make use of plug-in h
y
brid elec
t
owner behavior should be
p
ro
p
erl
y
modelle
d
In this
p
a
p
er stochastic nature of electric v
e
modelled and the
p
arkin
g
lots needed for
s
islanding situation are optimall
y
p
laced in t
h
p
ro
p
osed method is ex
p
erimented on a 3
5
results of simulation reveal that o
p
timal usa
g
e
has notable effect on the im
p
rovement of s
y
s
t
islanding mode which consecutivel
y
results i
n
of power system reliability.
Keywords: Self-Healing,
S
ervice Restorati
o
Electric Vehicles Smart parking lot, Reliability.
I. INTRODUCTION
In recent years, with drastic increase i
n
and people's expectations, importance of
p
service reliability is more accentuated [1].
O
as a result of notable increase in number
a
distribution systems, their management and
delegated to operators as before. Having the
in recent years the grids with capability of
w
management have become vastly sought-aft
e
named as energy smart grids. Diverse typ
e
such as online measurement, real-tim
e
monitoring and control, fault detection, f
a
automatic service restoration are utilized i
n
l
f-Healing Property of S
m
U
sing Electric Vehicles
a
Load Control
F
ereidunia
n
o
f Electrical and
r
Engineering
i
University of
h
nology
r
an, Iran
n
@eetd.kntu.ac.i
r
Hamid Lesani
Department of Electrical and
Computer Engineering
University of Tehran
Tehran, Iran
Lesani@ut.ac.ir
M
ment is one of the
m
art. Self-healing,
n, reconfi
g
uration
e
ature of the smart
o
n of distributed
m
ore facilitated the
nd, the si
g
nificant
m
anufacturin
g
, has
ric vehicles in the
e
less, in order to
t
ric vehicles, their
d
in the first
p
lace.
e
hicle's behavior is
s
elf-healing in the
h
e
p
ower
g
rid. The
5
-bus system. The
of electric vehicles
t
em self-healing in
n
the enhance ment
o
n, Plug-in Hybrid
n
living standards
p
ower quality and
O
n the other hand,
a
nd complexity of
control cannot be
se points in mind,
w
ide and automatic
e
r. These grids are
e
s of technologies
e
and automatic
a
ult isolation and
n
smart grids. So
using smart grids, can offer n
u
increase in system security, reducti
o
nature-friendly power provision, e
c
p
ower grid, enhancement of reli
a
p
resent power grid is relatively sl
faults, which elongates the black
-
especially digital consumers, to an u
n
Hence, self-healing property of
s
fault detection, isolation, grid re
c
restoration, is its most outstanding f
a
reliability [4]-[5]. Shahsavari et al.
h
framework for creating a self-hea
l
They have studied self-healing stru
c
System and Devices [6]. In traditio
n
conducted by an operator with
switching events. This lengthy
reliability and consecutively it
satisfaction level. On the contrary, i
n
increased number of distributed
capabilities like demand response
a
such as online monitoring and co
m
healing process is hastened. But as
p
have such a capability, a healin
g
approach is
p
resented, which is
a
needed to improve self-healing
p
Furthermore, the stochastic nature o
f
accompanied by the necessity of re
in the output power of these sourc
e
different energy storage devices
supercapacitors, superconductors an
d
developments in the battery manufa
c
the increasing trend of environment
a
fuel price, have paved the way f
o
vehicles to drag themselves to the c
e
m
art Grid in
a
nd Direct
Mirjavad Hashemi Gavgani
Department of Electrical and
Computer Engineering
Shahid Beheshti University
Tehran, Iran
M
irjavad_hashemi@ace.sbu.ac.ir
u
merous advantages like:
o
n of system energy loss,
c
onomic improvement of
a
bility and etc. [2]. The
ow when reacting to the
-
out time for consumers,
n
desirable extent [3].
s
mart grid, which involves
c
onfiguratio
n
and service
a
cto
r
for increasing system
h
ave proposed a conceptual
l
ing property in systems.
c
ture of grid in two layers:
al grids, healing process is
a significant number of
process reduces system
will degrade consumer
n
smart grids, as a result of
generation sources and
a
nd advanced technologies
m
munication devices, self-
resent power grid does not
g
property reinforcement
a
set of practical actions
p
roperty of the grid [6].
f
renewable power sources
duction of the uncertainty
e
s, has made the usage of
like batteries, flywheels,
d
etc. inevitable [7]. Latest
c
turing industry along with
a
l concerns and increase in
o
r plug-in hybrid electric
e
nter of attention.

In this regard, considerable amount of research is
conducted to investigate plug-in hybrid electric vehicles in
smart grids. In the literature, plug-in hybrid electric vehicles
have been utilized for diverse purposes such as: economic
operation of the power system [8]-[9], reduction of pollutant
emissions yield by fuel vehicles [10], smart grid voltage and
frequency control [11]-[13], diminishing the black-out
probability in the peak load hours [14], active power loss
reduction [15], reduction of congestion is distribution lines
[16] and service restoration process improvement [17].
In recent researchers, variety of approaches have been
adopted for modelling the behavior of plug-in hybrid electric
vehicles. A stochastic approach for modelling EV owner's
behavior is presented in [18]. In [19] the availability of EVs is
modelled by using traffic data, while authors of [20] have
devised a deterministic model for EVs' behavior. In this paper,
the stochastic behavior of electric vehicles is properly
modelled. The aforementioned model has been employed for
optimal placement of parking lots, in order to improve self-
healing property of power grid in islanding mode. Eventually,
the effect of using plug-in hybrid electric vehicles on power
system reliability is evaluated. The rest of this paper is
organized as follows:
The stochastic modelling method for behavior of electric
vehicles is presented in section II, the applied formulation is
elaborated in section III. Subsequently, the simulation results
are demonstrated in section IV and conclusions are derived
and presented in section V.
II. MODELLING OF THE BEHAVIOR OF PLUG-IN HYBRID
ELECTR IC VEHICLES
The behavior of electric vehicles has been modelled in variety
of ways in past researches. In [21], a new approach is
suggested for estimation of the availability of EVs which
besides considering driver's behavior, makes use of stochastic
models that are mainly based on Monte Carlo Simulations. In
[22], a mathematical model for estimating the electric power
capacity of a parking lot is described. Authors in [23]
proposed an analytical way to derive the probability
distribution of available power capacity of EVs taking
into account drivers’ plug-in probability.
In this paper we aim to use plug-in hybrid electric vehicles
for providing an immediate support, when a black-out happens
after a fault. In this regard, a contract is signed between the
vehicle owners and the aggregator, so when an emergency
situation comes up, declare for necessity of vehicles'
participation is sent from parking lot to the vehicle owners.
Regarding the unpredictable nature of fault occurrence, the
amount of available charge in vehicles, when they receive the
parking's message, will be of a stochastic nature. We use the
model depicted in Fig. 2 for an EV emergency participation
system: the information about required power (active and
reactive) and service restoration duration (This data is
collected and sent by system operator) is received by EV's
contracted aggregator. This data is broadcasted within the
aggregator, received by contracted PHEVs. Vehicles on their
situation (like driving to some destination area or on parking in
a garage) can then use it to make their decision amongst acting
according to their contract (fast plug-in into parking lot) or
paying for lack of participation in restoration process.
So each parking lot, in each hour of service restoration
process, will have stochastic number of vehicles with
stochastic amount of charge in each, which can be formulated
as follows:
2/
() ( () ())
jjj
VG Cap chdch dep pre
P
tP t t
ηψψ
=× × − (1)
Occurance Time
() ( , )
pre t Islanding
tNormal
ψψ
ψμ
σ
== (2)
ji
22
2V2GV2G
() ( 1)
2
( ) P (t) P (t)
() ( , )
VG VG
VG
jN t iN t
VG N N
Pt
NtNormal
μσ
∈∈−
=+
=
∑
∑
(3)
min max
,
pre dep
Ψ≤ΨΨ≤Ψ (4)
Where, j
Cap
Pis the capacity of jth vehicle battery, /j
ch dch
η
is
the efficiency of the charge and discharge process for jth
vehicle battery, ()
pre t
ψ
and ()
dep t
ψ
are respectively the
amount of charge in the beginning and at the end of the hour t
after fault occurrence time and finally, 2()
j
VG
P
t is the amount
of the power, exchanged between the electric vehicle and the
corresponding aggregator. Eq. (2) demonstrates the stochastic
nature of the initial amount of charge available in the battery
of each vehicle in each hour. Additionally, the number of
participating vehicles in each hour is stochastic, which is
formulated in (3). All in all, in each hour of islanding mode of
the grid, a number of vehicles are available in parking lots,
while some others participate in power provision in each hour
with different levels of available charge in their batteries.
That's why the Eq. (3) has two expressions for exchanged
power. It should be noted that, due to the limitations such as
battery degradation and personal and emergency usage of
vehicles, limited amount of battery charge will be available.
Also, charging rate of PHEVs is assumed to be so fast that it
takes about 15 minutes to charge or discharge the batteries.
Figure 1. Information flow in etimating PHEV participation.

III. PROBLEM FORMULATION
Service restoration problem has been formulated with a
wide variety of objective functions such as minimization of
restoration time and power losses, or maximization of restored
loads with customer prioritization and etc. [17]. In this paper,
in order to evaluate the effects of presence of electric vehicles
on the improvement of self-healing subsequently on the
reliability of power system, smart parking lots are optimally
placed and the charging and discharging behavior of the
vehicles are planned. The constraints like load flow equations,
operation and available renewable energy sources are taken
into consideration.
A. Optimal Placement of Charging/Discharging Parking
Lots for EVs
In the past researches different objective functions are
proposed for charging/discharging parking lots of EVs. In [18]
the optimal placement has targeted the minimization of power
loss and voltage profile improvement. In this paper, we pursue
the minimization of lost load during the islanding process
(reliability improvement). Minimization of the lost load is the
direct result of reducing the average consumer black-out time
index and loss of energy. Hence the objective function is
defined as:
(,) (1 (,)): Min { }
P
kt Lsikt
d
tk
Obj ×−∑∑ (5)
In which, Lsi indicates the load supply index and Pd shows
the power demands of the loads. The consumers will be
different in their types, number of vehicles and their behavior.
Eventually, regarding the number of vehicles and maximum
capacity of the smart parking lot, a proper limit for the number
of parking lots is predicted.
B. Service Restoration Formulation
The optimal setting of charging/discharging of electric
vehicles, aims to supply maximum number of loads, with the
least power loss possible:
{
}
Mi n ( , ) (1 ( , )) ( ) ( ):
P
kt Lsikt Lpik Lsf P t
dloss
tt
k
Objective ×− × + ×∑∑ ∑ (6)
()
()
2
2
() () ()
() 3 ( ,)
() 3 ( , )
loss Br loss Tl loss Br
loss Tl Tl
k
loss Br lBr
k
P
tP tP t
Pt RIkt
P
tRILkt
−−−
−
−
=+
=× ×
=× ×
∑
∑
(7)
In which, Lpi represents the loss of load penalty, which also
indicates the priority or importance of that bus. Apparently,
penalty coefficient is greater than the power loss coefficient.
C. Operational Constraints
Operational constraints include the constraints of voltage
drop in load buses, current limitations, and output power of
controllable power sources.
max
bb
II≤ (8)
0.95 ( , ) 1.02
p
upu
Vkt≤≤ (9)
1
30 ( ) 30
x
v
r
V
tg V
ϕ
−
−°≤ = ≤ °
(10)
Equations (8), (9) and (10) indicate constraints related to
allowable current, acceptable voltage drop and voltage phase
angles respectively. The constraints for increasing or
decreasing the output power of controllable power sources are
presented as follows:
(,) (,1) (,1)
(( ,) ( , 1)) ( ) (1 ( ,))
CHP CHP k
max
kCHP
PktPkt RUvkt
SU v k t v k t P k v k t
≤−+×−
+× − −+ ×− (11)
(, 1) (,) (,)
((,1) (,)) ()(1(,1))
CHP CHP k
max
kCHP
Pkt PktRDvkt
SD v k t v k t P k v k t
−− ≤ ×
+× −− + ×− −
(12)
Where v indicates whether the CHP unit installed on bus k
is on or off in the hour t.
D. Development of Teng Load Flow
As islanding happens in a power grid, the grid loses its
radial nature, and due to the presence of smart parking lots, it
turns into a multi-directionally fed network. In such cases, the
method used for power flow must fit the studied grid. Teng's
approach for power flow is used in radial and grid connected
networks, so it can be altered as follows to suit the above
mentioned grid:
2
2
(,) [( (,) ( ,) (,) (,) (,))
( ( , ) ( , ) ( , ))) ( , )]/ ( , )
dVGWindCHP
dComp
I
bt P bt Lsi k t P k t P bt P bt
jQ bt Lsi k t Q bt V bt V bt
=×+ − −
−×− × (13)
(,) (, ) ( ,)
k
I
Lbt Brb k I k t=×
∑
(14)
()
(,) (,) (,)
lB
k
I
LF B k V k t Z IL B t×=×
∑
(15)
Where in (13) and (14) load currents and main line
currents are calculated respectively and Br is the injection
current matrix. In (15) the Kirchhoff's equations are restated,
where ILF represents the topology of the grid. Power balance
constraints can be stated as follows:
()
2
(,) (,) (,) ( ( ,) (,)) 0
dVGWCHP
k
Pkt Lsikt P kt P kt P kt×+ − + =
∑
(16)
()
(,) (,) (,) 0
dComp
k
Qkt Lsikt Q kt×− =
∑
(17)
Both problems of optimal placement of parking lots and
service restoration are formulated as mixed integer nonlinear
programming (MINLP). In this paper, GAMS software and
SBB algorithm are utilized for mathematical modelling,
optimization and solving the MINLP problem.
IV. SIMULATION RESULTS
The method presented in previous sections is
experimented on a 35-bus medium voltage grid [24] which is
shown in Fig. 1. Voltage level of the studied grid is 11 kVs
and the duration considered for an islanding mode is 3 hours.
One of the major defects in the studies related to the
calculation of reliability indices like system average
interruption frequency index and system average interruption
duration index, is that these coefficients are obtained using
the average load, while due to uncertain nature of fault
occurrence (relay trips, lightning, line outages, load increase
and etc.) the number of black-outs may not be accurately
predicted. Among possible failures, those emergent from the
sharp rise in system loads in peak-load period are more
predictable than others. So in this study, it is supposed that
the system load increases during the service restoration
process and Fig.3 shows this supposition. Also, reactive

power demand miss-consideration in restoration process may
lead to out of band bus voltages in actual condition [29]. Thus
in this paper, both active and reactive power consumption are
considered [24].
Figure 2 . Studied 35-bus power grid [24]
Figure 3 . Medium load of test system [24]
The important data about power sources in this grid,
including the type, place and capacity of distributed
generation units, is presented in Table 1. Regarding the
stochastic nature of wind behavior, the output power from
wind turbines is non-dispatchable. So in recent years, a
considerable number of researches have focused on the
prediction of wind speed and resultantly the output power of
wind farms. The accuracy for this prediction has reached 10%
in some cases. But the stochastic behavior of wind power is
beyond the scope of this paper, so the prediction error for
output power of wind generation sources is not considered
here.
On the other hand, Advanced Metering Infrastructure (AMI)
is one the major requirements for creation of smart grids. This
infrastructure can offer distant load control (DLC) as one of
its main features. In this paper, it is supposed that owing to
the devices installed on side branches, DLC can be put into
action. Additionally, the buses 5, 14, 16, 21, 30 and 34 (which
include hospitals and industrial centers) are put in top priority
for service restoration process. That is, their penalty for loss
of load is significantly higher compared to the other less
important loads.
TABLE I. PLACE AND CAPACITY OF DISTRIBUTED GENERATION
SOURCE S DATA
In case, that EVs are not used, the index of electric energy not
supplied (EENS) equals 12406 kWh and the plan for switching
the loads will be as presented in Table 2. As it is shown, the
buses 5, 14, 21, 30 and 35 are the most important loads and
have to be supplied even if the EVs are not present. In this
case, the necessity for supplying the prior loads, may increase
the total ENS index along with improving ENS index for prior
loads.
TABLE II. STATUS OF LOADS DURING ISLANDING MODE WITHOUT EVS
When EVs are added to the grid, if charge/discharge process is
properly managed, the EENS index will optimally decrease.
This decrease is most desirable when the parking lots are
optimally placed in the grid. As it mentioned because of
different consumer types (industrial, commercial, residential
and sensitive loads) for each feeder, number of available
PHEVs varies for each candidate parking lot. Table III
presents PHEV's availability for each candidate parking lot. As
this table depicts, five candidate parking lot with different
parking size are considered. When EVs are optimally placed,
the EENS index equals 1072.22 kWh which shows 90% of
decrease compared to the grid without EVs. The load
switching plan for this case is presented in Table IV, and
injecting power along with optimal place for each parking lot
is shown in Table V.
0
100
200
300
400
500
600
700
800
1 4 7 101316192225283134
Active Load (kW)
Bus Number
H1
H2
H3
Bus#
CHP Installation Bus 2 4 7 12 17 20 22 24 30 32 34
Capacity of CHP Unit 400 320 340 320 420 340 260 320 320 300 340
Wind Turbine Installation Bus 3 7 131719232933
Capacity of Wind Turbine Unit 200 160 200 160 240 120 160 240
VAR Comp. Installation Bus 2 4 10222434
Capacity of VAR Compensator 500 550 600 550 700 800
Bus#
Hour 23456789101112131415161718
First 11111111111110111
Second 00110010101010101
Third 11111100100111100
Hour 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
First 01100010111111111
Second 00101110100101111
Third 00100111001101110

TABLE III. AVAILABLE PHEV NUMBER FOR EACH CANDIDATE PARKING
LOT
TABLE IV. STATUS OF LOADS DURING ISLANDING MODE IN PRESENCE
OF EVS
TABLE V. POWER INJECTION OF SMART PARKING LOTS
In case that EVs and renewable sources are not used, the
ENS will be 27.677 MWh. This is while, if just renewable
sources are utilized, the ENS will be reduced to 12.406 MWh,
which shows a reduction of 15.271 MWh in ENS. On the
other hand, if only EVs are involved in power supply, ENS
index will become 18.153 MWh, that reveals a 9.524 MWh of
decrease. Accumulatively, we can expect a 24.695 MWh of
decrease in ENS index, for these two sources, if used
simultaneously. But if we implement EVs and DG sources
together, the ENS factor will decrease to only 1072.22 kWh,
which shows 26.605 MWh decrease compared to the base
case and 1.91 MWh more decrease than what we expected to
have in simultaneous usage of these two sources. The reason
for this difference is that in the first hours, the available EVs
are charged in the feeders which are fully supplied with
enough energy (non-dispatchable wind power overgeneration
will be saved in EV batteries) and then they are discharged in
the last hours. Practically, when EVs and DG sources are put
in use simultaneously, a virtual smart garage will assist the
grid in service restoration, by injecting 1.91 MWh of energy.
Therefore, self-healing ability of grid in the islanding mode of
operation is improved by coordinating different kinds of
power sources (in this paper, DGs and PHEVs) and thus
reliability is enhanced.
Figure 4 . EV State of charge in first participating group in parking lot 1
Figure 5 . EV injected power by the first group of participants in firt
candidate parking lot
In Fig.2 an example of optimal management of
charge/discharge of EVs for first group of participants in
restorations process, that is the smart parking lot No. 1, is
shown in details. As it is obvious, in the first hour, regarding
the light load on the grid, EVs will charge and in second and
third hours they will discharge mainly. The amounts of
charging and discharging of these vehicles is shown in Fig. 3.
Actually this figure describes a specific parking lot and other
one will have a different behaviors. For example, the smart
parking lot no. 2 will only discharge its vehicles in first hours,
while smart parking lot no. 1 tries to supply the load in third
hour. Consequently, the behavior of each parking lot, depends
on different features such as system loading, parking lot place,
grid topology, feeder type and number of consumers of the bus
on which the parking lot is installed.
EV Number Group1 Group2 Group3
Participation Hour
First Second Third
Smart Parking 1
100 60 40
Smart Parking 2
150 50 40
Smart Parking 3
170 40 40
Smart Parking 4
150 60 50
Smart Parking 5
200 50 40
Bus#
Hour 2 3 4 5 6 7 8 9 101112131415161718
First 11111111111111111
Second 11111111111111111
Third 11111111111111111
Hour 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
First 01101111111111111
Second 11111111111111111
Third 01111111111111111
Injected Power Group1 Group3
Injection Hour
First Second Third Second Third Third
Opt. Pl.
S. P. 1
-730.64 285.546 1359.55 399.92 139.824 339.06 4
S. P. 2
1579.84 0 0 448.718 0 360.383 27
S. P. 3
1285 -265.664 522.312 799.718 0 537.828 28
S. P. 4
155.55 951.867 434.234 274.884 264.861 473.836 21
S. P. 5
-69.649 1333.86 890.7 473.836 0 306.595 19
Group2
0
0.2
0.4
0.6
0.8
1
1
7
13
19
25
31
37
43
49
55
61
67
73
79
85
91
97
EV State of charge (%)
EV Number
Initial SOC
-25
-20
-15
-10
-5
0
5
10
15
20
25
1
6
11
16
21
26
31
36
41
46
51
56
61
66
71
76
81
86
91
96
EV injected power (kW)
EV Number
H1 H2

V. CONCLUSION
Boosting power system reliability is one of the main
reasons why smart grid has gone so widespread. Self-healing
as one the most outstanding features of smart grid, including
fault detection, isolation, grid reconfiguration and service
restoration, is of great help in this regard. Service restoration
as last stage of self-healing is done by maneuvering on
controllable (manually or automatically). In the smart grids,
distributed generations have a high penetration. Thus, using
distributed generation will facilitate and hasten service
restoration process by satisfying some loads that are non-
restorable when DG sources are not involved. Besides, Battery
manufacturing technology has paved the way for electric
vehicles to participate in self-healing process and increasing
the reliability of power grids. In this paper along with
stochastic modelling of behavior of plug-in hybrid electric
vehicles, effect of PHEVs on self-healing property
improvement in the islanding mode of operation of grid is
evaluated. The results show that, by using PHEVs, ENS index
as one of reliability indices will decrease notably. Therefore by
improving reliability indices, PHEVs optimal usage leads to
increasing the self-healing property of smart grid.
REFERENCES
[1] Sargolzaei, Arman, Kang K. Yen, and M.N. Abdelghani, “Time-Delay
Switch Attack on Load Frequency Control in Smart Grid.” Advances in
Communication Technology, vol. 5, pp. 55-64.
[2] Du Xin-Wei, Ye Qiang, “Review of Smart Grid and its Development
Prospect in Sichuan,” IEEE power & energy magazine, 2005.
[3] Amin S.M., Bruce F.W., “Toward a smart grid,” IEEE power & energy
magazine, vol. 34, No. 5, pp. 34-41, 2005.
[4] A. Kostlov and P. M. S. Carvalho, “Towards Self-Healing in
Distribution Networks Operation: Bipartite Graph Modelling for
Automated Switching,” Electric Power Syst. Res. 81, pp. 51–56, May
2011.
[5] X. Mamo, S. Mallet, T. Coste, and S. Grenard, “Distribution
automation: The cornerstone for smart grid development strategy,” in
Proc. IEEE Power Energy Soc. Gen. Meeting, pp. 1–6, Calgary, AB,
Canada, 2009.
[6] A. Shahsavari, A. Fereidunian, S.M. Mazhari, and H. Lesani, “A Healer
Reinforcement Approach to Smart Distribution Grids by Improving
Fault Location Function in FLISR”, 13th Int. Conf. on Env. and Elect.
Eng., Wroclaw, Poland, Aug. 2013.
[7] Z. Liu, W. Gao Z. Liu, and W. Gao, “Wind Power Plant Prediction by
Using Neural Networks,” IEEE Energy Conversion Conf. and Expo.,
Raleigh, North Carolina, Sep. 15–20, 2012.
[8] Yu Zheng et al., “Electric Vehicle Battery Charging/Swap Stations in
Distribution Systems: Comparison Study and Optimal Planning,” IEEE
Trans. Power Syst., vol. 29, no. 1, pp. 221–229, Jan. 2014.
[9] Cong Liu et al., “Assessment of Impacts of PHEV Charging Patterns on
Wind-Thermal Scheduling by Stochastic Unit Commitment,” IEEE
Trans. Smart Grid, vol. 3, no. 2, pp. 678–683, Jun. 2012.
[10] A. Y. Saber and G. K. Venayagamoorthy, “Resource Scheduling Under
Uncertainty in a Smart Grid with Renewables and Plug-in Vehicles,”
IEEE Syst. J., vol. 6, no. 1, pp. 103–109, Mar. 2012.
[11] Chenye Wu, Hamed Mohsenian-Rad, Jianwei Huang, and Juri
Jatskevich, “PEV–Based Combined Frequency and Voltage Regulation
for Smart Grid,” Innovative Smart Grid Technologies (ISGT), pp. 1–6,
Washington, DC, USA, Jan. 16–20, 2012.
[12] Koichiro Shimizu, Taisuke Masuta, Yutaka Ota, and Akihiko
Yokoyama, “A New Load Frequency Control Method in Power System
Using Vehicle-to-Grid System Considering Users’ Convenience,” 17th
Power Syst. Comp. Conf., Stockholm, Sweden, Aug. 22–26, 2011.
[13] Tulga Ersal, Changsun Ahn, Ian A. Hiskens, Huei Peng, and Jeffrey L.
Stein, “Impact of Controlled Plug-In EVs on Microgrids: A Military
Microgrid Example,” in Proc. IEEE Power Energy Soc. Gen. Meet.,
pp. 1-7, San Diego, CA, USA, Jul. 2011.
[14] C. Pang, P. Dutt, and M. Kezunovic, “BEVs/PHEVs as Dispersed
Energy Storage for V2B Uses in the Smart Grid,” IEEE Trans. Smart
Grid, vol. 3, no. 1, pp. 473–482, Mar. 2012.
[15] Lidan Chen, C. Y. Chung, Yongquan Nie, and Rongrong Yu,
“Modeling and Optimization of Electric Vehicle Charging Load in a
Parking Lot,” in 5
th IEEE PES Asia–Pacific Power and Energy Eng.
Conf., Hong Kong, Dec. 8–11, 2013.
[16] Junjie Hu, Shi You, Morten Lind, and Jacob Ostergaard, “Coordinated
Charging of Electric Vehicles for Congestion Prevention in the
Distribution Grid,” IEEE Trans. Smart Grid, vol. 5, no. 2, pp. 703–711,
Mar. 2014.
[17] Clara Gouvela et al., “Microgrid service restoration: The Role of
Plugged-In Electric Vehicles,” IEEE Ind. Electron. Mag., pp. 26–41,
Dec. 2013.
[18] Wencong Su and Mo-Yuen Chow, “Performance Evaluation of an
EDA-Based Large-Scale Plug-In Hybrid Electric Vehicle Charging
Algorithm,” IEEE Trans. Smart Grid., vol. 3, no. 1, pp. 308–315, Mar.
2012.
[19] Bei Zhang and Meladen Kezunovic, “Impact of Available Electric
Vehicle Battery Power Capacity on Power System Reliability,” in Proc.
IEEE Power Energy Soc. Gen. Meet., pp.1–5, Jul. 2011.
[20] Y. M. Atwa and E. F. El-Saadany, “Reliability Evaluation for
Distribution System With Renewable Distributed Generation During
Islanded Mode of Operation,” IEEE Trans. Power Syst., vol. 24, no. 2,
pp. 572–581, May. 2009.
[21] C. Sandels, U. Franke, N. Ingvar et al., “Vehicle to Grid ; Monte Carlo
simulations for optimal Aggregator strategies,” in Int. Conf. on
Power System Technology (POWERCON), Oct. 24-28, 2010.
[22] pp. 1-8. U. C. Chukwu, and S. M. Mahajan, “V2G electric power
capacity estimation and ancillary service market evaluation,” in
IEEE Power and Energy Soc. Gen. Meet., pp. 1-8, Jul. 24-29, 2011.
[23] H. Sekyung, H. Soohee, and K. Sezaki, “Estimation of Achievable
Power Capacity From Plug-in Electric Vehicles for V2G
Frequency Regulation: Case Studies for Market Participation,” IEEE
Trans. on Smart Grid, vol. 2, no. 4, pp. 632-641, Dec. 2011.
[24] Y. Kumar, B. Das, and J. Sharma, “Multiobjective, Multiconstraint
Service Restoration of Electric Power Distribution System With Priority
Customers,” IEEE Trans. Power Del., vol. 23, no. 1, pp. 261–270, Jan.
2008.
[25] R. K. Karn, Yogendra Kumar, and Gayatri Agnihotri, “Multiobjective
Service Restoration Considering Primary Customers using Hybrid GA–
ACO Algorith m,” in. J. of Comput. Applicat., vol. 64, no. 3, pp. 1–10,
Feb. 2013.
[26] Y. Xu, Z. Y. Dong, K. P. Wong, E. Li, and B. Yue, “Optimal Capacitor
Placement to Distribution Transformers for Power Loss Reduction in
Radial Distribution Systems ,” IEEE Trans. Power Syst., vol. 28, no. 4,
pp. 4072 - 4079, Nov. 2013.
[27] M. Moradijoz, A. Ghazanfarimeymand, M.P. Moghaddam, and M.
Haghifam, “Optimum Placement of Distributed Generation and Parking
Lots for Loss Reduction in Distribution Networks,” Proc. of 17th Conf.
on Elect Power Dist. Networks (EPDC), pp. 1-5, May 2-3, 2012.
[28] J. H. Teng, “A Direct Approach for Distribution System Load Flow
Solutions,” IEEE Trans. Power Del., vol. 18, no. 3, pp. 882 - 887, Jul.
2003.
[29] Paul A. Trodden, Waqquas Ahmed Bukhsh, Andreas Grothey, and Ken
I. M. McKinnon, “Optimization-Based Islanding of Power Networks
Using Piecewise Linear AC Power Flow,” IEEE Trans. Power syst.,
vol. 29, no. 3, pp. 1212–1220, May 2014.