In
te
r
n
ation
a
l Jou
rn
al
o
f Po
we
r
Elec
tron
ic
s an
d
D
r
ive S
y
stem
(IJ
PED
S
)
V
o
l.
10, N
o.
1, Mar
ch 20
19,
p
p.
160~
1
6
9
IS
S
N
: 2088-
86
94,
D
O
I
:
10.11
59
1
/ij
ped
s
.
v10
.
i
1.pp
1
60-
16
9
160
Jou
rn
a
l
h
o
me
pa
ge
:
ht
tp:
//i
a
e
score
.
com
/
j
o
u
r
na
l
s
/
i
n
d
e
x
.
p
hp/IJ
PED
S
Smart database concept
for power management
in an electrical veh
i
cle
Ma
hmo
udi
C
hok
r
i,
F
la
h Ay
men
,
Sbita
L
a
s
sa
a
d
ENI
G
Na
t
i
on
a
l
S
c
hool
o
f
E
ngi
neeri
n
g
o
f
G
abès
, U
n
iv
ersity O
f
G
a
b
è
s
,
Tu
n
isi
a
Art
i
cl
e In
fo
ABSTRACT
A
r
tic
le hist
o
r
y
:
R
e
c
e
i
v
e
d
M
ay
2
4
,
2
018
Re
vise
d Ju
l 2
7
,
201
8
Ac
ce
p
t
ed
Au
g
1
6
,
2
018
As
w
orld
p
op
u
l
at
io
n
contin
ues
to
g
row
and
t
h
e
limited
amount
o
f
f
oss
il
f
u
e
l
s
beg
i
n
t
o
d
imin
is
h,
i
t
may
not
b
e
po
ss
ib
l
e
t
o
aff
o
rd
t
he
n
eeded
am
ou
nt
o
f
energ
y
d
em
and
e
d
by
t
he
w
orl
d
b
y
o
n
l
y
u
si
ng
fo
ss
il
f
u
e
ls.
M
eanw
h
ile,
the
a
b
un
da
n
t
n
a
t
ure
of
r
e
n
e
w
a
b
le
e
n
e
rgy
sou
r
c
e
s
b
ring
s
ne
w
b
e
ginn
in
g
f
o
r
n
e
xt
gen
e
ratio
ns
.
Greater
p
en
et
rati
on
o
f
elect
ric
veh
i
cles
w
ill
play
a
n
i
m
port
a
nt
rol
e
i
n
buil
d
i
n
g
g
r
een
a
nd
h
ealt
hy
w
o
rld
.
T
h
e
m
a
i
n
rem
a
in
i
n
g
is
sue
to
m
ak
e
th
e
switch
f
r
om
c
on
vent
io
nal
t
o
e
l
e
c
t
ric
v
e
hicl
e
is
p
erf
o
rm
ance
c
o
s
t
;
E
f
fic
i
e
n
t
E
V
s
t
h
a
t
c
a
n
d
riv
e
f
o
r
l
on
g
di
sta
n
c
e
s,
o
n
sing
le
c
ha
r
ge,
are
still
exp
e
ns
ive
f
o
r
ordin
a
ry
c
on
su
mer.
T
o
add
r
ess
t
h
is
r
ang
e
p
rob
l
em
,
m
a
n
y
att
e
m
p
ts
h
av
e
been
m
ade
du
rin
g
l
as
t
decad
e.
T
he
g
oal
w
a
s
to
c
on
cei
ve
a
po
wer
ef
ficient
elect
ric
vehi
cle,
capab
le
o
f
m
a
nag
i
ng
its
e
n
e
rg
y
an
d
reach
lo
ng
er
d
i
s
t
a
nce
s
.
It
d
ep
ends
o
n
the
elect
ri
cal
a
rchit
ectu
r
es
a
n
d
used
alg
o
rithms
.
Th
i
s
p
ap
er
a
d
d
s
new
p
e
rspect
iv
e
f
o
r
po
wer
M
a
nagem
e
n
t
in
E
Vs;
Th
e
p
r
opo
sed
m
e
t
hod
ol
og
y
int
r
od
uces
a
n
ew
p
o
w
er
m
an
agem
ent
archi
t
ectu
r
e
b
a
sed
on
c
om
m
u
n
i
catio
n
and
car
l
e
a
rn
in
g.
T
h
e
c
on
ve
ntional
so
ftw
a
re
l
evel
i
n
EV
h
as
b
een
r
e
p
l
aced
w
i
t
h
s
e
lf
r
ead
jus
t
ab
le
s
oftw
are.
E
Vs
are
co
nnect
ed
t
h
r
o
u
g
h
a
d
at
ab
as
e
,
a
n
d
c
an
u
plo
a
d
or
dow
nl
oad
ad
ju
st
me
n
t
param
e
ters
w
hi
le
s
o
f
tw
are
i
s
r
un
ni
ng
.
To
t
ake
advan
t
ag
e
o
f
t
he
new
archi
t
ectu
r
e,
a
n
ew
l
earni
ng
tech
ni
qu
e
con
cept
i
s
i
nt
rod
u
ced
t
o
o,
b
ased
o
n
Clo
u
d
exp
e
rienc
e
e
x
c
han
g
e
b
e
t
w
een
E
l
ectric
Vehi
cles
.
Th
is
e
nh
an
ce
me
n
t
aim
s
t
o
buil
d
a
b
et
ter
EV
e
x
p
erien
ce
in
p
ower
m
an
agem
en
t
t
h
ro
ug
h
Clou
d
sha
r
ing
a
n
d
de
fini
te
ly
c
ut
w
ith
c
on
ve
n
t
io
na
l
a
r
c
h
ite
c
t
u
r
e
tha
t
m
a
y
hav
e
reach
ed
it
s
b
ou
n
d
a
r
i
e
s.
K
eyw
ord
s
:
Com
m
uni
c
a
t
ion
HEV
Ma
nage
me
n
t
Po
wer
Co
pyri
gh
t © 2
019 In
stit
u
t
e
of Advanced
En
gi
neeri
n
g
an
d
S
c
ien
ce.
All
rights
res
e
rv
ed.
Corres
pon
d
i
n
g
Au
th
or:
Ma
hm
oud
i
Ch
okr
i,
S
P
EG
Rese
a
rch U
n
i
t
, E
N
I
G
N
a
tiona
l S
c
ho
ol
o
f E
ngi
nee
r
i
ng o
f
G
abè
s,
U
n
i
v
ersi
ty O
f
G
a
bè
s,
Tuni
sia
.
Em
ail:
cho
k
ri.
m
ahm
oud
i@
gm
ail.c
o
m
1.
I
N
TR
OD
U
C
TI
O
N
G
l
o
b
al w
ar
min
g
, clima
t
e
c
h
an
ge,
incr
easi
n
g
de
ma
nd on e
n
e
r
gy, an
d
n
a
ti
on
s’ desire to bec
ome
gre
e
n
a
n
d
resp
e
c
t
f
ul
t
o
n
e
xt
g
en
e
r
at
i
o
n
s
h
av
e
re
sult
ed
i
n
a
n
eed
t
o
c
h
a
ng
e
el
e
c
tri
c
i
t
y
p
rodu
c
t
io
n
a
n
d
co
nsump
t
ion
,
inc
l
ud
ing
a
r
a
dica
l
i
n
c
r
ea
s
e
i
n
rene
w
a
b
l
e
e
n
erg
y
s
o
u
rc
es
f
or
d
a
il
y
app
l
i
cat
i
ons.
In
f
o
rma
t
io
n
a
n
d
com
m
unic
a
t
io
n
t
e
c
h
n
o
l
og
y
(ICT)
i
s
a
l
rea
d
y
esse
n
tia
l
i
n
a
ut
om
oti
ve
i
n
d
us
try.
I
t
im
pro
v
es
d
r
i
v
i
n
g
safet
y
,
perform
ance
a
nd
com
f
ort.
M
ore
ove
r,
t
he
se
e
ffec
ts
g
o
fur
t
her
w
i
t
h
e
l
ect
ri
c
v
e
h
i
c
l
e
s
:
IC
T
b
e
came
t
h
e
bac
k
b
one
o
f
al
l re
leva
nt f
u
n
c
t
i
o
ns
i
n
the
veh
i
cle
,
o
pe
n
i
ng
ho
ri
zo
ns fo
r
ne
w
ele
ctr
i
cal
a
rc
h
itec
t
ures.
I
n
l
a
s
t
few
yea
r
s,
E
lec
t
r
i
c
V
e
hic
l
e
star
t
e
d
replac
i
n
g
c
o
nve
n
tio
n
a
l
v
e
h
i
c
l
e
i
n
m
a
n
y
c
o
u
n
t
r
i
e
s
,
a
ro
und
t
h
e
wo
rld
.
A
s
en
t
r
y
ra
ng
e
EVs
a
r
e
mo
re
a
f
f
o
rd
abl
e
n
o
w
ad
a
y
s,
t
h
e
se
c
ar
s
are
a
sm
ar
t
c
hoic
e
f
or
a
n
y
e
nd-
user
,
how
e
v
e
r
,
auto
nom
y
a
nd
a
p
p
l
ica
tio
ns
v
ar
i
e
ty
r
em
ain
u
n
so
lve
d
[
1].
M
a
ny
a
ttem
p
ts
t
o
im
prov
e
P
o
w
e
r
Ma
nage
me
n
t
i
n
EV
h
a
v
e
bee
n
d
o
n
e,
a
nd
m
a
ny
s
t
r
uc
tures
have
b
e
e
n
de
v
e
lo
pe
d.
T
o
d
a
y
,
i
t
i
s
c
l
ea
r
tha
t
c
urrent
st
a
t
e of t
he
a
rt is r
eac
hi
ng
a
p
oin
t
w
here
new
a
nd
sig
n
i
fica
n
t
l
y di
ffere
n
t
a
r
c
h
ite
ct
ures
a
re
nee
ded [
2
].
Evaluation Warning : The document was created with Spire.PDF for Python.
Int J
P
o
w
El
e
c
&
D
ri S
yst
I
S
S
N
:
2088-
86
94
Sm
ar
t
da
t
a
bas
e
c
oncep
t for po
wer m
a
n
agem
e
nt in a
n
elec
trica
l
veh
ic
le
(Mahm
ou
d
i
C
hok
ri)
16
1
Wit
h
a
b
e
tter
un
de
rsta
n
d
i
ng
for
o
u
r
needs,
a
nd
ba
se
d
o
n
s
imilar
a
p
p
roa
c
h
es
i
n
o
t
her
fi
eld
s
s
uc
h
a
s
rob
o
t
i
c
s
a
n
d
c
om
pute
r
s
c
i
e
n
ce,
w
e
have
a
n
ew
v
i
s
io
n
for
ele
c
tri
c
v
e
h
i
cle
in
t
he
n
e
x
t
de
cade
.
H
ere
,
w
e
form
ula
t
e
the
power
m
a
n
age
m
ent
co
n
t
ro
l
p
r
ob
lem
and
prese
n
t
t
h
e
co
n
t
rol
a
l
gori
t
h
ms
t
hat
ca
n
be
u
se
d
t
o
deri
ve
t
he
o
p
t
i
m
a
l
c
o
n
tr
ol
p
o
l
ic
y
[3].
I
n
our
p
er
spe
c
t
ive,
a
n
E
l
e
c
t
ri
c
Ve
hi
cl
e
eq
uip
p
e
d
wi
th
l
ea
rni
n
g
Po
we
r
Ma
nage
me
n
t
T
ech
ni
q
u
es,
ca
n
e
nha
nc
e
i
t
s
exper
i
e
n
ce
o
ve
r
the
t
i
m
e
by
le
arni
ng
n
e
w
c
on
d
iti
on
s
a
n
d
ma
kin
g
new
deci
si
o
n
s
[4].
B
u
t
i
f
th
e
kn
ow
le
dge
i
s
share
d
b
e
t
w
e
e
n
m
any
EV
s,
it
ca
n
be
a
m
ajor
l
e
a
p
f
o
rw
ar
d.
Re
fe
re
nce
s
o
f
va
ri
ous
r
esea
rc
h
e
fforts
that
h
ave
use
d
t
hese
a
l
g
or
it
hm
s
in
v
ario
us
c
o
n
fi
g
u
rati
o
n
s
a
r
e
pro
v
i
de
d
in
t
his
paper
as
[
5]-[7].
I
n
addi
tion,
w
e
are
pr
o
posing
a
new
a
ppr
oac
h
e
s
t
ha
t
ca
n
m
a
ke
i
nno
va
ti
on
i
n
P
ow
e
r
Ma
nage
me
n
t
A
rc
hi
tec
t
ur
e
and lea
d
o
ther
r
ese
a
rc
hers
t
o
u
n
ex
p
l
ore
d
i
d
eas
Th
is
n
ew
a
r
c
hi
tect
ure
is
b
a
s
e
d
o
n
a
sm
art
netw
ork
i
ng
bet
w
ee
n
EV
s
r
u
nni
n
g
a
n
en
ha
nc
ed
v
er
sio
n
o
f
the
lear
n
i
ng
P
o
w
e
r
m
a
nage
me
nt
c
o
n
t
r
o
l
a
l
g
o
r
i
t
h
m
and
c
o
m
m
unica
ti
ng
all
the
da
ta
t
o
a
Sm
a
r
t
D
a
tab
a
se.
The
DB
i
s
rei
n
fo
rced
w
i
t
h
ANNs
a
t
t
r
i
b
u
ti
on
an
d
sel
e
cti
o
n
p
r
o
g
r
ams
t
o
m
a
n
a
g
e
o
p
t
i
m
a
l
t
r
a
f
f
i
c
.
T
h
i
s
n
e
w
a
r
ch
it
ec
tu
re
t
ec
h
n
i
c
all
y
o
f
f
ers
c
o
mmu
ni
ca
t
i
o
n
,
e
xp
erie
n
c
e
sh
a
r
i
n
g
an
d
be
t
t
e
r
p
ower
m
a
n
agem
ent
ac
hi
e
v
em
en
ts. We
bel
i
e
v
e
it c
a
n
be a
ga
m
e
c
h
a
n
ger
i
n
A
u
t
o
m
oti
v
e
I
nd
ustry.
The
re
ma
inde
r
of
t
h
i
s
pa
per
proc
eeds
as
f
ol
l
o
w
s
.
A
f
ter
a
ge
nera
l
in
t
r
oduc
tio
n,
w
e
int
r
o
d
uc
e
in
Se
ctio
n
II
the
pow
e
r
m
a
n
age
m
ent
c
o
ntrol
a
l
g
o
ri
t
h
m
s
f
or
E
lec
t
ric
V
ehicles
to
d
ate.
I
n
S
ection
III
we
e
x
p
lain
the
tw
o
k
now
m
etho
ds
o
f
P
o
w
e
r
m
a
na
gem
e
nt
A
rchi
tec
t
ure
.
T
hen,
w
e
d
et
ail
o
u
r
pro
pose
d
c
on
tri
b
uti
o
n
i
n
t
h
e
ma
nage
me
nt
o
f
pow
er
i
n
a
n
e
lec
t
r
i
c
a
l
v
eh
ic
le.
F
i
nal
l
y,
w
e
conc
l
ud
e
t
h
e
e
xpo
se
d
wo
rk
w
it
h
so
me
f
u
ture
r
esearch
.
2.
E
V
AND
P
OW
ER
M
ANA
GEM
E
NT
Co
nt
rol
st
ra
t
e
gi
e
s
f
o
r
h
y
b
r
i
d
-e
l
e
ct
r
i
c
v
e
hi
cle
s
g
e
n
e
r
al
ly
t
arg
e
t
se
ver
a
l
s
i
multa
ne
ou
s
ob
jec
t
i
v
es.
T
h
e
prima
r
y
o
n
e
i
s
t
he
m
i
n
im
iz
ati
o
n
of
t
he
v
eh
ic
le
f
ue
l
c
onsum
p
t
ion
,
wh
il
e
a
l
so
a
t
t
e
mp
t
i
ng
t
o
mi
ni
mi
ze
em
issi
on
s
a
n
d
to
m
a
i
n
t
a
i
n
or
e
n
h
ance
d
r
i
va
b
i
l
ity.
T
o
d
a
t
e,
t
he
p
o
w
e
r
m
a
na
gem
e
n
t
(
P
M
)
syste
m
i
n
E
V
s
i
s
bas
i
ca
ll
y
form
ed
b
y
t
w
o
la
ye
rs;
H
i
g
h
l
e
v
e
l
s
o
f
tw
are
-
based
super
v
i
s
i
on
a
n
d
l
o
w
le
ve
l
har
d
w
a
re
-based
c
ontro
l
w
h
ic
h
ca
n
be
d
i
v
ide
d
i
n
t
o
tw
o
co
ntr
o
l
l
a
yer
s
l
ow
l
e
v
e
l
c
o
m
pon
e
n
t
a
n
d
l
o
w
leve
l
c
o
n
t
rol.
B
o
t
h
ha
rdw
a
r
e
a
nd
softw
a
re
c
ontr
o
l
l
a
yers
w
orks
toge
t
h
er
to o
p
t
i
mize
P
M
syst
e
m
i
n
E
V
s
a
s it
is
e
xp
lai
n
e
d
i
n
[8],
[9].
Ma
jor
c
h
a
l
l
e
n
g
e
o
f
e
nergy
ma
nage
me
nt
s
yst
e
m
(EMS
)
in
a
n
e
l
ec
t
r
i
c
v
e
hi
cl
e
i
s
t
o
a
s
su
re
o
p
t
i
m
al
u
se
and
re
gener
a
tion
o
f
t
he
t
o
t
a
l
e
nerg
y
i
n
t
he
v
e
h
i
c
le
[
1
0
].
R
e
g
a
r
dle
ss
of
numbe
r
of
s
o
u
r
ces,
the
p
o
w
e
rtrain
con
f
ig
ure
u
rat
i
on,
a
t
a
n
y
tim
e
and
f
o
r
any
ve
hic
l
e
spee
d,
t
he
c
o
n
t
ro
l
str
a
teg
y
h
as
t
o
deter
m
i
n
e
t
h
e
p
o
w
e
r
di
stri
b
u
t
i
on
b
e
t
w
e
en
d
iffere
n
t
e
ne
rgie
s.
W
he
n
tw
o
s
t
or
a
g
e
s
y
ste
ms
o
r
t
w
o
fue
l
c
on
ve
rters
ar
e
av
aila
b
l
e
add
i
tio
na
l
po
w
e
r
distri
b
u
t
i
o
n
b
e
t
w
e
en
t
he
R
ES
S
s
a
nd
b
e
tw
een
t
he
f
u
e
l
c
o
n
v
erter
s
h
as
t
o
be
d
e
t
e
r
m
i
ne
d.
These
decis
i
on
s
are
cons
tra
i
n
e
d
b
y
t
w
o
f
ac
t
o
rs.
First
of
a
l
l
,
t
he
m
ot
ive
pow
er
r
equeste
d
b
y
t
h
e
drive
r
m
ust
alw
a
ys
b
e
sat
i
s
f
ie
d
u
p
t
o
a
ma
ximum
pow
e
r
d
em
and
a
l
re
ad
y
know
n.
T
hen
,
c
ha
rge
stat
us
m
ust
be
m
ai
nt
aine
d
wi
t
h
in
, a
l
lo
wing
t
h
e
v
e
h
i
c
le to
be
c
harge
con
t
in
u
ousl
y
a
s it i
s
exp
la
ine
d
i
n [11]
Ma
ny
fac
t
ors
ca
n
a
f
fec
t
t
he
E
V
perform
an
ce
,
such
a
s
s
i
z
e
,
purp
o
se
o
f
use,
e
n
v
i
r
onm
e
n
t
,
d
ri
vin
g
style
(sp
o
rt
y,
s
oft,
m
ode
rate
o
r
com
b
ine
d
).
A
l
l
t
hese
f
ac
t
o
r
s
m
a
y
l
ead
t
o
a
deep
a
n
d
q
u
i
ck
d
i
s
char
ge
r
a
t
e
o
f
the ba
tte
ry a
n
d
i
t
s
d
am
age
[12
]
.
To
k
e
e
p
it
he
alt
h
y
an
d
gu
id
e
i
t
t
o
a
sl
ow
d
isc
h
ar
ge
e
ve
n
w
h
e
n
a
h
e
a
vy
l
o
ad
i
s
on
de
ma
nd,
t
he
elec
tr
ic ve
h
i
c
l
e
is pow
e
r
ed
by
a
com
b
ina
t
ion
of m
ul
tip
le
s
o
u
r
ce
s
[
13]-[15].
The
m
a
in
e
lem
e
nt
i
s
the
bat
t
e
r
ies.
M
os
t
of
t
he
e
l
e
c
t
ric
veh
i
cle
s
use
Lit
h
iu
m
ion
bat
t
e
r
y.
L
i
t
h
i
um
i
o
n
bat
t
eries
ar
e
e
n
v
i
ro
nm
en
tal
l
y
frien
dl
y
a
nd
ha
ve
h
ig
he
r
ener
g
y
d
e
n
si
t
y
,
lo
nger
l
i
fe
s
p
a
n,
a
nd
hi
g
h
er
pow
e
r
den
s
i
t
y
t
h
an
c
o
nve
n
t
i
o
nal
bat
t
ery
[1
6]
.
They
h
ave
w
i
de
a
p
p
l
i
c
at
io
n
i
n
e
l
e
ct
ri
c
v
e
hi
cl
e
s
a
nd
o
t
h
e
r
e
l
ect
ron
i
cs.
Since
la
rge
n
u
m
b
e
r
of
L
ith
iu
m
-
ion
ba
tte
ries
u
se
d
i
n
s
erie
s
in
e
l
e
ct
ri
c
v
e
h
i
cl
es
s
o
t
h
e
r
e
ari
s
e
s
t
h
e
p
robl
e
m
s
of
sa
f
e
t
y
,
d
u
r
a
b
ilit
y
,
t
h
e
rma
l
b
rea
k
do
wn
a
n
d
co
st
,
wh
i
c
h
l
i
m
i
t
s
t
h
e
a
p
plica
t
i
on
of
t
he
L
it
hi
um
i
o
n
b
a
t
tery
[
17]
.
S
o
m
e
e
lec
t
ric
ve
h
i
c
l
es
u
se
o
t
h
er
k
inds
o
f
ba
t
t
e
r
ies
suc
h
a
s
P
l
u
m
b
-A
cid,
N
ic
ke
l
-
Ca
dmium
,
a
nd
l
i
t
h
i
um
-
po
lym
e
r
.
T
he
s
elec
ti
o
n
o
f
a
b
a
t
t
e
r
y
is
b
ase
d
on
m
a
n
y
c
riter
i
a
,
s
u
ch
a
s
en
erg
y
,
w
ei
g
h
t
,
l
ife
t
i
m
e,
p
ri
c
e
,
v
o
l
t
a
g
e
,
s
i
ze
[
18
]
,
[
1
9
]
.
To
o
bt
ai
n
a
po
wer
b
o
o
s
t
,
s
up
e
r
c
a
p
a
c
i
t
o
r
is
u
sed
.
I
t
h
a
s
t
h
e
c
h
ar
ac
ter
i
s
tic
s
o
f
b
o
t
h,
c
a
p
ac
it
or
a
n
d
bat
t
ery.
I
t
ca
n
re
lease
a
large
c
h
arge
i
n
a
s
hor
t
peri
od.
A
s
up
e
r
c
a
p
a
c
i
t
o
r
ba
n
k
i
s
he
nc
e
ad
o
p
te
d
to
s
up
pl
y
in
sta
n
ta
ne
o
u
s
cha
r
ge
t
o
ass
i
s
t
t
he
m
a
i
n
bat
t
ery
in
h
e
a
vy
c
o
n
s
u
m
p
t
io
n.
T
he
s
up
er
capa
c
i
t
o
r,
u
n
d
er
ma
nage
me
nt,
can be
cha
r
ge
d
by t
h
e
ma
i
n
b
a
tter
i
es
[
2
0
],
[21].
R
e
cen
tl
y
,
m
an
y
ma
n
u
f
act
u
r
e
r
s
ac
c
o
rd
ed
m
ore
at
t
e
n
t
io
n
t
o
s
ol
ar
p
ane
l
s.
T
hey
w
i
l
l
p
rov
i
de
t
he
pow
e
r
ma
nage
me
nt
s
ys
tem
w
ith
a
n
au
xi
lia
ry
e
lec
t
r
i
c
e
n
e
r
g
y
t
o
b
e
u
se
d
later
for
ba
t
t
er
y
c
h
a
r
g
i
ng
o
r
e
l
ec
tron
ics
pow
er
sup
p
l
y
i
ng a
s
i
t is exp
l
a
ine
d
b
y [2
2]
, [2
3
].
I
n
o
rd
e
r
t
o
r
e
c
ove
r
k
i
ne
t
i
c
e
n
ergy
los
t
i
n
ve
hi
c
l
e
bra
k
i
n
g
e
l
ec
tr
ic
v
e
h
i
c
le
s
c
a
n
a
l
s
o
s
a
v
e
e
n
e
r
g
y
i
n
stop
a
nd
g
o
d
ri
vi
ng
t
h
ro
u
gh
r
e
ge
nera
t
i
ve
b
r
a
ki
ng.
I
n
t
h
is
t
echn
iq
u
e
,
t
h
e
El
ect
ri
c
mo
t
o
r
i
s
u
se
d
a
s
a
g
en
era
t
o
r
c
o
nv
e
r
t
i
n
g
t
h
e
k
in
et
i
c
o
f
th
e
v
e
hi
cl
e'
s
mot
i
o
n
b
a
ck
t
o
e
l
ectri
c
e
n
e
rg
y
,
r
ath
e
r
th
an
d
iss
i
pa
t
i
ng
i
t
a
s
h
eat
i
n
the
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN: 2088-
8694
I
nt
J
P
ow
Elec
& Dr
i
S
y
st, Vol. 10,
N
o.
1, Mar
c
h 2
0
1
9
:
16
0 –
16
9
16
2
brake
s
.
T
h
e
re
gener
a
tive
br
aki
n
g
c
a
n
r
e
c
over
50
%
to
8
0
%
o
f
t
h
e
ki
net
i
c
e
n
er
g
y
f
or
l
a
t
er
u
s
e
.
Th
is
i
s
espec
i
al
ly
v
al
u
a
ble
for
veh
i
c
l
es tha
t
sto
p
a
n
d
start
f
re
q
u
e
n
tly
l
ike
buses and in-city BEV
s [
4
]
.
Fo
r
B
E
Vs
a
nd
P
HEVs
,
mo
re
t
h
a
n
sou
r
ce
o
f
en
ergy
c
an
b
e
u
s
ed
but
G
r
id
P
ow
e
r
i
s
the
ma
i
n
e
nerg
y
source
.
It
a
ll
o
w
s
char
ging
b
a
t
t
er
i
e
s
an
d
su
per
c
a
pac
i
t
o
rs.
M
a
n
y
c
har
g
i
n
g
m
odes
are
avai
la
bl
e
w
i
t
h
e
nha
nce
d
cha
r
g
i
ng
tim
e.
A
pow
e
r
m
a
n
age
m
e
n
t
uni
t
base
d
o
n
s
ma
rt
a
lg
ori
t
h
m
s
m
a
na
ges
t
h
e
sourc
e
s
a
nd
pe
rfor
ms
com
b
i
n
a
t
i
o
ns
o
r
timi
n
g
be
tw
e
e
n
the
m
t
o
ob
ta
in
o
p
t
i
m
al
v
e
h
icle
r
es
po
ns
ive
n
es
s
and
ba
ttery
h
ea
lt
h.
T
h
i
s
pow
er
is trans
f
e
rre
d to a
re
gula
r
m
otor
c
on
tr
ol
u
n
i
t
,
w
hic
h
d
ri
ves the
ve
h
i
cle.
3.
GE
N
E
RALITY
O
N
O
FFL
INE AND
O
N
L
INE POWER
M
A
N
A
G
E
MENT CONTR
O
L
ALG
O
RIT
H
M
G
e
nera
lly,
tw
o
cate
g
o
r
ies
of
a
lg
ori
t
h
ms
a
re
p
r
e
sent
i
n
t
h
e
p
o
w
e
r
ma
nage
me
n
t
f
or
t
he
E
V
app
l
ica
t
i
o
n.
T
he
o
ffl
i
n
e
p
o
w
e
r
m
a
nage
me
n
t
a
lg
ori
t
hm,
w
h
i
c
h
is
b
a
s
ed
o
n
an
o
p
t
i
m
iz
ati
o
n
Cri
t
eria
a
nd
t
h
e
on
line
al
g
o
ri
th
m, w
hich
i
s
ba
sed
o
n
a
p
re
d
i
c
t
i
v
e
co
ntr
o
l
l
er.
The
offl
i
n
e
a
l
g
o
r
ithm
w
a
s
ba
se
d
on
a
s
toc
h
ast
i
c
o
p
t
ima
l
c
on
t
r
o
l
o
f
c
o
mpl
e
x
dyn
a
m
i
c
s
y
s
t
e
ms.
Th
e
pro
b
lem
is
f
or
m
u
l
a
t
e
d
as
s
e
q
ue
nt
ial
de
c
i
si
o
n
m
aki
n
g
u
nde
r
unc
er
ta
in
t
y
,
w
h
e
r
e
a
contr
o
ller
is
f
a
ced
w
ith
t
he
t
a
sk
o
f
sel
ecti
n
g
act
ion
s
i
n
s
e
v
e
ral
ti
me
s
t
e
ps
t
o
e
f
fi
ci
en
tl
y
achie
ve
t
he
s
ys
tem
’
s
l
o
n
g
-te
r
m
goa
ls.
T
h
is
i
s
w
a
s
exp
l
a
i
ne
d
b
y
[24]-
[
26].
Effec
tive
l
y,
t
h
e
s
ys
t
e
m
is
o
b
s
erve
d
dur
in
g
a
com
p
l
e
t
e
d
p
e
rio
d
o
f
t
i
m
e
T
k
now
n
a
s
“
the
dec
i
s
i
on-
ma
kin
g
h
or
i
z
o
n
”
an
d
c
a
n
be
e
it
her
fi
nite
o
r
i
n
f
i
n
i
te.
F
o
r
the
f
in
ite
d
ec
isi
on-m
a
ki
n
g
h
oriz
on
prob
le
m
,
t
h
e
ob
jec
t
i
v
e
is
t
o
dr
i
v
e
the
o
p
t
i
m
al
c
on
tro
l
p
olic
y
tha
t
m
in
i
m
ize
s
t
ot
al
e
x
p
ect
ed
c
ost
cri
t
eri
o
n
,
but
i
n
ma
j
o
r
ca
se
s,
t
he
horiz
on
lim
i
t
i
s
p
u
s
h
t
o i
n
fi
n
i
te
as
i
t
i
s
exp
o
se
d i
n (1
)
.
l
i
m
→
∞
1
1
∑
,
0
(1)
I
n
t
he o
ffl
i
ne a
lg
ori
t
hm
, w
e
c
a
n
a
l
s
o fou
n
d
t
he D
ynam
i
c program
m
i
n
g
(DP
),
w
hich is ge
ne
ral
i
ze
d
a
s
the
ma
i
n
m
et
ho
d
t
o
a
nal
y
z
e
s
eque
n
t
ia
l
d
e
c
i
sio
n
-m
aki
n
g
pro
b
l
em
s,
s
uch
as
d
eter
min
i
st
ic
a
n
d
s
t
o
cha
s
ti
c
op
tim
iza
t
i
o
n
and
c
o
n
t
ro
l
pro
b
lem
s
,
mini-m
ax
p
ro
b
l
em
s,
a
nd
ot
her
v
a
r
ie
d
pro
b
l
em
s
[2
7].
While
t
he
n
a
t
ure
o
f
t
h
ese
p
r
ob
l
e
m
s
m
a
y
v
a
r
y
wi
d
e
l
y
,
t
h
ei
r
u
nde
rl
yi
ng
s
t
r
u
c
tu
re
i
s
s
imilar
t
o
e
a
c
h
othe
r
an
d
has
t
w
o
pri
n
cipa
l
fe
at
ures:
a
n
u
nder
l
y
i
ng
dis
c
re
te
t
im
e
dyna
mic
syst
e
m
w
hos
e
st
a
t
e
evol
v
e
s
a
c
c
o
rdin
g
t
o
g
iv
en
t
ra
nsit
i
on
p
r
ob
a
b
il
i
t
i
e
s
th
at
d
ep
end
on
t
h
e
d
eci
si
on
t
ake
n
a
t
e
ach
t
i
m
e
a
n
d
a
c
o
st
f
u
n
c
ti
o
n
t
ha
t
is
a
d
d
i
t
i
ve
ove
r
tim
e
as
it
is
c
i
t
ed
i
n
[28],
[24].
The
se
c
o
n
d
c
ateg
ory
of
t
h
e
powe
r
m
a
n
agem
ent
al
g
o
rithm
s
c
an
b
e
c
alle
d
t
h
e
O
n
li
ne
P
ow
e
r
Ma
nage
me
n
t
C
ontro
l
al
g
o
ri
t
h
m
.
T
he
M
od
el
p
re
dic
t
ive
c
o
n
t
ro
l
(
M
P
C
)
is
o
ne
o
f
the
on
l
i
ne
a
l
gor
ith
m
s
it
is
relies
o
n
p
re
di
cti
o
n
m
ode
ls
t
o
o
b
ta
i
n
a
c
on
tr
ol
a
ct
ion
by
sol
v
i
ng
a
n
o
n
li
ne
o
pt
i
m
i
z
a
t
io
n
p
r
ob
l
e
m
ov
e
r
a
f
i
n
ite
hor
izo
n
.
It
i
s
ofte
n
u
s
e
d
i
n
con
s
trai
ne
d
re
gu
la
tor
y
r
ela
t
e
d
c
on
tr
ol
p
rob
l
em
s
o
f
l
ar
ge
s
ca
le
m
u
l
tiva
r
i
a
ble
syste
m
s,
w
he
re
t
he
o
b
j
e
c
t
i
v
e
is
t
o
oper
a
te
t
he
s
ystem
in
a
c
e
r
t
ai
n
de
s
i
r
e
d
w
a
y
[2
5].
A
t
a
s
pec
i
fic
t
i
me
“
t”,
the
a
c
t
u
a
l
s
t
a
t
e
o
f
t
h
e
s
y
s
t
e
m
i
s
s
a
m
p
l
e
d
,
a
n
d
c
o
n
t
r
o
l
s
t
r
a
t
e
g
y
i
s
c
a
lc
ula
t
e
d
f
or
a
r
e
l
ati
v
e
l
y
shor
t
t
i
me
hori
z
on
N
h
e
r
e
,
l
(
X
t
,
U
t
)
i
s
a
co
s
t
f
un
c
t
ion
,
a
s
f
o
l
l
o
w
ing:
mi
n
∑
,
(2)
,
,
,
∈
∶
∈
(3)
A
l
so,
P
ontr
y
ag
in’s
M
i
n
im
u
m
P
r
i
nc
i
p
le
a
nd
ECMS
a
re
u
sed.
T
he
p
ri
nci
p
le
i
s
re
l
a
te
d
to
r
eso
l
v
i
ng
a
n
op
tim
iza
t
i
o
n p
r
ob
lem
by der
i
vi
n
g
a
s
et
of ne
ce
ssa
r
y
c
o
n
d
it
i
ons
t
h
at
m
ust
be
s
a
t
i
s
fi
e
d
b
y
an
y
opti
m
al
s
ol
u
tio
n
.
These
c
o
n
d
iti
ons
b
ec
om
e
suffi
c
i
e
n
t
un
de
r
ce
rtain
c
o
nv
exi
t
y
c
o
n
d
i
t
i
on
s
on
t
h
e
obj
ec
t
i
v
e
a
nd
c
on
st
rai
n
t
fu
nc
ti
o
n
s
[2
9]
.
We
c
an
f
o
und
als
o
,
the
Rule
-
B
a
s
e
d
P
ow
er
M
anage
m
e
n
t
Con
t
ro
l
A
l
g
o
rith
ms.
Ef
f
e
c
t
iv
ely
thi
s
me
tho
d
r
e
l
ays
on
e
x
pert
e
x
p
e
r
i
e
nce
ba
se
t
o
dete
rmi
n
e
f
i
ne
a
dj
us
t
m
e
n
t
s
t
o
b
e
a
p
p
l
i
e
d
i
n
P
M
C
a
l
g
o
r
i
t
h
m
.
T
h
e
P
M
C
str
a
teg
y
c
a
n
b
e
base
d
o
n
f
uzz
y
l
o
g
ic,
dece
n
t
ral
i
ze
d
a
d
ap
t
i
v
e
l
o
g
i
c
,
o
r
e
v
e
n
n
e
w
s
e
t
o
f
r
u
l
e
b
a
s
e
d
P
M
C
alg
o
ri
t
h
ms [30
],
[
31].
To
o
p
t
im
ize
E
V
e
ffic
i
e
n
cy,
PMC
a
l
g
o
r
ith
m
s
i
ncl
u
de
a
l
ea
rni
ng
m
e
ch
an
is
m
t
h
a
t
a
llows
i
mp
ro
vi
ng
perform
ance
o
ver
time
,
e
ve
ry
s
i
n
gle
rea
c
tion
of
t
he
d
ri
v
e
r
is
c
on
side
red
inc
l
ud
in
g
d
r
ivi
n
g
st
y
l
e,
s
pri
n
t,
brea
k
i
ng
s
t
yle,
a
n
d
d
is
ta
nce
s
d
ri
ven.
A
l
l
t
h
e
se
c
o
l
lec
t
e
d
i
nfor
mat
i
o
n
buil
d
a
d
a
t
ab
a
s
e
sp
eci
fi
c
t
o
t
h
e
u
se
r
dri
v
in
g
s
t
y
l
e
a
n
d
t
h
e
r
e
ar
e
PM
a
d
j
us
t
m
en
ts
c
om
mun
i
ca
t
e
d
to
d
r
i
v
ing
pa
r
a
m
e
ters.
T
h
is
h
as
a
m
ajor
i
m
p
ac
t
o
n
fue
l
e
c
o
n
o
m
y
a
nd
s
y
s
t
e
m
r
espo
ns
ive
n
e
ss
[30].
Th
is
p
r
i
nc
ip
l
e
i
s
i
n
c
l
u
d
e
d
i
n
t
h
e
S
m
a
r
t
/
L
e
a
r
n
i
n
g
P
o
w
e
r
Ma
nage
me
n
t
C
on
tro
l
A
l
g
ori
t
h
m
s.
Evaluation Warning : The document was created with Spire.PDF for Python.
Int J
P
o
w
El
e
c
&
D
ri S
yst
I
S
S
N
:
2088-
86
94
Sm
ar
t
da
t
a
bas
e
c
oncep
t for po
wer m
a
n
agem
e
nt in a
n
elec
trica
l
veh
ic
le
(Mahm
ou
d
i
C
hok
ri)
16
3
GPS
e
n
h
a
n
c
ed
P
o
w
er
M
an
ag
e
m
e
n
t
Co
nt
ro
l
Algo
rit
h
ms.
Th
e
s
e
a
l
g
o
r
i
t
hms
are
to
e
n
h
ance
P
M
C
alg
o
ri
t
h
ms
u
si
ng
i
n
f
orm
a
tio
n
re
ceive
d
from
a
G
lo
b
a
l
P
o
si
ti
on
ing
S
ys
tem
(
G
PS
).
T
he
a
lg
ori
t
hm
u
ses
da
ta
a
nd
loa
d
s
c
o
rre
spond
i
ng
t
opo
gr
aph
y
o
f
the
roa
d
a
n
d
o
per
a
tes
acc
ord
i
ng
t
o
p
r
e
c
on
fi
gu
re
(
u
r
e
d
d
ri
vi
ng
s
ty
l
e
t
o
minim
i
z
e
f
u
e
l
cons
ump
t
io
n.
T
he
se
e
nha
n
cem
ent
a
l
g
o
ri
t
h
m
s
a
re
u
si
n
g
d
ri
v
i
ng
pa
tt
ern
r
ecog
n
i
t
i
on
t
o
a
u
t
o
m
a
t
i
c
a
l
l
y
s
e
l
e
c
t
a
c
o
n
t
r
o
l
a
l
g
o
r
i
t
h
m
f
r
o
m
a
b
a
n
k
o
f
s
i
x
o
p
t
imiz
e
d
r
epr
e
sen
t
at
i
v
e
dri
v
in
g
m
ode
s
usin
g
art
i
f
i
c
i
a
l
n
e
u
r
al
n
e
t
w
o
rk
s
(
ANNs)
[
3
2
]
.
Man
y
wo
r
k
s
cont
rib
u
t
e
d
to
g
i
v
e
a
bet
t
er
u
n
d
e
r
s
ta
ndi
n
g
f
or
P
ow
e
r
Ma
nage
me
n
t
i
n
El
ec
t
r
i
c
V
e
h
icle.
D
e
pe
nd
in
g
o
n
pow
er
t
r
ai
n
arc
h
i
t
e
c
t
ure,
v
ari
ous
P
M
C
a
lg
ori
t
hm
s
have
b
ee
n
in
t
r
od
uc
ed
[
20
],
[
33],
[21
]
,
[22].
Th
is
w
or
k
foc
u
ses
i
n
i
m
p
ro
vi
n
g
l
ear
nin
g
a
lgor
it
hm
s
an
d
e
n
ha
nci
n
g
E
V
a
bi
l
i
t
y
t
o
m
a
ster
pow
e
r
ma
nage
me
n
t
o
ve
r
ti
m
e
a
n
d
e
xper
i
e
n
ce.
D
riv
i
n
g
f
ac
tor
s
a
nd
dri
v
er
b
e
h
avior
has
m
a
jor
im
pact
o
n
f
u
el
efficie
n
c
y
,
ve
h
i
c
l
e
res
p
o
n
s
i
ve
ness
a
n
d
dri
v
i
n
g
a
u
t
o
n
o
m
y
[
2
9
]
.
U
p
to now
,
t
h
e
r
e
sea
r
ch
r
e
por
t
e
d
few
p
o
te
ntia
l
works
in
t
his
category
;
a
m
u
l
t
i
-m
odes
P
M
C
stra
t
e
g
y
f
or
p
ara
l
l
e
l
H
E
V
p
r
o
p
o
se
d
i
n
2
0
0
2
b
y
Je
on
us
in
g
dri
v
in
g
p
a
t
t
e
r
n
re
cog
n
i
tio
n
t
o
s
e
l
ec
t
a
co
n
t
ro
l
al
g
o
ri
thm
a
u
t
o
m
at
ica
l
l
y
from
si
x
o
p
t
i
m
ize
d
d
riv
i
ng
m
odes
us
i
n
g
A
r
ti
fic
i
a
l
N
eur
a
l
N
e
tw
o
r
ks
(
A
N
N
s
)
[34
]
,
[35],
[14].
O
n
l
i
n
e
pre
d
i
c
ti
on
of
f
u
t
ure
dr
iv
i
ng
c
y
c
l
e
bas
e
d
o
n
rec
o
rded
d
a
t
a
was intro
d
u
ce
d
by
[3
6].
The
lear
nin
g
s
t
r
ateg
y
w
a
s
ex
p
l
ai
ne
d
b
y
C
he
n
[37]
a
nd
S
a
lm
an
[
3
8
]
f
o
r
paral
l
el
H
EV
s
to
m
axim
i
z
e
fue
l
e
c
o
n
o
m
y
.
A
l
ear
ni
n
g
t
ec
hn
i
que
i
s
a
p
p
l
i
e
d
t
o
t
he
c
os
t
func
tio
n
in
o
r
d
e
r
t
o
ad
jus
t
p
a
r
a
m
eter
s
in
r
ea
l
t
i
me
.
N
e
ur
o-dy
nam
i
c
pr
ogr
am
ming
w
a
s
als
o
u
s
e
d
i
n
pow
er
m
ana
g
em
en
t
c
o
n
t
r
o
l
a
lg
ori
t
h
ms.
F
e
w
years
la
ter
,
K
o
lma
n
ovsk
y
a
nd
D
e
x
t
rei
t
i
ntr
o
d
u
c
e
d
G
am
e
The
o
ry
A
l
gor
ithm
s
i
n
p
o
w
er
m
a
n
a
g
emen
t
and
e
x
p
e
ri
me
n
t
al
resul
t
s o
n
La
n
d R
over
F
r
e
e
lande
r
HEV
m
a
rk-
2
wer
e rem
a
rkab
le [
10
].
Th
us,
i
t
is
c
l
e
a
r
tha
t t
o
ob
t
a
i
n
the
be
st
ener
g
y
co
ns
um
pti
on, a
rea
l-t
i
m
e
co
n
t
ro
ller mus
t
a
d
a
pt
i
tse
l
f to
vary
in
g
dr
ivi
n
g
c
i
rc
um
st
a
n
ce
s
an
d
c
o
ndi
t
i
o
n
s.
I
n
2
0
1
1
,
in
c
l
u
d
i
n
g
dri
v
i
n
g
co
n
d
i
t
ions
i
n
pow
er
m
a
n
ag
em
ent
has
been
t
h
ough
o
u
t
[
3
0
]
.
G
l
o
b
a
l
P
os
i
t
i
o
n
i
ng
S
y
stem
(
G
P
S
)
r
e
t
r
i
e
v
ed
i
nf
orma
tion
w
e
r
e
u
se
d
t
o
d
e
t
erm
i
ne
u
p
c
o
m
i
n
g
topo
g
r
ap
hy
o
f
the
ro
ad
a
nd
a
d
j
ust
P
M
C
p
a
ramet
e
rs.
An
e
nv
iro
n
m
e
n
t
f
r
i
end
l
y
a
r
ch
i
t
ect
ure
is
pro
pose
d
f
or
H
EV
s
by
L
i
in
2
01
2
t
o
i
m
p
r
o
v
e
e
ffic
ienc
y
[3
2].
Thi
s
c
o
n
c
e
p
t
i
n
t
e
g
r
a
t
e
s
t
h
r
e
e
m
o
d
u
l
e
s
;
a
c
l
e
a
n
ene
r
g
y
pow
e
r
tr
ain,
a
nd
el
ectri
fie
d
cha
ss
is a
n
d
in
t
e
lli
ge
nt
i
n
f
o
r
m
a
t
i
o
n i
n
ter
acti
o
n dev
i
c
e
s.
4.
PROPOS
E
D
CONTRIBU
T
ION
IN POWE
R
M
ANA
G
EMEN
T
The
c
h
oic
e
of
t
he
a
ppro
p
ria
t
e
t
o
po
l
o
g
y
r
e
q
u
i
re
s
prelim
inar
y
u
n
d
e
r
st
a
ndi
ng
o
f
ve
hi
cl
e
u
s
e
pu
rpo
s
e
s
stud
y
o
f
d
ri
v
i
ng
c
y
c
l
es,
ve
h
i
c
l
e
s
i
ze
a
nd
w
e
ig
ht,
de
sire
d
perf
o
rma
n
ce
,
and
t
ype
o
f
app
l
ica
t
i
o
n.
O
nc
e
t
h
e
to
pol
o
g
y
ha
s
b
e
en
s
e
t
,
the
se
c
o
n
d
s
te
p
is
t
he
d
es
ig
n
of
a
n
en
e
r
gy
ma
na
gem
e
nt
c
o
n
t
ro
l
(E
M
C
)
st
r
a
t
e
g
y
,
w
h
i
c
h
is
a
n
e
ssen
t
i
a
l
key
for
a
n
e
ffic
ie
n
t
e
lec
t
r
i
c
veh
i
c
l
e
[3
9].
We
can
d
e
v
i
s
e
the
P
o
w
e
r
m
a
na
gem
e
n
t
c
on
t
r
ol
in
to
t
ow
le
v
e
l
.
Low
l
e
ve
l
P
M
c
on
tro
l
,
w
h
i
c
h
offer
s
a
r
i
c
h
ra
nge
o
f
arc
h
itec
t
ure
s
a
s
:
S
e
r
i
e
s
H
EV
i
s
c
onve
n
i
e
n
t
for
stop-
an
d-run
u
s
e
,
s
uc
h
as
c
i
t
y
dr
i
v
i
ng.
I
t
c
a
n
r
e
c
o
v
er
e
ne
rgy
f
r
o
m
rege
nera
tive
brea
k
i
ng
a
n
d
fe
e
d
b
a
t
ter
i
e
s
.
Mea
n
w
h
ile,
Pa
ralle
l
H
E
V
ha
s
a
w
e
a
k
b
at
tery
c
apac
i
t
y
[3
8].
The
I
C
E
a
n
d
E
M
c
o
m
p
l
e
m
e
n
t
e
a
c
h
o
t
h
e
r
w
h
i
l
e
dri
v
in
g.
T
hus,
it
ca
n
be
r
el
i
a
ble
i
n
e
i
t
her
c
ity
o
r
hi
ghw
a
y
.
Th
is
k
in
d
o
f
s
truc
ture
g
et
s
a
be
tter
eff
i
c
i
enc
y
b
e
c
a
u
s
e
o
f
t
h
e
r
e
d
u
c
e
d
b
a
t
t
e
r
y
p
a
c
k
a
n
d
s
m
a
l
l
e
l
e
c
t
r
i
c
m
o
t
o
r
.
T
he
m
ain
are
a
,
bot
h
pre
v
io
us
a
r
c
hi
te
cture
s
ca
nno
t
c
over
is
t
he
p
rec
i
se
c
on
tro
l
s
trate
g
y.
T
h
u
s
t
o
w
co
m
p
lex
co
n
f
ig
u
r
a
tion
a
r
e
use
d
;
Se
ries-par
al
le
l
HEV
and
Comp
le
x
H
E
V
.
P
H
E
V
sus
t
ai
ns
l
onge
r
in
E
M
m
ode
t
ha
n
ICE
m
ode.
I
t
is
s
uita
b
l
e
for
b
o
t
h
c
i
t
y
an
d
hi
ghw
ay,
and
shar
es
t
he
s
a
m
e
adva
nta
g
e
s
a
nd
di
sad
v
a
n
ta
g
e
s
of
a
regul
a
r
HEV
[
29][21][
4
0].
For
BEVs,
in-
w
h
ee
l
dr
ive
c
o
n
f
ig
ure
(urat
i
on
is
m
os
t
su
i
t
ab
le
f
o
r
c
it
y
u
s
e
du
e
t
o
l
i
g
htw
e
ig
h
t
a
n
d
fre
que
nt
s
top-a
nd-ru
n
situa
t
io
ns.
BE
V
s
a
r
e
d
e
s
i
g
n
e
d
ma
inly
f
or
s
hort
di
st
a
n
c
e
a
ut
onom
y
d
e
sp
ite
o
f
mi
ni
m
a
l
ene
r
gy
lo
ss
i
n
transm
i
s
s
i
on.
H
and
lin
g
of
B
EV
s w
i
ll be
a
ffec
ted b
y
t
he
ne
w
w
hee
l conf
igur
a
tio
n
and i
n
cr
ease
of
i
t
s
w
eig
h
t
.
I
n
h
ig
h
leve
l,
t
he
p
ow
er
m
ana
g
em
en
t
c
o
ntr
o
l
l
e
r
w
ou
l
d
t
a
k
e
a
d
va
n
t
a
ge
o
f
di
ffe
ren
t
a
l
g
ori
t
hm
s
deve
l
ope
d
for
th
is
p
ur
pose
,
b
u
t
a
ls
o
ta
ke
s
e
v
en
m
ore
a
dva
n
t
age
s
fro
m
enhanc
i
ng
a
l
g
o
ri
t
h
m
s
,
wea
t
her
con
d
i
t
i
on
s,
w
e
a
th
e
r
f
orec
as
t,
G
PS
positi
o
n
a
nd dr
iv
i
ng ex
p
e
rien
ce
[3
1
]
.
4.1.
Int
e
llige
n
t powe
r
mana
gem
e
nt arc
hite
cture
The
i
d
ea
is
sim
p
l
e
;
lear
n
i
n
g
P
M
C
a
lg
orit
hm
c
a
n
b
e
im
pro
v
e
d;
E
V
s
wo
u
l
d be ab
l
e t
o
l
ea
r
n
from
ea
ch
ot
her
t
h
rou
g
h
c
om
munic
a
t
i
o
n;
a
u
ser
ex
pe
rience
e
xc
ha
nge
d
a
t
aba
s
e
,
e
nc
ry
pte
d
t
o
re
spec
t
dri
v
er
s’
p
r
i
va
cy
[3
1]
[
4
1
].
B
y
pro
v
i
d
ing
m
o
re
a
c
c
ura
t
e
an
d
up
t
o
d
a
t
e
da
ta
t
o
p
o
w
e
r
m
a
na
gem
e
n
t
s
yst
e
m,
f
uel
e
c
o
n
o
m
y
c
a
n
be
i
m
p
ro
ve
d
,
reduc
i
ng p
o
l
lu
ta
nt
e
m
i
ssio
n
s,
a
s w
e
ll
a
s e
x
te
nd
ing
b
a
t
tery
l
i
f
e
t
i
m
e
and r
a
n
g
e.
I
n
E
V
P
o
w
e
r
Ma
nage
me
n
t
,
ma
in
i
ntro
d
u
c
e
d
a
l
gori
t
hm
s,
w
he
the
r
t
he
y
ar
e
Off
line
or
O
nl
i
n
e
,
a
r
e
gene
ra
ll
y
a
p
pli
e
d
loca
lly
t
o
o
p
t
i
mize
e
nerg
y
effic
i
enc
y
.
N
o
p
ot
ent
i
a
l
st
ud
ie
s
we
re
i
n
t
ro
du
ce
d
i
n
t
h
i
s
ca
te
g
o
r
y
exc
e
p
t
s
a
f
et
y
t
h
ro
ug
h
V
e
h
i
c
l
e
to
V
e
h
ic
le
c
omm
unica
ti
on
[4
2].
Th
e
new
a
r
chi
t
ec
tur
e
i
n
t
rod
u
c
e
d
i
n
t
h
i
s
work
aim
s
t
o
e
x
p
l
ore
veh
i
c
l
es’
co
m
m
unic
a
tio
n
to
i
m
p
ro
ve
p
o
w
er
m
anage
me
nt
,
t
h
e
ge
ner
a
l
pri
n
c
i
p
l
e
of
t
hi
s
new
pro
pos
iti
o
n
c
an be
sum
m
a
r
ized
i
n F
i
g
u
re
1.
It is
based
on
t
w
o l
eve
l
s.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN: 2088-
8694
I
nt
J
P
ow
Elec
& Dr
i
S
y
st, Vol. 10,
N
o.
1, Mar
c
h 2
0
1
9
:
16
0 –
16
9
16
4
Fi
g
u
re
1
. Smar
t
l
e
a
r
ni
ng
arc
h
i
t
e
c
t
u
r
e co
n
cept
The
f
i
r
s
t
le
vel
;
w
here
v
e
h
ic
le
o
pt
imiz
es
p
ow
e
r
m
a
n
age
m
ent
usin
g
s
e
l
f-lea
r
ni
ng
tec
hni
que
s
a
n
d
bu
ild
s
its
o
w
n
E
nerg
y
E
x
p
e
rience
(
E
.
EX
).
T
his
u
n
i
q
u
e
E
.EX
w
i
l
l
t
a
ke
a
dva
n
t
a
g
e
of
v
ar
i
o
us
d
rivin
g
situa
t
io
ns.
The
way
th
e
ve
hic
l
e
ad
justs
PM
C
al
gor
it
hm
p
ar
am
eter
s
is
r
elate
d
t
o
i
t
s
dri
v
i
ng
ex
pe
rie
n
c
e
;
t
he
veh
i
c
l
e
lea
r
ns
f
rom
dri
v
er
’s
b
eha
v
ior
a
n
d
m
ood,
G
P
S
pos
it
ion,
r
o
a
d
c
o
nd
i
tio
ns
,
ti
me
c
on
di
t
i
on
s,
w
e
a
t
h
e
r
con
d
i
t
i
on
s.
A
ll
t
hese
i
n
f
orm
a
t
i
o
n
w
ill
hel
p
t
o
b
u
i
l
d
a
rich
e
ne
rg
y
ex
per
i
e
n
ce
a
n
d
to
s
et
f
i
n
e
sele
c
t
ion
c
r
it
e
r
ia
f
o
r
t
h
e
n
e
xt
l
e
v
e
l
.
No
w
ac
cord
i
n
g
to
eac
h
d
r
iv
ing
si
t
u
atio
n
,
a
new
set
of
a
d
j
us
tm
ent
para
me
t
e
rs
i
s
pro
v
ide
d
.
Thro
ug
h
t
i
me
,
V
e
hic
l
e
w
i
l
l
d
eve
l
op
a
u
n
i
que
u
n
d
e
r
st
a
n
din
g
f
or
e
nerg
y
ma
na
gem
e
nt
.
Its
ex
per
i
e
n
c
e
w
i
ll
grow
,
w
h
ile a
d
a
pta
t
io
n an
d
respo
n
s
i
ve
ne
ss w
i
l
l
get
b
e
t
t
e
r.
Th
e
se
con
d
l
ev
el
i
n
th
is
n
ew
a
rc
h
i
t
e
ct
u
r
e
wi
ll
t
a
k
e
ad
van
t
a
g
e
o
f
si
ngl
e
v
e
hi
cl
e
a
c
hie
v
emen
ts
i
n
e
n
e
r
g
y
e
x
p
e
r
i
e
n
c
e
t
o
b
u
i
l
d
a
S
m
a
r
t
L
e
a
r
n
i
n
g
D
a
t
a
b
a
s
e
(
S
L
D
B
)
.
I
n
t
h
i
s
a
r
c
h
it
e
c
t
u
r
e,
a
c
l
oud
b
ased
D
at
ab
ase
w
ill
col
l
ec
t EV
Ener
gy Ex
per
i
e
n
ces
f
rom
diffe
rent
c
o
nnec
t
e
d
veh
ic
le
s
t
o
b
e
sort by e
v
e
n
t
s
.
Ene
r
g
y
e
x
p
eri
e
nces
a
re
u
p
l
oade
d
to
t
he
b
a
s
e
a
nd
sor
t
by
ve
ry
f
i
ne
e
ven
t
c
ri
ter
i
o
n
.
Thro
ug
h
a
com
p
rehe
nsive
power
m
ana
g
em
ent
da
ta
b
as
e
sol
u
t
i
o
n
,
a
vehic
l
e
de
tec
t
i
n
g
simi
lar
co
n
d
iti
o
n
s
d
u
ri
n
g
its
p
at
h
sen
d
s
an
a
ss
i
s
t
reque
st
t
o
t
h
e
D
B
.
This
r
e
q
u
e
st
w
ill
be
i
de
n
t
i
f
ie
d
a
nd
fe
e
dbac
k
w
i
l
l
be
dow
nl
oa
de
d
t
o
v
eh
ic
le
if
a
va
il
a
b
l
e
.
There
f
ore
,
i
t
ge
t
s
a
dva
n
t
age
o
f
v
a
l
ua
b
l
e
pre
v
io
us
e
ner
gy
exper
i
e
n
ce
o
f
a
no
the
r
v
e
h
i
c
le
.
By
dow
nl
o
a
din
g
a
dj
us
tme
n
t
al
g
o
r
it
hm
p
ar
am
eters
for
p
o
w
e
r
m
a
na
gem
e
n
t
,
i
t
obt
a
i
n
s
o
p
timal,
know
t
o
d
a
te
,
en
er
gy
eco
no
my
a
n
d
v
e
hi
c
l
e
re
s
p
o
n
s
i
ve
nes
s
i
m
m
e
d
i
a
t
e
l
y
.
4.2.
Des
c
ript
i
o
n
o
f
a
seq
uence
I
n
t
hi
s
ar
ch
ite
cture
,
w
e
exa
m
ine
a
l
l
i
n
for
m
atio
n
c
o
l
l
e
c
t
ed
by
v
e
h
icle.
T
h
is
c
o
n
tri
b
u
t
e
s
i
n
bu
i
l
d
i
n
g
veh
i
c
l
e
dec
i
si
o
n
.
These
i
n
for
m
atio
n
c
o
ns
t
i
t
u
te
a
n
eve
n
t;
e
.g.
A
n
e
l
e
c
t
ri
c
ve
hic
l
e
be
in
g
drive
n
b
y
a
fem
a
le
dri
v
er,
ha
v
i
ng
a
h
a
p
p
y
moo
d
,
from
a
G
P
S
p
o
sit
i
on
A
t
o
B,
i
n
s
p
e
c
i
fic
road
c
onditions,
i
n
s
u
m
m
er
,
in
s
pecific
w
e
a
t
he
r
c
o
n
d
itio
n
a
n
d
t
e
mpe
r
ature
,
t
he
r
ide
w
a
s
i
n
a
s
pec
i
fic
d
a
t
e
&
T
i
m
e
a
w
e
e
k
e
n
d
,
a
f
e
a
s
t
d
a
y
,
i
n
t
h
e
m
o
r
n
i
n
g
,
t
r
a
f
f
i
c
c
o
n
d
i
t
i
o
n
s
d
u
r
i
n
g
t
h
a
t
d
a
y
a
t
r
a
f
f
i
c
j
a
m
o
c
c
u
r
r
e
d.
T
he
v
ehic
le
s
ta
t
u
s
i
n
dica
tes
ful
l
y
c
h
a
r
ge
d
bat
t
eries,
w
it
h
hal
f
f
ue
l
t
a
n
k
l
eft
.
A
ccor
d
i
ng
t
o
t
his
eve
n
t
i
n
s
pac
e
a
nd
tim
e,
w
ith
t
his
g
e
nder
of
d
ri
ver,
i
n
t
h
is
mood,
w
ork
i
n
g
w
it
h
al
l
m
e
n
t
io
ne
d
co
nd
i
tio
ns,
w
e
g
e
n
era
t
e
a
uniq
u
e
s
et
o
f
ad
ju
st
ment
p
ara
m
e
t
e
r
s
.
I
n
v
e
hi
cl
e
rec
o
rded
d
a
t
a
th
i
s
e
ve
nt
m
ay
o
c
c
u
rs
p
rob
a
bl
y
i
n
t
he
f
u
t
ure
.
D
r
i
ve
rs
g
ene
r
al
ly
s
elec
t
unc
onsc
i
o
u
s
l
y
sam
e
iti
ner
a
ry
t
o
li
n
k
b
e
t
w
e
e
n
t
w
o
p
o
i
nts
A
and
B.
s
ome
dri
v
ers
m
a
y
kn
ow
s
h
o
rtc
u
t
s
d
uri
ng
p
e
ak
t
i
m
e
in
o
r
d
er
t
o
avo
i
d
j
a
m
a
nd
im
pro
v
e
fuel
e
c
o
n
o
my.
A
ll
t
h
e
s
e
inf
o
rm
atio
n
c
a
n
b
ri
n
g
a
ss
is
tance
to
n
ew
v
eh
ic
le
s
i
n
t
h
i
s
are
a
to
d
e
f
ine,
i
n
a
com
p
rehe
nsi
v
e
w
a
y,
t
he
d
ec
is
io
n
t
o
m
a
k
e,
i
ti
n
e
r
ar
y
t
o
s
ug
ges
t
a
n
d
p
a
r
am
eter
s
t
o
s
e
t
i
n
i
t
s
P
M
C a
l
gor
it
h
m
.
4.3.
Nove
l
ene
r
gy
expe
rien
ce
con
cept
A
d
a
p
t
i
ve
a
n
d
S
m
a
rt Lear
nin
g
a
lg
ori
t
hms e
s
t
i
m
a
t
e
t
h
e
ir o
pt
ima
l
v
al
ues
in
r
ea
l-ti
m
e
i
n orde
r t
o
sa
t
i
s
fy
t
h
e
c
h
arg
e
-su
s
ta
i
n
i
n
g
con
s
t
r
ai
n
t
a
nd
t
o
a
c
h
i
ev
e
b
e
st
p
erfo
rman
c
e
[25]
.
Intel
l
ige
n
t
de
c
i
sion
t
a
k
e
n
s
e
e
m
s
t
o
be
on
ly
r
e
l
a
t
e
d
t
o
veh
i
c
l
e.
H
o
w
e
v
er,
differe
nt
c
irc
u
mstanc
es
m
a
y
l
ea
d
t
o
c
omple
t
e
d
i
ffe
ren
t
d
ec
isi
o
ns
a
n
d
vari
ous r
esu
lts.
In
a
s
ec
on
d
st
ep
,
En
e
r
g
y
Ex
p
e
ri
en
ce
l
ay
ers
are
i
n
t
r
od
uce
d
;
In
fo
rm
ati
on
s
u
c
h
a
s
e
ngi
ne
s
ta
tus,
bat
t
eries’
r
ate,
e
t
c
.
are
pr
ov
i
d
e
d
a
s
c
o
n
v
e
n
ti
ona
l
fr
om
t
he
E
le
c
t
ric
ve
hic
l
e.
T
o
bu
i
l
d
a
rich
e
xpe
rience
,
in
form
ation
from
d
rive
r
a
n
d
env
i
ro
nme
n
t
suc
h
a
s
dri
v
er
g
en
der
,
m
ood
,
GPS
locat
i
o
n,
w
eathe
r
an
d
roa
d
con
d
i
t
i
on
s a
r
e adde
d
t
o
cr
eate
an
E
ven
t
.
Evaluation Warning : The document was created with Spire.PDF for Python.
Int J
P
o
w
El
e
c
&
D
ri S
yst
I
S
S
N
:
2088-
86
94
Sm
ar
t
da
t
a
bas
e
c
oncep
t for po
wer m
a
n
agem
e
nt in a
n
elec
trica
l
veh
ic
le
(Mahm
ou
d
i
C
hok
ri)
16
5
The
S
m
a
r
t
le
ar
ning
P
o
w
e
r
Ma
na
gem
e
n
t
C
on
t
r
o
l
A
l
gor
ithm
run
s
s
e
p
a
ra
t
e
ly
t
o
pred
ic
t
new
adj
u
s
t
m
e
n
t
p
a
r
am
eter
s
which
a
r
e
sui
t
ab
le
f
or
t
h
i
s
eve
n
t.
T
he
c
o
mb
i
n
at
i
on
betw
ee
n
the
eve
n
t
a
nd
the
para
me
ters
c
on
stit
u
t
es the
e
ne
r
gy ex
pe
rienc
e
.
To
g
e
t
D
ata
b
a
s
e
a
s
sis
t
ance
,
a
re
quest
m
en
tio
ni
n
g
t
he
c
u
rre
nt
e
v
en
t
is
f
or
mula
te
d
an
d
sen
t
.
Th
e
v
e
hi
cl
e
sho
u
ld
h
av
e
an
i
n
e
xi
st
ent
o
r
poo
r
e
x
p
e
ri
en
c
e
r
el
at
ed
t
o
t
he
o
cc
urred
si
t
u
a
t
i
on.
B
as
ica
l
l
y
,
tw
o
sc
e
n
ario
s
are
fa
ce
d
.
I
n
f
i
rst
case
,
a
cc
o
r
di
ng
t
o
t
h
e
se
nt
r
equ
e
st,
n
o
s
im
ila
r
sit
u
a
t
i
o
ns a
re
f
ound.
The
Dat
a
-Base
ca
nno
t
affor
d
a
dj
ustm
en
t
p
a
r
a
m
e
te
rs
f
or
t
h
e
E
lec
t
ric
Veh
i
cle.
T
hus,
no
assista
n
ce
is
o
ffere
d
.
As
p
r
e
sen
t
ed
in Fig
ure
2.
F
i
gure
2.
P
M
w
itho
u
t
sm
a
r
t
data-
base
a
ssi
s
t
anc
e
The
ve
h
i
cl
e
ge
ne
rates
its
o
w
n
a
dap
t
i
v
e
para
me
t
e
rs
f
or
t
he
f
a
c
e
d
s
i
t
ua
t
i
on
a
i
m
in
g
to
o
p
t
im
ize
p
o
w
e
r
ma
nage
me
nt.
These
P
a
ra
me
t
e
rs,
c
oup
l
e
d
w
i
t
h
t
he
e
ve
n
t
d
e
s
c
r
ip
tio
n
ar
e
up
l
o
ade
d
a
s
a
n
e
x
p
erie
nce
for
fu
t
u
re
use.
T
h
e
D
ata
-
B
a
se
p
ro
vi
des
Electr
i
c
V
e
hi
cle
wi
th
a
dj
ust
m
ent
pa
ra
me
te
rs.
The
vehic
l
e
ut
i
l
iz
es
o
p
t
i
m
ized
para
me
ters
f
or
t
he
f
ac
e
d
s
i
t
u
at
ion.
T
he
se
P
a
r
ame
t
e
r
s
w
ill
a
ll
ow
t
o
ob
tai
n
i
m
m
e
d
i
a
te
o
p
tima
l
P
M,
b
e
t
ter
veh
i
c
l
e
r
e
spo
n
s
i
v
ene
s
s
a
n
d
t
o
i
mpr
o
ve
i
m
m
e
diat
el
y
fue
l
e
con
o
m
y
f
o
r
a
lo
nger
ride
r
a
nge.
Th
i
s
p
he
n
o
m
e
no
n
is
e
x
pose
d
i
n
Fig
u
re
3.
Figure
3.
P
M
with
s
ma
rt dat
a
b
ase
assi
s
t
a
n
c
e
A
s
w
ell
a
s
E
l
e
c
t
ric
V
e
h
i
c
l
e
s
i
n
t
e
nd
to
i
m
p
r
ove
k
now
le
dge
t
hrou
g
h
l
o
ca
l
a
nd
cl
o
u
d
e
x
p
erie
nc
e
sh
a
r
in
g
,
t
h
e
D
at
a-B
a
se
o
p
e
ra
t
e
s
id
ent
i
call
y
.
Ea
ch
t
i
m
e
th
e
u
p
l
o
ad
ed
p
a
r
a
m
e
t
e
r
s
d
e
fi
n
e
a
b
e
t
t
e
r
ef
fi
ci
en
cy
f
o
r
an
e
xi
sti
ng e
x
p
e
rienc
e
c
ompa
r
e
d
to t
he sa
v
e
d
da
t
a
,
t
he
ne
w
par
a
m
e
ter
s
r
epla
ce
t
h
e
ol
d o
n
e
s.
V
a
rious
i
n
f
or
m
a
tio
n
can
b
e
del
i
ver
e
d
by
v
eh
ic
l
e
s
e
n
sor
s
t
o
cre
a
t
e
t
h
e
e
v
e
n
t
.
I
n
o
r
d
e
r
t
o
s
i
m
p
l
i
f
y
syste
m
s
an
d
r
e
duce
com
p
l
e
xit
y
,
w
e
u
se
s
truc
ture
s
hari
ng
s
t
r
a
t
eg
y.
W
h
i
c
h
i
nv
ol
v
e
s
t
h
at
s
am
e
ph
ys
ic
a
l
st
r
u
c
t
ure
or
s
e
n
sor
ca
n
be
s
h
a
red
b
y
d
iffer
e
nt
f
u
n
c
t
i
o
n
s
a
nd
pr
o
v
i
d
e
ne
e
d
ed
i
n
f
or
ma
t
i
on
.
E.g.
s
am
e
cam
er
a
ca
n
be
u
se
d
in
D
row
s
ines
s
D
e
tec
tio
n
S
y
s
t
e
m
(
D
D
S
)
t
o
ale
r
t
dri
v
e
r
sl
eep
in
ess
an
d
as
a
f
ac
e
a
n
al
y
z
e
r
t
o
deter
m
i
n
e
dr
i
v
e
r
m
ood.
S
ame
sens
ors
use
d
i
n
La
ne
K
e
e
p
i
ng
A
s
s
i
s
t
S
y
s
t
e
m
(LK
A
S
)
[41]
t
o
o
b
lige
driver
t
o
kee
p
h
a
nds
o
n
st
e
e
ri
ng
w
h
ee
l,
c
a
n
p
ro
v
i
de
h
ea
rt
b
e
ats
r
a
te
a
nd
d
et
ermi
ne
i
f
t
h
e
d
r
iv
er
i
s
n
e
rv
ou
s.
P
re
v
i
ou
s
rese
arc
h
es a
s [
39]
, [43] ha
v
e hi
g
h
l
i
gh
te
d
t
h
e
prom
ine
n
ce
of
str
uct
u
re
s
hari
ng i
n
E
l
e
c
t
ric
V
e
hic
l
e.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN: 2088-
8694
I
nt
J
P
ow
Elec
& Dr
i
S
y
st, Vol. 10,
N
o.
1, Mar
c
h 2
0
1
9
:
16
0 –
16
9
16
6
The
r
ef
ore,
f
or
c
omple
x
s
ys
t
e
m
desi
gn,
f
ew
s
t
u
d
i
e
s
w
e
r
e
i
n
t
r
o
d
uc
e
d
.
In
2
0
12,
K
eq
ia
ng Li
p
ro
pose
d
a
com
p
rehe
nsive
d
e
sig
n
f
or
s
t
r
uct
u
re
s
har
i
n
g
;
In
o
rde
r
t
o
mi
nimiz
e
se
ns
ors
on
boar
d
a
n
d
s
im
pli
f
y
a
r
ch
i
t
e
c
ture
,
sam
e
p
h
y
sic
a
l
e
q
uipm
en
t
ca
n
prov
i
d
e
in
fo
rm
atio
n
f
o
r
di
ffere
n
t
f
u
n
c
t
i
o
ns
o
r
fea
t
ur
es
i
n
a
veh
i
c
l
e.
T
h
i
s
i
m
p
r
ov
es
e
ffi
ci
e
n
c
y
f
o
r
c
o
m
p
l
ex
a
rc
hi
t
e
cture
at
t
h
e
l
o
w
est
c
o
st
.
I
n
our
case
,
w
e
a
r
e
m
o
re
i
nter
est
e
d
i
n
mu
l
t
i
s
en
so
ry
p
ri
n
c
i
p
l
e
,
wh
i
c
h
ai
ms
t
o
red
u
ce
t
h
e
nu
mb
er
o
f
on
bo
ard
se
n
s
ors.
A
s
show
n
in
F
ig
ure
4,
s
ensors
are
share
d
a
m
o
n
g
o
rd
inary
func
tio
ns
s
uc
h
as
c
omfor
t
o
r
sec
u
rity
a
n
d
P
o
w
e
r
M
a
n
a
g
e
m
e
n
t
L
e
a
r
n
i
n
g
p
r
o
c
e
s
s
.
E.g.
,
GP
S
ante
nna
p
r
ovi
de
s
l
o
ca
ti
on
inf
o
rmatio
n
t
o
onb
oa
rd
n
av
i
ga
t
i
on
sys
t
em
f
or
t
ur
n-b
y
-tur
n
assistanc
e
and,
s
im
u
l
ta
ne
ous
ly,
calc
u
lat
e
s
c
o
ord
i
n
a
tes
w
h
i
c
h
are
use
f
u
l
t
o
b
uil
d
t
h
e
e
n
e
rgy
ex
pe
ri
e
n
c
e
o
r
t
o
r
et
ri
eve
assista
n
ce
fro
m
the
S
m
a
r
t
Da
t
a
ba
se.
Infor
m
a
t
io
n
com
e
f
rom
vehic
l
e
a
s
w
e
l
l
a
s
d
r
i
v
e
r
a
n
d
e
n
v
i
r
o
n
m
e
n
t
.
T
h
e
mult
ise
n
s
o
ry
s
truc
ture
s
hari
ng
s
ho
u
l
d
be
c
o
n
tr
ol
le
d.
D
e
p
en
di
ng
on
v
a
l
ua
t
i
o
n
,
t
o
r
e
duce
c
o
st,
fur
t
h
e
r
redu
nda
nc
y or
re
l
ia
b
i
l
ity,
the
st
r
u
c
t
ure
w
ill
b
e
de
f
i
n
e
d
a
nd d
e
si
gne
d
[4
4].
Fi
g
u
r
e 4
.
M
ul
t
i
-sen
so
r st
ru
ct
ure
sh
a
r
i
n
g
4.4.
N
ove
l
A
l
g
o
r
i
t
h
m S
t
r
u
ctu
r
e
an
d
Layou
t
A
clo
u
d
d
a
t
a
b
ase
ty
pic
a
l
l
y
r
uns
o
n
a
c
l
oud
-
c
om
puti
n
g
pl
atfor
m
,
such
a
s
S
a
lesf
orce,
G
o
G
rid,
a
nd
M
i
c
r
os
oft
A
z
u
r
e
.
A
s
de
ploy
m
e
nt,
the
ch
o
i
ce
o
f
C
lo
u
d
d
ata
b
a
s
e
s
i
s
t
h
e
i
nde
pe
nde
n
c
y
d
u
e
to
d
iffe
re
nt
man
u
f
a
c
tu
re
rs’
c
o
mp
eti
t
io
n
an
d
p
r
o
f
e
s
si
o
n
a
l
s
e
c
r
e
t
a
n
onymi
t
y
.
B
y
us
ing
se
c
u
re
d
an
d
i
n
d
e
pe
n
d
e
n
t
vir
t
u
a
l
ma
chine
s
,
de
ci
sion
a
b
o
u
t
d
a
t
a
m
odel
w
i
l
l
b
e
take
n
i
n
f
ur
th
er
w
o
rk
w
heth
er
it
sh
ou
l
d
b
e
S
Q
L-
based
N
o
-S
Q
L
d
a
t
a
b
a
s
e
m
od
el
.
As
a
g
e
n
eral
l
a
you
t
,
E
nerg
y
exp
e
ri
en
c
e
s
a
r
e
u
p
l
oa
de
d
sys
t
e
m
a
tic
al
ly
t
o
t
h
e
ba
se.
Whe
n
in
form
ation
is
r
eque
ste
d
i
n
n
o
rm
al
o
r
pred
i
c
ti
ve
a
ppr
oa
ch
,
da
t
a
ba
se
s
ee
k
s
t
he
c
o
nve
n
i
e
n
t
e
x
perie
n
c
e
b
ase
o
n
si
m
i
lari
ty
i
n
e
v
en
ts.
Re
sea
r
ch
a
n
d
e
xper
i
e
n
c
e
s
ele
c
tio
n
w
i
ll
be
o
p
t
i
m
ize
d
w
it
h
A
r
ti
fic
i
a
l
N
e
u
rona
l
N
e
t
w
orks
(A
NN
s). T
h
e
se
S
oftw
ar
e
func
ti
o
n
s
aim
to ta
k
e
adva
n
t
a
g
es of C
l
ou
d
Com
p
u
tin
g r
e
vo
l
u
ti
on.
I
n
h
i
g
h
s
u
pe
r
v
i
s
ory
P
o
w
e
r
Ma
nagem
e
n
t
L
aye
r
(
P
M
L),
ma
ny
a
l
g
o
r
i
t
h
m
s
h
a
v
e
bee
n
d
e
v
e
l
ope
d.
D
e
pe
n
d
in
g
on
p
ow
er
t
r
ain
a
r
c
h
itec
t
ure,
m
a
i
n
l
y
fi
ve
t
ec
hn
ique
s
d
e
li
v
e
r
e
d
i
n
t
e
n
d
e
d
r
esul
ts
a
n
d
p
ro
ved
relia
bi
l
ity.
In
t
hi
s
sect
i
o
n,
w
e
w
i
l
l
go
t
h
ro
ug
h
a
brief
m
e
n
t
i
o
n
of
r
e
c
e
nt
e
ff
o
r
t
s
i
nt
rod
u
ced
i
n
bo
th
o
ff
line
a
nd
on
line
a
l
gor
it
hm
s.
A
nd
w
i
ll
intr
od
uce
a
n
e
n
h
anc
e
d
a
lg
ori
t
h
m
str
u
c
ture
t
o
g
o
w
it
h
t
h
e
n
o
ve
l
arc
h
i
t
e
c
t
ure,
a
s
prese
n
t
e
d i
n
F
i
gur
e 5.
Wh
en
w
e
ai
med
t
o
r
e
bui
l
d
t
h
e
p
o
w
er
m
a
n
a
g
emen
t
e
x
p
e
ri
e
n
ce
,
we
s
e
t
an
e
n
h
a
n
ce
d
o
n
li
ne
P
M
C
alg
o
ri
t
h
m
tha
t
i
n
h
eri
t
s
ma
rt
l
ear
ning
feat
ure
from
e
xis
t
i
n
g
Le
a
rn
in
g
PM
C
a
l
go
rith
m
an
d
l
o
cali
z
a
tio
n
f
eat
u
r
e
from
“
G
P
S
Enhance
d
A
lgor
i
t
hm
”.
T
he
N
ov
el
A
lg
ori
t
h
m
ta
kes
ad
va
nt
a
g
e
o
f
t
h
e
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
,
b
u
t
s
e
t
s
the
o
p
tima
l
p
a
r
a
m
e
t
ers
in
a
v
e
c
t
o
r
c
o
lle
c
ting
al
l
t
h
e
c
o
n
d
it
io
n
s
in
w
h
i
ch
t
h
e
opt
i
m
al
p
a
r
amet
ers
are
f
ound
.
Th
is
p
ara
m
e
t
e
r’s
v
e
c
tor
is
c
alle
d
O
p
t
i
ma
l
Ex
perienc
e
V
e
c
tor.
I
n
stea
d
o
f
e
nha
nc
i
ng
j
u
st
o
ne
v
e
h
ic
le,
the
vec
t
or
i
n
t
e
g
ra
t
e
s
t
h
e
EV
i
n
f
orm
a
ti
on
suc
h
a
s
bran
d,
m
ode
l
,
p
ow
e
rtrain,
manuf
ac
turer
to
b
e
a
referenced
Ex
perie
n
c
e
V
ec
t
o
r
, the
n
sen
t
to o
u
r
S
m
a
r
t
clou
d da
t
a
ba
se
f
or
s
a
vi
ng.
A
s
w
e
use
diff
ere
n
t
EV
s,
d
iffe
rent
E
ne
r
gy
e
xpe
r
i
e
n
ce
s
are
com
m
u
n
i
cat
ed
t
o
th
e
Da
t
a
b
a
se
b
uil
d
i
n
g
and
im
por
tan
t
d
ec
isi
o
n
ce
n
t
er
r
eady
for
use
.
T
he
r
un
ni
n
g
a
lg
ori
th
m
o
f
t
he
p
ro
po
se
d
met
h
od
i
s
p
r
e
s
e
n
te
d
in
F
i
gure
7.
T
he
i
ni
t
i
al
iz
i
ng str
u
cture
of
t
he
ve
h
ic
le
is als
o
e
x
p
ose
d
i
n F
i
gur
e
6
.
Evaluation Warning : The document was created with Spire.PDF for Python.
Int J
P
o
w
El
e
c
&
D
ri S
yst
I
S
S
N
:
2088-
86
94
Sm
ar
t
da
t
a
bas
e
c
oncep
t for po
wer m
a
n
agem
e
nt in a
n
elec
trica
l
veh
ic
le
(Mahm
ou
d
i
C
hok
ri)
16
7
Fi
g
u
re
5
. Smar
t
l
e
a
r
ni
ng
arc
h
i
t
e
c
t
u
r
e co
n
cept
Fi
g
u
r
e
6
.
Veh
i
c
l
e
in
i
tia
liz
in
g struc
t
ur
e
F
i
gure
7.
F
low
c
ha
rt or
a
sta
r
t seq
u
e
n
ce
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN: 2088-
8694
I
nt
J
P
ow
Elec
& Dr
i
S
y
st, Vol. 10,
N
o.
1, Mar
c
h 2
0
1
9
:
16
0 –
16
9
16
8
5.
CONCL
U
S
ION
C
o
n
n
ec
ti
n
g
v
e
h
i
c
les
r
e
pre
s
e
n
t
s
a
n
imp
o
rta
n
t
t
r
en
d
f
o
r
t
h
e
nex
t
ge
ner
a
t
i
on.
T
he
k
e
y
t
e
c
h
n
o
lo
g
y
i
n
t
h
e
Smart
Le
a
r
n
i
n
g
Arc
h
it
ec
t
u
re
i
s
e
x
pe
ri
e
n
ce
s
h
a
ri
ng
b
et
we
en
v
e
h
i
c
l
e
s
t
o
i
mp
ro
ve
S
e
l
f
l
e
a
r
ni
ng
a
nd
enha
nc
e
p
o
w
e
r
m
a
nage
me
n
t
.
P
r
o
v
i
d
i
n
g
gree
n
s
o
lu
t
i
o
n
s
w
i
t
h
e
nv
iro
nm
en
t
a
l
fr
i
e
n
d
l
y
g
o
a
l
s
s
h
ou
l
d
not
i
n
a
n
y
w
a
y
com
p
rom
i
se
i
n
v
e
h
i
c
l
e
respo
n
s
i
v
e
ness
o
r
dri
v
in
g
s
t
y
l
e.
T
he
pre
s
e
n
t
p
a
pe
r
de
scri
b
e
s
a
new
me
t
hod
o
f
power
m
ana
g
ing
in
E
lectr
i
c
V
e
hic
l
es.
The
work
c
a
rried
o
ut
d
eter
m
i
ne
t
he
p
ossib
i
lit
y
of
p
r
o
v
i
di
n
g
a
n
ew
arc
h
i
t
e
c
t
u
re
t
h
a
t
ca
n
af
ford
p
ower
m
anage
m
e
n
t,
n
e
t
work
i
n
g
a
n
d
c
o
n
n
e
c
t
i
v
i
t
y
b
e
t
w
e
e
n
E
V
s
.
A
s
a
n
e
x
t
s
t
e
p
i
n
th
i
s
w
ork,
s
i
m
ula
t
ions
r
esu
lts
w
ill
b
e
pub
l
i
s
h
ed i
n or
der
t
o
r
e
infor
ce
our
A
rchitec
t
ure
.
REFE
RENCES
[1]
E.
H
elm
e
rs
a
n
d
P
.
Marx
,
"El
e
ct
ric
cars
:
t
ech
ni
cal
c
haract
eris
ti
cs
a
nd
e
nv
iron
m
e
nt
al
i
m
p
acts
,
"
Helm
ers
and
Marx
En
viro
n
m
ent
a
l S
c
i
e
nces
Eur
o
p
e
,
v
ol.
2
4
,
p
p.
1
-15
20
12
.
[2]
M
.
A
.
Hann
an,
F
.
A
.
Azi
d
i
n
,
and
A
.
M
o
h
am
ed,
"Hy
b
rid
el
e
c
tric
v
e
hic
l
e
s
a
nd
t
he
ir
c
h
a
lle
n
ge
s:
A
r
e
v
ie
w,"
Ren
e
wabl
e an
d Sus
tain
abl
e
E
n
e
r
g
y
Revi
ews
,
vo
l.
2
9,
p
p
.
1
3
5
–
150,
J
an.
20
14.
[3]
J.
Z
.
P
e
i,
Y
.
X.
S
u,
a
nd
D
.
H.
Z
han
g
,
"F
uzzy
e
nerg
y
man
a
g
e
m
e
nt
strat
e
gy
f
o
r
paral
l
el
H
EV
b
as
ed
on
pi
geo
n
-
in
spired
optim
i
z
ati
on
a
l
g
o
rit
h
m
,
"
Sci.
C
h
ina
Technol.
Sci,
vol.
60,
pp
42
5-4
3
3
,
2
0
1
7
.
[4]
C.
M
ahm
o
udi,
A.
F
lah,
and
P
.
L
a
ss
aa
d
S
B
ITA,
"No
vel
con
cept
of
P
ower M
an
a
g
e
m
ent
Arch
itect
ure b
a
sed
on
S
m
a
rt
EV
Learn
i
ng
D
a
t
aBase," Re
cent
Ad
v.
E
l
ectros
c
ie
n
ce Com
put
. J.
,
pp
.
1
91–
19
6,
2016.
[5]
Z.
W
e
i
,
J.
X
u
a
n
d
D.
H
a
lim
,
"
H
E
V
E
n
e
rgy
M
a
na
g
e
me
nt
F
uz
z
y
L
og
ic
Co
ntrol
Ba
s
e
d
on
D
y
n
a
m
i
c
P
r
og
rammi
ng
,
"
20
15
IEEE Vehicl
e Po
wer a
nd
Pr
op
ul
sio
n
Co
nf
e
r
en
ce
(
V
PP
C)
,
Mon
t
real
,
QC
, p
p. 1-5
,
20
15
.
[6]
C.
K
.
Saman
t
a,
S
.
K
.
P
ad
hy
,
S.
P
.
Pan
i
g
r
ahi
an
d
B.
K
.
Pan
i
g
r
ah
i
,
"
H
ybrid
swarm
inte
lligence
met
h
ods
f
o
r
energy
ma
n
a
ge
me
nt
i
n
hy
brid
e
le
c
t
r
i
c
v
e
hic
l
e
s
,"
i
n
I
E
T Elect
ri
ca
l System
s in Tr
a
n
sport
a
tion
,
vol.
3
,
p
p
.
2
2-29
,
Ma
rc
h
20
13
.
[7]
M.
I
r
f
a
n
,
Ma
c
h
mu
d
Effe
n
d
y
,
N
u
r
Alif
,
La
i
l
i
s
S
,
Ilha
m
Pa
ka
y
a
,
Am
ru
l
Faru
q
,
"P
erf
o
rman
ce
Co
m
p
aris
on
o
f
F
u
zzy
Lo
gi
c
and
P
r
op
o
r
t
i
o
n
a
l
-i
n
t
egral
for
an
Electro
ni
c
Lo
ad Co
n
t
r
o
l
l
er,"
In
tern
atio
na
l Jo
ur
na
l of
Power Electr
onics
an
d
Dr
ive S
y
s
t
e
m
(
I
JPEDS)
, v
ol
. 8
,
pp. 1
88
6-1
8
9
3
, Decem
ber 2
017
.
[8]
F
.
R
.
S
a
lm
asi,
"
Con
t
ro
l
S
t
rat
e
gies
f
o
r
H
yb
rid
Elect
ric
Vehi
cles
:
Ev
oluti
o
n
,
C
lassif
i
catio
n,
C
om
paris
o
n
,
a
n
d
F
u
t
ure
T
r
e
n
d
s
,"
i
n
IE
EE
Transac
t
ion
s
on Vehicu
l
a
r
Technol
ogy
,
vo
l.
5
6,
p
p
.
2
393
-24
0
4
,
S
ep
t
.
2
00
7.
[9]
L.
X
u
,
J
.
Hua,
X
.
L
i
,
Q.
M
en
g,
J
.
L
i
a
nd
M
.
O
u
y
a
ng
,
“Co
n
tro
l
s
t
rateg
y
optimizat
io
n
o
f
a
h
y
b
rid
f
u
el
cell
v
e
hi
c
l
e
with
brak
i
ng en
e
r
g
y
r
eg
enerat
ion
,
”
2
0
0
8
IE
EE Ve
h
i
cle Power
an
d
Pro
pulsion
Confer
ence
,
Harb
i
n
, p
p
.
1
-6,
20
0
8
.
[10]
C.
D
ext
r
eit
an
d
I.
K
ol
m
a
nov
sky,
"
App
r
oach
es
t
o
energ
y
m
ana
g
em
en
t
of
h
y
b
rid
el
ectri
c
v
e
hicl
es:
E
x
p
e
rim
e
nt
a
l
com
p
ari
s
o
n
," i
n
P
roc
.
UK
A
CC I
n
t
.
Conf
.
Control,
pp
.
1–
6,
2
0
1
0
.
[11]
P.
P
isu
a
n
d
G.
R
iz
z
o
ni,
"
A
C
o
m
p
a
ra
tive
S
tu
dy
O
f
Su
pe
rv
isory
Con
tro
l
S
t
r
ategies
f
o
r
H
ybri
d
Elect
ric
V
e
hicl
es
,
"
in
I
E
EE Tra
n
s
ac
tion
s
o
n
Co
nt
ro
l S
y
ste
m
s Te
c
h
no
lo
gy
, vo
l
.
1
5
, pp
. 5
0
6
-5
18
, May 2
00
7.
[12]
S
i
t
i
K
halidah
R
ahi
m
i,
Z
arafi
A
h
m
a
d
,
E
rwan
S
ulaim
a
n,
E
nw
elu
m
M
ba
diwe
I
,
S
y
ed
M
uh
am
m
a
d
Nau
f
a
l
S
yed
Ot
h
m
an,
"P
erf
o
rm
an
ce
A
naly
sis
o
f
1
2Sl
o
t
wit
h
V
ari
o
u
s
R
o
t
or
P
ol
e
N
u
m
b
e
r
s
H
E
-
F
S
M
f
o
r
H
E
V
A
p
p
l
i
c
a
t
i
o
n
,
"
Int
e
rna
t
i
o
n
a
l
Jo
u
r
n
a
l of Po
wer
E
l
ectr
onics an
d
D
r
i
ve System (
I
JPED
S
)
,
vol. 8, pp.
188
6-1
8
9
3
, Decem
ber 2
017
.
[13]
E.
B
io
ndi,
C.
B
o
l
dri
n
i
an
d
R.
B
run
o
,
"Opt
im
al
c
harg
in
g
o
f
e
l
e
ct
ric
veh
i
cle
f
l
e
e
t
s
f
o
r
a
car
s
h
a
ring
s
ystem
with
po
wer
s
h
arin
g,
"
20
16
IEEE In
te
r
n
a
t
io
na
l Ene
r
gy
Co
nfe
r
e
n
c
e
(
E
N
E
R
G
Y
C
ON)
,
Le
uv
en, p
p.
1-6
,
2
0
16
.
[14]
P
.
P
isu,
K
.
Ko
pru
b
as
i
and
G
.
R
i
zzon
i
,
"Energ
y
M
a
nag
e
m
e
n
t
a
nd
D
r
ivab
ilit
y
Contro
l
P
r
o
b
l
e
m
s
f
or
H
y
b
ri
d
Elect
ric
Veh
i
cles
,"
Pr
oceed
ing
s
of
th
e
44
th
IEEE
Co
nf
eren
ce on
Deci
si
on
an
d Co
nt
rol
,
Sev
i
l
l
e,
S
pai
n
,
pp
.
1
824
-18
3
0
, 2
005
.
[15]
C
.
M
a
h
m
o
u
d
i
,
A
.
F
l
a
h
a
n
d
L
.
S
b
i
t
a
,
"
P
r
o
t
o
t
y
p
e
d
e
s
i
g
n
o
f
a
c
o
m
p
a
ct
p
l
u
g
-
in
s
ol
ar
e
lectri
c
vehicle
ap
plicati
o
n
f
o
r
sm
art
power
m
an
agem
ent
arch
itect
ure,"
2
017
In
ter
natio
nal
Conf
eren
ce on
G
r
een
En
ergy
Con
v
er
si
on
Sys
t
ems
(G
EC
S
)
,
H
a
mmamet
,
p
p
.
1
-4
,
2017
.
[16]
N
.
H
.
F
.
I
s
m
a
i
l
a
n
d
S
.
F
.
T
o
h
a
,
"
S
t
a
t
e
o
f
c
h
a
r
g
e
e
s
t
i
m
a
t
i
o
n
o
f
a
Lithiu
m-i
o
n
battery
f
o
r
e
lectri
c
vehi
cle
b
a
sed
o
n
part
icle
s
war
m
o
ptimizati
on,"
20
13
IEEE Interna
t
i
onal
Conferen
ce o
n
S
m
a
r
t
Ins
t
r
u
men
t
ati
on,
M
e
a
s
u
r
em
ent
and
Applications (
I
C
S
IMA)
,
Ku
al
a
L
u
mp
u
r
, p
p.
1-4
,
2
0
1
3
.
[17]
D.
N
ott
e
r
et a
l
.,
“
Con
t
ri
buti
on
of
L
i
-Ion
Batteries
t
o
th
e
E
nvironm
ent
a
l
Imp
act
o
f
E
l
ectric
Vehicles,”
Envi
ron. Sci
.
Te
chn
o
l.
,
vo
l.
44,
p
p.
6
55
0–
65
56,
2
0
1
0
.
[18]
Bi.
Em
i
l
,
M
.
M
.
Ju
an,
an
d
M
.
H
enri
k,
“
Optim
a
l
c
hargi
n
g
o
f
a
n
e
l
ectri
c
v
e
hi
cle
us
ing
a
M
a
rk
ov
d
e
ci
si
on
p
roces
s,”
Applied Energy
v
o
l
.
12
3,
p
p.
1-1
2, 20
1
4
.
[19]
C.
T
sa
i
and
C.
T
i
ng,
"
E
v
alu
a
ti
o
n
o
f
a
m
u
lti
-
pow
er
s
ys
tem
f
o
r
an
e
lect
ri
c
vehi
cle,"
ICCAS 2010
,
Gye
o
ng
gi-d
o,
pp
.
1
308
-13
1
1
, 2
010
.
[20]
M
.
S
a
l
azar
a
nd
N.
E
rtu
g
rul,
"
Po
ten
t
i
a
l
enh
a
nce
m
ent
s
f
or
v
eh
icle
e
l
ectrical
p
ower
m
anagement
syste
m
s
i
n
m
i
litary
veh
i
cles
,"
20
13
Au
st
ralasia
n
Un
i
vers
i
ties Po
wer
En
gi
neer
ing Confer
ence
(
A
UPEC)
,
H
obart
,
T
A
S,
p
p.
1
-6
,
20
13
.
[21]
M
.
C
h
o
i,
J
.
Lee
an
d
S
.
S
eo
,
"
R
ea
l-Tim
e
O
pt
imi
zati
on
f
o
r
P
o
w
e
r
M
anag
em
e
n
t
S
y
s
t
ems
of
a
B
at
tery/
S
up
ercapacit
o
r
Hy
brid
E
n
e
rgy
S
t
orage
S
y
stem
i
n
Elect
ric
Vehicl
es,"
i
n
IEEE Tra
n
s
a
cti
o
n
s
on
Vehi
cul
a
r
T
ech
no
log
y
,
vo
l.
6
3,
n
o
.
8
,
p
p.
3
60
0-3
6
1
1
, Oc
t
.
20
1
4
.
[22]
L
.
R
o
s
a
r
i
o
,
P
.
C
.
K
.
L
u
k
,
J
.
T
.
E
c
o
n
o
m
o
u
a
n
d
B
.
A
.
W
h
i
t
e
,
"
A
M
o
d
u
lar
P
o
w
e
r
an
d
Energ
y
M
anagem
en
t
St
ruc
t
u
r
e
f
o
r
Dual
-En
e
rgy
S
o
u
r
ce
E
l
ectri
c
V
e
hicl
es,"
2
0
0
6
IE
EE
V
e
hic
l
e P
o
wer
an
d
Pr
op
ulsio
n
Conferen
ce
,
W
i
nd
sor,
p
p
. 1-6
,
20
06
.
Evaluation Warning : The document was created with Spire.PDF for Python.
Int J
P
o
w
El
e
c
&
D
ri S
yst
I
S
S
N
:
2088-
86
94
Sm
ar
t
da
t
a
bas
e
c
oncep
t for po
wer m
a
n
agem
e
nt in a
n
elec
trica
l
veh
ic
le
(Mahm
ou
d
i
C
hok
ri)
16
9
[23]
F
.
R
.
S
a
lm
asi,
"
Con
t
ro
l
S
t
rat
e
gies
f
o
r
H
yb
rid
Elect
ric
Vehi
cles
:
Ev
oluti
o
n
,
C
lassif
i
catio
n,
C
omp
a
ris
o
n
,
a
n
d
F
u
t
ure
T
r
e
n
d
s
,"
i
n
IE
EE
Transac
t
ion
s
on Vehicu
l
a
r
Technol
ogy
,
vo
l.
5
6,
n
o
.
5
,
pp
.
2
393-2
404
,
S
ept.
2
0
0
7.
[24]
A.
A
.
M
a
lik
opou
lo
s,
"
S
uperv
is
ory
P
o
wer
M
a
nagem
e
n
t
C
on
tro
l
A
l
gor
ithm
s
f
or
H
ybri
d
E
l
ectri
c
V
e
hi
cles:
A
Su
rvey
," in
IEEE T
r
an
sa
c
t
i
ons
o
n
In
tellig
ent
T
r
a
n
s
p
o
r
tati
on
System
s
,
v
o
l
.
1
5,
no.
5
,
p
p
.
1
869
-1885
,
O
c
t
.
2
0
1
4
.
[25]
N
.
C
h
e
n
,
T
.
Q
.
S
.
Q
u
e
k
a
n
d
C
.
W
.
T
a
n
,
"
O
p
t
i
m
a
l
c
h
a
r
g
i
n
g
o
f
e
l
e
c
t
r
ic
v
eh
icles
in
s
mart
g
rid:
C
h
a
ract
erizati
o
n
an
d
val
l
ey-
f
illing
al
gorit
h
ms,"
20
12 IE
EE
T
h
ird In
ter
n
a
t
io
na
l
Con
f
eren
c
e
o
n
Sm
ar
t Grid Com
m
u
n
i
c
ation
s
(
S
m
a
rtGridComm)
,
T
a
inan
,
pp
.
1
3
-1
8
,
2
01
2.
[26]
K.
C
le
me
n
t
,
E.
H
a
e
s
e
n
a
nd
J
.
Dr
ie
se
n,
"
Co
ord
i
na
te
d
c
h
a
r
gin
g
o
f
mult
iple
p
lu
g-i
n
h
ybri
d
e
lectri
c
veh
i
cl
es
i
n
resi
dent
ia
l
d
i
strib
u
tio
n
grids,
"
200
9 IEE
E/P
ES
P
o
wer Sys
t
ems
Co
nf
erence an
d Expo
si
ti
o
n
,
S
eat
tl
e
,
W
A,
p
p
. 1-7
,
20
09
.
[27]
I.
K
oohi
a
nd
V
.
Z
.
G
roza,
"
Optim
i
z
ing
Par
tic
l
e
S
warm
O
pt
imizat
i
o
n
a
lg
or
it
hm,"
20
14
IEEE 27
th Can
adia
n
Con
f
eren
ce on
El
ectrica
l a
nd Co
m
put
er Engin
eerin
g (CCE
CE)
,
T
oro
n
t
o
,
ON,
pp.
1
-5,
2
01
4.
[28]
Y.
H
e,
B
.
Ven
k
at
esh
an
d
L
.
G
u
a
n,
"
Op
tim
al
S
ched
ul
ing
f
o
r
Ch
arg
i
n
g
a
n
d
D
i
s
c
h
a
r
g
i
n
g
o
f
E
l
e
c
t
r
i
c
V
e
h
i
c
l
e
s
,
"
in
IEEE
T
r
ansa
c
t
i
o
n
s
on Sm
art Gri
d
,
vol.
3
,
p
p.
1
0
95-1
1
0
5
,
Sept.
20
12
.
[29]
X.
L
i
a
n
d
S
.
S
.
W
i
l
l
i
a
m
s
on
,
"
A
ss
e
s
sm
en
t
o
f
E
ff
ici
e
ncy
Im
p
r
ovem
e
nt
T
e
c
h
ni
qu
e
s
f
or
F
utu
r
e
Po
we
r
Ele
c
t
ron
i
c
s
Int
e
nsi
v
e
Hy
brid
E
lectri
c
Veh
i
cl
e
Dri
v
e
Train
s
,"
200
7 IE
EE
Can
a
d
a
E
l
ect
r
i
cal
Power Co
nf
e
r
ence
,
M
o
n
t
real,
Qu
e.,
pp
.
2
68-2
7
3
,
2
0
0
7.
[30]
A.
A
.
M
a
li
kopou
lo
s,
"
Real-T
im
e,
S
elf
-
L
earn
i
n
g
I
d
e
n
t
i
f
ication
an
d
Sto
c
ha
sti
c
O
ptima
l
C
o
n
t
r
ol
o
f
Ad
v
a
nc
e
d
Pow
e
rtrain Syste
m
s
,"
i
n
Ann
Arbor,
MI,
US
A,
20
1
1
.
[31]
M
.
S
al
eem,
G
.
A
.
Di
C
aro,
a
nd
M
.
F
a
rooq
,
"
S
w
a
rm
i
ntellig
e
nce-b
a
sed
rou
ting
pro
t
o
c
ol
f
or
w
i
r
eless
s
e
ns
or
n
e
two
r
ks
: Su
rve
y
a
nd
f
uture
d
i
re
c
tion
s
,"
I
n
f
.
S
c
i.
(
N
y)
.,
vol.
18
1,
pp.
4
5
97–
46
24
,
2011
.
[32]
Li
G
u
o
y
ong
a
nd
Y
a
n
F
a
n
g
,
"N
N-bas
e
d
fu
el
i
nj
ecti
o
n
con
t
rol
s
y
ste
m
f
o
r
h
y
bri
d
f
u
e
l
en
gine,
"
20
12
IEE
E
Sym
p
o
s
iu
m
o
n
Elect
r
i
cal
&
El
ec
tr
on
ics Engin
eering
(
E
EE
SY
M
)
, Ku
a
la
L
u
m
p
u
r,
pp.
3
3
6
-3
40,
2012
.
[33]
M
.
T
s
a
i,
C
.
T
s
e
n
g
and
Y.
L
in,
"
P
o
w
er
m
anagem
en
t
and
con
t
rol
of
an
e
l
ectric
v
e
hi
c
l
e
with
a
u
x
iliary
f
uel
cell
an
d
wi
nd
e
nergi
e
s,
"
2
0
1
3
IEEE Int
e
rn
ati
o
n
a
l Con
f
erence o
f
IEEE Regi
on
10
(
T
ENCON 20
13)
,
Xi
'an,
p
p.
1
-4,
20
13.
[34]
YLa
v
a
n
ya
D
h
a
n
e
sn
,
P.
M
u
r
ug
e
s
a
n
,
"
A
N
ov
e
l
A
pp
roa
c
h
i
n
Sc
he
du
ling
O
f
th
e
Real
-
T
i
m
e
T
asks
I
n
Het
e
rog
e
n
e
o
u
s
Multicore
Processor
w
i
th
F
uzzy
L
ogi
c
T
echni
q
u
e
For
Mi
cro-gr
i
d
P
ower
M
a
n
agem
en
t,"
In
ternat
io
nal Jo
ur
na
l
o
f
Po
wer
E
l
ectr
o
n
i
cs
an
d Dri
ve System (
I
JPED
S
)
,
v
o
l.
9
,
p
p
.
80-8
8
,
M
a
rch
2
018
.
[35]
J.
L
a
i
a
n
d
D
.
J.
N
elso
n,
"
E
n
ergy
M
anagem
en
t
P
o
w
e
r
Conv
erters
i
n
H
yb
rid
Elect
ric
and
F
u
el
C
el
l
V
e
hic
l
es,
"
in
Pr
oceed
ing
s
of th
e
IEEE
,
v
o
l.
9
5,
n
o.
4
,
pp.
766-7
7
7
,
A
p
r
il
2
0
0
7
.
[36]
G.
P
agan
e
l
li,
M
.
Tat
e
no,
A
.
Brahm
a
,
G.
R
izzon
i
a
nd
Y.
G
u
e
z
e
n
n
ec
,
"
C
o
n
t
r
ol
d
e
v
elop
m
e
nt
f
or
a
h
y
b
ri
d-elect
ric
sp
ort
-
u
tili
t
y
v
e
h
i
c
l
e:
s
t
r
ategy
,
i
m
p
l
e
m
e
nt
atio
n
and
f
i
el
d
test
results,"
Pr
oceeding
s
of
th
e 2
0
0
1
Amer
ica
n
Co
ntro
l
Con
f
eren
ce.
(
C
a
t
.
No
.
0
1
C
H3
7148)
,
A
rling
t
o
n
,
V
A
,
US
A
,
p
p
.
5
064-5
0
6
9
v
ol.
6
,
2001.
[37]
C.
K
.
Chan
,
K.
W
.
E.
C
heng
a
nd
S
.
L
.
H
o
,
"
D
evel
opm
en
t
of
P
a
c
kag
in
g
an
d
Electri
cal
I
nter
f
acing
fo
r
Electri
cal
Veh
i
cles
,"
20
06
2
n
d
In
ter
n
a
tiona
l Con
f
er
ence on
P
o
wer E
l
ectr
o
n
i
cs S
y
stems an
d Ap
pl
ications
,
Ho
ng
K
ong,
pp
.
2
34-2
4
0
,
2
0
0
6.
[38]
M
.
S
a
l
m
a
n
,
N
.
J
.
S
c
h
o
u
t
e
n
,
a
n
d
N
.
A
.
K
h
e
i
r
,
"
C
o
n
t
r
o
l
s
t
r
a
t
e
g
i
e
s
f
o
r
p
a
r
a
l
l
e
l
h
y
b
r
i
d
v
e
h
i
c
l
e
s
,
"
P
r
oceed
in
gs
o
f
t
h
e
20
00
Am
erica
n
Co
nt
rol Co
n
f
erence.
ACC
(
I
EEE
Ca
t.
No
.
0
0CH
3
6
334
)
,
v
o
l.
1
,
pp.
5
24–
52
8,
200
0.
[39]
S
.
H
ajf
o
ro
o
s
h,
M
.
A.
S
.
M
a
s
oum,
and
S.
M
.
Islam
,
"
Real-tim
e
cha
rgi
n
g
coo
r
di
nati
on
o
f
p
l
ug
-in
elect
ric
v
e
h
i
cl
es
bas
e
d
on
hy
bri
d
f
u
zzy
d
is
crete
parti
c
le
swarm
optimi
zati
on,
"
El
ectr.
Power
Sys
t
. Res
.
,
v
o
l.
1
28
,
p
p
.
19
–29,
No
vem
b
er
201
5.
[40]
P
.
E
l
b
ert,
T
.
N
ü
esch,
A.
R
itt
e
r,
N
.
Mu
rgo
v
s
k
i
,
a
nd
L
.
G
u
zzel
la,
"
En
gi
ne
O
n/Off
Contro
l
fo
r
the
E
n
erg
y
M
a
nag
e
m
e
n
t
o
f
a
S
e
rial
H
y
b
rid
El
ectri
c
Bus
vi
a
Conv
ex
O
pt
imi
z
a
t
ion
,
"
I
E
E
E T
r
an
sa
c
t
i
ons o
n
V
e
h
i
cu
lar
T
echn
o
lo
gy
,
v
o
l
.
63
,
p
p
. 3
54
9–
35
59
, 20
1
4
.
[41]
Y.
C
ao
e
t
al.,
"
A
n
O
p
timized
E
V
Ch
argi
ng
M
o
d
e
l
Cons
id
ering
TOU
Pri
c
e
and
SO
C
Curve,"I
EE
E
Tr
ansac
t
ion
s
on
Sm
ar
t Gri
d
,
v
o
l
. 3
,
p
p.
3
88
–3
93,
20
1
2
.
[42]
A.
F
o
t
o
u
h
i
,
D
.
J
.
Au
g
e
r,
K
.
P
r
op
p,
S
.
Lo
ng
o,
a
nd
M
.
Wild
,
"
A
r
e
vi
ew
o
n
el
ectri
c
v
e
hi
c
l
e
batt
ery
m
o
d
e
lli
ng:
F
ro
m
Lithium-
i
o
n
tow
a
rd Lithium-
Sul
p
hur,
" R
enewab
le a
n
d
Sust
a
i
nable E
n
er
gy
R
eviews
,
vol.
56,
pp.
1
00
8–1
02
1,
2016.
[43]
A.
P
an
day
and
H.
O
.
Ban
s
al,
"A
R
ev
iew
of
O
p
t
im
al
E
n
e
rgy
Man
a
gem
ent
S
t
rateg
i
es
f
o
r
H
yb
rid
Elect
ri
c
Vehi
cle,"
In
t
.
J. Veh
.
Tech
n
o
l.,
v
o
l
. 2
01
4,
p
p. 1-9
,
2
0
1
4
.
[44]
Z.
C
h
e
n,
C
.
C
.
M
i
,
R
.
X
i
on
g,
J
.
Xu
,
a
n
d
C.
Y
o
u
,
"
E
ne
rg
y
ma
n
a
ge
m
en
t
of
a
p
o
w
er-sp
l
i
t
p
lu
g-i
n
hyb
rid
elect
ric
veh
i
cle
bas
e
d
o
n
g
eneti
c
a
lg
ori
t
h
m
a
nd
quad
r
ati
c
p
ro
gram
m
i
n
g
,
"
J
o
u
r
na
l o
f
Powe
r So
urc
e
s
,
vo
l.
248,
pp
.
4
16-4
2
6
,
2
0
1
4.
Evaluation Warning : The document was created with Spire.PDF for Python.