I
nte
rna
t
io
na
l J
o
urna
l o
f
Appl
ied P
o
wer
E
ng
i
neer
ing
(
I
J
AP
E
)
Vo
l.
1
5
,
No
.
3
,
Sep
tem
b
er
20
2
6
,
p
p
.
1
0
7
2
~
1
093
I
SS
N:
2252
-
8
7
9
2
,
DOI
:
1
0
.
1
1
5
9
1
/ijap
e
.
v
1
5
.
i
3
.
pp
1
0
7
2
-
1
0
9
3
1072
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//
ija
p
e.
ia
esco
r
e.
co
m/
Adv
a
nced AI
-
dri
v
en battery s
tate
estima
tion usin
g
deep
lea
rning
and ense
mble mo
dels
N
.
Su
m
a
na
K
ee
rt
hi
1
, B
.
J
y
o
t
hi
1
,
M
.
Sh
a
ra
ny
a
2
,
CH
.
Srini
v
a
s
3
,
K
.
Sh
ra
v
a
ni
4
,
V
.
Sa
nje
e
v
a
Ra
o
5
,
T
ee
rda
la
Ra
k
esh
4
,
Nit
t
a
la
R
a
m
cha
nd
ra
4
,
M
a
llig
un
t
a
K
ira
n K
um
a
r
1
, K
.
V
.
G
o
v
a
rdha
n Ra
o
4
1
D
e
p
a
r
t
me
n
t
o
f
El
e
c
t
r
i
c
a
l
a
n
d
El
e
c
t
r
o
n
i
c
s E
n
g
i
n
e
e
r
i
n
g
,
K
o
n
e
r
u
L
a
k
sh
m
a
i
a
h
E
d
u
c
a
t
i
o
n
F
o
u
n
d
a
t
i
o
n
,
G
u
n
t
u
r
,
I
n
d
i
a
2
D
e
p
a
r
t
me
n
t
o
f
El
e
c
t
r
i
c
a
l
a
n
d
El
e
c
t
r
o
n
i
c
s E
n
g
i
n
e
e
r
i
n
g
,
B
V
R
I
T
H
y
d
e
r
a
b
a
d
C
o
l
l
e
g
e
o
f
E
n
g
i
n
e
e
r
i
n
g
f
o
r
W
o
me
n
,
H
y
d
e
r
a
b
a
d
,
I
n
d
i
a
3
D
e
p
a
r
t
me
n
t
o
f
El
e
c
t
r
i
c
a
l
a
n
d
El
e
c
t
r
o
n
i
c
s E
n
g
i
n
e
e
r
i
n
g
,
JN
TU
H
-
U
n
i
v
e
r
si
t
y
C
o
l
l
e
g
e
o
f
E
n
g
i
n
e
e
r
i
n
g
,
M
a
h
a
b
u
b
a
b
a
d
,
I
n
d
i
a
4
D
e
p
a
r
t
me
n
t
o
f
El
e
c
t
r
i
c
a
l
a
n
d
El
e
c
t
r
o
n
i
c
s E
n
g
i
n
e
e
r
i
n
g
,
S
t
.
M
a
r
t
i
n
’
s
En
g
i
n
e
e
r
i
n
g
C
o
l
l
e
g
e
,
S
e
c
u
n
d
e
r
a
b
a
d
,
I
n
d
i
a
5
A
ssi
st
a
n
t
D
i
v
i
si
o
n
a
l
En
g
i
n
e
e
r
(
A
D
E)
,
TSG
EN
C
O
,
H
y
d
e
r
a
b
a
d
,
I
n
d
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
u
l 3
0
,
2
0
2
5
R
ev
is
ed
Ma
r
1
7
,
2
0
2
6
Acc
ep
ted
J
u
n
9
,
2
0
2
6
Ac
c
u
ra
te
e
stim
a
ti
o
n
o
f
t
h
e
sta
te
o
f
c
h
a
r
g
e
(S
OC)
a
n
d
sta
te
o
f
h
e
a
lt
h
(
S
OH
)
o
f
li
t
h
iu
m
-
i
o
n
b
a
tt
e
ries
is
e
ss
e
n
t
ial
fo
r
imp
ro
v
in
g
t
h
e
p
e
rfo
rm
a
n
c
e
,
sa
fe
ty
,
a
n
d
li
fe
sp
a
n
o
f
e
lec
tri
c
v
e
h
icle
s
(
EVs).
Trad
it
io
n
a
l
e
stim
a
ti
o
n
m
e
th
o
d
s
o
ften
fa
c
e
c
h
a
ll
e
n
g
e
s
su
c
h
a
s
h
i
g
h
c
o
m
p
u
tati
o
n
a
l
c
o
m
p
lex
it
y
,
l
imited
a
d
a
p
tab
il
it
y
to
b
a
tt
e
ry
a
g
in
g
,
a
n
d
re
d
u
c
e
d
a
c
c
u
ra
c
y
u
n
d
e
r
v
a
ry
i
n
g
o
p
e
ra
ti
n
g
c
o
n
d
it
io
n
s.
To
o
v
e
rc
o
m
e
th
e
se
li
m
it
a
ti
o
n
s,
th
is
p
a
p
e
r
p
re
se
n
ts
a
h
y
b
rid
d
a
ta
-
d
riv
e
n
fra
m
e
wo
rk
th
a
t
c
o
m
b
i
n
e
s
m
a
c
h
in
e
lea
rn
in
g
a
n
d
d
e
e
p
lea
rn
i
n
g
tec
h
n
iq
u
e
s
fo
r
S
OC
a
n
d
S
OH
p
re
d
ictio
n
.
F
o
r
S
OC
e
stim
a
ti
o
n
,
l
in
e
a
r
re
g
re
ss
io
n
,
re
c
u
rre
n
t
n
e
u
ra
l
n
e
two
r
k
s
(RNN
),
g
a
ted
re
c
u
rre
n
t
u
n
it
(
G
RU
)
,
a
n
d
s
tac
k
e
d
l
o
n
g
sh
o
r
t
-
term
m
e
m
o
ry
(
LS
TM
)
m
o
d
e
ls
a
re
e
m
p
lo
y
e
d
.
F
o
r
S
OH
p
re
d
ictio
n
,
e
n
se
m
b
le
lea
rn
in
g
m
e
th
o
d
s,
in
c
l
u
d
in
g
sta
c
k
in
g
re
g
re
ss
o
r
,
tree
-
b
a
se
d
p
ip
e
li
n
e
o
p
ti
m
iza
ti
o
n
to
o
l
(
TP
OT
)
re
g
re
ss
o
r
,
a
n
d
h
y
b
ri
d
G
RU
-
LS
TM
m
o
d
e
ls,
a
re
u
ti
li
z
e
d
.
Th
e
p
r
o
p
o
se
d
m
o
d
e
ls
a
re
e
v
a
lu
a
ted
u
sin
g
p
u
b
li
c
ly
a
v
a
il
a
b
le
li
th
i
u
m
-
io
n
b
a
tt
e
ry
d
a
tas
e
ts
u
n
d
e
r
d
iffere
n
t
c
h
a
rg
e
–
d
isc
h
a
rg
e
c
o
n
d
it
io
n
s.
Re
su
lt
s
sh
o
w
th
a
t
d
e
e
p
lea
rn
in
g
a
p
p
ro
a
c
h
e
s
a
c
h
iev
e
su
p
e
ri
o
r
p
e
rf
o
rm
a
n
c
e
,
with
G
RU
-
LS
TM
a
n
d
s
tac
k
e
d
LS
TM
m
o
d
e
ls p
r
o
v
i
d
i
n
g
h
ig
h
ly
a
c
c
u
ra
te S
OC
e
stim
a
ti
o
n
(R²
≈
0
.
9
9
3
,
RM
S
E
≈
0
.
0
1
5
),
w
h
il
e
t
h
e
TP
OT
-
b
a
se
d
e
n
se
m
b
le
m
o
d
e
l
d
e
li
v
e
rs
n
e
a
r
-
p
e
rfe
c
t
S
OH
p
re
d
ictio
n
(R
²
≈
1
.
0
)
.
A
we
b
-
b
a
se
d
imp
lem
e
n
tatio
n
f
u
rth
e
r
e
n
a
b
les
r
e
a
l
-
ti
m
e
b
a
tt
e
ry
m
o
n
it
o
ri
n
g
,
d
e
m
o
n
stra
ti
n
g
th
e
fra
m
e
wo
rk
’s
p
ra
c
ti
c
a
li
ty
f
o
r
a
d
v
a
n
c
e
d
b
a
tt
e
ry
m
a
n
a
g
e
m
e
n
t
sy
ste
m
s
(BM
S
)
i
n
EV
a
p
p
li
c
a
ti
o
n
s.
K
ey
w
o
r
d
s
:
Dee
p
lear
n
in
g
GR
U
-
L
STM
L
ith
iu
m
-
io
n
b
atter
y
State
o
f
c
ha
r
ge
State
o
f
h
ea
lth
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
K
.
V
.
Go
v
ar
d
h
an
R
ao
Dep
ar
tm
en
t o
f
E
lectr
ical
an
d
E
lectr
o
n
ics E
n
g
in
ee
r
i
n
g
,
St.
Ma
r
tin
’
s
E
n
g
in
ee
r
i
n
g
C
o
lleg
e
Secu
n
d
er
ab
a
d
,
T
elan
g
an
a
-
5
0
0
1
0
0
,
I
n
d
ia
E
m
ail:
k
v
.
g
o
v
ar
d
h
an
r
a
o
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
His
to
r
ical
s
tatis
tics
r
ev
ea
l
th
at
th
e
v
eh
icu
lar
tr
an
s
p
o
r
t'
s
co
n
tr
ib
u
tio
n
to
ca
r
b
o
n
-
r
elate
d
em
i
s
s
io
n
s
h
as
b
ee
n
clim
b
in
g
p
r
o
g
r
ess
iv
ely
to
r
o
u
g
h
ly
th
ir
ty
-
s
ev
en
p
er
ce
n
t
af
ter
tr
an
s
p
o
r
t
-
r
elate
d
lim
itatio
n
s
r
elate
d
to
th
e
C
OVI
D
p
an
d
em
ic
g
o
t
r
elax
e
d
[
1
]
.
T
h
e
tr
a
n
s
p
o
r
t
in
d
u
s
tr
y
is
m
o
r
e
r
elian
t
o
n
en
er
g
y
f
r
o
m
p
etr
o
leu
m
t
h
an
an
y
th
in
g
else,
an
d
tr
an
s
p
o
r
tati
o
n
o
n
r
o
ad
s
ac
co
u
n
ts
f
o
r
th
e
m
ajo
r
ity
o
f
th
e
em
is
s
io
n
s
o
f
g
r
ee
n
h
o
u
s
e
g
ases
[
2
]
.
T
o
ad
d
r
ess
s
u
ch
clim
atic
im
p
ac
ts
,
in
ter
n
atio
n
al
g
r
o
wth
-
ce
n
tr
ic
s
tr
ateg
ies
n
ee
d
t
o
ch
a
n
g
e
th
eir
f
o
cu
s
to
a
ca
r
b
o
n
-
f
r
ee
s
o
ciety
[
3
]
.
T
h
e
ad
v
an
ce
m
e
n
t
o
f
b
atter
y
-
wid
e
tech
is
f
u
n
d
am
en
tal
to
th
is
c
h
an
g
e,
s
er
v
in
g
as
a
v
ital
f
ac
ilit
ato
r
o
f
a
f
u
tu
r
e
th
at
is
m
o
r
e
en
v
ir
o
n
m
en
tally
f
r
ien
d
ly
[
4
]
.
W
ith
th
is
f
o
cu
s
ch
an
g
e
in
clin
ed
to
war
d
s
a
m
o
r
e
s
u
s
tain
ab
le
r
eso
u
r
ce
a
n
d
elec
tr
ic
m
o
b
ilit
y
,
b
o
t
h
th
e
d
ep
en
d
a
b
ilit
y
an
d
ef
f
icac
y
o
f
b
atter
y
p
ac
k
s
h
av
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ap
p
l Po
wer
E
n
g
I
SS
N:
2252
-
8
7
9
2
A
d
va
n
ce
d
A
I
-
d
r
iven
b
a
tter
y
s
ta
te
esti
ma
tio
n
u
s
in
g
d
ee
p
lea
r
n
in
g
a
n
d
…
(
N
.
S
u
ma
n
a
K
ee
r
th
i
)
1073
b
ec
o
m
e
c
r
itical
[
4
]
,
[
5
]
.
L
ith
i
u
m
-
io
n
co
m
p
o
s
itio
n
-
b
ased
b
a
tter
y
p
ac
k
s
h
o
ld
th
e
wid
e
r
a
n
g
e
o
f
p
o
ten
tial
o
f
s
er
v
in
g
th
e
elec
tr
ic
-
o
p
er
ated
v
eh
icles
an
d
r
en
ewa
b
le
s
o
u
r
c
e
p
o
wer
-
b
ased
f
ac
ilit
ies
ac
r
o
s
s
th
e
n
atio
n
[
6
]
-
[
8
]
.
Su
ch
k
in
d
s
o
f
b
atter
y
p
ac
k
s
r
eq
u
ir
e
p
er
i
o
d
ical,
ac
cu
r
ate
m
o
n
ito
r
in
g
o
f
th
e
cr
itical
p
ar
am
eter
s
in
o
r
d
er
to
p
r
ev
en
t
it
f
r
o
m
u
n
f
o
r
eseen
u
n
n
ec
ess
ar
y
p
er
f
o
r
m
a
n
ce
-
ce
n
tr
ic
d
eg
r
ad
atio
n
,
an
d
t
o
ass
u
r
e
o
f
f
s
af
er
o
p
er
atio
n
al
l
th
e
tim
e.
T
h
e
g
lo
b
al
tr
a
n
s
itio
n
to
war
d
s
elec
tr
ic
m
o
b
ilit
y
a
n
d
s
u
s
tain
ab
le
en
er
g
y
s
y
s
tem
s
h
as
led
to
wid
esp
r
ea
d
ad
o
p
tio
n
o
f
lith
iu
m
-
i
o
n
b
atter
y
p
ac
k
s
ac
r
o
s
s
elec
tr
ic
v
e
h
icles
(
E
Vs).
an
d
g
r
i
d
-
co
n
n
ec
ted
s
to
r
ag
e
f
ac
ilit
ies
[
9
]
,
[
1
0
]
.
T
h
ese
b
atter
y
p
ac
k
s
s
er
v
e
as
th
e
b
ac
k
b
o
n
e
o
f
m
o
d
er
n
en
er
g
y
ap
p
licatio
n
s
,
n
ec
ess
itatin
g
ac
cu
r
ate
an
d
r
ea
l
-
tim
e
m
o
n
ito
r
in
g
to
en
s
u
r
e
s
af
ety
,
ef
f
icien
cy
,
an
d
lo
n
g
e
v
ity
[
1
1
]
-
[
1
3
]
.
Am
o
n
g
th
e
co
r
e
p
ar
am
eter
s
u
s
ed
in
b
atter
y
d
iag
n
o
s
tics
ar
e
s
ta
te
o
f
ch
ar
g
e
(
SOC
)
,
wh
ich
r
ef
lects
th
e
av
ailab
le
ch
ar
g
e,
an
d
s
tate
o
f
h
ea
lth
(
SOH)
,
wh
ich
in
d
icate
s
th
e
a
g
in
g
a
n
d
u
s
ab
le
life
s
p
a
n
o
f
th
e
b
atter
y
[
1
4
]
.
Ho
we
v
er
,
esti
m
atin
g
th
ese
m
etr
ics
ac
cu
r
ately
is
an
o
n
g
o
i
n
g
ch
al
len
g
e
d
u
e
to
th
e
c
o
m
p
lex
,
d
y
n
am
ic
n
atu
r
e
o
f
b
atter
y
b
e
h
a
v
io
r
u
n
d
er
v
ar
y
in
g
o
p
er
atio
n
al
a
n
d
en
v
ir
o
n
m
en
tal
co
n
d
itio
n
s
[
1
5
]
.
I
n
r
ec
e
n
t
y
ea
r
s
,
th
e
g
r
o
win
g
em
b
r
ac
em
en
t
o
f
p
o
r
tab
le
elec
tr
o
n
ic
cir
c
u
itrie
s
,
s
u
s
tain
ab
le
E
Vs
,
an
d
r
en
ewa
b
le
r
eso
u
r
ce
-
p
o
wer
e
d
s
y
s
tem
s
h
as
s
ig
n
if
ican
tly
in
cr
e
ased
th
e
r
eq
u
ir
e
m
en
t
f
o
r
d
ep
e
n
d
ab
le
b
atter
y
-
p
ac
k
m
an
ag
em
en
t
f
ac
ilit
ies
[
1
6
]
,
[
1
7
]
.
Am
o
n
g
th
e
co
r
e
p
a
r
am
eter
s
th
at
d
ef
i
n
e
b
atter
y
p
er
f
o
r
m
an
ce
,
b
o
th
th
e
m
ea
s
u
r
es
lik
e
SOC
an
d
SOH
tak
e
u
p
a
v
ital
p
ar
t
[
1
8
]
.
Ac
cu
r
ate
esti
m
atio
n
o
f
th
ese
m
e
tr
ics
is
n
ee
d
ed
f
o
r
en
h
an
cin
g
th
e
life
s
p
an
o
f
b
att
er
y
p
ac
k
,
ass
u
r
in
g
o
f
f
th
e
im
p
r
o
v
e
d
p
r
o
tectio
n
,
an
d
o
p
tim
izin
g
en
er
g
y
u
s
ag
e
[
1
9
]
.
T
r
a
d
itio
n
al
esti
m
atio
n
tech
n
iq
u
es
o
f
ten
f
all
s
h
o
r
t
d
u
e
to
th
eir
r
elian
ce
o
n
p
h
y
s
ical
m
o
d
els,
s
en
s
itiv
ity
to
en
v
ir
o
n
m
en
tal
c
o
n
d
itio
n
s
,
an
d
in
ab
ilit
y
to
g
en
er
alize
ac
r
o
s
s
b
atter
y
ty
p
es
an
d
u
s
ag
e
p
atte
r
n
s
[
2
0
]
,
[
2
1
]
.
T
h
e
ad
v
en
t
o
f
d
ee
p
lear
n
in
g
,
with
its
p
o
ten
tial
to
co
m
p
ile
s
o
p
h
is
ticated
n
o
n
-
lin
ea
r
ass
o
ciatio
n
s
f
r
o
m
r
aw
in
f
o
,
o
f
f
er
s
a
p
r
o
m
is
in
g
alter
n
ativ
e
.
W
e
d
ev
elo
p
an
in
tellig
en
t,
r
e
al
-
tim
e,
an
d
a
m
u
ch
ad
ap
tab
le
s
o
lu
tio
n
f
o
r
b
atter
y
s
tate
co
m
p
u
tatio
n
th
at
tak
es
ad
v
an
tag
e
t
h
e
p
r
o
s
p
ec
ts
o
f
h
y
b
r
i
d
d
ee
p
lea
r
n
in
g
m
o
d
e
ls
.
B
y
co
m
b
in
in
g
alg
o
r
ith
m
s
s
u
ch
as
lin
ea
r
r
e
g
r
ess
io
n
,
r
ec
u
r
r
en
t
n
e
u
r
al
n
e
two
r
k
(
R
NN
)
,
s
tack
e
d
lo
n
g
s
h
o
r
t
-
ter
m
m
em
o
r
y
(
L
STM
)
,
an
d
t
r
ee
-
b
ased
p
ip
el
in
e
o
p
tim
izatio
n
to
o
l
(
T
POT
)
r
eg
r
ess
o
r
,
we
ai
m
to
s
o
r
t
o
u
t
th
e
c
o
n
s
tr
ain
ts
o
f
co
n
v
en
tio
n
al
ap
p
r
o
ac
h
es
an
d
im
p
r
o
v
e
p
r
e
d
ictiv
e
ac
cu
r
a
cy
.
T
h
e
u
ltima
te
g
o
al
is
to
c
o
n
tr
i
b
u
te
to
war
d
s
af
er
an
d
m
o
r
e
ef
f
icien
t
b
atter
y
u
s
ag
e
in
cr
itical
ap
p
licatio
n
s
,
t
h
er
eb
y
a
d
v
an
ci
n
g
th
e
p
r
o
g
r
ess
io
n
o
f
in
tellig
en
t,
s
u
s
tain
ab
le
en
er
g
y
a
d
m
in
is
tr
atio
n
f
ac
ilit
ies.
W
h
en
it
co
m
es
to
th
e
b
atter
y
-
ce
n
tr
ic
m
o
n
ito
r
in
g
,
two
o
f
th
e
m
etr
ics
p
lay
a
k
ey
r
o
le,
wh
ich
ar
e
as
f
o
llo
ws:
SOC
an
d
SOH
[
2
2
]
.
T
h
e
lev
el
o
f
th
e
cu
r
r
e
n
t
ch
ar
g
e
co
r
r
esp
o
n
d
in
g
t
o
th
e
f
u
ll
ca
p
ac
ity
o
f
a
b
atter
y
p
ac
k
is
d
en
o
ted
b
y
SOC
[
2
3
]
.
L
ik
ewise,
th
e
ag
in
g
an
d
r
em
ain
in
g
u
s
ab
le
life
with
r
eg
ar
d
to
th
e
b
atter
y
p
ac
k
ar
e
d
en
o
ted
b
y
SOH
[
2
4
]
.
I
f
th
e
esti
m
atio
n
o
f
SO
C
is
m
ad
e
in
ac
cu
r
ately
,
th
en
it
ca
n
ca
u
s
e
u
n
ex
p
ec
ted
s
h
u
td
o
wn
s
.
Similar
ly
,
if
th
e
SOH
h
as
b
ee
n
m
is
ju
d
g
ed
,
t
h
en
it
ca
n
ca
u
s
e
ea
r
ly
b
atte
r
y
r
ep
lace
m
e
n
ts
o
r
f
ailu
r
es.
T
h
e
ex
is
tin
g
m
eth
o
d
s
av
ailab
le
f
o
r
s
u
ch
esti
m
atio
n
s
f
r
eq
u
en
tly
d
e
p
en
d
u
p
o
n
ce
r
ta
in
p
h
y
s
ical
m
o
d
els
an
d
m
an
u
al
ca
lib
r
atio
n
,
lack
in
g
th
e
ad
a
p
tab
le
ca
p
ab
ilit
y
a
g
ain
s
t th
e
v
ar
iab
le
o
p
er
atio
n
al
c
o
n
d
itio
n
s
.
Nu
m
er
o
u
s
m
eth
o
d
o
lo
g
ies h
av
e
b
ee
n
p
r
o
p
o
s
ed
to
ad
d
r
ess
SOC
an
d
SOH
esti
m
atio
n
,
a
f
e
w
h
av
e
b
ee
n
d
is
cu
s
s
ed
b
elo
w.
Fan
k
h
au
s
er
et
a
l
.
[
3
]
clar
if
ied
th
e
co
n
ce
p
t
o
f
“n
et
ze
r
o
”
em
is
s
io
n
s
,
em
p
h
asizin
g
th
at
it is
n
o
t
ju
s
t
a
n
u
m
er
ical
tar
g
et
b
u
t
a
s
tr
ateg
ic
f
r
am
ewo
r
k
r
eq
u
ir
in
g
d
ee
p
,
r
ea
l
em
is
s
io
n
s
cu
ts
alo
n
g
s
id
e
ca
r
b
o
n
r
em
o
v
al.
T
h
e
a
u
th
o
r
s
ca
u
tio
n
ag
ain
s
t
o
v
er
-
r
elian
ce
o
n
o
f
f
s
ets
an
d
u
r
g
e
p
o
licy
m
ak
er
s
to
en
s
u
r
e
cr
ed
ib
ilit
y
,
tr
an
s
p
ar
en
cy
,
an
d
eq
u
it
y
in
n
et
-
ze
r
o
c
o
m
m
itm
en
ts
.
T
h
ey
ad
v
o
ca
te
f
o
r
s
ec
to
r
-
s
p
ec
if
ic
p
ath
way
s
,
r
o
b
u
s
t
ac
co
u
n
tin
g
m
ec
h
an
is
m
s
,
an
d
c
o
o
r
d
in
ate
d
g
o
v
er
n
a
n
ce
to
alig
n
clim
ate
ac
tio
n
with
s
cien
tifi
c
g
o
als.
T
h
eir
w
o
r
k
s
tr
ess
e
s
th
e
im
p
o
r
tan
ce
o
f
tr
an
s
f
o
r
m
in
g
th
e
p
la
n
s
in
to
ac
tio
n
ab
le
p
r
o
g
r
ess
,
wh
ich
h
elp
s
to
p
r
o
v
id
e
th
e
clar
ity
o
n
d
ef
in
itio
n
s
,
tim
elin
es,
an
d
r
esp
o
n
s
ib
ilit
ies.
T
h
e
co
n
tr
ib
u
tio
n
m
ad
e
b
y
th
e
a
u
th
o
r
s
h
elp
s
to
p
r
o
m
o
te
t
h
e
d
ec
ar
b
o
n
izatio
n
,
o
f
f
er
in
g
a
co
m
p
r
eh
en
s
iv
e
u
n
d
e
r
s
tan
d
in
g
o
f
clim
ate
p
o
licy
d
esig
n
.
Gan
d
o
m
an
et
a
l
.
[
2
5
]
u
n
d
er
ta
k
e
an
ex
p
l
o
r
atio
n
o
f
f
u
n
d
am
e
n
tals
an
d
a
d
v
an
ce
m
e
n
ts
co
n
c
er
n
in
g
th
e
ass
es
s
m
en
t
o
f
r
eliab
ilit
y
an
d
s
af
ety
o
f
lith
iu
m
-
io
n
b
atter
y
p
a
ck
s
in
E
Vs
.
T
h
eir
ex
p
l
o
r
atio
n
co
v
er
s
th
e
cr
itical
f
ailu
r
e
m
o
d
es,
d
ia
g
n
o
s
tics
,
a
n
d
test
m
eth
o
d
s
th
at
ai
d
s
th
e
ev
alu
atio
n
o
f
b
atter
y
h
ea
lth
an
d
p
r
e
v
en
tio
n
o
f
ac
cid
en
ts
.
T
h
ey
s
u
cc
ess
f
u
lly
r
ev
iewe
d
th
e
m
o
d
ellin
g
tec
h
n
iq
u
es,
ag
in
g
m
ec
h
an
is
m
s
,
an
d
th
er
m
al
-
ce
n
t
r
ic
is
s
u
es
w
ith
em
p
h
asis
o
n
th
e
lack
o
f
s
tan
d
ar
d
izatio
n
an
d
th
e
r
eq
u
ir
em
en
t
o
f
r
ea
l
-
tim
e
d
at
a
in
teg
r
atio
n
.
T
h
ei
r
wo
r
k
h
ig
h
lig
h
ts
th
e
s
ig
n
if
ican
ce
o
f
m
u
ltid
is
cip
lin
ar
y
tec
h
n
iq
u
es
th
at
u
s
ed
a
co
m
b
in
atio
n
o
f
elec
tr
ical,
th
er
m
al,
an
d
m
ec
h
an
ical
m
o
d
e
ls
f
o
r
o
b
tain
in
g
ac
cu
r
ate
ass
es
s
m
en
ts
.
L
iu
et
a
l
.
[
2
6
]
m
ad
e
u
s
e
o
f
a
s
im
p
lifie
d
p
s
eu
d
o
-
two
-
d
im
e
n
s
io
n
al
(
P2
D)
elec
tr
o
ch
em
ica
l
m
o
d
el
t
o
p
r
esen
t
a
d
u
al
esti
m
atio
n
m
et
h
o
d
f
o
r
SOC
an
d
SOH
o
f
a
b
atter
y
p
ac
k
.
T
h
is
wo
r
k
d
em
o
n
s
tr
ates
th
e
b
en
ef
its
th
at
th
e
in
teg
r
atio
n
b
etwe
en
th
e
elec
tr
o
ch
em
ical
m
o
d
ellin
g
an
d
esti
m
atio
n
alg
o
r
ith
m
s
y
ield
b
y
p
r
o
v
i
d
in
g
h
ig
h
ly
ac
cu
r
ate
in
s
ig
h
ts
o
n
t
h
e
in
ter
n
al
d
y
n
am
ics
o
f
a
b
a
tter
y
p
ac
k
f
r
o
m
tim
e
to
tim
e.
As
th
e
wo
r
k
ca
n
ac
co
u
n
t
f
o
r
b
atter
y
d
eg
r
a
d
atio
n
ef
f
ec
ts
(
lik
e
lith
iu
m
p
latin
g
an
d
ca
p
ac
ity
f
ad
e
)
,
it
ca
n
em
p
o
wer
th
e
r
ea
l
-
tim
e
h
ea
lth
m
o
n
ito
r
i
n
g
o
f
a
b
atter
y
p
ac
k
.
T
h
e
d
e
v
elo
p
e
d
m
o
d
el
h
o
ld
s
th
e
p
o
ten
tial
to
b
alan
ce
b
etwe
en
co
m
p
lex
ity
an
d
co
m
p
u
tatio
n
al
ef
f
icien
c
y
,
wh
ich
m
ak
es
it
id
ea
l
to
m
a
n
a
g
e
th
e
b
atter
y
p
ac
k
s
in
a
n
y
ele
ctr
ical
f
ac
ilit
y
.
T
h
e
au
th
o
r
s
in
th
is
wo
r
k
h
av
e
co
n
tr
ib
u
ted
m
u
ch
to
war
d
s
th
e
en
h
an
ce
m
en
t
o
f
p
r
ed
ictiv
e
m
ain
ten
an
ce
an
d
r
eliab
ilit
y
o
f
lith
iu
m
-
i
o
n
b
atter
y
p
ac
k
s
in
s
talled
with
in
E
Vs
a
n
d
g
r
i
d
s
to
r
ag
e.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
9
2
I
n
t J Ap
p
l Po
wer
E
n
g
,
Vo
l.
1
5
,
No
.
3
,
Sep
tem
b
er
20
2
6
:
1
0
7
2
-
1
0
9
3
1074
L
iu
et
a
l
.
[
2
7
]
p
er
f
o
r
m
a
d
is
cu
s
s
io
n
o
n
th
e
c
r
itical
tech
n
o
l
o
g
ies
th
at
s
h
ap
e
th
e
b
atter
y
m
a
n
ag
em
en
t
s
y
s
tem
s
(
B
M
S)
in
elec
tr
ic
v
eh
icles.
T
h
e
m
ain
em
p
h
asis
o
f
th
e
wo
r
k
was
o
n
ce
r
tain
co
r
e
f
u
n
ctio
n
s
lik
e
ce
ll
b
alan
cin
g
,
th
e
r
m
al
m
an
ag
em
e
n
t,
ch
ar
g
e
co
n
tr
o
l,
an
d
f
au
lt
d
etec
tio
n
.
T
h
is
wo
r
k
id
en
tifie
s
th
e
im
p
o
r
tan
t
r
o
l
e
p
lay
ed
b
y
cr
itical
f
ac
to
r
s
lik
e
ac
cu
r
ate
SOC
an
d
SOH
esti
m
atio
n
,
co
m
m
u
n
icatio
n
p
r
o
t
o
co
ls
,
an
d
in
teg
r
atio
n
with
v
eh
icle
co
n
tr
o
l
p
latf
o
r
m
s
.
T
h
e
a
u
th
o
r
s
d
is
cu
s
s
th
e
s
ch
em
atic
o
f
b
o
t
h
th
e
h
ar
d
war
e
an
d
s
o
f
twar
e
co
m
p
o
n
en
ts
with
atten
tio
n
g
iv
en
to
em
er
g
in
g
tr
en
d
s
lik
e
clo
u
d
-
b
ased
m
o
n
ito
r
in
g
an
d
AI
-
d
r
iv
en
an
al
y
tics
.
T
h
is
r
ev
iew
in
v
esti
g
atio
n
m
a
d
e
b
y
th
em
s
tr
ess
es
th
e
ex
is
tin
g
c
o
n
s
tr
ain
ts
lik
e
c
o
s
t
r
ed
u
c
tio
n
,
d
ata
s
ec
u
r
ity
,
an
d
th
e
in
teg
r
atio
n
o
f
ad
v
a
n
ce
d
s
en
s
o
r
s
.
T
h
e
co
n
tr
ib
u
tio
n
m
ad
e
b
y
th
e
wo
r
k
o
f
f
er
s
s
o
u
n
d
tech
n
ical
k
n
o
wled
g
e
t
h
at
th
e
f
u
tu
r
e
r
ese
ar
ch
er
s
an
d
e
n
g
in
ee
r
s
ca
n
lev
e
r
ag
e
to
d
esig
n
r
o
b
u
s
t BMS p
latf
o
r
m
s
.
An
ex
ten
s
iv
e
o
v
er
v
iew
o
f
th
e
r
m
al
m
an
a
g
em
en
t
s
tr
ateg
ies
was
p
r
esen
ted
b
y
Din
g
et
a
l.
[
2
8
]
to
s
o
r
t
o
u
t
th
e
tech
n
ical
co
n
s
tr
ain
ts
r
ea
lized
i
n
b
atter
y
p
ac
k
s
o
f
E
Vs
.
T
h
e
au
th
o
r
s
ca
teg
o
r
iz
e
m
eth
o
d
s
in
to
air
co
o
lin
g
,
liq
u
id
co
o
lin
g
,
p
h
ase
ch
an
g
e
m
ater
ials
,
a
n
d
th
er
m
o
elec
tr
ic
tech
n
o
lo
g
ie
s
,
an
aly
zin
g
th
eir
ef
f
ec
tiv
en
ess
,
co
s
t,
an
d
f
ea
s
i
b
ilit
y
.
T
h
e
p
ap
er
also
ex
p
l
o
r
es
n
o
v
el
tech
n
iq
u
es
lik
e
n
an
o
f
lu
id
s
an
d
h
y
b
r
id
s
y
s
tem
s
th
at
co
m
b
in
e
p
ass
iv
e
an
d
ac
tiv
e
co
o
lin
g
.
I
t
em
p
h
asizes
th
e
s
ig
n
if
ican
ce
o
f
th
er
m
al
u
n
if
o
r
m
ity
in
en
h
an
cin
g
b
atter
y
p
er
f
o
r
m
an
ce
an
d
life
s
p
an
.
Ad
d
itio
n
all
y
,
it
d
is
cu
s
s
es
co
n
tr
o
l
alg
o
r
ith
m
s
an
d
d
esig
n
o
p
tim
izatio
n
tech
n
i
q
u
es
f
o
r
e
f
f
icien
t
h
ea
t
d
is
s
ip
atio
n
.
T
h
is
r
ef
er
en
ce
is
cr
u
cial
f
o
r
en
s
u
r
in
g
b
atter
y
s
af
ety
,
im
p
r
o
v
in
g
en
e
r
g
y
e
f
f
icien
cy
,
an
d
s
u
p
p
o
r
tin
g
t
h
e
ev
o
lu
tio
n
o
f
h
ig
h
-
p
er
f
o
r
m
an
ce
E
V
s
y
s
tem
s
.
Sh
r
iv
astav
a
et
a
l.
[
2
9
]
r
e
v
iew
s
tate
-
of
-
ch
ar
g
e
esti
m
atio
n
tech
n
iq
u
es
f
o
r
lith
iu
m
-
i
o
n
b
atter
i
es,
with
a
f
o
cu
s
o
n
th
e
Kalm
an
f
ilter
f
am
ily
(
e
x
ten
d
ed
Kalm
an
f
ilter
(
E
KF)
,
u
n
s
ce
n
ted
Kalm
an
f
ilter
(
UKF)
,
an
d
c
u
b
atu
r
e
Kalm
an
f
ilter
(
C
KF)
)
.
I
t
co
m
p
a
r
es
th
e
s
tr
en
g
th
s
a
n
d
lim
itatio
n
s
o
f
ea
c
h
v
ar
ian
t
an
d
d
is
cu
s
s
es
th
eir
ap
p
licab
ilit
y
in
r
ea
l
-
tim
e
B
MS.
T
h
e
s
tu
d
y
h
ig
h
lig
h
ts
h
o
w
m
o
d
el
-
b
ased
esti
m
atio
n
im
p
r
o
v
es
r
o
b
u
s
tn
ess
u
n
d
er
v
ar
y
in
g
o
p
er
atin
g
co
n
d
itio
n
s
a
n
d
b
atter
y
a
g
in
g
.
T
h
e
au
t
h
o
r
s
also
ex
am
in
e
im
p
lem
en
tatio
n
ch
allen
g
es,
s
u
ch
as
m
o
d
ellin
g
ac
cu
r
ac
y
,
co
m
p
u
ta
tio
n
al
co
s
t,
an
d
s
en
s
o
r
n
o
is
e.
T
h
is
p
ap
er
is
ess
en
tial
f
o
r
u
n
d
er
s
tan
d
i
n
g
h
o
w
ad
v
an
ce
d
f
ilter
in
g
tech
n
iq
u
e
s
ca
n
en
h
a
n
ce
th
e
r
eliab
ilit
y
o
f
b
atter
y
ch
ar
g
e
esti
m
atio
n
,
esp
ec
ially
in
ap
p
licatio
n
s
d
em
an
d
in
g
p
r
ec
is
io
n
,
s
u
ch
as a
u
to
n
o
m
o
u
s
E
Vs
an
d
en
er
g
y
s
to
r
a
g
e
s
y
s
tem
s
.
L
i
et
a
l.
[
3
0
]
u
s
ed
m
u
lti
-
s
o
u
r
ce
d
ata
d
ep
en
d
in
g
u
p
o
n
a
u
n
i
q
u
e
d
ee
p
tr
an
s
f
er
lear
n
i
n
g
tec
h
n
iq
u
e
to
tack
le
th
e
p
r
o
b
lem
o
f
h
ea
lth
-
c
o
n
s
cio
u
s
b
atter
y
d
esig
n
i
n
g
.
T
o
im
p
r
o
v
e
th
e
b
atter
y
d
esig
n
'
s
tr
ain
in
g
p
r
o
ce
s
s
,
a
d
ee
p
tr
an
s
f
er
lear
n
i
n
g
tech
n
i
q
u
e
was
cr
ea
ted
th
at
allo
ws
s
tate
esti
m
ato
r
s
to
b
e
estab
li
s
h
ed
u
s
in
g
o
p
er
atin
g
d
ata
f
r
o
m
v
ar
io
u
s
E
V
k
in
d
s
.
Data
o
n
b
atter
y
f
u
n
ctio
n
in
g
g
ath
er
ed
f
r
o
m
two
d
if
f
e
r
en
t
k
i
n
d
s
o
f
elec
tr
ic
b
u
s
es
was
u
s
ed
to
v
alid
ate
th
e
a
p
p
r
o
ac
h
.
T
h
e
b
atter
y
m
o
d
el
er
r
o
r
m
ay
b
e
k
e
p
t
to
2
.
4
3
%
an
d
1
.
2
7
%
th
r
o
u
g
h
o
u
t
th
e
life
cy
cle
with
th
e
h
elp
o
f
th
e
tr
an
s
f
er
lear
n
in
g
tech
n
i
q
u
e
an
d
th
e
estab
lis
h
ed
h
ea
lth
y
s
tate
p
er
ce
p
tio
n
d
esig
n
.
A
d
ata
-
d
r
i
v
en
ap
p
r
o
ac
h
b
ased
o
n
s
p
ar
s
e
Gau
s
s
ian
p
r
o
ce
s
s
r
eg
r
ess
io
n
(
GPR
)
an
d
th
e
r
an
d
o
m
p
a
r
tial
ch
ar
g
in
g
p
r
o
ce
s
s
was
d
ev
elo
p
ed
b
y
[
3
1
]
.
I
n
itially
,
th
e
p
ar
ti
al
ch
ar
g
in
g
p
r
o
ce
s
s
'
s
r
an
d
o
m
ca
p
ac
ity
in
cr
em
en
t
s
eq
u
en
ce
s
(
Q)
at
v
a
r
io
u
s
v
o
lt
ag
e
s
eg
m
en
ts
wer
e
tak
e
n
o
u
t.
Q'
s
av
er
ag
e
v
al
u
e
a
n
d
s
tan
d
ar
d
d
ev
iatio
n
wer
e
em
p
lo
y
ed
as
ch
a
r
ac
ter
is
tics
to
s
h
o
w
th
e
co
n
d
itio
n
o
f
th
e
b
atter
y
.
Hig
h
c
o
r
r
elat
io
n
s
b
etwe
en
th
e
ch
ar
ac
ter
is
tics
an
d
b
atter
y
SOH
wer
e
co
n
f
ir
m
e
d
at
v
ar
io
u
s
tem
p
er
atu
r
es
an
d
d
is
ch
ar
g
in
g
cu
r
r
en
t
r
ates
af
ter
a
co
r
r
elatio
n
s
tu
d
y
was
p
er
f
o
r
m
ed
f
o
r
th
r
ee
d
is
tin
ct
b
atter
y
ty
p
es.
T
h
e
f
in
al
s
tep
in
v
o
lv
ed
b
u
ild
in
g
s
p
ar
s
e
GPR
m
o
d
els to
esti
m
ate
th
e
SOH
u
s
in
g
th
e
s
u
g
g
ested
ch
a
r
ac
ter
is
tics
as in
p
u
ts
.
T
ian
e
t
a
l
.
[
3
2
]
s
u
g
g
ested
a
v
er
s
atile
ap
p
r
o
ac
h
th
at
ad
d
r
ess
es
th
e
is
s
u
es
o
f
co
n
d
itio
n
o
f
h
ea
lth
an
d
SOC
e
s
tim
ates
at
th
e
s
am
e
ti
m
e
b
y
esti
m
atin
g
m
ax
im
al
an
d
r
em
ain
in
g
ca
p
ac
ities
u
s
in
g
ju
s
t
b
r
ief
ch
ar
g
in
g
d
ata.
T
h
e
s
u
g
g
ested
ap
p
r
o
ac
h
was
b
ased
o
n
a
co
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
(
C
NN)
th
at
ca
n
esti
m
ate
two
s
tates
u
s
in
g
ju
s
t
s
h
o
r
t
-
ter
m
ch
ar
g
in
g
d
ata.
E
ig
h
t
0
.
7
4
Ah
b
a
tter
ies
'
d
eter
io
r
atio
n
d
ata
s
er
v
ed
as
th
e
f
ir
s
t
b
asis
f
o
r
v
alid
atin
g
th
e
s
u
g
g
ested
ap
p
r
o
ac
h
.
Su
b
s
eq
u
en
t
test
s
o
n
two
m
o
r
e
b
atter
y
ty
p
es
d
e
m
o
n
s
tr
ated
th
at
th
e
s
u
g
g
ested
ap
p
r
o
ac
h
m
ay
g
u
ar
an
tee
ac
cu
r
ate
esti
m
ate
u
n
d
er
v
ar
i
o
u
s
o
p
er
atin
g
s
itu
atio
n
s
an
d
b
atter
y
ch
em
is
tr
ies.
A
v
er
s
atile
an
d
s
im
p
le
m
ec
h
a
n
is
m
f
o
r
ac
cu
r
atel
y
esti
m
atin
g
m
an
y
s
tates
ac
r
o
s
s
b
atter
y
life
was
p
r
o
v
id
e
d
b
y
th
e
m
eth
o
d
.
A
s
tu
d
y
o
f
b
atter
y
s
tate
esti
m
ate
tech
n
iq
u
es
d
ep
en
d
i
n
g
o
n
s
ev
er
al
ML
tech
n
iq
u
es
an
d
Ham
m
in
g
n
etwo
r
k
s
was
p
r
esen
ted
b
y
[
3
3
]
.
T
h
e
ap
p
r
o
ac
h
es
wer
e
co
m
p
ar
ed
b
ased
o
n
d
ec
lar
e
d
ac
cu
r
ac
y
,
test
s
ettin
g
s
,
b
atter
y
ty
p
es,
in
p
u
ts
an
d
o
u
tp
u
ts
,
an
d
d
ata
q
u
ality
to
p
r
o
v
i
d
e
r
ea
d
er
s
with
a
b
r
o
ad
er
p
er
s
p
ec
tiv
e
o
f
th
e
ML
lan
d
s
ca
p
e
f
o
r
SOC
an
d
SOH
ca
lcu
latio
n
.
W
h
en
co
m
p
ar
in
g
esti
m
ate
m
eth
o
d
o
lo
g
ies,
it
was
s
u
g
g
ested
th
at
th
e
n
etwo
r
k
s
u
n
d
e
r
co
m
p
ar
is
o
n
s
h
o
u
ld
b
e
t
r
ain
ed
an
d
ev
alu
at
ed
u
s
in
g
th
e
s
am
e
d
ata
an
d
h
av
e
a
co
m
p
ar
ab
le
n
u
m
b
er
o
f
lear
n
ab
le
p
a
r
am
ete
r
s
.
Yan
g
et
a
l.
[
3
4
]
u
s
ed
th
e
d
is
ch
ar
g
e
cu
r
v
es
to
s
u
g
g
est
a
n
ew
h
ea
lth
in
d
icato
r
.
I
n
v
esti
g
a
tio
n
s
wer
e
co
n
d
u
cte
d
o
n
h
o
w
tem
p
er
atu
r
e,
cu
r
r
e
n
t,
an
d
d
ep
th
-
of
-
d
is
c
h
ar
g
e
af
f
ec
te
d
th
e
h
ea
lth
in
d
icato
r
.
T
h
e
u
n
iq
u
e
h
ea
lth
in
d
icato
r
was
th
e
n
u
s
e
d
to
ch
ar
ac
te
r
ize
th
e
h
ea
lth
c
o
n
d
itio
n
o
f
b
atter
ies
th
at
h
ad
a
g
ed
at
v
ar
y
in
g
r
ates
an
d
d
ep
t
h
s
o
f
d
is
ch
ar
g
e
.
B
atter
ies
o
p
er
atin
g
in
v
ar
io
u
s
r
eg
i
m
es
h
av
e
th
eir
SOH
ev
alu
ated
u
s
in
g
an
em
p
ir
ical
m
o
d
el.
T
h
e
n
ew
h
ea
lth
in
d
icat
o
r
p
r
o
v
e
d
s
u
cc
ess
f
u
l
an
d
p
r
ac
tical
f
o
r
s
atellite
b
atter
y
SOH
esti
m
ate,
ac
co
r
d
in
g
to
th
e
s
atis
f
ac
to
r
y
esti
m
atio
n
f
in
d
in
g
s
with
a
m
ax
im
u
m
er
r
o
r
o
f
4
%.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ap
p
l Po
wer
E
n
g
I
SS
N:
2252
-
8
7
9
2
A
d
va
n
ce
d
A
I
-
d
r
iven
b
a
tter
y
s
ta
te
esti
ma
tio
n
u
s
in
g
d
ee
p
lea
r
n
in
g
a
n
d
…
(
N
.
S
u
ma
n
a
K
ee
r
th
i
)
1075
Fo
r
esti
m
atin
g
SOH,
W
an
g
et
a
l
.
[
3
5
]
alo
n
e
to
o
k
a
n
ew
h
ea
lth
in
d
icato
r
f
r
o
m
th
e
co
n
s
tan
t
v
o
ltag
e
p
r
o
f
ile'
s
m
o
n
ito
r
in
g
p
ar
a
m
eter
s
.
A
n
o
v
el
in
d
icatio
n
was c
r
ea
ted
af
ter
th
e
eq
u
iv
alen
t c
ir
c
u
it m
o
d
el
was u
s
ed
to
f
ir
s
t
d
ef
in
e
th
e
b
atter
y
ag
in
g
p
r
o
ce
s
s
es
d
u
r
in
g
th
e
co
n
s
tan
t
v
o
ltag
e
p
r
o
f
ile.
I
t
was
s
u
g
g
ested
to
u
s
e
a
f
r
am
ewo
r
k
f
o
r
ex
tr
ac
tin
g
th
is
in
d
icatio
n
o
n
lin
e.
T
h
e
p
er
f
o
r
m
an
ce
ev
alu
atio
n
an
d
co
r
r
elatio
n
an
aly
s
is
f
u
r
t
h
er
d
em
o
n
s
tr
ate
th
e
f
le
x
ib
ilit
y
a
n
d
e
f
f
icac
y
o
f
th
e
s
u
g
g
ested
ap
p
r
o
ac
h
f
o
r
d
eter
m
in
in
g
t
h
e
h
ea
lth
s
tatu
s
o
f
lith
iu
m
-
io
n
b
atter
ies.
A
h
ier
ar
ch
ical
esti
m
atin
g
ap
p
r
o
ac
h
th
at
tak
es
th
e
cu
r
r
e
n
t
r
ate
in
to
ac
co
u
n
t
was
d
ev
elo
p
ed
b
y
[
3
6
]
.
B
atter
y
s
tates
wer
e
esti
m
ated
u
s
in
g
a
m
u
ltis
ca
le
d
u
al
ex
ten
d
ed
Kalm
a
n
f
ilter
(
DE
KF
)
,
p
a
r
am
eter
s
wer
e
id
en
tifie
d
u
s
in
g
th
e
d
ata
-
d
r
i
v
en
tech
n
iq
u
e,
an
d
th
e
f
r
ac
tio
n
al
-
o
r
d
e
r
m
o
d
el
was
em
p
lo
y
ed
f
o
r
b
atter
y
m
o
d
elin
g
.
C
o
m
p
a
r
in
g
th
e
s
u
g
g
ested
tech
n
iq
u
e
to
class
ic
D
E
KF
u
n
d
er
two
s
itu
atio
n
s
,
th
e
ac
cu
r
ac
y
o
f
SOC
esti
m
ate
in
cr
ea
s
ed
b
y
3
5
.
8
% a
n
d
3
6
.
5
%,
r
esp
ec
tiv
ely
.
Un
d
e
r
two
cu
r
r
en
t c
ir
cu
m
s
tan
ce
s
,
th
e
ac
cu
r
ac
y
o
f
SOH
esti
m
ate
was in
cr
ea
s
ed
b
y
3
4
.
8
% a
n
d
4
3
.
1
%,
r
esp
ec
tiv
ely
,
u
s
in
g
th
e
s
u
g
g
ested
tech
n
iq
u
e.
Kh
an
et
a
l.
[
3
7
]
s
u
g
g
ested
MCC
Ps
B
MS
to
u
s
e
th
e
DL
id
ea
to
esti
m
ate
th
e
h
ea
lth
an
d
c
ap
ac
ity
o
f
b
atter
ies,
wh
er
e
th
e
p
atter
n
s
i
n
th
ese
C
Ps
alter
a
s
th
e
b
atter
y
m
atu
r
es
o
v
er
tim
e
an
d
c
y
cles.
T
h
is
led
to
a
th
o
r
o
u
g
h
i
n
v
esti
g
atio
n
o
f
b
o
t
h
ML
an
d
DL
-
d
e
p
en
d
e
n
t
ap
p
r
o
ac
h
es
in
o
r
d
er
to
o
f
f
er
a
tan
g
ib
le
co
m
p
ar
is
o
n
o
f
th
e
two
ap
p
r
o
ac
h
es.
Su
p
p
o
r
t
v
ec
to
r
r
e
g
r
ess
io
n
(
SVR
)
an
d
ad
ap
tiv
e
b
o
o
s
tin
g
(
AB
)
wer
e
ex
ten
s
iv
ely
test
ed
with
C
NN,
m
u
lti
-
lay
er
p
e
r
ce
p
tr
o
n
s
(
ML
P
)
,
b
i
-
d
ir
ec
tio
n
al
L
STM
(
B
iLST
M
)
,
an
d
L
STM
i
n
o
r
d
er
t
o
d
eter
m
in
e
th
e
b
est
m
eth
o
d
f
o
r
esti
m
atin
g
b
atter
y
ca
p
ac
ity
SOH.
B
iLST
M
s
u
r
p
ass
es
all
o
t
h
er
tech
n
iq
u
es,
as
d
em
o
n
s
tr
ated
em
p
ir
ically
u
s
in
g
th
e
NASA
b
atter
y
d
ataset
to
co
n
f
ir
m
a
n
d
v
er
if
y
t
h
e
s
u
g
g
est
ed
s
tr
ateg
y
.
Hu
et
a
l.
[
3
8
]
c
r
ea
ted
an
d
ex
p
er
im
en
tally
v
er
if
ie
d
a
u
n
iq
u
e
co
-
esti
m
atio
n
h
ier
ar
ch
y
f
o
r
lith
iu
m
-
io
n
b
atter
y
SOC
,
SOH,
an
d
s
tate
o
f
p
o
wer
(
SOP
)
.
Fo
r
ac
cu
r
ate
esti
m
ates
an
d
m
o
d
est
co
m
p
u
tin
g
co
s
t,
a
m
u
lti
-
tim
e
-
s
ca
le
esti
m
atin
g
f
r
am
ewo
r
k
was
d
esig
n
ed
tak
i
n
g
in
to
ac
co
u
n
t
th
e
u
n
d
er
l
y
in
g
c
o
n
n
ec
tio
n
an
d
f
ea
t
u
r
es
am
o
n
g
v
a
r
io
u
s
s
tates.
A
n
u
m
b
er
o
f
o
p
er
atio
n
al
r
estrictio
n
s
wer
e
r
esp
ec
ted
b
y
th
e
m
o
d
el
-
d
ep
en
d
e
n
t
r
ea
l
-
tim
e
SOP
esti
m
ate,
wh
ich
ac
cu
r
at
ely
p
r
o
jecte
d
th
e
b
atter
y
p
ea
k
p
o
wer
g
iv
e
n
u
p
d
ated
s
tates
an
d
p
a
r
am
eter
s
.
L
astl
y
,
m
an
y
test
s
wer
e
co
n
d
u
cted
to
s
h
o
w
th
e
co
m
b
in
a
tio
n
SOC
/S
OH/SO
P
est
im
ati
o
n
'
s
d
u
r
ab
ilit
y
an
d
ef
f
icac
y
.
W
h
en
co
m
p
a
r
ed
t
o
in
d
ep
en
d
e
n
t
esti
m
atio
n
s
o
lu
tio
n
s
,
ex
p
e
r
im
en
tal
f
in
d
in
g
s
d
em
o
n
s
tr
ate
th
e
s
ig
n
if
ican
t
ad
v
a
n
tag
es
o
f
th
e
s
u
g
g
ested
co
-
esti
m
atio
n
h
ier
a
r
ch
y
.
B
o
th
n
ew
a
n
d
o
ld
ce
lls
h
ad
SOC
,
v
o
ltag
e,
an
d
ca
p
ac
ity
esti
m
ate
er
r
o
r
s
o
f
less
th
an
3
%,
2
5
m
V,
a
n
d
3
%,
r
esp
ec
tiv
ely
.
T
r
ad
itio
n
al
p
h
y
s
ics
-
b
ased
m
o
d
els,
s
u
ch
as
eq
u
iv
alen
t
cir
cu
it
m
o
d
els
(
E
C
Ms)
an
d
elec
tr
o
ch
em
ical
m
o
d
els
,
h
av
e
b
ee
n
e
x
ten
s
iv
ely
u
s
ed
f
o
r
SOC
/SOH
est
im
ati
o
n
,
as
r
ev
iewe
d
in
[
2
6
]
,
[
2
9
]
.
W
h
ile
ef
f
ec
tiv
e
in
co
n
tr
o
lled
s
ettin
g
s
,
th
ese
m
o
d
els
o
f
ten
d
em
an
d
d
etailed
b
atter
y
ch
ar
ac
ter
izatio
n
an
d
ar
e
co
m
p
u
tatio
n
ally
ex
p
en
s
iv
e.
R
ec
en
t r
esear
ch
h
a
s
s
h
if
ted
to
war
d
s
d
ata
-
d
r
i
v
en
m
eth
o
d
s
:
-
L
iu
et
a
l.
[
2
6
]
p
r
o
p
o
s
ed
a
jo
i
n
t
SOC
/
SOH
est
im
atio
n
tech
n
i
q
u
e
lev
er
a
g
in
g
a
p
s
eu
d
o
-
2
D
el
ec
tr
o
ch
em
ical
m
o
d
el,
wh
ich
en
ab
led
en
h
a
n
c
ed
r
ea
l
-
tim
e
m
o
n
ito
r
in
g
b
y
ca
p
tu
r
in
g
i
n
ter
n
al
d
y
n
am
ics.
-
T
ian
et
al
.
[
3
2
]
i
n
tr
o
d
u
ce
d
a
C
NN
-
b
ased
m
eth
o
d
f
o
r
s
im
u
ltan
eo
u
s
ly
p
r
ed
ictin
g
SOC
an
d
SOH
f
r
o
m
s
h
o
r
t
-
ter
m
ch
a
r
g
in
g
d
ata,
o
f
f
er
in
g
a
lig
h
tweig
h
t
y
et
ac
cu
r
ate
f
r
am
ewo
r
k
.
-
Xu
et
a
l.
[
3
6
]
em
p
lo
y
ed
a
h
ier
ar
ch
ical
m
u
ltis
ca
le
Kalm
an
f
ilter
-
b
ased
ap
p
r
o
ac
h
th
at
s
ig
n
if
ican
tly
im
p
r
o
v
e
d
SOC
an
d
SOH
esti
m
atio
n
ac
cu
r
ac
y
u
n
d
er
v
ar
iab
l
e
cu
r
r
en
t
r
ates.
-
Kh
an
et
a
l.
[
3
7
]
e
v
alu
ated
tr
a
d
itio
n
al
ML
(
e.
g
.
,
SVR
,
Ad
aBo
o
s
t)
ag
ain
s
t
d
ee
p
lear
n
in
g
te
ch
n
iq
u
es
(
e.
g
.
,
L
STM
,
B
iLST
M)
,
r
ev
ea
lin
g
B
iLST
M
'
s
s
u
p
er
io
r
p
er
f
o
r
m
an
ce
f
o
r
SOH
p
r
ed
ictio
n
u
s
in
g
NASA
b
atter
y
d
atasets
.
-
T
POT
r
eg
r
ess
o
r
,
as e
x
p
lo
r
e
d
in
[
3
3
]
,
u
tili
ze
s
au
to
m
ated
m
ac
h
in
e
lear
n
in
g
(
Au
to
ML
)
f
o
r
m
o
d
el
s
elec
tio
n
an
d
h
y
p
e
r
p
ar
a
m
eter
tu
n
in
g
,
p
r
o
v
id
in
g
p
r
o
m
is
in
g
o
u
tco
m
es
f
o
r
SOH
p
r
ed
ictio
n
s
with
o
u
t
m
an
u
al
in
ter
v
en
tio
n
.
-
Kalm
an
f
ilter
-
b
ased
ap
p
r
o
ac
h
es
lik
e
E
KF
an
d
UKF
[
2
9
]
o
f
f
er
r
elia
b
le
p
er
f
o
r
m
a
n
ce
u
n
d
e
r
s
tead
y
co
n
d
itio
n
s
b
u
t
ten
d
to
f
alter
wh
en
b
atter
y
d
y
n
am
ics
ch
an
g
e
d
u
e
to
a
g
in
g
,
tem
p
er
atu
r
e
v
ar
iatio
n
,
o
r
ir
r
eg
u
lar
u
s
ag
e.
-
E
lectr
o
ch
em
ical
m
o
d
els
[
2
6
]
,
wh
ile
ac
cu
r
ate,
d
em
an
d
co
m
p
lex
p
h
y
s
ical
m
o
d
elin
g
an
d
ar
e
n
o
t
well
-
s
u
ited
f
o
r
r
ea
l
-
tim
e,
s
ca
lab
le
im
p
lem
en
tatio
n
s
.
-
ML
m
eth
o
d
s
lik
e
SVR
,
Ad
a
B
o
o
s
t,
an
d
XGBo
o
s
t
[
3
7
]
h
an
d
le
n
o
n
lin
ea
r
p
atter
n
s
b
u
t
o
f
ten
d
ep
e
n
d
h
ea
v
ily
o
n
m
an
u
al
f
ea
tu
r
e
e
n
g
in
ee
r
in
g
an
d
h
y
p
er
p
ar
am
eter
t
u
n
in
g
.
-
C
NN
-
b
ased
d
u
al
esti
m
atio
n
[
3
2
]
ad
d
r
ess
es
b
o
th
SOC
an
d
SOH,
b
u
t
m
ay
u
n
d
er
p
er
f
o
r
m
wh
en
th
e
ch
ar
g
in
g
d
ata
lack
s
v
ar
iatio
n
o
r
g
r
an
u
lar
ity
.
-
L
iu
et
a
l.
[
2
6
]
an
d
Sh
r
iv
ast
av
a
e
t
a
l
.
[
2
9
]
h
ig
h
lig
h
t
th
e
lim
itatio
n
s
o
f
m
o
d
el
-
b
ase
d
ap
p
r
o
ac
h
es,
estab
lis
h
in
g
th
e
m
o
tiv
atio
n
f
o
r
d
ata
-
d
r
iv
e
n
s
o
lu
tio
n
s
.
-
L
i
et
a
l.
[
3
0
]
an
d
T
ian
e
t
a
l
.
[
3
2
]
d
e
m
o
n
s
tr
ate
th
e
g
r
o
win
g
s
u
cc
ess
o
f
d
ee
p
lear
n
in
g
in
b
atter
y
d
iag
n
o
s
tics
,
lay
in
g
th
e
f
o
u
n
d
at
io
n
f
o
r
u
s
in
g
R
NN
an
d
L
STM
in
o
u
r
SOC
m
o
d
elin
g
.
-
Vid
al
et
a
l.
[
3
3
]
co
n
d
u
ct
a
s
u
r
v
ey
c
o
m
p
ar
i
n
g
ML
m
eth
o
d
s
,
ju
s
tify
in
g
o
u
r
d
ec
is
io
n
to
test
m
u
ltip
le
m
o
d
els s
id
e
-
by
-
s
id
e
u
n
d
er
th
e
s
am
e
d
ataset
an
d
co
n
d
itio
n
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
9
2
I
n
t J Ap
p
l Po
wer
E
n
g
,
Vo
l.
1
5
,
No
.
3
,
Sep
tem
b
er
20
2
6
:
1
0
7
2
-
1
0
9
3
1076
-
Kh
an
et
a
l.
[
3
7
]
em
p
ir
ically
p
r
o
v
e
t
h
at
B
iLST
M
o
u
tp
er
f
o
r
m
s
o
th
er
m
o
d
els
f
o
r
SOH
—
th
is
s
u
p
p
o
r
ts
o
u
r
r
atio
n
ale
f
o
r
u
s
in
g
s
tack
ed
L
STM
f
o
r
SOC
d
u
e
to
s
im
ilar
te
m
p
o
r
al
d
e
p
en
d
e
n
cies.
-
Den
g
et
a
l.
[
3
1
]
an
d
Yan
g
et
a
l,
d
is
cu
s
s
f
ea
tu
r
e
e
n
g
in
e
er
in
g
f
r
o
m
p
ar
tial
c
h
ar
g
e/
d
is
ch
ar
g
e
c
y
cles,
r
ein
f
o
r
cin
g
th
e
id
ea
th
at
ef
f
ec
t
iv
e
m
o
d
els ca
n
b
e
tr
ain
ed
f
r
o
m
co
m
m
o
n
o
p
er
atio
n
al
d
ata
.
-
Xu
et
a
l.
[
3
6
]
in
tr
o
d
u
ce
a
h
i
er
ar
ch
ical
esti
m
atio
n
m
o
d
el,
wh
ich
in
s
p
ir
es
o
u
r
m
o
d
u
lar
a
p
p
r
o
ac
h
with
d
ec
o
u
p
le
d
p
ip
elin
es f
o
r
SOC
an
d
SOH.
T
h
ese
s
tu
d
ies
co
n
f
ir
m
th
e
g
r
o
win
g
in
ter
est
in
m
ac
h
in
e
l
ea
r
n
in
g
a
n
d
d
ee
p
lear
n
in
g
f
o
r
b
atter
y
d
iag
n
o
s
tics
.
Ho
wev
er
,
m
o
s
t
o
f
th
em
tr
ea
t
SOC
an
d
SOH
as
is
o
lated
p
r
o
b
lem
s
o
r
r
ely
h
ea
v
ily
o
n
co
m
p
le
x
tr
ain
in
g
s
etu
p
s
,
lack
in
g
r
ea
l
-
tim
e
ad
ap
tab
ilit
y
an
d
u
n
i
f
ied
s
y
s
tem
d
esig
n
.
Als
o
,
r
e
liab
le
an
d
h
ig
h
ly
co
m
p
u
tatio
n
al
SOC
an
d
SOH
esti
m
ate
alg
o
r
ith
m
s
ar
e
r
eq
u
ir
ed
,
as
well
as
th
e
ab
ilit
y
to
m
o
d
if
y
in
r
esp
o
n
s
e
to
th
e
p
r
ac
tical
-
s
en
s
e
o
p
er
atin
g
s
itu
atio
n
s
(
tem
p
er
atu
r
e
v
ar
iati
o
n
s
,
ag
e,
an
d
s
o
o
n
)
with
m
i
n
im
u
m
in
f
o
r
m
atio
n
u
s
ed
f
o
r
tr
ai
n
in
g
.
Desp
ite
n
o
tab
le
ad
v
an
ce
m
e
n
t
s
,
cu
r
r
en
t
m
eth
o
d
s
f
o
r
esti
m
at
in
g
SOC
an
d
SOH
s
u
f
f
er
f
r
o
m
m
u
ltip
le
lim
itatio
n
s
:
-
L
ac
k
o
f
r
ea
l
-
tim
e
ad
ap
tab
ilit
y
:
Ma
n
y
m
eth
o
d
s
f
ail
u
n
d
er
d
y
n
am
ic
o
p
e
r
atio
n
al
co
n
d
itio
n
s
d
u
e
to
th
eir
s
tatic
m
o
d
elin
g
ass
u
m
p
tio
n
s
.
-
Ab
s
en
ce
o
f
u
n
if
ied
s
y
s
tem
s
:
Few
f
r
am
ewo
r
k
s
s
im
u
ltan
e
o
u
s
ly
ad
d
r
ess
b
o
th
SOC
an
d
SO
H
co
m
p
u
tatio
n
u
s
in
g
d
is
tin
ct
b
u
t in
teg
r
ated
m
o
d
els.
-
Ma
n
u
al
m
o
d
el
tu
n
i
n
g
:
C
o
n
v
e
n
tio
n
al
tech
n
iq
u
es
o
f
ten
r
eq
u
ir
e
d
o
m
ain
-
s
p
ec
if
ic
e
x
p
er
tis
e
f
o
r
ca
lib
r
atio
n
,
lim
itin
g
s
ca
lab
ilit
y
.
-
I
n
ad
eq
u
ate
en
d
-
u
s
er
in
teg
r
ati
o
n
:
Mo
s
t
r
esear
ch
i
m
p
lem
en
tatio
n
s
ar
e
n
o
t
d
ep
lo
y
ab
le
t
h
r
o
u
g
h
u
s
er
-
f
r
ien
d
ly
i
n
ter
f
ac
es,
m
ak
in
g
th
e
m
in
ac
ce
s
s
ib
le
to
n
o
n
-
s
p
ec
iali
s
ts
.
T
h
ese
g
ap
s
h
ig
h
lig
h
t
th
e
n
ee
d
f
o
r
a
m
o
d
u
lar
,
in
tellig
en
t
s
y
s
tem
th
at
ca
n
p
r
ed
ict
b
o
th
SOC
an
d
SOH
u
s
in
g
h
y
b
r
id
,
ad
a
p
tab
le
m
eth
o
d
o
lo
g
ies
wh
ile
b
ein
g
d
ep
lo
y
a
b
le
in
r
ea
l
-
wo
r
ld
s
ettin
g
s
.
Acc
u
r
ate
an
d
r
ea
l
-
tim
e
esti
m
atio
n
o
f
b
atter
y
p
ac
k
-
ce
n
tr
ic
m
ea
s
u
r
es
-
v
iz.
,
SOC
an
d
SOH
r
em
ain
s
a
s
u
b
s
tan
tial
c
h
allen
g
e
in
en
er
g
y
f
ac
ilit
ies,
p
ar
ticu
lar
ly
in
elec
tr
ic
-
o
p
er
ated
v
eh
icles
an
d
s
m
ar
t
g
r
id
s
.
C
o
n
v
en
tio
n
al
esti
m
atio
n
tech
n
iq
u
es,
s
u
ch
as
Kalm
an
f
ilter
s
an
d
elec
tr
o
ch
em
ical
m
o
d
els,
o
f
ten
s
u
f
f
er
f
r
o
m
h
ig
h
co
m
p
lex
ity
,
l
o
w
ad
ap
tab
ilit
y
,
an
d
s
en
s
itiv
ity
to
en
v
ir
o
n
m
en
tal
c
h
an
g
es
[
3
9
]
,
[
4
0
]
.
T
h
ese
m
o
d
els
m
ay
n
o
t
g
en
er
alize
s
p
an
n
i
n
g
u
p
o
n
th
e
d
is
tin
ct
b
atter
y
p
ac
k
v
ar
ian
ts
,
o
p
e
r
atio
n
al
cir
cu
m
s
tan
ce
s
,
an
d
d
e
g
r
ad
atio
n
p
atter
n
s
,
lead
in
g
t
o
in
ac
cu
r
ate
p
r
ed
ictio
n
s
.
Fu
r
th
er
m
o
r
e
,
u
n
if
ied
m
o
d
els
f
o
r
b
o
th
SOC
an
d
SOH
a
r
e
d
if
f
ic
u
lt
to
co
n
s
tr
u
ct
d
u
e
to
d
if
f
er
in
g
d
ata
ch
ar
ac
ter
is
tics
an
d
d
ep
en
d
en
c
ies.
T
h
is
lim
itat
io
n
p
o
s
es
r
is
k
s
s
u
ch
as
in
ef
f
icien
t
b
atter
y
u
tili
za
tio
n
,
r
ed
u
ce
d
o
p
er
atio
n
al
s
af
et
y
,
an
d
in
cr
ea
s
ed
m
ain
ten
an
ce
co
s
ts
.
As
b
atter
y
s
y
s
tem
s
s
ca
le
in
s
ize
an
d
im
p
o
r
tan
ce
,
t
h
e
n
ee
d
f
o
r
a
f
lex
ib
le,
in
tellig
e
n
t,
an
d
p
r
ec
is
e
esti
m
atio
n
m
eth
o
d
h
as
b
ec
o
m
e
cr
itical
[
1
5
]
.
T
h
e
p
r
o
b
lem
is
f
u
r
th
er
co
m
p
o
u
n
d
ed
b
y
th
e
l
ac
k
o
f
ac
ce
s
s
ib
le
to
o
ls
th
at
i
n
teg
r
ate
m
ac
h
in
e
lea
r
n
in
g
-
b
a
s
ed
p
r
ed
ictio
n
s
in
to
p
r
ac
tical
ap
p
licatio
n
s
f
o
r
en
d
-
u
s
er
s
.
T
h
e
k
ey
is
s
u
es
in
clu
d
e:
i)
n
on
-
lin
ea
r
ity
o
f
b
atter
y
b
eh
av
io
r
u
n
d
e
r
d
iv
er
s
e
o
p
er
atin
g
co
n
d
itio
n
s
;
ii)
v
ar
iab
ilit
y
in
d
eg
r
ad
atio
n
r
ates
d
u
e
to
tem
p
er
atu
r
e
,
u
s
ag
e
p
atter
n
s
,
an
d
ce
ll
ch
em
is
tr
y
;
iii)
l
ac
k
o
f
u
n
iv
er
s
al
m
o
d
elin
g
a
p
p
r
o
ac
h
es
—
wh
at
wo
r
k
s
f
o
r
SOC
m
ay
n
o
t
wo
r
k
f
o
r
SOH
;
iv
)
n
ee
d
f
o
r
r
ea
l
-
tim
e
in
f
e
r
en
ce
with
h
i
g
h
ac
cu
r
ac
y
in
s
af
ety
-
c
r
itical
ap
p
licatio
n
s
;
an
d
v
)
d
if
f
ic
u
lty
in
co
llectin
g
h
ig
h
-
q
u
ality
lab
eled
d
atasets
,
esp
ec
ially
f
o
r
SOH.
T
h
er
ef
o
r
e,
th
is
wo
r
k
a
d
d
r
ess
es
th
e
u
r
g
en
t
n
e
ed
to
d
e
v
elo
p
a
r
o
b
u
s
t,
r
ea
l
-
tim
e,
a
n
d
u
s
er
-
f
r
ien
d
ly
d
u
al
-
m
o
d
el
f
r
am
ewo
r
k
u
s
in
g
d
ee
p
lear
n
in
g
tech
n
i
q
u
es
to
p
r
ec
is
ely
co
m
p
u
te
b
o
th
SOC
an
d
SOH,
ass
u
r
in
g
b
etter
en
e
r
g
y
ad
m
i
n
is
tr
atio
n
a
n
d
b
atter
y
lo
n
g
e
v
ity
.
I
n
o
u
r
w
o
r
k
,
we
m
a
k
e
u
s
e
o
f
d
ata
-
ce
n
t
r
ic
an
d
h
y
b
r
i
d
d
ee
p
lear
n
in
g
m
eth
o
d
o
lo
g
ies
with
t
h
e
p
r
im
e
aim
o
f
ad
d
r
ess
in
g
th
e
af
o
r
esaid
lim
itatio
n
s
ex
p
er
ien
ce
d
in
th
e
b
atter
y
p
ac
k
o
f
E
Vs
.
W
e
tak
e
ad
v
a
n
ta
g
e
o
f
h
is
to
r
ical
an
d
r
ea
l
-
tim
e
d
ata
co
llected
with
wid
e
r
an
g
e
o
f
s
en
s
o
r
s
,
wh
ich
em
p
o
wer
s
th
e
lear
n
i
n
g
o
f
co
m
p
lex
b
atter
y
b
eh
a
v
io
r
p
atter
n
s
to
o
f
f
er
v
ig
o
r
o
u
s
p
r
ed
ictio
n
s
o
f
SOC
an
d
SOH.
I
n
o
u
r
wo
r
k
,
we
m
ak
e
u
s
e
o
f
l
in
ea
r
r
eg
r
ess
io
n
,
R
NN,
s
tack
ed
L
STM
,
g
ate
d
r
ec
u
r
r
en
t
u
n
it
(
GR
U)
,
an
d
GR
U
-
L
STM
f
o
r
SOC
,
an
d
Stack
in
g
an
d
T
POT
r
eg
r
ess
o
r
,
lin
ea
r
r
eg
r
ess
io
n
,
GR
U,
an
d
GR
U
-
L
STM
f
o
r
SOH.
T
h
e
en
tire
s
y
s
tem
co
n
s
tr
u
cted
b
y
u
s
u
s
in
g
th
e
af
o
r
esaid
m
eth
o
d
o
lo
g
ies
h
as
b
ee
n
im
p
lem
en
te
d
as
a
u
s
er
-
f
r
ien
d
ly
web
ap
p
licatio
n
.
W
e
h
a
v
e
u
s
ed
m
u
ltip
le
to
o
ls
s
er
v
in
g
u
n
iq
u
e
r
o
les.
W
e
m
a
k
e
u
s
e
o
f
Py
th
o
n
to
d
o
th
e
m
o
d
el
d
ev
elo
p
m
e
n
t
an
d
i
n
teg
r
atio
n
.
Als
o
,
we
m
ak
e
u
s
e
o
f
o
t
h
er
to
o
ls
lik
e
HT
ML
,
C
SS
,
an
d
J
av
aScr
ip
t
f
o
r
p
o
wer
in
g
t
h
e
in
ter
f
ac
e
t
o
o
f
f
e
r
s
m
o
o
th
er
u
s
er
in
ter
ac
tio
n
.
T
h
e
d
ev
elo
p
ed
h
y
b
r
i
d
f
r
am
ew
o
r
k
ca
n
s
u
cc
ess
f
u
lly
o
f
f
er
a
s
ca
lab
le
s
o
lu
tio
n
,
p
er
m
itti
n
g
th
e
r
ea
l
-
tim
e
b
atter
y
s
tate
m
o
n
ito
r
in
g
to
s
er
v
e
v
ar
i
o
u
s
in
d
u
s
tr
ies.
T
h
is
ca
n
ev
en
tu
ally
im
p
r
o
v
e
th
e
e
n
er
g
y
ef
f
icien
cy
an
d
life
cy
cle
m
an
ag
em
en
t
o
f
p
o
wer
s
y
s
tem
s
.
T
h
is
wo
r
k
in
tr
o
d
u
ce
s
a
h
y
b
r
id
d
ee
p
lear
n
i
n
g
f
r
am
ewo
r
k
th
at
ad
d
r
ess
es th
e
ab
o
v
e
ch
allen
g
e
s
b
y
:
-
Dev
elo
p
in
g
s
ep
ar
ate,
o
p
tim
ize
d
m
o
d
els
f
o
r
SOC
an
d
SOH
p
r
ed
ictio
n
,
tailo
r
ed
to
t
h
e
r
esp
e
ctiv
e
d
atasets
(
f
etch
ed
f
r
o
m
a
s
in
g
le
s
o
u
r
ce
)
an
d
b
e
h
av
io
r
al
d
y
n
am
ics.
-
E
m
p
lo
y
in
g
a
co
m
p
ar
ativ
e
m
et
h
o
d
o
lo
g
y
,
wh
e
r
ein
th
r
ee
tec
h
n
iq
u
es
—
lin
ea
r
r
eg
r
ess
io
n
,
R
NN,
an
d
s
tack
ed
L
STM
—
ar
e
u
s
ed
f
o
r
SOC
co
m
p
u
tatio
n
,
an
d
two
—
s
tack
in
g
r
eg
r
ess
o
r
an
d
T
POT
r
eg
r
es
s
o
r
—
ar
e
u
s
ed
f
o
r
SOH.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ap
p
l Po
wer
E
n
g
I
SS
N:
2252
-
8
7
9
2
A
d
va
n
ce
d
A
I
-
d
r
iven
b
a
tter
y
s
ta
te
esti
ma
tio
n
u
s
in
g
d
ee
p
lea
r
n
in
g
a
n
d
…
(
N
.
S
u
ma
n
a
K
ee
r
th
i
)
1077
-
I
n
co
r
p
o
r
atin
g
GR
U
an
d
GR
U
-
L
STM
m
o
d
els
to
en
h
an
c
e
SOC
/
SOH
e
s
tim
atio
n
th
r
o
u
g
h
im
p
r
o
v
ed
s
eq
u
en
ce
lear
n
in
g
ca
p
a
b
ilit
ies
.
GR
U
o
f
f
er
s
a
lig
h
tweig
h
t
an
d
ef
f
icien
t
s
o
lu
tio
n
with
f
ewe
r
p
ar
am
eter
s
,
id
ea
l
f
o
r
f
ast
SOC
esti
m
atio
n
,
wh
ile
th
e
GR
U
-
L
STM
h
y
b
r
id
m
o
d
el
co
m
b
i
n
es
th
e
m
e
m
o
r
y
r
ete
n
tio
n
s
tr
en
g
th
s
o
f
L
STM
with
th
e
co
m
p
u
tatio
n
al
ef
f
icien
cy
o
f
GR
U
to
p
r
o
v
id
e
r
o
b
u
s
t
p
r
e
d
ictio
n
s
f
o
r
l
o
n
g
-
ter
m
d
eg
r
a
d
atio
n
p
atter
n
s
.
-
I
n
teg
r
atin
g
Au
to
ML
(
T
POT)
f
o
r
au
to
m
atic
p
ip
elin
e
o
p
tim
iz
atio
n
,
en
s
u
r
in
g
m
in
im
al
h
u
m
a
n
in
ter
v
en
tio
n
f
o
r
m
o
d
el
tu
n
i
n
g
.
-
Dep
lo
y
in
g
a
f
u
ll
-
s
tack
web
ap
p
licatio
n
u
s
in
g
Py
th
o
n
(
Flas
k
)
f
o
r
b
ac
k
en
d
m
o
d
elin
g
an
d
H
T
ML
/C
SS
/J
S
f
o
r
th
e
f
r
o
n
te
n
d
in
ter
f
ac
e,
th
u
s
en
ab
lin
g
r
ea
l
-
tim
e,
u
s
er
-
ac
ce
s
s
ib
le
p
r
ed
ictio
n
s
.
-
E
v
alu
atin
g
an
d
b
en
c
h
m
ar
k
in
g
all
tech
n
iq
u
es
with
in
d
u
s
tr
y
-
s
tan
d
ar
d
m
etr
ics
(
R
²,
MA
E
,
MSE
,
R
MSE
)
,
v
alid
atin
g
th
e
r
o
b
u
s
tn
ess
an
d
r
eliab
ilit
y
o
f
th
e
m
o
d
els.
T
h
is
d
u
al
-
m
o
d
el
ap
p
r
o
ac
h
allo
ws
u
s
er
s
to
in
p
u
t
b
atter
y
d
ata
an
d
in
s
tan
tly
r
ec
eiv
e
SOC
an
d
SOH
esti
m
atio
n
s
,
m
ak
in
g
it
s
u
itab
le
f
o
r
p
r
ac
tical
d
ep
l
o
y
m
en
t
i
n
E
Vs,
s
m
ar
t
g
r
id
s
,
an
d
B
MS
p
latf
o
r
m
s
.
T
h
e
p
r
in
cip
al
o
b
jectiv
e
o
f
o
u
r
wo
r
k
is
to
b
u
ild
a
s
tr
o
n
g
,
r
ea
l
-
ti
m
e,
an
d
ac
cu
r
ate
f
ac
ilit
y
f
o
r
th
e
co
m
p
u
tatio
n
o
f
b
o
th
m
ea
s
u
r
es
-
v
iz.
,
SOC
an
d
SOH
o
f
b
atter
y
p
ac
k
s
m
ad
e
o
f
lith
iu
m
-
io
n
co
m
p
o
s
itio
n
u
s
in
g
h
y
b
r
id
m
ac
h
in
e
lear
n
in
g
m
o
d
els.
Sp
ec
if
ically
i
n
o
u
r
wo
r
k
,
we
lo
o
k
to
:
-
Gen
er
ate
a
s
ep
ar
ate
d
ee
p
lear
n
in
g
m
eth
o
d
o
lo
g
y
to
war
d
s
th
e
co
m
p
u
tatio
n
o
f
SOC
u
s
in
g
lin
ea
r
r
eg
r
ess
io
n
,
R
NN,
an
d
s
tack
ed
L
STM
,
en
s
u
r
in
g
ac
cu
r
ate
p
r
e
d
ictio
n
o
f
b
atter
y
ch
a
r
g
e
le
v
el
u
n
d
er
v
ar
y
in
g
o
p
er
atio
n
al
c
o
n
d
itio
n
s
.
-
C
r
ea
te
an
ad
v
an
ce
d
en
s
em
b
l
e
-
b
ased
m
o
d
el
u
s
in
g
s
tack
in
g
r
eg
r
ess
o
r
an
d
T
POT
r
eg
r
e
s
s
o
r
f
o
r
SOH
esti
m
atio
n
,
f
o
cu
s
in
g
o
n
ca
p
t
u
r
in
g
lo
n
g
-
ter
m
b
atter
y
d
e
g
r
ad
at
io
n
p
atter
n
s
.
-
I
n
teg
r
ate
b
o
th
m
o
d
els
in
to
a
Py
th
o
n
-
b
ased
b
ac
k
e
n
d
an
d
c
o
n
n
ec
t
th
em
with
a
lig
h
tweig
h
t,
r
esp
o
n
s
iv
e
f
r
o
n
ten
d
in
ter
f
ac
e
u
s
in
g
HT
M
L
,
C
SS
,
an
d
J
av
aScr
ip
t.
-
Allo
w
u
s
er
s
to
in
p
u
t
b
atter
y
p
ar
am
eter
s
o
r
s
en
s
o
r
d
ata
th
r
o
u
g
h
a
web
in
ter
f
ac
e
a
n
d
r
ec
eiv
e
p
r
ac
tical
-
s
en
s
e
esti
m
atio
n
s
o
f
b
atter
y
p
a
ck
-
ce
n
tr
ic
m
ea
s
u
r
es
-
v
iz.
,
SOC
an
d
SOH.
-
E
n
s
u
r
e
th
at
th
e
f
ac
ilit
y
is
m
o
d
u
lar
,
in
ter
p
r
etab
le,
an
d
ca
p
a
b
le
o
f
a
d
ap
tin
g
to
n
ew
d
ata
o
r
b
atter
y
t
y
p
es
with
m
in
im
al
r
etr
ain
in
g
.
B
y
f
u
lf
illi
n
g
th
ese
o
b
jectiv
es,
th
e
wo
r
k
co
n
tr
ib
u
tes
to
s
m
ar
te
r
b
atter
y
m
an
ag
em
e
n
t,
im
p
r
o
v
ed
s
af
ety
,
an
d
lo
n
g
e
r
b
atter
y
life
s
p
an
—
esp
ec
ially
in
g
r
id
s
to
r
ag
e
f
ac
ilit
ies
an
d
elec
tr
ic
-
p
o
wer
ed
v
eh
icles
wh
er
e
b
atter
y
p
er
f
o
r
m
an
ce
is
m
u
ch
-
c
r
itical.
T
h
e
r
em
ain
in
g
s
ec
tio
n
s
o
f
th
e
m
an
u
s
cr
ip
t
h
av
e
b
ee
n
o
r
g
a
n
ized
as
f
o
llo
ws:
i)
Sectio
n
2
d
etails
th
e
p
r
o
p
o
s
e
d
m
eth
o
d
o
lo
g
y
,
in
clu
d
in
g
m
o
d
el
ar
ch
itectu
r
es,
o
p
er
atio
n
al
f
lo
w,
an
d
al
g
o
r
ith
m
s
elec
tio
n
.
ii)
Sectio
n
3
p
r
ese
n
ts
ex
p
er
im
en
tal
r
esu
lts
u
s
in
g
v
is
u
aliza
tio
n
p
lo
ts
an
d
a
c
o
m
p
ar
ativ
e
tab
le
o
f
ev
alu
atio
n
m
etr
ics.
iii)
Sectio
n
4
co
n
clu
d
es
t
h
e
s
tu
d
y
b
y
h
ig
h
lig
h
tin
g
th
e
ac
h
iev
e
d
r
esu
lts
,
p
r
ac
tical
r
elev
an
ce
,
a
n
d
d
ir
ec
tio
n
s
f
o
r
f
u
tu
r
e
wo
r
k
—
in
clu
d
in
g
r
ea
l
-
tim
e
I
o
T
in
teg
r
atio
n
an
d
m
o
d
el
g
e
n
er
aliza
tio
n
u
s
in
g
tr
an
s
f
er
lear
n
in
g
.
2.
P
RO
P
O
SE
D
CO
M
P
U
T
A
T
I
O
N
M
E
T
H
O
DO
L
O
G
Y
T
h
e
d
ec
lin
e
i
n
th
e
ca
p
ac
ity
o
f
b
atter
y
p
ac
k
s
a
n
d
d
im
in
is
h
in
g
p
o
wer
is
ca
u
s
ed
b
y
a
n
u
m
b
er
o
f
in
ter
r
elate
d
f
ac
to
r
s
,
r
en
d
e
r
in
g
th
e
in
v
esti
g
atio
n
d
if
f
icu
lt
an
d
d
iv
er
s
e.
Acr
o
s
s
th
e
ag
e
o
f
th
e
b
atter
y
p
ac
k
,
its
ab
ilit
y
to
o
p
er
ate
d
im
in
is
h
e
s
,
h
en
ce
,
co
r
r
ec
t
SOC
ca
lcu
latio
n
r
eq
u
ir
es
a
m
eth
o
d
o
l
o
g
y
th
at
tak
es
in
to
co
n
s
id
er
atio
n
o
f
d
eter
io
r
atio
n
o
f
th
e
b
atter
y
p
ac
k
.
Similar
ly
,
an
in
co
r
r
ec
t
SOC
esti
m
ate
m
i
g
h
t
ca
u
s
e
m
is
tak
es
in
ad
ju
s
tin
g
th
e
SOH
o
f
a
b
atter
y
p
ac
k
.
As
a
r
esu
lt,
co
n
tem
p
o
r
an
eo
u
s
ca
lcu
latio
n
o
f
SOC
an
d
SOH
is
wid
ely
ac
k
n
o
wled
g
e
d
as
cr
itical
f
o
r
s
u
cc
ess
f
u
l
b
atter
y
p
ac
k
a
d
m
in
is
tr
atio
n
f
ac
ilit
ies.
Desp
ite
th
e
f
ac
t
th
at
d
ee
p
lear
n
in
g
m
eth
o
d
o
lo
g
ies
h
a
v
e
d
em
o
n
s
tr
ated
f
av
o
r
ab
le
o
u
tco
m
es
in
th
e
c
o
m
p
u
tatio
n
o
f
SOC
an
d
ar
e
r
eg
u
lar
l
y
cited
in
liter
atu
r
e,
th
ese
ar
e
ty
p
ically
r
esear
ch
e
d
s
ep
ar
ately
,
with
n
o
co
m
p
lete
co
m
p
ar
is
o
n
s
tu
d
y
.
T
h
is
lim
its
th
eir
co
m
p
r
e
h
en
s
io
n
o
f
th
eir
p
ar
ticu
lar
ab
ilit
ies
an
d
co
n
s
tr
a
in
ts
in
p
ar
ticu
lar
u
s
e
s
ettin
g
s
,
wh
en
s
elec
tin
g
th
e
b
est
s
tr
ateg
y
is
cr
itical
f
o
r
p
ea
k
f
u
n
ctio
n
al
ab
ilit
y
.
T
h
e
in
v
es
tig
atio
n
p
r
o
v
id
es
an
in
-
d
ep
th
s
tr
ateg
y
,
co
m
p
ar
in
g
5
wid
ely
u
s
ed
m
eth
o
d
o
lo
g
ies
ac
r
o
s
s
th
e
id
en
tical
s
e
ttin
g
s
.
Ou
t
o
f
th
e
s
ev
en
m
eth
o
d
o
lo
g
ies
u
s
ed
,
we
u
s
e
f
iv
e
m
eth
o
d
o
l
o
g
ies f
o
r
SOC
co
m
p
u
tatio
n
,
an
d
we
u
s
e
f
iv
e
m
eth
o
d
o
lo
g
ies f
o
r
SOH
co
m
p
u
tatio
n
.
C
o
n
s
eq
u
en
tly
,
we
a
r
e
in
tr
o
d
u
cin
g
s
ev
en
s
tate
-
of
-
th
e
-
ar
t
m
e
th
o
d
o
lo
g
ies
with
th
e
two
d
if
f
er
en
t
aim
s
as
s
h
o
wn
in
F
ig
u
r
e
1
.
T
h
o
s
e
two
aim
s
ar
e
th
e
co
m
p
u
tatio
n
o
f
b
o
th
th
e
b
atter
y
p
ac
k
-
ce
n
tr
ic
m
ea
s
u
r
es
lik
e
SOC
an
d
SOH
b
y
s
eg
r
eg
atin
g
th
r
ee
m
eth
o
d
o
lo
g
ies
f
o
r
th
e
f
o
r
m
er
a
n
d
two
m
et
h
o
d
o
lo
g
i
es
f
o
r
th
e
latter
as
d
en
o
ted
in
F
ig
u
r
e
1
.
As
f
ar
as
th
e
d
ata
co
n
s
id
er
ed
is
co
n
ce
r
n
ed
,
we
s
ep
ar
ate
th
e
d
ata
ac
co
r
d
in
g
ly
to
co
m
p
u
te
b
atter
y
p
ac
k
-
ce
n
tr
ic
SOC
u
s
in
g
f
iv
e
m
eth
o
d
o
lo
g
ies
-
v
iz.
,
lin
ea
r
r
eg
r
ess
io
n
,
R
NN,
an
d
s
tack
ed
L
STM
,
GR
U,
an
d
GR
U
-
L
STM
GR
U,
an
d
G
R
U
-
L
STM
,
GR
U,
an
d
GR
U
-
L
STM
an
d
s
im
u
ltan
eo
u
s
ly
,
co
m
p
u
te
b
atter
y
p
ac
k
-
ce
n
tr
ic
SOH
u
s
in
g
f
iv
e
m
eth
o
d
o
lo
g
ies
v
iz.
,
s
tack
in
g
r
eg
r
ess
o
r
,
T
POT
r
eg
r
ess
o
r
,
lin
ea
r
r
eg
r
ess
io
n
,
GR
U,
an
d
GR
U
-
L
STM
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
9
2
I
n
t J Ap
p
l Po
wer
E
n
g
,
Vo
l.
1
5
,
No
.
3
,
Sep
tem
b
er
20
2
6
:
1
0
7
2
-
1
0
9
3
1078
T
h
e
n
o
tab
le
c
o
n
tr
ib
u
tio
n
s
o
f
o
u
r
wo
r
k
ar
e
as f
o
llo
ws:
-
B
y
in
teg
r
atin
g
a
r
an
g
e
o
f
m
o
d
els
—
lin
ea
r
r
eg
r
ess
io
n
f
o
r
in
ter
p
r
etab
ilit
y
,
R
NN
an
d
s
tack
ed
L
STM
f
o
r
tem
p
o
r
al
p
atter
n
lear
n
in
g
,
an
d
Au
to
ML
to
o
ls
lik
e
T
POT
an
d
s
tack
in
g
r
eg
r
ess
o
r
f
o
r
e
n
s
em
b
le
lear
n
in
g
—
we
o
f
f
er
a
d
iv
er
s
if
ied
m
o
d
elin
g
ap
p
r
o
ac
h
.
-
I
m
p
o
r
ta
n
tly
,
we
in
tr
o
d
u
ce
G
R
U
an
d
GR
U
-
L
STM
m
o
d
els
to
en
h
an
ce
th
e
f
r
am
ewo
r
k
'
s
ca
p
ac
ity
f
o
r
lear
n
in
g
b
o
t
h
s
h
o
r
t
-
an
d
lo
n
g
-
ter
m
d
ep
en
d
en
cies
in
b
atter
y
b
eh
av
io
r
.
GR
U
p
r
o
v
id
es
lig
h
tweig
h
t,
f
ast
-
co
n
v
er
g
in
g
p
r
ed
ictio
n
s
with
f
ewe
r
p
ar
am
eter
s
,
id
ea
l
f
o
r
r
e
al
-
tim
e
SOC
esti
m
atio
n
.
T
h
e
GR
U
-
L
STM
h
y
b
r
id
c
o
m
b
in
es
th
e
co
m
p
u
t
atio
n
al
ef
f
icien
cy
o
f
GR
U
with
th
e
d
ee
p
m
em
o
r
y
r
eten
tio
n
o
f
L
STM
,
o
f
f
er
in
g
r
o
b
u
s
t p
er
f
o
r
m
an
ce
f
o
r
b
o
t
h
SOC
an
d
SOH
task
s
.
-
Ou
r
ar
ch
itectu
r
e
m
ain
tain
s
a
m
o
d
u
lar
d
esig
n
th
at
s
ep
ar
at
es
SO
C
an
d
SOH
p
ip
elin
es
wh
ile
u
s
in
g
a
u
n
if
ied
p
r
ep
r
o
ce
s
s
in
g
an
d
d
ep
lo
y
m
en
t stru
ctu
r
e
—
an
ap
p
r
o
a
ch
r
ar
ely
s
ee
n
i
n
p
r
i
o
r
wo
r
k
s
.
T
h
is
co
m
p
r
eh
en
s
iv
e
in
teg
r
atio
n
o
f
tr
ad
itio
n
al,
ML
,
a
n
d
a
d
v
an
ce
d
d
e
ep
lear
n
in
g
m
o
d
e
ls
en
ab
les
g
r
ea
ter
f
lex
ib
ilit
y
,
ac
cu
r
ac
y
,
an
d
s
ca
lab
ilit
y
f
o
r
r
ea
l
-
wo
r
ld
b
atter
y
d
i
ag
n
o
s
tics
.
Fig
u
r
e
1
.
B
lo
ck
d
iag
r
am
o
f
c
o
m
p
u
tin
g
b
atter
y
SOC
an
d
SOH
with
s
tate
-
of
-
th
e
-
ar
t m
eth
o
d
o
lo
g
ies
2
.
1
.
O
bje
c
t
iv
es
I
n
o
u
r
wo
r
k
,
we
in
tr
o
d
u
ce
a
h
y
b
r
id
d
ee
p
lear
n
in
g
f
r
a
m
ewo
r
k
f
o
r
th
e
p
r
ec
is
e
co
m
p
u
tatio
n
o
f
m
ea
s
u
r
es
lik
e
SOC
an
d
SOH
in
th
e
b
atter
y
p
ac
k
s
m
ad
e
u
p
o
f
th
e
lith
iu
m
-
io
n
co
m
p
o
s
itio
n
.
Un
lik
e
t
r
ad
itio
n
al
m
o
d
el
-
b
ased
ap
p
r
o
ac
h
es,
th
is
s
y
s
tem
u
s
es
s
ep
ar
ate,
s
p
ec
ialized
m
ac
h
in
e
lear
n
in
g
m
o
d
els
tr
ain
ed
o
n
h
is
to
r
ical
d
atasets
(
f
etch
ed
f
r
o
m
a
s
in
g
le
s
o
u
r
ce
[
4
1
]
)
t
o
p
r
e
d
ict
two
m
ea
s
u
r
es
lik
e
SOC
an
d
SOH.
I
n
ca
s
e
o
f
SOC
p
r
ed
ictio
n
,
m
eth
o
d
o
lo
g
ies
lik
e
lin
ea
r
r
eg
r
ess
io
n
,
R
NN,
an
d
s
tack
ed
L
STM
ar
e
im
p
lem
en
ted
to
ca
p
tu
r
e
b
o
th
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ap
p
l Po
wer
E
n
g
I
SS
N:
2252
-
8
7
9
2
A
d
va
n
ce
d
A
I
-
d
r
iven
b
a
tter
y
s
ta
te
esti
ma
tio
n
u
s
in
g
d
ee
p
lea
r
n
in
g
a
n
d
…
(
N
.
S
u
ma
n
a
K
ee
r
th
i
)
1079
lin
ea
r
an
d
tem
p
o
r
al
r
elatio
n
s
h
ip
s
in
b
atter
y
ch
ar
g
e
p
atter
n
s
.
I
n
ca
s
e
o
f
SOH,
wh
ich
d
e
p
e
n
d
s
o
n
d
e
g
r
ad
atio
n
tr
en
d
s
o
v
er
tim
e,
en
s
em
b
le
-
b
ased
m
eth
o
d
o
lo
g
ies
s
u
ch
a
s
s
tack
in
g
r
eg
r
ess
o
r
an
d
T
POT
r
eg
r
ess
o
r
ar
e
em
p
lo
y
ed
to
e
n
s
u
r
e
r
o
b
u
s
t
g
en
er
aliza
tio
n
.
T
h
e
e
n
tire
b
ac
k
en
d
is
d
ev
elo
p
ed
u
s
in
g
Py
th
o
n
,
allo
win
g
in
teg
r
atio
n
o
f
p
r
ep
r
o
ce
s
s
ed
d
a
tasets
,
m
o
d
el
tr
ain
in
g
,
an
d
r
ea
l
-
tim
e
in
f
er
en
ce
API
s
.
T
h
e
f
r
o
n
ten
d
p
ar
t h
as b
ee
n
co
m
p
iled
with
th
e
d
ep
l
o
y
m
e
n
t
o
f
to
o
ls
lik
e
J
av
aScr
ip
t,
C
SS
,
an
d
HT
ML
,
p
r
o
v
i
d
in
g
a
clea
n
u
s
er
in
ter
f
ac
e
wh
er
e
u
s
er
s
ca
n
i
n
p
u
t
r
elev
a
n
t
s
en
s
o
r
d
ata
o
r
b
atter
y
p
ar
am
eter
s
.
T
h
e
s
y
s
tem
th
en
r
etu
r
n
s
r
ea
l
-
tim
e
SOC
an
d
SOH
v
alu
es.
T
h
is
m
o
d
u
lar
an
d
d
u
al
-
m
o
d
el
d
esig
n
en
s
u
r
e
s
s
ca
lab
ilit
y
,
in
ter
p
r
etab
ilit
y
,
an
d
ad
ap
tab
ilit
y
to
d
if
f
er
en
t
b
atter
y
ch
e
m
is
tr
ies
an
d
u
s
e
-
ca
s
es.
Ov
er
all,
it
en
h
a
n
ce
s
B
MS
an
d
en
er
g
y
s
to
r
a
g
e
r
eliab
ilit
y
in
E
Vs
,
d
r
o
n
es,
s
m
ar
t g
r
id
s
,
an
d
co
n
s
u
m
er
elec
tr
o
n
ics.
E
s
tim
atin
g
th
e
SOC
s
er
v
es
as
th
e
f
u
n
d
a
m
en
tal
r
ea
l
-
tim
e
i
n
d
icato
r
,
e
n
ab
lin
g
a
s
af
er
an
d
ef
f
icien
t
b
atter
y
o
p
er
atio
n
th
r
o
u
g
h
o
u
t
d
aily
u
s
e
cy
cles.
T
h
e
c
o
m
p
u
t
atio
n
o
f
SOC
is
cr
itical
f
o
r
im
m
ed
iate
o
p
er
atio
n
al
d
ec
is
io
n
s
b
ec
au
s
e
it
tells
u
s
er
s
h
o
w
m
u
ch
en
er
g
y
r
e
m
ain
s
av
ailab
le
at
an
y
g
iv
en
m
o
m
en
t.
T
h
is
p
r
ev
en
ts
o
v
er
-
d
is
ch
ar
g
e,
ca
u
s
in
g
p
e
r
m
an
en
t
d
am
ag
e
to
th
e
ce
lls
o
f
a
b
atter
y
p
ac
k
.
T
h
e
in
n
o
v
ativ
e
id
ea
b
eh
in
d
th
is
wo
r
k
is
to
d
eter
m
in
e
th
e
b
atter
y
p
ac
k
'
s
So
H
u
s
in
g
d
ata
ac
q
u
ir
ed
at
th
e
tim
e
o
f
th
e
co
u
r
s
e
o
f
ch
a
r
g
e
o
p
er
atio
n
.
T
h
e
en
d
in
g
o
f
a
ch
ar
g
in
g
p
r
o
ce
d
u
r
e
is
a
p
a
r
ticu
lar
ly
cr
itical
t
im
e
to
o
b
tain
So
H
d
ata.
W
ith
th
is
in
f
o
r
m
atio
n
,
co
n
s
u
m
er
s
ar
e
a
b
le
to
f
i
g
u
r
e
o
u
t
h
o
w
t
h
e
b
atter
y
p
ac
k
wo
u
ld
g
et
u
tili
ze
d
.
T
o
ac
co
m
p
lis
h
s
o
,
we
d
ep
lo
y
tw
o
m
eth
o
d
o
l
o
g
ies,
wh
ich
ar
e
tau
g
h
t a
n
d
ass
ess
ed
u
s
in
g
d
ata
co
ll
ec
ted
wh
ile
tak
in
g
c
h
ar
g
e.
2
.
2
.
O
utline
o
f
e
x
perim
ent
a
l
/
m
et
ho
d
s
ec
t
io
n
T
h
is
s
ec
tio
n
o
u
tlin
es
t
h
e
co
m
p
lete
ex
p
er
i
m
en
tal
d
esig
n
,
d
at
a
s
o
u
r
ce
s
,
p
r
ep
r
o
ce
s
s
in
g
s
tep
s
,
m
o
d
elin
g
tech
n
iq
u
es,
p
er
f
o
r
m
an
ce
ev
alu
atio
n
s
tr
ateg
ies,
an
d
web
-
b
ased
im
p
lem
e
n
tatio
n
em
p
lo
y
e
d
f
o
r
th
e
esti
m
atio
n
o
f
SOC
an
d
SOH
in
lith
iu
m
-
io
n
b
atter
y
p
ac
k
s
u
s
in
g
h
y
b
r
id
m
a
ch
in
e
lear
n
in
g
an
d
d
ee
p
lea
r
n
i
n
g
m
eth
o
d
s
.
2
.
2
.
1
.
Da
t
a
s
et
d
escript
io
n
T
wo
s
ep
ar
ate
d
atasets
wer
e
f
etch
ed
f
r
o
m
a
s
in
g
le
s
o
u
r
ce
[
4
1
]
an
d
s
u
b
s
eq
u
en
tly
u
s
ed
f
o
r
SOC
an
d
SOH
p
r
ed
ictio
n
task
s
:
-
T
h
e
f
o
llo
win
g
ar
e
th
e
k
ey
tr
ait
s
o
f
th
e
SOC
d
ataset:
i)
T
y
p
e:
Pu
b
licly
av
aila
b
le
E
V
b
atter
y
cy
clin
g
d
ataset.
ii)
Featu
r
es:
v
o
ltag
e
(
V)
,
c
u
r
r
en
t
(
A)
,
t
em
p
er
atu
r
e
(
°C
)
,
t
im
e
(
s
)
.
iii)
T
ar
g
et:
SOC
(
%),
u
p
d
ated
p
er
cy
cle.
-
T
h
e
f
o
llo
win
g
ar
e
th
e
k
ey
tr
ait
s
o
f
th
e
SOH
d
ataset:
i)
T
y
p
e:
NASA
b
atter
y
p
r
o
g
n
o
s
tics
d
ataset.
ii)
Featu
r
es: C
y
cle
n
u
m
b
er
,
ter
m
i
n
al
v
o
ltag
e,
c
h
ar
g
e
c
ap
ac
ity
,
t
em
p
er
atu
r
e,
an
d
i
n
ter
n
al
r
esis
tan
ce
.
iii)
T
ar
g
et:
SOH
(
%),
d
ef
in
ed
as th
e
r
atio
o
f
cu
r
r
en
t c
ap
ac
ity
t
o
r
ated
ca
p
ac
ity
.
No
te:
B
o
th
d
ataset
s
wer
e
clea
n
ed
an
d
f
o
r
m
atted
in
to
C
SV f
i
les an
d
wer
e
s
u
b
s
eq
u
en
tly
im
p
o
r
ted
u
s
in
g
Pan
d
as
(
v
1
.
5
.
3
)
in
Py
th
o
n
3
.
1
0
f
o
r
p
r
o
ce
s
s
in
g
.
2
.
2
.
2
.
Da
t
a
p
re
pro
ce
s
s
ing
T
h
e
p
r
e
p
r
o
ce
s
s
in
g
p
ip
elin
e
was im
p
lem
en
ted
as f
o
llo
ws:
-
Data
c
lean
in
g
:
First,
we
r
em
o
v
ed
r
o
ws
with
m
is
s
in
g
o
r
co
r
r
u
p
ted
s
en
s
o
r
r
ea
d
i
n
g
s
.
T
h
en
,
we
ex
clu
d
e
d
o
u
tlier
s
b
ey
o
n
d
3
s
tan
d
ar
d
d
e
v
iatio
n
s
u
s
in
g
Z
-
s
co
r
e
n
o
r
m
al
izatio
n
.
Fin
ally
,
we
f
ilter
ed
S
OH
d
ataset
to
r
etain
f
u
ll c
h
ar
g
e
-
d
is
ch
ar
g
e
cy
cles o
n
ly
.
-
Featu
r
e
s
ca
lin
g
:
Stan
d
ar
d
Scal
er
f
r
o
m
Scik
it
-
lear
n
was
u
s
ed
to
n
o
r
m
alize
all
co
n
tin
u
o
u
s
f
ea
tu
r
es
(
ze
r
o
m
ea
n
an
d
u
n
it v
a
r
ian
ce
)
b
ef
o
r
e
f
ee
d
in
g
i
n
to
th
e
m
o
d
els.
-
T
r
ain
-
test
s
p
lit
:
E
ac
h
d
ataset
was
s
p
lit
in
to
8
0
%
tr
ain
in
g
an
d
2
0
%
test
in
g
p
ar
titi
o
n
s
u
s
in
g
s
tr
atif
ied
s
am
p
lin
g
to
m
ain
tain
d
is
tr
ib
u
t
io
n
co
n
s
is
ten
cy
.
2
.
2
.
3
.
E
x
perim
ent
a
l
s
et
up
-
Har
d
war
e:
I
n
tel
C
o
r
e
i7
,
1
6
G
B
R
AM
,
5
1
2
GB
SS
D.
-
So
f
twar
e:
i)
Py
th
o
n
3
.
1
0
,
ii)
T
en
s
o
r
Flo
w
2
.
1
1
.
0
,
iii)
Scik
it
-
lear
n
1
.
2
.
2
,
iv
)
T
POT
0
.
1
1
,
v
)
Flas
k
2
.
2
.
2
,
v
i)
Pan
d
as,
Nu
m
Py
,
Ma
t
p
lo
tlib
,
Seab
o
r
n
.
2
.
3
.
Su
bs
t
a
ntia
l
c
o
ntr
ibu
t
io
ns
a
nd
n
o
v
elt
y
T
h
is
wo
r
k
in
tr
o
d
u
ce
s
s
ev
er
al
n
o
v
el
co
n
tr
ib
u
tio
n
s
th
at
s
et
it
ap
ar
t
f
r
o
m
ex
is
tin
g
liter
atu
r
e
i
n
th
e
f
ield
o
f
b
atter
y
d
iag
n
o
s
tics
.
2
.
3
.
1
.
Dua
l
-
m
o
del
a
rc
hite
ct
u
re
Un
lik
e
p
r
e
v
io
u
s
s
tu
d
ies
th
at
eith
er
f
o
cu
s
o
n
State
o
f
C
h
ar
g
e
(
SOC
)
o
r
State
o
f
He
alth
(
SOH)
in
d
ep
en
d
en
tly
,
o
r
attem
p
t
to
u
s
e
a
s
in
g
le
m
o
d
el
f
o
r
b
o
th
,
th
is
wo
r
k
ad
o
p
ts
a
m
o
d
u
lar
d
u
al
-
p
ip
elin
e
ar
ch
itectu
r
e.
SOC
an
d
SOH
ar
e
esti
m
ated
u
s
in
g
d
is
tin
ct,
s
p
ec
ialized
lear
n
in
g
alg
o
r
ith
m
s
tailo
r
ed
to
th
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
9
2
I
n
t J Ap
p
l Po
wer
E
n
g
,
Vo
l.
1
5
,
No
.
3
,
Sep
tem
b
er
20
2
6
:
1
0
7
2
-
1
0
9
3
1080
r
esp
ec
tiv
e
ch
ar
ac
ter
is
tics
o
f
ch
ar
g
e
b
e
h
av
io
r
a
n
d
d
e
g
r
ad
a
tio
n
p
atter
n
s
.
T
h
is
m
o
d
u
lar
d
esig
n
en
ab
les
ea
ch
p
r
ed
ictio
n
task
t
o
b
e
o
p
tim
iz
ed
in
d
ep
e
n
d
en
tl
y
,
r
esu
ltin
g
in
im
p
r
o
v
ed
esti
m
atio
n
ac
cu
r
ac
y
,
g
r
ea
te
r
m
o
d
el
f
lex
ib
ilit
y
,
an
d
e
n
h
an
ce
d
ad
a
p
tab
ilit
y
to
d
iv
er
s
e
b
atter
y
o
p
e
r
atin
g
co
n
d
itio
n
s
.
2
.
3
.
2
.
M
et
ho
do
lo
g
ica
l
i
nn
o
v
a
t
io
n
Fo
r
SOC
esti
m
atio
n
,
f
iv
e
co
m
p
ar
ativ
e
m
o
d
els
ar
e
ev
alu
ate
d
:
lin
ea
r
r
e
g
r
ess
io
n
(
as
a
b
aselin
e)
,
R
NN,
GR
U,
s
tack
ed
L
STM
,
an
d
a
h
y
b
r
i
d
GR
U
-
L
STM
m
o
d
e
l.
T
h
is
ex
p
an
d
ed
s
u
ite
o
f
m
o
d
els
en
s
u
r
es
th
e
f
r
am
ewo
r
k
ca
p
tu
r
es
a
wid
e
r
an
g
e
o
f
tem
p
o
r
al
d
ep
e
n
d
en
ci
es
—
f
r
o
m
s
h
o
r
t
b
u
r
s
ts
to
lo
n
g
-
r
an
g
e
p
atter
n
s
.
Fo
r
SOH
esti
m
at
io
n
,
two
en
s
em
b
l
e
-
b
ased
ap
p
r
o
ac
h
es
ar
e
im
p
le
m
en
ted
:
s
tack
in
g
r
eg
r
ess
o
r
an
d
T
POT
r
eg
r
ess
o
r
.
T
POT
lev
er
ag
es
Au
t
o
ML
v
ia
g
en
etic
p
r
o
g
r
am
m
in
g
to
au
t
o
m
atica
lly
co
n
s
tr
u
ct
an
d
o
p
ti
m
ize
p
ip
elin
es.
T
h
e
co
m
b
in
atio
n
o
f
m
a
n
u
ally
c
o
n
f
ig
u
r
e
d
tem
p
o
r
al
m
o
d
els
(
e.
g
.
,
GR
U,
L
STM
)
with
Au
to
ML
p
ip
elin
es
(
e.
g
.
,
T
POT)
in
a
u
n
i
f
ied
f
r
am
ewo
r
k
f
o
r
d
u
al
-
s
tate
esti
m
atio
n
r
ep
r
esen
ts
a
u
n
iq
u
e
co
n
t
r
ib
u
ti
o
n
n
o
t
p
r
ev
io
u
s
ly
ex
p
lo
r
ed
in
r
elate
d
w
o
r
k
.
2
.
3
.
3
.
P
er
f
o
r
m
a
nce
v
a
lid
a
t
io
n
All
m
o
d
els
wer
e
r
ig
o
r
o
u
s
ly
tr
ain
ed
an
d
test
ed
u
n
d
er
co
n
s
is
ten
t
co
n
d
itio
n
s
u
s
in
g
p
u
b
licly
av
ailab
le
r
ea
l
-
wo
r
ld
b
atter
y
d
atasets
,
an
d
ev
alu
ate
d
u
s
in
g
in
d
u
s
tr
y
-
s
ta
n
d
ar
d
m
etr
ics
—
R
²,
MA
E
,
MSE
,
an
d
R
MSE
.
T
h
e
GR
U
an
d
GR
U
-
L
STM
m
o
d
els
ac
h
iev
ed
ex
c
ellen
t
ac
cu
r
ac
y
f
o
r
SOC
esti
m
atio
n
,
with
th
e
GR
U
-
L
STM
m
o
d
el
m
atch
in
g
th
e
to
p
-
p
er
f
o
r
m
in
g
s
tack
ed
L
STM
(
R
²
=
0
.
9
9
3
5
,
R
MSE
=
0
.
0
1
4
7
)
.
T
h
e
T
POT
r
eg
r
ess
o
r
ac
h
ie
v
ed
p
er
f
ec
t p
r
e
d
ictio
n
f
o
r
SOH
(
R
²
=
1
.
0
)
,
estab
lis
h
in
g
a
n
ew
b
e
n
ch
m
ar
k
o
n
t
h
e
d
ataset.
2
.
3
.
4
.
Web
-
b
a
s
ed
d
eplo
y
m
ent
T
h
is
wo
r
k
in
clu
d
es
a
co
m
p
let
e
f
u
ll
-
s
tack
d
ep
lo
y
m
en
t,
co
n
s
is
tin
g
o
f
a
b
ac
k
en
d
p
o
wer
e
d
b
y
Py
th
o
n
(
Flas
k
)
an
d
a
f
r
o
n
ten
d
d
ev
elo
p
ed
u
s
in
g
HT
ML
,
C
SS
,
an
d
J
av
aScr
ip
t.
User
s
ca
n
in
ter
ac
t
with
th
e
s
y
s
tem
in
r
ea
l
tim
e
to
in
p
u
t
b
atter
y
d
a
ta
an
d
r
ec
eiv
e
in
s
tan
t
SOC
an
d
SOH
p
r
ed
ictio
n
s
,
t
h
er
eb
y
b
r
id
g
in
g
t
h
e
g
ap
b
etwe
en
m
o
d
el
d
ev
el
o
p
m
en
t
a
n
d
p
r
ac
tical
u
tili
ty
.
T
h
is
d
ep
lo
y
m
en
t
d
e
m
o
n
s
tr
ates
th
e
f
ea
s
ib
ilit
y
o
f
in
teg
r
atin
g
ad
v
an
ce
d
b
atter
y
esti
m
atio
n
m
o
d
els
in
to
u
s
er
-
f
r
ien
d
l
y
ap
p
l
icatio
n
s
,
m
ak
in
g
th
e
p
r
o
p
o
s
ed
f
r
am
ewo
r
k
s
u
itab
le
f
o
r
r
ea
l
-
tim
e
B
MS
ap
p
licatio
n
s
in
elec
tr
ic
v
eh
icles.
2
.
3
.
5
.
F
uture
-
r
ea
dy
d
esig
n
T
h
e
f
r
a
m
ewo
r
k
’
s
m
o
d
u
lar
s
t
r
u
ctu
r
e
f
ac
ilit
ates
s
ea
m
less
e
x
ten
s
io
n
s
,
in
clu
d
in
g
I
o
T
-
e
n
a
b
led
B
MS
in
teg
r
atio
n
,
clo
u
d
-
h
o
s
ted
an
al
y
tics
,
an
d
m
o
b
ile
d
ep
lo
y
m
en
t
.
L
ig
h
tweig
h
t
m
o
d
els,
s
u
ch
a
s
th
e
GR
U,
en
s
u
r
e
ef
f
icien
t
im
p
lem
en
tatio
n
o
n
ed
g
e
d
ev
ices
wh
ile
m
ain
tain
in
g
h
ig
h
p
r
ed
ictio
n
p
er
f
o
r
m
an
ce
.
Fu
r
th
er
m
o
r
e
,
s
ca
lab
le
Au
to
ML
co
m
p
o
n
en
t
s
p
r
ep
ar
e
th
e
s
y
s
tem
to
ac
co
m
m
o
d
ate
f
u
tu
r
e
d
atasets
an
d
ev
o
lv
in
g
b
atter
y
d
iag
n
o
s
tic
r
eq
u
ir
em
en
ts
,
th
er
e
b
y
en
h
an
cin
g
its
lo
n
g
-
ter
m
a
d
ap
tab
ilit
y
an
d
p
r
ac
tical
ap
p
lic
ab
ilit
y
.
2
.
4
.
G
o
a
ls
o
f
t
he
d
esig
n a
nd
d
ev
elo
pm
ent
a
pp
ro
a
ch
T
h
e
d
esig
n
an
d
d
ev
elo
p
m
en
t
o
f
th
e
p
r
o
p
o
s
ed
f
r
am
ewo
r
k
wer
e
g
u
id
ed
b
y
th
e
f
o
llo
win
g
co
r
e
o
b
jectiv
es:
-
Acc
u
r
ac
y
:
T
o
en
s
u
r
e
h
ig
h
p
r
ed
ictiv
e
p
er
f
o
r
m
a
n
ce
f
o
r
b
o
th
SOC
an
d
SOH
ac
r
o
s
s
d
i
v
er
s
e
b
atter
y
co
n
d
itio
n
s
,
lev
e
r
ag
in
g
r
o
b
u
s
t
m
o
d
elin
g
s
tr
ateg
ies
s
u
ch
as
GR
U,
GR
U
-
L
STM
,
an
d
s
tack
ed
L
STM
th
at
ca
n
ca
p
tu
r
e
b
o
th
s
h
o
r
t
-
ter
m
d
y
n
am
ics an
d
lo
n
g
-
ter
m
d
eg
r
a
d
a
tio
n
tr
en
d
s
.
-
Ad
ap
tab
ilit
y
:
T
o
b
u
ild
a
f
lex
i
b
le,
m
o
d
el
-
ag
n
o
s
tic
ar
ch
itectu
r
e
th
at
s
u
p
p
o
r
ts
q
u
ick
ad
ap
tati
o
n
to
d
if
f
er
en
t
b
atter
y
ch
em
is
tr
ies,
co
n
f
ig
u
r
a
tio
n
s
,
an
d
o
p
er
atio
n
al
en
v
ir
o
n
m
en
ts
—
en
ab
led
b
y
m
o
d
u
lar
p
ip
elin
es
an
d
tr
an
s
f
er
ab
le
lear
n
in
g
m
ec
h
an
is
m
s
.
-
Scalab
ilit
y
:
T
o
cr
ea
te
an
ef
f
ici
en
t
y
et
ex
ten
s
ib
le
f
r
am
ewo
r
k
,
wh
er
e
lig
h
tweig
h
t
m
o
d
els
lik
e
GR
U
ca
n
b
e
d
ep
lo
y
e
d
o
n
e
d
g
e
d
e
v
ices
f
o
r
r
ea
l
-
tim
e
SOC
est
im
atio
n
,
an
d
m
o
r
e
c
o
m
p
lex
m
o
d
els
lik
e
GR
U
-
L
STM
an
d
T
POT
ca
n
b
e
s
ca
led
to
ce
n
tr
alize
d
clo
u
d
s
y
s
tem
s
o
r
f
lee
t
-
lev
el
b
atter
y
m
o
n
ito
r
in
g
.
-
Usab
ilit
y
:
T
o
d
ev
elo
p
a
r
esp
o
n
s
iv
e
an
d
in
t
u
itiv
e
web
in
ter
f
ac
e
th
at
allo
ws
d
o
m
ain
ex
p
er
t
s
—
en
g
in
ee
r
s
,
r
esear
ch
er
s
,
o
r
tech
n
ician
s
—
to
in
p
u
t
b
atter
y
p
ar
am
ete
r
s
a
n
d
r
etr
ie
v
e
SOC
/SOH
p
r
ed
ictio
n
s
with
o
u
t
n
ee
d
in
g
ML
ex
p
er
tis
e.
-
Au
to
m
atio
n
:
T
o
in
te
g
r
ate
A
u
to
ML
v
ia
T
POT
f
o
r
d
y
n
a
m
ic
m
o
d
el
d
is
co
v
e
r
y
an
d
tu
n
in
g
,
r
e
d
u
cin
g
m
an
u
al
co
n
f
ig
u
r
atio
n
a
n
d
e
n
ab
lin
g
th
e
s
y
s
tem
to
co
n
tin
u
o
u
s
ly
ev
o
lv
e
a
n
d
im
p
r
o
v
e
as
m
o
r
e
d
ata
b
ec
o
m
es a
v
ailab
le.
2
.
5
.
M
et
ho
do
lo
g
ies
Her
e,
we
b
r
ief
ly
ex
p
lain
th
e
s
ev
en
m
eth
o
d
o
lo
g
ies
u
s
ed
in
o
u
r
wo
r
k
co
n
ce
r
n
e
d
with
co
m
p
u
tin
g
b
o
th
th
e
b
atter
y
p
ac
k
’
s
SOC
an
d
S
OH.
T
h
e
f
o
llo
win
g
s
ev
en
m
eth
o
d
o
lo
g
ies
wer
e
u
s
ed
ac
r
o
s
s
th
e
two
tar
g
et
task
s
,
an
d
th
eir
r
esp
ec
tiv
e
h
y
p
er
p
ar
a
m
eter
s
wer
e
eith
er
tu
n
e
d
m
an
u
ally
o
r
o
p
tim
ized
u
s
in
g
T
POT.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ap
p
l Po
wer
E
n
g
I
SS
N:
2252
-
8
7
9
2
A
d
va
n
ce
d
A
I
-
d
r
iven
b
a
tter
y
s
ta
te
esti
ma
tio
n
u
s
in
g
d
ee
p
lea
r
n
in
g
a
n
d
…
(
N
.
S
u
ma
n
a
K
ee
r
th
i
)
1081
2
.
5
.
1
.
L
inea
r
r
eg
re
s
s
io
n
L
in
ea
r
r
eg
r
ess
io
n
is
a
f
o
u
n
d
at
io
n
al
s
u
p
er
v
is
ed
lea
r
n
in
g
m
et
h
o
d
o
lo
g
y
u
s
ed
to
m
o
d
el
th
e
a
s
s
o
ciatio
n
b
etwe
en
o
n
e
o
r
in
d
e
p
en
d
e
n
t
p
ar
am
eter
s
an
d
a
co
n
tin
u
o
u
s
d
e
p
en
d
en
t
v
ar
iab
le.
I
n
s
im
p
le
lin
ea
r
r
eg
r
ess
io
n
,
th
e
r
elatio
n
s
h
ip
is
r
ep
r
esen
ted
b
y
(
1
)
.
=
₀
+
₁
+
(
1
)
W
h
er
e
x
d
en
o
tes
th
e
in
p
u
t
f
ea
tu
r
e
(
i.e
.
,
v
o
ltag
e,
cu
r
r
e
n
t)
,
y
d
en
o
tes
th
e
f
o
r
ec
asted
r
esu
lta
n
t
(
i.e
.
,
SOC
)
,
β₀
is
th
e
in
ter
ce
p
t,
β₁ is
th
e
s
lo
p
e
(
c
o
ef
f
icien
t)
,
a
n
d
ε
is
th
e
er
r
o
r
t
er
m
.
I
t
is
a
k
ey
ap
p
r
o
ac
h
in
th
e
a
r
ea
o
f
m
ac
h
in
e
lea
r
n
in
g
.
L
in
ea
r
r
eg
r
ess
io
n
is
a
f
o
r
m
o
f
s
u
p
er
v
is
ed
lear
n
in
g
in
wh
ic
h
th
e
f
r
a
m
ewo
r
k
ex
am
i
n
es
tr
ain
ed
d
ata
wit
h
lab
els
(
s
u
ch
as
d
ata
lo
g
s
)
to
d
etec
t
tr
en
d
s
b
ef
o
r
e
u
s
in
g
th
o
s
e
tr
en
d
s
to
g
en
e
r
a
te
f
o
r
ec
asts
ab
o
u
t
f
r
esh
d
ata
.
T
h
e
wo
r
d
"lin
ea
r
"
i
n
d
icate
s
th
e
s
tr
aig
h
t
p
ath
s
tr
u
ctu
r
e
o
f
th
e
m
o
d
eled
c
o
n
n
ec
tio
n
,
wh
er
ea
s
th
e
wo
r
d
"
r
eg
r
ess
io
n
"
in
d
icate
s
th
e
p
r
o
ce
d
u
r
e
o
f
ca
lcu
latin
g
th
e
m
an
n
er
in
wh
ich
p
a
r
am
eter
s
r
etu
r
n
to
th
ei
r
m
ea
n
r
ea
d
in
g
s
.
T
h
e
m
o
d
el
wo
r
k
s
b
y
r
e
d
u
cin
g
th
e
m
ea
n
s
q
u
ar
ed
er
r
o
r
(
M
SE)
b
etwe
en
th
e
ac
tu
ally
ex
is
tin
g
v
alu
es
an
d
p
r
e
d
icted
v
al
u
es
u
s
in
g
m
eth
o
d
s
s
u
ch
as
o
r
d
in
ar
y
least
s
q
u
ar
es
(
OL
S).
Fo
r
b
atter
y
S
OC
esti
m
atio
n
,
th
is
m
eth
o
d
o
l
o
g
y
p
r
esu
m
es
a
li
n
e
ar
ly
-
ex
is
tin
g
ass
o
ciatio
n
am
o
n
g
s
t
s
en
s
o
r
r
ea
d
i
n
g
s
an
d
ch
a
r
g
e
lev
el,
wh
ich
is
v
alid
in
s
tead
y
-
s
tate
r
eg
io
n
s
o
f
th
e
ch
ar
g
e
-
d
is
ch
ar
g
e
cu
r
v
e.
T
h
o
u
g
h
s
im
p
le
a
n
d
in
t
er
p
r
etab
le,
lin
ea
r
r
eg
r
ess
io
n
ca
n
n
o
t
m
o
d
el
n
o
n
-
lin
ea
r
o
r
tim
e
-
d
ep
en
d
en
t
b
a
tter
y
b
eh
av
i
o
r
s
.
Ho
wev
er
,
it’
s
o
f
ten
u
s
ed
as
a
b
aselin
e
o
r
f
o
r
s
y
s
tem
s
wh
er
e
q
u
ick
,
r
ea
l
-
tim
e
esti
m
atio
n
s
ar
e
m
o
r
e
im
p
o
r
tan
t
t
h
an
u
ltra
-
h
ig
h
ac
cu
r
ac
y
.
I
t
’
s
f
ast,
lig
h
tweig
h
t,
an
d
r
e
q
u
ir
es
v
er
y
litt
le
p
r
ep
r
o
ce
s
s
in
g
,
m
a
k
in
g
it
s
u
itab
le
f
o
r
in
itial
m
o
d
el
d
ep
lo
y
m
en
t
o
r
in
teg
r
atio
n
in
to
r
eso
u
r
ce
-
co
n
s
t
r
ain
ed
s
y
s
tem
s
.
I
n
s
im
p
ler
ter
m
s
,
it
id
en
tifi
es
tr
en
d
s
in
d
ata
b
y
id
e
n
tif
y
in
g
lin
k
s
am
o
n
g
p
ar
a
m
eter
s
.
As
an
illu
s
tr
atio
n
,
a
b
u
s
in
ess
th
at
d
ea
ls
with
r
ea
l
estate
m
ay
u
tili
ze
ass
et
s
ca
le
f
o
r
p
r
ed
ictio
n
o
f
h
o
u
s
e
v
al
u
es,
wh
er
ea
s
a
m
er
ch
an
d
is
er
m
ay
em
p
lo
y
p
r
e
v
io
u
s
d
ata
o
n
s
ales f
o
r
p
r
ed
ictio
n
o
f
ea
r
n
i
n
g
s
f
o
r
th
e
f
u
tu
r
e.
T
o
m
a
k
e
ex
tr
em
ely
r
eliab
le
f
o
r
ec
asts
,
th
e
ap
p
r
o
ac
h
esti
m
ates sim
u
ltan
eo
u
s
ly
th
e
in
te
r
ce
p
t
an
d
th
e
s
lo
p
e.
W
e
ap
p
ly
li
n
ea
r
r
eg
r
ess
io
n
f
o
r
th
e
SOC
esti
m
atio
n
b
y
lev
er
a
g
in
g
th
e
r
elatio
n
s
h
i
p
s
b
etwe
en
m
ea
s
u
r
ab
le
b
atter
y
p
a
r
am
eter
s
an
d
ac
tu
al
ch
ar
g
e
le
v
els.
T
h
is
lear
n
in
g
ap
p
r
o
ac
h
is
b
ein
g
tr
a
in
ed
u
p
o
n
d
iv
er
s
e
r
an
g
e
o
f
h
is
to
r
ical
d
ata
lik
e
v
o
ltag
e
m
ea
s
u
r
em
en
ts
,
c
u
r
r
en
t
p
atter
n
s
,
tem
p
er
at
u
r
e
r
ea
d
in
g
s
,
an
d
th
eir
co
r
r
esp
o
n
d
in
g
k
n
o
wn
SOC
v
a
lu
es.
W
ith
th
e
id
en
tific
atio
n
o
f
lin
ea
r
tr
en
d
s
b
etwe
en
th
e
i
n
p
u
t
p
ar
am
eter
s
an
d
ch
ar
g
e
s
tate,
th
e
r
eg
r
ess
io
n
m
o
d
el
ca
n
estab
lis
h
m
ath
em
atica
l
r
elatio
n
s
h
ip
s
f
o
r
p
r
ed
ictin
g
SOC
f
o
r
s
u
b
s
eq
u
en
t n
ew
d
ata.
W
e
ap
p
ly
lin
ea
r
r
eg
r
ess
io
n
f
o
r
th
e
SOH
esti
m
atio
n
with
t
h
e
ca
p
tu
r
e
o
f
g
r
a
d
u
al
d
e
g
r
ad
atio
n
p
atter
n
s
o
win
g
to
b
atter
ies
ag
in
g
.
T
h
e
m
eth
o
d
ca
n
an
aly
ze
lo
n
g
-
t
er
m
tr
en
d
s
in
p
er
f
o
r
m
an
ce
in
d
icato
r
s
s
u
ch
as
ca
p
ac
ity
f
ad
e
,
in
ter
n
al
r
esis
tan
ce
ch
an
g
es,
an
d
v
o
ltag
e
r
es
p
o
n
s
e
ch
a
r
ac
ter
is
tics
.
T
h
en
,
it
ca
n
c
o
r
r
elate
th
o
s
e
in
d
icato
r
s
with
k
n
o
wn
SOH
b
en
ch
m
ar
k
s
ag
ain
s
t
th
e
test
in
g
o
r
s
p
ec
if
icatio
n
s
.
W
ith
th
is
SOH
esti
m
atio
n
,
th
e
o
v
er
all
h
ea
lth
c
o
n
d
itio
n
an
d
r
e
m
ain
in
g
u
s
ef
u
l lif
e
ca
n
b
e
f
o
r
e
ca
s
ted
.
T
h
e
cr
itical
f
ac
ts
o
f
lin
ea
r
r
eg
r
ess
io
n
ar
e
as
f
o
llo
ws:
i)
T
o
o
l:
Scik
it
-
lear
n
’
s
L
in
ea
r
R
eg
r
ess
io
n
()
,
ii)
e
q
u
atio
n
:
SOC
=
β₀
+
β₁x
+
ε
,
iii)
p
u
r
p
o
s
e:
B
aselin
e
co
m
p
ar
is
o
n
f
o
r
SOC
p
r
ed
ictio
n
u
s
in
g
a
lin
ea
r
m
o
d
el
,
iv
)
l
im
itatio
n
: I
n
ab
ilit
y
to
ca
p
t
u
r
e
n
o
n
-
lin
ea
r
ity
in
b
atter
y
b
e
h
av
i
o
r
.
2
.
5
.
2
.
RNN
R
NNs
ar
e
a
ty
p
e
o
f
n
eu
r
al
s
y
s
tem
th
at
is
cr
ea
ted
ex
clu
s
i
v
ely
to
s
er
v
e
d
ata
th
at
is
s
e
q
u
en
tial
in
n
atu
r
e.
On
th
e
co
n
tr
ar
y
t
o
s
tan
d
ar
d
f
ee
d
f
o
r
war
d
n
eu
r
al
s
y
s
tem
s
,
th
is
m
eth
o
d
o
lo
g
y
f
ea
t
u
r
es
lo
o
p
s
,
allo
win
g
d
ata
to
b
e
s
av
ed
.
At
ea
ch
d
u
r
atio
n
s
tag
e,
an
R
NN
ac
q
u
ir
es
an
in
p
u
t
v
ec
to
r
alo
n
g
with
t
h
e
h
id
d
e
n
lev
el
f
r
o
m
th
e
p
r
ec
ed
ed
s
tag
e,
p
r
o
d
u
cin
g
a
n
ew
h
id
d
en
le
v
el
an
d
p
o
s
s
ib
ly
an
o
u
tp
u
t.
Ma
th
em
atica
lly
,
th
e
r
elatio
n
s
h
ip
is
ex
p
r
ess
ed
in
(
2
)
.
ℎₜ
=
(
ₓ
ₜ
+
ₕℎₜ₋₁
+
)
(
2
)
T
h
is
d
esig
n
en
ab
les
R
NNs
to
"r
em
em
b
er
"
p
ast
in
f
o
r
m
atio
n
,
m
ak
in
g
th
em
id
ea
l
f
o
r
m
o
d
e
lin
g
tim
e
-
s
er
ies
d
ata
lik
e
b
atter
y
s
en
s
o
r
r
ea
d
in
g
s
o
v
er
tim
e.
Ho
wev
er
,
b
en
ch
m
a
r
k
R
NNs
ex
p
er
ien
ce
d
im
in
is
h
in
g
o
r
b
lo
win
g
u
p
g
r
ad
ie
n
t
p
r
o
b
l
em
s
,
esp
ec
ially
f
o
r
lo
n
g
s
eq
u
en
ce
s
,
wh
ic
h
h
a
m
p
er
s
l
ea
r
n
in
g
lo
n
g
-
ter
m
d
ep
en
d
e
n
cies.
Fo
r
b
atter
y
SOC
p
r
ed
ictio
n
,
R
NNs
ar
e
p
o
w
er
f
u
l
b
ec
au
s
e
th
e
y
co
n
s
id
er
p
r
ev
io
u
s
tim
e
s
tep
s
wh
ile
esti
m
atin
g
th
e
cu
r
r
en
t
c
h
ar
g
e
le
v
el.
Fo
r
ex
am
p
le,
if
a
b
atter
y
h
as
b
ee
n
d
is
ch
ar
g
in
g
s
tead
ily
f
o
r
th
e
last
f
ew
m
in
u
tes,
th
at
tr
en
d
is
ca
p
t
u
r
ed
an
d
u
s
ed
to
r
ef
in
e
th
e
cu
r
r
en
t SOC
esti
m
ate.
I
n
th
is
wo
r
k
,
R
NN
is
u
s
ed
to
m
o
d
el
th
e
tem
p
o
r
al
b
eh
av
io
r
o
f
b
atter
ies
d
u
r
in
g
ch
ar
g
in
g
an
d
d
is
ch
ar
g
in
g
c
y
cles.
T
h
is
in
clu
s
io
n
ca
n
aid
th
e
ca
p
tu
r
e
o
f
c
er
tain
p
atter
n
s
lik
e
v
o
ltag
e
d
r
o
p
r
ates
o
r
cu
r
r
en
t
v
ar
iatio
n
s
co
r
r
esp
o
n
d
i
n
g
t
o
th
e
SOC
-
ce
n
tr
ic
ch
an
g
es.
Desp
i
te
th
e
f
ac
t
th
at
R
NNs
ca
n
tak
e
m
o
r
e
tim
e
to
tr
ai
n
an
d
p
r
ep
r
o
ce
s
s
th
e
d
ata,
it
ca
n
s
till
b
o
o
s
t
th
e
p
r
ed
ictio
n
ac
cu
r
ac
y
in
d
y
n
am
ic
en
v
ir
o
n
m
en
ts
o
win
g
to
its
Evaluation Warning : The document was created with Spire.PDF for Python.