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
.
1
,
Ma
r
ch
20
2
6
,
p
p
.
2
5
7
~
2
7
4
I
SS
N:
2252
-
8
7
9
2
,
DOI
:
1
0
.
1
1
5
9
1
/ijap
e
.
v
1
5
.
i
1
.
pp
257
-
274
257
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//
ija
p
e.
ia
esco
r
e.
co
m/
M
a
chine learning
-
driv
en prog
no
sti
cs for
l
i
thium
-
io
n
ba
tt
eries
:
e
nha
ncing
RUL
p
rediction a
nd
per
forma
nce in
sm
a
r
t
energy
sto
ra
g
e sy
stems
B
o
da
pa
t
i V
enk
a
t
a
Ra
j
a
nn
a
1
,
Aa
luri See
nu
2
,
K
o
nd
ra
g
un
t
a
Ra
ma
K
rish
na
ia
h
3
,
Ana
ntha
Sra
v
a
nthi P
eddin
t
i
4
,
Nela
t
uri N
a
nd
a
P
ra
ka
s
h
5
,
B
a
nd
re
d
di Ve
nk
a
t
a
Ses
hu
k
um
a
ri
6
,
G
iripra
s
a
d Am
ba
t
i
7
,
Sh
a
ik
H
a
s
a
ne
Aha
m
ma
d
8
,
Cha
k
ra
pa
ni Sriv
a
rdha
n K
um
a
r
9
,
Alla
m
ra
j
u Shu
bh
a
ng
i R
a
o
10
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
,
M
LR
I
n
st
i
t
u
t
e
o
f
Te
c
h
n
o
l
o
g
y
,
H
y
d
e
r
a
b
a
d
,
I
n
d
i
a
2
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
r
i
n
g
,
S
h
r
i
V
i
s
h
n
u
E
n
g
i
n
e
e
r
i
n
g
C
o
l
l
e
g
e
f
o
r
W
o
me
n
,
B
h
i
ma
v
a
r
a
m
,
I
n
d
i
a
3
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
r
i
n
g
,
R
K
C
o
l
l
e
g
e
o
f
E
n
g
i
n
e
e
r
i
n
g
,
K
e
t
h
a
n
a
k
o
n
d
a
,
I
n
d
i
a
4
D
e
p
a
r
t
me
n
t
o
f
El
e
c
t
r
o
n
i
c
s a
n
d
C
o
m
mu
n
i
c
a
t
i
o
n
En
g
i
n
e
e
r
i
n
g
,
A
d
i
t
y
a
U
n
i
v
e
r
si
t
y
,
S
u
r
a
m
p
a
l
e
m,
I
n
d
i
a
5
D
e
p
a
r
t
me
n
t
o
f
El
e
c
t
r
o
n
i
c
s a
n
d
C
o
m
mu
n
i
c
a
t
i
o
n
En
g
i
n
e
e
r
i
n
g
,
C
h
a
l
a
p
a
t
h
i
I
n
st
i
t
u
t
e
o
f
E
n
g
i
n
e
e
r
i
n
g
a
n
d
Te
c
h
n
o
l
o
g
y
,
G
u
n
t
u
r
,
I
n
d
i
a
6
D
e
p
a
r
t
me
n
t
o
f
I
n
f
o
r
mat
i
o
n
Te
c
h
n
o
l
o
g
y
,
V
N
R
V
i
g
n
a
n
a
J
y
o
t
h
i
I
n
st
i
t
u
t
e
o
f
En
g
i
n
e
e
r
i
n
g
a
n
d
Te
c
h
n
o
l
o
g
y
,
H
y
d
e
r
a
b
a
d
,
I
n
d
i
a
7
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
,
V
N
R
V
i
g
n
a
n
a
J
y
o
t
h
i
I
n
st
i
t
u
t
e
o
f
E
n
g
i
n
e
e
r
i
n
g
a
n
d
Te
c
h
n
o
l
o
g
y
,
H
y
d
e
r
a
b
a
d
,
I
n
d
i
a
8
D
e
p
a
r
t
me
n
t
o
f
El
e
c
t
r
o
n
i
c
s a
n
d
C
o
m
mu
n
i
c
a
t
i
o
n
En
g
i
n
e
e
r
i
n
g
,
K
o
n
e
r
u
La
k
s
h
ma
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
9
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
c
h
o
o
l
o
f
E
n
g
i
n
e
e
r
i
n
g
,
M
a
l
l
a
R
e
d
d
y
U
n
i
v
e
r
si
t
y
,
H
y
d
e
r
a
b
a
d
,
I
n
d
i
a
10
D
e
p
a
r
t
me
n
t
o
f
E
l
e
c
t
r
i
c
a
l
a
n
d
E
l
e
c
t
r
o
n
i
c
s E
n
g
i
n
e
e
r
i
n
g
,
P
r
e
si
d
e
n
c
y
U
n
i
v
e
r
si
t
y
,
B
a
n
g
a
l
o
r
e
,
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
Ap
r
1
0
,
2
0
2
5
R
ev
is
ed
Sep
2
5
,
2
0
2
5
Acc
ep
ted
No
v
2
8
,
2
0
2
5
In
t
h
e
e
v
o
lv
i
n
g
lan
d
sc
a
p
e
o
f
e
n
e
r
g
y
sy
ste
m
s,
b
a
tt
e
ries
p
lay
a
c
rit
ic
a
l
ro
le
i
n
e
n
a
b
li
n
g
h
y
b
ri
d
a
n
d
sta
n
d
-
a
l
o
n
e
r
e
n
e
wa
b
le
e
n
e
rg
y
st
o
ra
g
e
so
lu
ti
o
n
s
.
P
re
c
ise
l
y
e
stim
a
ti
n
g
b
a
tt
e
ry
li
fe
a
n
d
re
m
a
in
i
n
g
u
se
fu
l
o
p
e
ra
ti
o
n
a
l
li
fe
will
g
o
a
lo
n
g
wa
y
in
e
n
h
a
n
c
i
n
g
t
h
e
e
fficie
n
c
y
o
f
t
h
e
sy
ste
m
with
a
ss
u
re
d
re
li
a
b
il
it
y
in
sm
a
rt
p
o
we
r
st
o
ra
g
e
d
e
v
ice
s.
T
h
is
r
e
p
o
rt
c
o
m
p
re
h
e
n
si
v
e
ly
su
r
v
e
y
s
a
d
v
a
n
c
e
d
a
p
p
ro
a
c
h
e
s
i
n
t
h
e
m
a
n
a
g
e
m
e
n
t
o
f
b
a
tt
e
ries
t
h
ro
u
g
h
sta
te
-
of
-
t
h
e
-
a
rt
a
rti
ficia
l
in
telli
g
e
n
c
e
to
o
ls
-
su
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
s,
re
lev
a
n
c
e
v
e
c
to
r
m
a
c
h
in
e
s
(RVM),
lo
n
g
sh
o
rt
-
term
m
e
m
o
ry
(LS
TM
)
m
o
d
e
ls,
a
n
d
b
a
y
e
sia
n
f
il
ters
-
th
a
t
a
re
b
e
in
g
u
se
d
wi
th
a
v
iew
t
o
e
n
h
a
n
c
in
g
re
m
a
in
in
g
u
se
fu
l
li
fe
(RUL)
e
stim
a
tes
a
n
d
m
a
k
i
n
g
re
a
l
-
ti
m
e
sy
ste
m
h
e
a
lt
h
m
o
n
it
o
ri
n
g
c
a
p
a
b
il
it
ies
p
o
ss
ib
le
.
M
o
d
e
li
n
g
a
p
p
ro
a
c
h
e
s
su
r
v
e
y
e
d
in
c
lu
d
e
sta
te
e
stim
a
ti
o
n
,
c
a
p
a
c
it
y
,
a
n
d
t
h
e
rm
a
l
m
a
n
a
g
e
m
e
n
t,
wh
il
e
d
isc
u
ss
in
g
t
h
e
ir
a
p
p
li
c
a
b
il
it
y
to
li
th
i
u
m
-
io
n
b
a
t
terie
s.
Th
e
re
v
iew
a
lso
e
x
p
l
o
re
s
p
u
b
li
c
ly
a
v
a
il
a
b
le
b
a
tt
e
r
y
d
a
tas
e
ts,
fe
a
tu
re
e
n
g
i
n
e
e
rin
g
stra
teg
ies
,
a
n
d
h
y
b
rid
d
ia
g
n
o
stic
f
ra
m
e
wo
r
k
s.
A
tec
h
n
o
e
c
o
n
o
m
ic p
e
rsp
e
c
ti
v
e
is
p
ro
v
i
d
e
d
to
a
ss
e
ss
sy
ste
m
p
e
rfo
rm
a
n
c
e
in
re
n
e
wa
b
le
-
in
teg
ra
t
e
d
p
o
we
r
g
rid
s.
T
h
i
s
p
a
p
e
r
a
i
m
s
t
o
c
o
n
s
o
l
i
d
a
t
e
c
u
r
re
n
t
k
n
o
w
l
e
d
g
e
,
p
r
o
v
i
d
e
c
o
m
p
a
r
a
t
i
v
e
i
n
s
i
g
h
t
s
i
n
t
o
t
h
e
s
t
r
e
n
g
t
h
s
a
n
d
l
i
m
i
t
a
t
i
o
n
s
o
f
d
i
f
f
e
r
e
n
t
a
p
p
r
o
a
c
h
e
s
,
a
n
d
h
i
g
h
l
i
g
h
t
o
p
e
n
r
e
s
e
a
r
c
h
c
h
a
l
le
n
g
e
s
t
o
g
u
i
d
e
f
u
t
u
r
e
d
e
v
e
l
o
p
m
e
n
t
s
i
n
s
m
a
r
t
A
I
-
e
n
a
b
l
e
d
b
a
t
t
e
r
y
s
y
s
t
e
m
s
t
h
a
t
s
u
p
p
o
r
t
s
u
s
t
a
in
a
b
l
e
a
n
d
r
e
s
i
l
i
e
n
t
e
n
e
r
g
y
i
n
f
r
a
s
t
r
u
c
t
u
r
e
.
K
ey
w
o
r
d
s
:
B
atter
y
m
an
ag
em
en
t sy
s
tem
L
ith
iu
m
-
io
n
b
atter
ies
Ma
ch
in
e
lear
n
in
g
Pro
g
n
o
s
tics
an
d
h
ea
lth
m
an
ag
em
en
t
R
em
ain
in
g
u
s
ef
u
l lif
e
State
o
f
ch
ar
g
e
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
:
B
o
d
ap
ati
Ven
k
ata
R
ajan
n
a
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
,
ML
R
I
n
s
itu
te
o
f
T
ec
h
n
o
lo
g
y
Hy
d
er
ab
ad
,
I
n
d
ia
E
m
ail:
r
ajan
n
ab
v
2
0
1
2
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
Du
e
to
d
ec
lin
e
in
f
o
s
s
il
f
u
els,
t
h
er
e
is
n
ee
d
f
o
r
u
s
ag
e
o
f
m
o
r
e
r
en
ewa
b
le
en
er
g
y
r
eso
u
r
ce
s
with
s
to
r
ag
e
d
ev
ices
to
m
ee
t
th
e
lo
ad
d
em
an
d
d
u
r
i
n
g
p
ea
k
a
n
d
o
f
f
-
p
ea
k
tim
es.
Hen
ce
,
b
atter
ies
ar
e
u
s
ed
f
o
r
s
to
r
in
g
th
e
ex
ce
s
s
en
er
g
y
s
u
p
p
lied
b
y
r
e
n
ewa
b
le
s
o
u
r
ce
s
in
h
y
b
r
id
s
y
s
tem
s
[
1
]
.
B
atter
ies
w
ith
h
ig
h
en
er
g
y
d
e
n
s
ity
,
f
ast
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
.
1
,
Ma
r
ch
20
2
6
:
257
-
2
7
4
258
ch
ar
g
i
n
g
,
wi
d
e
o
p
e
r
a
ti
n
g
t
em
p
er
at
u
r
e
r
a
n
g
e
,
l
o
w
s
el
f
-
d
is
c
h
a
r
g
i
n
g
r
a
te
,
an
d
li
f
e
c
y
cle
l
o
n
g
e
v
ity
c
a
n
b
e
e
m
p
l
o
y
e
d
in
e
lec
tr
ic
v
e
h
i
cles
(
E
Vs
)
,
p
o
r
ta
b
l
e
el
ec
t
r
o
n
ics
,
b
io
m
e
d
i
ca
l
d
e
v
i
ce
s
,
a
n
d
f
o
r
a
p
p
lic
ati
o
n
s
i
n
i
n
d
u
s
t
r
i
al
a
n
d
tr
a
n
s
p
o
r
tati
o
n
s
ec
t
o
r
s
wi
th
r
is
i
n
g
c
o
n
ce
r
n
s
i
n
cli
m
a
te
ch
an
g
e,
en
er
g
y
s
ec
u
r
it
y
,
a
n
d
s
u
s
tai
n
a
b
le
d
e
v
e
lo
p
m
e
n
t
[
2
]
.
L
ith
iu
m
-
io
n
b
atter
ies
ar
e
wid
e
ly
u
s
ed
in
in
d
u
s
tr
y
d
u
e
to
th
eir
ex
ten
d
ed
life
s
p
an
,
h
ig
h
en
er
g
y
d
en
s
ity
,
a
n
d
s
m
all
s
ize,
h
o
wev
er
th
ey
also
h
av
e
d
is
ad
v
an
tag
es
lik
e
d
eter
io
r
atio
n
an
d
s
af
ety
co
n
ce
r
n
s
.
R
ec
en
t
s
tu
d
ies
h
av
e
s
tr
ess
ed
th
e
n
ee
d
o
f
ac
cu
r
ate
R
UL
esti
m
ates in
ap
p
licatio
n
s
s
u
ch
as
elec
tr
ic
v
eh
icles a
n
d
r
e
n
ewa
b
l
e
en
er
g
y
s
y
s
tem
s
.
Me
th
o
d
s
o
f
b
atter
y
life
m
o
d
el
in
g
,
h
ea
lth
m
o
n
ito
r
in
g
,
an
d
R
UL
f
o
r
ec
asti
n
g
ar
e
d
is
cu
s
s
ed
in
ter
m
s
o
f
s
tatis
t
ical,
elec
tr
o
ch
em
ical,
an
d
m
ac
h
i
n
e
lear
n
in
g
[
3
]
.
T
h
is
r
ev
iew
ad
d
r
ess
es
b
atter
y
s
tate
e
s
tim
atio
n
,
d
ata
s
ets,
an
d
r
esid
u
al
u
s
ef
u
l
life
p
r
ed
ictio
n
s
m
eth
o
d
s
;
d
ev
el
o
p
in
g
b
atter
y
m
an
a
g
em
en
t
s
y
s
tem
s
(
B
MS)
an
d
R
UL
p
r
ed
ictio
n
m
eth
o
d
s
f
o
r
e
n
h
a
n
cin
g
th
e
u
n
d
er
s
tan
d
in
g
o
f
b
atter
y
ag
in
g
p
atter
n
s
in
ac
tu
al
o
p
er
atio
n
.
Data
ac
q
u
is
itio
n
an
d
co
m
p
r
eh
e
n
s
iv
e
Li
-
io
n
b
atter
y
d
ata
r
eso
u
r
ce
s
f
o
cu
s
es
o
n
th
e
co
llectio
n
o
f
b
atter
y
d
ata
f
o
r
th
e
p
u
r
p
o
s
e
o
f
cr
ea
tin
g
a
m
ac
h
in
e
lear
n
in
g
-
b
ased
p
r
o
g
n
o
s
tics
an
d
h
ea
lth
m
an
a
g
em
en
t
s
y
s
tem
s
m
o
d
el
r
ev
o
lv
i
n
g
ar
o
u
n
d
p
u
b
lic
d
ata
r
e
p
o
s
ito
r
ie
s
[
4
]
.
Hea
lth
m
o
n
ito
r
in
g
a
n
d
f
ea
tu
r
e
ex
tr
ac
tio
n
in
v
o
l
v
es
m
u
ch
m
o
r
e
th
an
co
llectin
g
d
ata,
as
it
b
o
ls
ter
s
th
e
f
u
n
ctio
n
ality
an
d
p
r
ec
is
io
n
o
f
m
ac
h
in
e
lear
n
in
g
m
o
d
els
b
y
p
ar
s
in
g
r
aw
d
ata
an
d
r
ef
in
in
g
t
h
e
cr
iticality
o
f
th
e
ex
tr
ac
ted
h
ea
lth
in
d
icato
r
s
.
Ma
n
y
r
esear
ch
er
s
h
av
e
atte
m
p
ted
to
d
ev
el
o
p
d
ep
e
n
d
ab
l
e
m
eth
o
d
s
f
o
r
h
e
alth
i
n
d
i
c
a
t
o
r
d
e
te
c
t
i
o
n
t
o
es
ti
m
a
t
e
t
h
e
r
e
m
ai
n
i
n
g
u
s
e
f
u
l
li
f
e
(
R
UL
)
o
f
L
i
-
i
o
n
b
at
t
e
r
y
c
e
l
ls
a
c
c
u
r
a
t
e
l
y
[
5
]
-
[
7
]
.
T
h
is
v
a
l
u
e
r
e
f
l
e
c
ts
t
h
e
t
o
t
a
l
n
u
m
b
e
r
o
f
c
h
a
r
g
e
-
d
i
s
c
h
a
r
g
e
c
y
c
l
es
a
b
at
t
e
r
y
c
e
ll
c
a
n
s
u
s
t
ai
n
b
e
f
o
r
e
r
e
q
u
i
r
i
n
g
r
e
p
l
a
c
e
m
e
n
t
.
T
h
e
r
em
ain
d
er
o
f
th
is
p
ap
e
r
i
s
o
r
g
an
ized
as
f
o
llo
ws:
Sectio
n
2
r
ev
iews
B
MS
p
ar
am
eter
esti
m
atio
n
m
eth
o
d
s
,
s
ec
tio
n
3
d
is
cu
s
s
es
R
UL
p
r
ed
ictio
n
a
p
p
r
o
a
ch
es,
s
ec
tio
n
4
p
r
esen
ts
in
tellig
en
t
ML
t
ec
h
n
iq
u
es
f
o
r
R
UL
an
d
SOH,
s
ec
tio
n
5
in
clu
d
es
d
ataset
r
eso
u
r
ce
s
a
n
d
f
ea
tu
r
e
en
g
in
ee
r
in
g
s
tr
ateg
ies,
s
ec
t
io
n
6
s
y
n
th
esizes
co
m
p
ar
ativ
e
in
s
ig
h
ts
an
d
cr
iti
ca
l
s
y
n
th
esis
,
s
ec
tio
n
7
an
aly
ze
s
p
er
f
o
r
m
an
ce
o
p
tim
izatio
n
s
tr
ateg
ies
in
B
MS,
s
ec
tio
n
8
o
u
tlin
es
f
u
tu
r
e
r
esear
ch
d
ir
ec
tio
n
s
an
d
o
p
e
n
ch
allen
g
es,
an
d
s
ec
tio
n
9
in
clu
d
es
co
n
clu
s
io
n
an
d
f
u
tu
r
e
wo
r
k
o
f
th
e
p
a
p
er
.
Fu
r
th
er
m
o
r
e,
th
is
p
a
p
er
r
ev
iews
s
o
m
e
e
x
is
tin
g
tech
n
iq
u
es
an
d
p
r
o
v
id
es
a
co
m
p
a
r
is
o
n
o
f
d
if
f
er
en
t
m
ac
h
in
e
lear
n
in
g
m
eth
o
d
o
l
o
g
ies
b
ased
o
n
ad
v
an
tag
es,
d
is
ad
v
an
ta
g
es,
an
d
is
s
u
es
th
at
r
em
ain
u
n
s
o
lv
ed
,
t
h
u
s
en
co
u
r
ag
in
g
f
u
r
th
er
co
n
tr
ib
u
tio
n
s
in
AI
f
ield
s
.
2.
B
AT
T
E
RY
M
ANAG
E
M
E
N
T
SYS
T
E
M
(
B
M
S)
P
ARA
M
E
T
E
R
E
ST
I
M
A
T
I
O
N
B
e
c
a
u
s
e
o
f
t
h
e
c
o
m
p
l
e
x
d
e
s
i
g
n
a
n
d
n
o
n
l
i
n
e
a
r
b
e
h
a
v
i
o
r
o
f
t
h
e
s
e
c
e
l
l
s
,
s
o
p
h
i
s
t
i
c
a
t
e
d
a
l
g
o
r
i
t
h
m
i
c
t
e
c
h
n
i
q
u
e
s
ar
e
n
ec
ess
ar
y
in
lith
iu
m
-
i
o
n
b
atter
y
m
an
a
g
em
en
t sy
s
tem
s
en
h
an
ce
m
en
t.
C
alcu
latin
g
SOC
an
d
SOH
with
g
o
o
d
ac
cu
r
ac
y
is
g
r
ea
tly
im
p
o
r
tan
t
in
ef
f
ec
tiv
e
B
MS
o
p
er
atio
n
s
.
T
h
e
cu
r
r
en
t
s
eg
m
en
t
s
tar
ts
b
y
in
tr
o
d
u
cin
g
b
asic
id
ea
s
r
eg
ar
d
in
g
SOC
an
d
SOH
esti
m
atio
n
s
an
d
later
p
r
o
v
id
es
an
in
s
ig
h
t
in
to
s
o
m
e
s
tate
-
of
-
th
e
-
ar
t
d
ev
el
o
p
m
en
ts
b
ein
g
m
ad
e
u
s
in
g
m
ac
h
in
e
lea
r
n
in
g
tech
n
iq
u
es
to
g
eth
e
r
with
ad
v
an
ce
d
s
en
s
o
r
in
n
o
v
atio
n
s
.
T
h
e
em
p
h
asis
is
o
n
esti
m
atin
g
SO
C
an
d
SOH,
h
i
g
h
lig
h
tin
g
an
y
s
ig
n
if
ican
t
ch
a
llen
g
es
an
d
p
o
s
s
ib
le
ad
v
an
tag
es
id
en
tifie
d
f
o
r
an
in
f
o
r
m
ativ
e
u
n
d
e
r
s
tan
d
in
g
o
f
h
o
w
th
ese
co
m
p
o
n
en
ts
en
h
an
c
e
b
atter
y
ef
f
icie
n
cy
,
s
af
ety
,
an
d
life
s
p
an
.
2
.
1
.
E
x
a
m
ini
ng
t
he
SO
C
a
nd
SO
H
estim
a
t
e
T
h
e
p
o
p
u
lar
SOC
esti
m
atio
n
m
eth
o
d
s
in
clu
d
e
lo
o
k
u
p
t
ab
les,
co
u
lo
m
b
c
o
u
n
tin
g
,
th
e
ar
tific
ial
in
tellig
en
ce
alg
o
r
ith
m
s
o
f
n
eu
r
al
n
ets,
s
tati
s
tical
lear
n
in
g
th
r
o
u
g
h
SVMs,
elec
tr
ical
s
im
u
l
atio
n
s
u
s
in
g
E
L
Ms,
p
h
y
s
ical
th
eo
r
ies
u
tili
ze
d
in
E
MF
m
o
d
els,
th
e
ex
p
er
im
en
t
al
m
eth
o
d
o
lo
g
ies
o
f
I
C
A
an
d
DVA
an
aly
s
es,
an
d
m
ac
h
in
e
-
lear
n
i
n
g
-
b
ased
s
tr
ateg
ies
in
v
o
lv
in
g
d
atasets
to
m
a
k
e
p
r
ed
ictio
n
s
[
8
]
.
State
-
of
-
c
h
ar
g
e
an
d
s
tate
-
of
-
h
ea
lth
esti
m
atio
n
b
ased
o
n
th
e
n
ewe
s
t
m
eth
o
d
o
lo
g
y
ap
p
lie
s
ac
o
u
s
tic
-
u
ltra
s
o
n
ic
s
tr
ess
w
av
es
in
co
m
b
in
at
io
n
with
p
iezo
elec
tr
ic
s
en
s
o
r
s
an
d
s
tr
ain
g
au
g
es
in
o
r
d
e
r
to
c
lo
s
ely
m
o
n
ito
r
an
d
u
n
d
e
r
s
tan
d
th
e
in
ter
r
elatio
n
b
etwe
en
th
ese
two
p
ar
am
eter
s
b
etter
.
T
h
e
d
eg
r
ad
atio
n
m
e
ch
an
is
m
s
af
f
ec
t
th
e
life
s
p
an
an
d
th
e
r
eten
tio
n
ca
p
ac
ity
o
f
th
e
b
atter
y
;
in
tr
i
n
s
ic
q
u
alitie
s
ar
e
r
ed
u
ce
d
b
y
ac
tiv
e
m
ater
ial
l
o
s
s
,
ag
in
g
o
f
th
e
elec
tr
ical
co
n
d
u
cta
n
ce
,
an
d
ex
h
a
u
s
ted
lith
iu
m
in
v
en
t
o
r
y
,
p
r
o
b
ab
ly
in
itiatin
g
d
en
d
r
itic
g
r
o
wth
th
at
m
ay
lead
to
s
elf
-
d
is
ch
ar
g
e
[
9
]
.
Dif
f
er
e
n
tial a
n
a
ly
s
is
,
v
o
ltag
e
f
itti
n
g
alg
o
r
ith
m
s
,
an
d
AI
f
o
r
s
en
s
o
r
d
ata
in
ter
p
r
etatio
n
a
r
e
am
o
n
g
th
e
v
ar
io
u
s
tech
n
iq
u
es
ap
p
lied
to
E
-
So
H,
wh
ic
h
p
r
o
v
id
es
im
p
o
r
tan
t
in
f
o
r
m
atio
n
ab
o
u
t
b
atter
y
d
eg
r
ad
atio
n
wh
ile
en
h
an
cin
g
p
r
e
d
ictiv
e
m
ain
ten
an
ce
in
f
u
tu
r
e
b
atter
y
c
o
n
tr
o
l m
ec
h
a
n
is
m
s
.
2
.
2
.
SO
P
,
SO
E
,
a
nd
SO
T
est
im
a
t
io
n
r
e
v
i
e
w
State
o
f
ch
a
r
g
e
(
SOC
)
,
s
tate
o
f
h
ea
lth
(
SOH)
,
s
tate
o
f
p
o
wer
(
SOP),
s
tate
o
f
en
e
r
g
y
(
SOE)
,
an
d
s
tate
o
f
tim
e
(
SOT)
s
h
o
u
ld
all
b
e
esti
m
ated
b
y
a
th
o
r
o
u
g
h
B
MS
[
1
0
]
.
SOP
s
tan
d
s
f
o
r
u
s
ea
b
le
p
o
wer
,
SOE
f
o
r
a
v
a
i
l
a
b
l
e
e
n
e
r
g
y
,
a
n
d
SO
T
e
s
t
im
a
t
i
o
n
w
h
i
c
h
is
r
e
l
at
e
d
t
o
b
a
t
te
r
y
t
e
m
p
e
r
a
t
u
r
e
is
l
es
s
s
t
u
d
i
e
d
.
W
h
i
l
e
m
u
l
t
i
-
s
t
at
e
j
o
i
n
t
esti
m
atin
g
is
s
til
l a
cr
u
cial
r
esear
ch
to
p
ic,
h
y
b
r
id
esti
m
atio
n
s
th
at
in
clu
d
e
SOE
an
d
SOP h
av
e
b
ee
n
e
x
am
in
ed
.
2
.
3
.
SO
S
estim
a
t
io
n
r
e
v
i
e
w
W
ith
s
af
ety
-
o
r
ien
ted
s
tate
(
SO
S)
esti
m
atio
n
im
p
r
o
v
in
g
e
f
f
ici
en
cy
an
d
r
eliab
ilit
y
,
s
y
s
tem
s
af
ety
is
n
o
w
cr
u
cial
in
b
atter
y
m
an
a
g
em
e
n
t
s
y
s
tem
s
(
B
MS)
.
R
i
s
k
s
in
clu
d
in
g
f
ir
es,
e
x
p
lo
s
io
n
s
,
an
d
e
lectr
o
ly
te
leak
s
ar
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
Ma
ch
in
e
lea
r
n
in
g
-
d
r
iven
p
r
o
g
n
o
s
tics
fo
r
lith
iu
m
-
io
n
b
a
tter
ies:
e
n
h
a
n
cin
g
…
(
B
o
d
a
p
a
ti V
e
n
ka
ta
R
a
j
a
n
n
a
)
259
ad
d
r
ess
ed
b
y
im
p
r
o
v
em
e
n
ts
in
SOS
tech
n
iq
u
es,
s
u
ch
as
th
er
m
al
r
u
n
awa
y
ass
ess
m
en
ts
.
I
n
ter
n
al
s
h
o
r
t
cir
cu
it
test
s
u
s
ed
in
th
er
m
al
r
u
n
awa
y
in
v
esti
g
atio
n
s
id
en
tify
im
p
ac
ts
an
d
elec
tr
ical
f
ailu
r
es
as
th
e
p
r
im
ar
y
ca
u
s
es
o
f
th
er
m
al
r
u
n
awa
y
.
B
etter
s
af
ety
an
d
r
elia
b
ilit
y
p
er
f
o
r
m
an
ce
ar
e
m
ad
e
p
o
s
s
ib
le
b
y
th
e
co
m
b
in
atio
n
o
f
SOS
with
SOC
,
SO
H,
SOP,
S
OT
,
an
d
SOE.
T
h
e
p
r
im
ar
y
co
n
tr
ib
u
tin
g
f
ac
to
r
s
ar
e
tem
p
er
atu
r
e,
m
ec
h
an
ical
d
ef
o
r
m
atio
n
,
v
o
ltag
e,
cu
r
r
en
t,
a
n
d
tem
p
er
at
u
r
e.
˗
T
em
p
er
atu
r
e:
B
ec
au
s
e
h
ig
h
t
em
p
er
atu
r
es
in
cr
ea
s
e
th
e
r
is
k
o
f
th
er
m
al
r
u
n
awa
y
in
ad
d
i
tio
n
to
f
ir
es
an
d
ex
p
lo
s
io
n
s
,
th
ey
h
ar
m
o
th
er
m
ater
ials
as
well
as
th
e
s
o
lid
elec
tr
o
ly
te
in
ter
p
h
ase
(
SEI
)
lay
er
.
L
o
w
tem
p
er
atu
r
es
ca
u
s
e
lith
iu
m
to
ac
cu
m
u
late
o
n
n
eg
ativ
e
elec
tr
o
d
es,
lead
in
g
to
in
ter
n
al
s
h
o
r
t c
i
r
cu
its
an
d
r
e
d
u
ce
d
ca
p
ac
ity
.
˗
C
u
r
r
en
t:
H
ig
h
cu
r
r
en
t
ca
u
s
es
J
o
u
le
h
ea
tin
g
i
n
th
e
b
atter
y
,
wh
i
ch
,
if
im
p
r
o
p
er
m
an
ag
e
m
en
t
s
tr
ateg
ies
ar
e
u
s
ed
,
ca
n
lead
to
t
h
er
m
al
r
u
n
awa
y
.
L
ith
iu
m
p
latin
g
an
d
in
te
r
n
al
s
h
o
r
t
cir
c
u
it
g
e
n
er
atio
n
ar
e
b
o
t
h
m
o
r
e
lik
ely
to
o
cc
u
r
d
u
r
in
g
th
e
ch
a
r
g
in
g
p
r
o
ce
s
s
.
˗
Vo
ltag
e:
H
ea
t
an
d
f
u
m
es
ar
e
p
r
o
d
u
ce
d
wh
en
th
e
elec
tr
o
ly
te
an
d
p
o
s
itiv
e
elec
tr
o
d
e
ar
e
o
v
e
r
ch
ar
g
e
d
,
ca
u
s
in
g
d
am
ag
e
t
o
b
o
th
.
T
h
e
p
r
o
ce
s
s
o
f
d
ee
p
d
is
ch
ar
g
in
g
ce
lls
r
aises
th
e
r
is
k
o
f
s
h
o
r
t
cir
c
u
its
b
y
en
c
o
u
r
ag
i
n
g
t
h
e
g
r
o
wth
o
f
co
p
p
er
d
en
d
r
ites
.
˗
State
of
ch
ar
g
e
(
SOC
):
W
h
en
a
b
atter
y
f
ails
,
h
ig
h
SOC
v
alu
es
in
cr
ea
s
e
th
e
lik
elih
o
o
d
o
f
h
ea
t
-
r
elate
d
en
er
g
y
r
elea
s
es.
˗
State
o
f
h
ea
lth
(
SOH)
:
Alth
o
u
g
h
a
g
in
g
p
r
o
ce
s
s
es
ca
u
s
e
s
tr
u
ctu
r
al
f
ailu
r
e
an
d
lith
i
u
m
p
latin
g
cir
c
u
m
s
tan
ce
s
th
at
in
cr
ea
s
e
f
ailu
r
e
p
o
ten
tial,
o
ld
er
b
atter
ies'
in
itial lo
wer
ch
ar
g
e
r
ed
u
ce
s
th
eir
h
az
ar
d
.
˗
E
lectr
o
m
ec
h
an
ical
im
b
ala
n
ce
:
T
h
e
elec
tr
o
m
ec
h
a
n
ical
im
b
ala
n
ce
o
cc
u
r
s
b
ec
a
u
s
e
o
f
th
e
c
h
a
n
g
e
in
p
r
ess
u
r
e
o
f
th
e
b
atter
ies d
u
e
to
d
if
f
er
e
n
t c
h
ar
g
e
cy
cles,
wh
ic
h
r
esu
lts
in
d
ec
r
ea
s
in
g
p
er
f
o
r
m
a
n
ce
an
d
s
af
ety
.
˗
I
n
ter
n
al
i
m
p
e
d
an
ce
:
T
h
e
in
c
r
e
asin
g
th
ick
n
ess
o
f
th
e
SEI
is
o
n
e
o
f
th
e
f
ac
t
o
r
s
th
at
co
n
tr
ib
u
t
es
to
th
e
in
ter
n
al
r
esis
tan
ce
with
in
b
atter
ies,
d
ec
r
ea
s
in
g
th
eir
en
er
g
y
s
to
r
a
g
e
ca
p
ac
ity
an
d
g
en
er
al
p
er
f
o
r
m
a
n
ce
.
T
h
is
p
r
o
ce
s
s
am
p
lifie
s
io
n
lo
s
s
with
in
b
a
tter
ies,
lo
wer
in
g
t
h
eir
s
to
r
ag
e
ca
p
ab
ilit
y
an
d
ef
f
icien
cy
l
ev
els.
T
h
e
m
o
s
t
s
ig
n
if
ican
t
ca
u
s
es
o
f
m
alf
u
n
c
tio
n
in
th
e
L
I
B
s
y
s
tem
s
ar
e
o
v
er
h
ea
tin
g
,
m
alf
u
n
ctio
n
o
f
e
lectr
o
n
ic
s
tab
ilit
y
co
n
tr
o
l
s
y
s
tem
s
,
an
d
b
r
ea
k
d
o
wn
o
f
in
s
u
latio
n
.
A
n
u
m
b
er
o
f
s
tate
-
of
-
th
e
-
a
r
t
m
eth
o
d
o
l
o
g
i
es
will
h
av
e
to
b
e
r
esear
ch
ed
,
d
iag
n
o
s
ed
,
a
n
d
s
im
u
lated
in
em
er
g
en
cy
r
esp
o
n
s
es
to
en
h
an
ce
s
ec
u
r
ity
a
n
d
lo
n
g
ev
ity
.
As
e
n
er
g
y
s
to
r
ag
e
d
em
a
n
d
s
co
n
tin
u
e
t
o
in
cr
ea
s
e,
b
atter
ies
will
en
s
u
r
e
en
h
a
n
ce
d
s
af
ety
an
d
ef
f
ici
en
cy
th
r
o
u
g
h
th
e
d
ev
elo
p
m
e
n
t o
f
p
r
o
t
o
co
ls
as c
h
an
g
in
g
cir
cu
m
s
tan
ce
s
ar
e
m
o
n
ito
r
ed
.
3.
AP
P
RO
ACH
E
S F
O
R
P
RE
D
I
CT
I
NG
RE
M
A
I
NING
U
SE
F
UL
L
I
F
E
(
RU
L
)
E
s
tim
atin
g
h
o
w
m
u
ch
lo
n
g
e
r
lith
iu
m
-
io
n
b
atter
ies
will
wo
r
k
s
af
ely
with
o
u
t
r
ep
lace
m
en
t
d
ef
in
es
a
s
ig
n
if
ican
t
r
o
le
in
r
is
k
m
itig
ati
o
n
d
u
r
in
g
u
s
ag
e,
r
e
d
u
ce
s
m
ai
n
ten
an
ce
co
s
ts
,
an
d
e
n
h
an
ce
s
o
v
er
all
p
er
f
o
r
m
a
n
ce
d
u
r
in
g
o
p
er
atio
n
s
.
T
h
is
s
ec
tio
n
ad
o
p
ts
a
s
tr
u
ctu
r
e
d
a
p
p
r
o
ac
h
b
y
f
ir
s
t
r
ev
iewin
g
t
h
e
b
asic
th
eo
r
ies
u
n
d
er
l
y
in
g
th
e
m
o
d
elin
g
o
f
r
eliab
ilit
y
u
n
d
er
u
n
ce
r
tain
ty
,
in
clu
d
i
n
g
m
o
d
el
-
b
ased
,
d
ata
-
d
r
iv
en
,
an
d
h
y
b
r
id
a
p
p
r
o
ac
h
es.
T
h
en
,
it
id
e
n
tifie
s
co
n
tem
p
o
r
ar
y
r
esear
ch
in
t
h
is
ar
ea
b
y
co
v
e
r
in
g
m
o
d
er
n
a
p
p
r
o
ac
h
es
s
u
ch
a
s
d
y
n
am
ic
f
ilter
in
g
tech
n
iq
u
es,
p
r
o
b
ab
ilit
y
m
o
d
el
ap
p
r
o
ac
h
es,
an
d
co
g
n
itiv
e
co
m
p
u
tin
g
-
b
ased
ap
p
r
o
ac
h
es.
T
h
is
s
ec
tio
n
r
ev
iews
th
e
g
ap
s
in
t
h
e
c
u
r
r
en
t
tech
n
i
q
u
es a
n
d
s
tates th
e
u
n
s
o
lv
ed
p
r
o
b
lem
s
th
at
n
ee
d
to
b
e
o
v
e
r
co
m
e
to
h
av
e
p
r
ac
tical
ap
p
licatio
n
s
in
r
ea
l
-
life
s
itu
at
io
n
s
.
Pre
cise
esti
m
atio
n
o
f
a
lith
iu
m
-
io
n
b
atter
y
s
y
s
tem
p
e
r
f
o
r
m
a
n
ce
in
v
o
lv
es
s
tate
-
of
-
h
ea
lth
an
d
R
UL
ass
es
s
m
en
t
th
at
h
elp
s
o
r
ig
i
n
al
eq
u
ip
m
en
t
m
an
u
f
ac
tu
r
er
s
d
ec
i
d
e
o
n
tim
ely
r
ep
lace
m
en
t
cy
cles
f
o
r
ass
o
ciate
d
d
ata.
T
h
e
ac
cu
r
ate
esti
m
ates
av
o
id
m
a
lf
u
n
ctio
n
in
g
o
f
ce
ll
p
h
o
n
es
an
d
im
p
r
o
v
e
o
v
er
all
d
ev
ice
p
e
r
f
o
r
m
an
ce
.
Var
io
u
s
p
r
ed
ictiv
e
m
o
d
elin
g
tec
h
n
i
q
u
es
ar
e
ad
o
p
ted
f
o
r
r
eso
u
r
ce
u
tili
za
tio
n
lev
el
esti
m
atio
n
,
in
clu
d
in
g
b
u
t
n
o
t
lim
ited
to
m
eth
o
d
o
lo
g
ical
f
r
a
m
ewo
r
k
s
lik
e
th
eo
r
y
-
b
ased
a
n
d
em
p
ir
ical
m
o
d
el
ap
p
r
o
ac
h
es d
ev
elo
p
ed
d
ir
ec
tly
f
r
o
m
d
atasets
,
an
d
h
y
b
r
id
tec
h
n
iq
u
es th
at
m
e
r
g
e
elem
en
ts
o
f
ea
ch
ca
teg
o
r
y
.
3
.
1
.
M
o
del
-
ba
s
ed
a
pp
ro
a
ches
T
h
e
m
o
d
el
-
d
r
iv
e
n
ap
p
r
o
ac
h
r
elies
o
n
th
e
u
s
e
o
f
eq
u
atio
n
s
m
o
d
elin
g
b
atter
y
b
e
h
av
io
r
to
p
r
ed
ict
th
e
r
em
ain
in
g
o
p
e
r
atio
n
al
life
tim
e.
T
h
ese
m
o
d
els
a
r
e
b
ased
eit
h
er
o
n
m
ec
h
an
ical
laws
o
r
ch
em
ical
r
ea
ctio
n
s
t
o
m
o
d
el
s
y
s
tem
wea
r
an
d
d
eg
r
a
d
atio
n
p
h
en
o
m
en
a,
wh
ic
h
d
en
o
te
p
er
f
o
r
m
an
ce
lo
s
s
.
T
h
e
u
p
d
ate
o
f
s
tate
v
ar
iab
les
with
s
en
s
o
r
in
f
o
r
m
atio
n
in
cr
ea
s
es
k
n
o
wled
g
e
o
f
s
y
s
tem
d
eg
r
ad
atio
n
b
y
an
al
y
s
is
.
Ho
wev
er
,
u
s
in
g
m
o
d
els
r
eq
u
ir
es
th
e
s
im
u
latio
n
o
f
co
m
p
lex
n
o
n
lin
ea
r
ch
em
ical
p
r
o
ce
s
s
es
q
u
ite
ac
cu
r
ately
,
wh
i
ch
is
o
f
ten
d
if
f
icu
lt
o
win
g
to
p
r
ac
tical
is
s
u
es.
W
h
en
th
er
e
is
a
d
is
cr
ep
an
cy
b
etwe
en
wh
at
th
e
m
o
d
el
p
r
ed
icts
ab
o
u
t
b
atter
ies
an
d
th
e
ac
tu
al
p
er
f
o
r
m
an
ce
t
h
at
lead
s
to
in
co
r
r
ec
t f
o
r
ec
asts
.
3
.
2
.
Da
t
a
-
driv
en
a
pp
ro
a
che
s
T
h
e
p
r
ed
ictio
n
o
f
r
em
ain
i
n
g
u
s
ef
u
l
life
,
u
s
in
g
s
tatis
tical
tech
n
iq
u
es
d
er
iv
ed
f
r
o
m
lar
g
e
d
atas
ets
o
f
p
ast
b
icy
cle
u
s
ag
e
p
atter
n
s
,
is
ac
h
i
ev
ed
b
y
u
s
in
g
ad
v
an
ce
d
ar
tific
ial
in
tellig
en
ce
m
o
d
els.
Me
th
o
d
s
s
u
ch
as a
r
tific
ial
in
tellig
en
ce
m
o
d
els,
wh
ich
ar
e
tr
ain
ed
o
n
d
ata,
in
clu
d
e
alg
o
r
ith
m
s
f
o
r
r
ec
o
g
n
izin
g
tr
en
d
s
in
m
ea
s
u
r
em
en
ts
o
f
elec
tr
ical
s
ig
n
als
th
at
in
clu
d
e
am
p
litu
d
e
an
d
f
r
eq
u
e
n
cy
a
n
d
th
e
r
m
al
c
o
n
d
itio
n
s
.
T
h
es
e
m
eth
o
d
s
id
en
tif
y
co
m
p
lex
n
o
n
lin
ea
r
p
h
en
o
m
e
n
a
with
o
u
t
r
eq
u
ir
in
g
a
n
ex
te
n
s
iv
e
u
n
d
er
s
tan
d
i
n
g
o
f
th
e
u
n
d
e
r
ly
in
g
ch
em
ical
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
.
1
,
Ma
r
ch
20
2
6
:
257
-
2
7
4
260
p
r
o
ce
s
s
es.
Su
cc
ess
cr
itically
r
elies
o
n
a
v
ailab
le
d
ata
s
ets
th
at
ar
e
a
d
eq
u
ate
an
d
d
iv
e
r
s
e.
L
ac
k
o
f
g
en
e
r
al
ap
p
licab
ilit
y
to
wid
e
r
an
g
es
o
f
ch
em
ical
s
y
s
tem
s
,
o
p
er
atin
g
co
n
d
itio
n
s
,
o
r
o
th
er
co
n
tex
ts
is
o
n
e
o
f
th
e
m
ain
lim
itatio
n
s
.
R
em
ain
in
g
u
s
ef
u
l
life
-
m
ain
t
en
an
ce
s
ch
e
d
u
lin
g
i
d
en
tifie
s
h
o
w
m
u
c
h
lo
n
g
er
ea
ch
b
atter
y
,
E
V,
an
d
m
ac
h
in
e
will
c
o
n
tin
u
e
f
u
n
ctio
n
in
g
b
ef
o
r
e
r
ep
lace
m
en
t
.
A
p
p
licatio
n
s
in
clu
d
e
AI
to
o
ls
,
s
tatis
t
ical
m
o
d
elin
g
,
p
h
y
s
ical
s
tr
ateg
y
ap
p
r
o
ac
h
es,
an
d
h
y
b
r
id
s
.
3
.
3
.
M
et
ho
ds
ba
s
ed
o
n
s
t
a
t
is
t
ica
l
mo
d
e
l
s
T
h
e
m
o
d
els
u
tili
zin
g
s
y
s
tem
i
n
f
o
r
m
atio
n
f
r
o
m
th
e
p
ast
p
r
ed
ict
th
e
in
itial
s
tag
es
an
d
p
r
o
g
r
ess
io
n
s
in
wear
-
o
u
t
s
ce
n
ar
io
s
.
T
h
eir
a
p
p
r
o
ac
h
p
r
esen
ts
an
an
aly
tical
tech
n
iq
u
e
b
ased
o
n
th
e
s
tatis
tical
an
aly
s
is
o
f
th
e
p
er
f
o
r
m
an
ce
m
etr
ics
f
r
o
m
th
e
p
ast,
with
n
o
s
o
le
r
elian
ce
o
n
ch
r
o
n
o
lo
g
ical
ass
ess
m
en
ts
b
u
t
in
s
tead
u
tili
za
tio
n
o
f
co
m
p
r
e
h
en
s
iv
e
d
atasets
th
r
o
u
g
h
tim
e.
Ho
wev
e
r
,
wh
ile
th
ese
s
tati
s
tical
m
o
d
els
ar
e
im
m
en
s
ely
ca
p
ab
le
o
f
d
ea
lin
g
with
ab
u
n
d
an
t
h
is
to
r
ical
in
f
o
r
m
atio
n
,
th
ey
lack
f
lex
ib
ilit
y
in
h
an
d
lin
g
u
n
e
x
p
ec
ted
o
p
er
atio
n
al
s
ce
n
ar
io
s
.
T
h
ey
ar
e
also
n
o
t
ca
p
ab
le
o
f
m
o
d
elin
g
c
o
m
p
lic
ated
n
o
n
lin
ea
r
ch
an
g
es
as
e
f
f
ec
tiv
ely
as
m
ac
h
in
e
lear
n
in
g
ap
p
r
o
ac
h
es.
3
.
4
.
H
y
brid
a
pp
ro
a
ches
T
h
eo
r
etica
l
k
n
o
wled
g
e
o
b
ta
in
ed
b
y
m
o
d
elin
g
,
to
g
et
h
er
with
th
e
em
p
ir
ical
o
b
s
er
v
atio
n
s
b
y
ex
p
er
im
en
tatio
n
,
en
h
an
ce
s
th
e
p
r
ec
is
io
n
in
th
e
p
r
ed
ictio
n
s
wh
ile
m
ak
in
g
th
em
m
o
r
e
r
eliab
l
e.
I
t
is
d
esira
b
le
to
co
n
s
id
er
th
e
m
etr
ics
d
er
i
v
ed
f
r
o
m
p
h
y
s
ical
m
o
d
els
as
in
p
u
t
v
ar
iab
les
f
o
r
a
r
tific
ial
in
tellig
en
ce
alg
o
r
ith
m
s
.
Un
d
er
s
tan
d
in
g
a
n
d
f
le
x
ib
ilit
y
in
p
r
o
b
le
m
-
s
o
lv
in
g
s
ce
n
a
r
io
s
ar
e
in
cr
ea
s
ed
wh
en
b
o
th
tr
a
d
itio
n
al
p
h
y
s
ics
-
b
ase
d
ap
p
r
o
ac
h
es
an
d
m
ac
h
i
n
e
lear
n
in
g
tech
n
iq
u
es
co
m
e
to
g
eth
e
r
.
T
h
eir
m
eth
o
d
s
o
f
ten
o
u
tp
er
f
o
r
m
th
o
s
e
f
r
o
m
p
u
r
e
m
o
d
els
alo
n
e
in
te
r
m
s
o
f
p
r
ec
i
s
io
n
an
d
d
ep
e
n
d
ab
ilit
y
.
Ho
we
v
er
,
th
ese
h
y
b
r
id
a
r
ch
itectu
r
es
r
eq
u
i
r
e
s
ig
n
if
ican
t
co
m
p
u
tatio
n
al
p
o
wer
an
d
r
eso
u
r
ce
-
in
te
n
s
iv
e
task
s
d
u
r
in
g
b
o
th
tr
ain
in
g
an
d
test
in
g
;
th
ey
ar
e,
th
er
ef
o
r
e
,
n
o
t
f
ea
s
ib
le
f
o
r
p
r
ac
tical
u
s
e
in
r
ea
l
-
tim
e
em
b
ed
d
e
d
b
atter
y
m
an
ag
em
en
t
s
y
s
tem
s
.
Sen
s
o
r
in
f
o
r
m
atio
n
an
d
d
eg
r
ad
atio
n
p
atter
n
s
ar
e
u
s
ed
to
esti
m
ate
r
em
ain
in
g
u
s
ef
u
l
life
.
Her
e,
th
e
m
eth
o
d
s
wo
r
k
b
y
s
elec
tin
g
r
elev
an
t
f
ea
tu
r
es,
in
itiatin
g
a
p
r
ed
ictio
n
wh
en
n
ec
ess
ar
y
,
an
d
ass
ess
i
n
g
th
e
c
o
n
d
itio
n
o
f
t
h
e
s
y
s
tem
.
Kalm
an
f
ilter
s
an
d
B
ay
esian
ap
p
r
o
ac
h
es r
esu
lt in
b
etter
p
er
f
o
r
m
an
ce
in
ter
m
s
o
f
ac
cu
r
ac
y
.
3
.
5
.
F
ilte
r
-
ba
s
ed
a
pp
ro
a
ches
R
UL
o
f
L
I
B
s
is
es
s
en
tial
to
av
o
id
b
r
ea
k
d
o
wn
s
an
d
en
s
u
r
e
s
ec
u
r
ity
s
in
ce
it
o
f
f
e
r
s
an
id
ea
at
wh
at
p
o
in
t
a
b
atter
y
will
s
tar
t
v
is
ib
le
d
e
g
r
ad
atio
n
in
its
p
er
f
o
r
m
an
ce
.
R
is
k
s
ar
e
at
a
m
in
im
u
m
wh
e
n
ap
p
r
o
p
r
iate
u
p
k
ee
p
s
tr
ateg
ies
ar
e
s
u
p
p
o
r
ted
b
y
c
o
r
r
ec
t
p
r
ed
ictio
n
s
f
r
o
m
p
r
ec
is
e
R
UL
m
o
d
els.
T
h
o
u
g
h
i
n
ca
p
ab
le
o
f
m
ea
s
u
r
in
g
th
e
d
ir
ec
t
wea
r
o
f
a
b
atter
y
,
r
em
ai
n
in
g
u
s
ef
u
l
life
esti
m
ates
g
iv
e
im
p
o
r
tan
t
in
s
ig
h
ts
in
to
th
e
s
tate
o
f
h
ea
lth
o
f
th
e
b
atter
y
.
Oth
er
v
alu
ab
le
tech
n
iq
u
es
th
at
ar
e
m
o
s
t
h
elp
f
u
l
in
f
u
r
th
er
p
e
r
f
ec
tin
g
t
h
e
m
o
d
els
o
f
r
em
ain
in
g
u
s
ef
u
l
life
p
r
ed
ictio
n
,
with
b
etter
a
cc
u
r
ac
y
i
n
s
ig
n
al
p
r
o
ce
s
s
in
g
th
r
o
u
g
h
f
ilter
in
g
o
p
e
r
atio
n
s
an
d
v
alid
atio
n
o
f
in
f
o
r
m
atio
n
in
th
e
is
o
lated
u
n
its
,
in
clu
d
e
ad
a
p
tiv
e
f
ilter
s
,
ar
tific
ial
in
tellig
en
ce
s
tr
ateg
ies,
p
r
o
b
a
b
ilis
tic
alg
o
r
ith
m
s
,
an
d
ce
ll
-
lev
el
d
a
ta
an
aly
s
is
,
am
o
n
g
o
t
h
er
m
e
th
o
d
s
[
1
1
]
.
T
h
ese
f
ilter
m
eth
o
d
s
,
in
clu
d
in
g
th
e
Un
s
ce
n
ted
Kalm
an
f
ilter
an
d
its
v
ar
ian
ts
,
en
h
an
ce
esti
m
atio
n
ac
cu
r
ac
y
d
u
e
t
o
th
e
lo
wer
lev
els
o
f
n
o
is
e
an
d
d
y
n
am
ic
ad
ap
tatio
n
ab
ilit
y
in
t
r
ea
tin
g
n
o
n
-
lin
ea
r
f
ea
tu
r
e
s
o
f
b
atter
y
b
eh
av
io
r
.
Ho
we
v
er
,
th
ese
s
y
s
tem
s
r
eq
u
ir
e
ca
r
ef
u
l
p
r
elim
in
ar
y
s
etu
p
an
d
p
ar
am
eter
tu
n
in
g
f
o
r
ef
f
icien
t
p
er
f
o
r
m
an
ce
an
d
m
i
g
h
t
also
h
in
d
er
ef
f
ec
tiv
e
u
s
e
with
in
lim
ited
m
icr
o
co
n
tr
o
ller
-
b
ased
b
atter
y
m
an
ag
em
e
n
t sy
s
tem
ap
p
licatio
n
s
.
3.
5
.
1
.
Ada
ptiv
e
f
ilte
r
t
e
c
h
n
i
qu
e
s
Fil
ter
s
th
at
ar
e
ca
p
ab
le
o
f
m
itig
atin
g
s
ig
n
al
d
is
tu
r
b
an
ce
s
f
o
r
m
th
e
b
asis
o
f
th
e
p
r
e
d
ictio
n
o
f
r
em
ain
in
g
u
s
ef
u
l
life
in
m
an
y
ap
p
licatio
n
s
.
Fo
r
th
e
p
u
r
p
o
s
e
o
f
L
I
B
s
R
UL
p
r
ed
ictio
n
,
t
h
e
f
ilter
ap
p
r
o
ac
h
h
as
u
n
d
e
r
g
o
n
e
s
ev
er
al
m
ajo
r
m
o
d
if
icatio
n
s
:
−
Un
s
ce
n
ted
Kalm
an
f
ilter
(
UK
F)
T
h
is
p
ap
er
p
r
o
p
o
s
es
a
n
o
v
el
a
p
p
r
o
ac
h
f
o
r
th
e
SOH
ass
ess
m
en
t
th
r
o
u
g
h
in
te
g
r
atin
g
im
p
ed
an
ce
h
ea
lth
f
ea
tu
r
es with
an
UKF.
A
s
ta
te
p
r
ed
ictio
n
m
o
d
el
co
m
p
u
tes th
e
SOH
b
ased
o
n
th
e
n
u
m
b
er
o
f
tim
es th
e
ce
lls
ar
e
cy
cled
,
wh
er
ea
s
a
m
ea
s
u
r
em
e
n
t
p
r
ed
ictio
n
m
o
d
el
f
o
r
ec
asts
th
e
ex
p
ec
ted
h
ea
lth
in
d
icato
r
(
HI
)
u
s
in
g
th
e
SOH
esti
m
atio
n
.
As
im
p
ed
an
ce
m
ea
s
u
r
em
en
ts
b
ec
o
m
e
av
ailab
l
e
o
v
er
th
e
b
atter
y
'
s
life
,
th
e
f
ilter
u
p
d
ates
th
e
esti
m
ates
d
ep
en
d
in
g
o
n
th
e
d
if
f
er
en
ce
b
etwe
en
ex
p
ec
ted
an
d
m
ea
s
u
r
e
d
HI
.
T
h
r
ee
m
o
d
el
s
tr
u
ctu
r
es
ar
e
ex
am
in
ed
an
d
co
n
tr
asted
:
f
ee
d
f
o
r
war
d
n
eu
r
al
n
etwo
r
k
s
(
F
NN)
,
p
o
ly
n
o
m
ial
r
eg
r
ess
io
n
(
PR
)
,
an
d
Gau
s
s
ian
p
r
o
ce
s
s
r
eg
r
ess
io
n
(
GPR
)
.
T
h
e
s
u
g
g
ested
ap
p
r
o
ac
h
is
v
alid
ated
u
s
in
g
two
o
p
en
-
s
o
u
r
ce
d
atasets
to
tal
in
g
5
1
ce
lls
.
Data
s
et
2
co
n
tain
s
2
1
NM
C
1
1
1
ce
lls
ag
ed
u
s
in
g
v
a
r
io
u
s
m
eth
o
d
s
,
w
h
er
ea
s
d
ataset
1
h
as
3
0
NM
C
8
1
1
ce
lls
ag
ed
u
s
in
g
a
p
r
eset p
r
o
to
co
l a
t six
u
n
iv
er
s
ities
.
Alth
o
u
g
h
th
e
esti
m
ato
r
ca
n
ac
cu
r
ately
esti
m
ate
th
e
SOH.
W
i
th
litt
le
d
ep
en
d
en
cy
o
n
th
e
SOC
lev
el
at
wh
ich
E
I
S
is
p
er
f
o
r
m
ed
,
th
e
G
PR
m
o
d
els
p
r
o
d
u
ce
th
e
lo
west
esti
m
atio
n
er
r
o
r
am
o
n
g
th
e
th
r
ee
m
o
d
el
s
tr
u
ctu
r
es,
with
1
.
4
%
f
o
r
d
ataset
1
a
n
d
1
.
7
%
f
o
r
d
ataset
2
.
T
h
is
d
em
o
n
s
tr
ates
h
o
w
ad
ap
ta
b
le
th
e
e
s
tim
ato
r
is
to
ag
in
g
m
ec
h
an
is
m
s
an
d
ce
ll
-
to
-
ce
ll
d
if
f
er
en
ce
s
[
1
2
]
.
T
o
im
p
r
o
v
e
th
e
p
r
ec
is
io
n
an
d
s
tab
ilit
y
o
f
b
atter
y
SOC
e
s
tim
atio
n
,
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
Ma
ch
in
e
lea
r
n
in
g
-
d
r
iven
p
r
o
g
n
o
s
tics
fo
r
lith
iu
m
-
io
n
b
a
tter
ies:
e
n
h
a
n
cin
g
…
(
B
o
d
a
p
a
ti V
e
n
ka
ta
R
a
j
a
n
n
a
)
261
th
e
UKF
m
o
d
el
is
a
u
g
m
e
n
ted
with
a
FOLPF
alg
o
r
ith
m
an
d
p
r
esen
ted
.
T
h
e
s
u
g
g
ested
m
e
th
o
d
'
s
ap
p
licab
ilit
y
an
d
r
esil
ien
ce
ar
e
e
x
am
in
ed
in
HPPC
in
f
o
r
m
atio
n
f
o
r
s
ev
er
a
l
b
atter
y
k
in
d
s
[
1
3
]
.
T
h
e
u
n
s
c
en
ted
Kalm
an
f
ilter
(
UKF)
is
wid
ely
u
s
ed
f
o
r
n
o
n
li
n
ea
r
s
tate
esti
m
atio
n
in
b
atter
y
m
an
ag
e
m
en
t.
T
h
e
s
tate
a
n
d
m
ea
s
u
r
em
en
t
m
o
d
els
ca
n
b
e
ex
p
r
ess
ed
as in
(
1
)
an
d
(
2
)
.
T
h
e
U
K
F
(
e
q
u
a
t
i
o
n
s
(
1
)
a
n
d
(
2
)
,
A
l
g
o
r
i
t
h
m
1
)
i
s
a
d
v
a
n
t
a
g
e
o
u
s
f
o
r
h
a
n
d
l
i
n
g
n
o
n
l
i
n
e
a
r
b
a
t
t
e
r
y
d
y
n
a
m
i
c
s
an
d
p
r
o
v
id
es a
cc
u
r
ate
SOC
/S
OH
esti
m
atio
n
co
m
p
ar
ed
to
li
n
ea
r
ized
f
ilter
s
s
u
ch
as th
e
e
x
t
en
d
ed
Kalm
an
f
ilter
.
+
1
=
(
,
)
+
(
1
)
=
ℎ
(
)
+
(
2
)
W
h
er
e
(
,
)
is
th
e
n
o
n
lin
ea
r
s
tate
tr
an
s
itio
n
f
u
n
ctio
n
,
is
th
e
s
tate
v
ec
to
r
o
f
th
e
b
atter
y
at
tim
e
s
t
ep
k
,
is
th
e
co
n
tr
o
l
in
p
u
t
v
ec
to
r
at
tim
e
s
tep
k
,
ℎ
(
)
n
o
n
lin
ea
r
m
ea
s
u
r
e
m
en
t
f
u
n
ctio
n
,
an
d
d
en
o
te
p
r
o
ce
s
s
an
d
m
ea
s
u
r
em
en
t n
o
is
e,
r
esp
ec
tiv
ely
.
Alg
o
r
ith
m
1
.
UKF
f
o
r
SOC
/S
OH
esti
m
atio
n
1)
I
n
itialize
s
tate
0
,
co
v
a
r
ian
ce
0
2)
Gen
er
ate
s
ig
m
a
p
o
in
ts
f
r
o
m
(
,
)
3)
Pro
p
ag
ate
s
ig
m
a
p
o
in
ts
th
r
o
u
g
h
f
(
,
)
4)
C
o
m
p
u
te
p
r
e
d
icted
m
ea
n
a
n
d
co
v
ar
ian
ce
5)
Up
d
ate
with
m
ea
s
u
r
em
en
t
u
s
in
g
ℎ
(
)
6)
R
etu
r
n
u
p
d
ate
d
s
tate
esti
m
ate
+
1
T
h
e
s
y
m
b
o
ls
u
s
ed
in
(
1
)
an
d
(
2
)
an
d
Alg
o
r
ith
m
1
a
r
e
s
u
m
m
a
r
ized
in
T
ab
le
1
f
o
r
clar
ity
.
I
n
t
h
e
co
n
tex
t
o
f
lith
iu
m
-
io
n
b
atter
y
m
an
a
g
em
en
t,
th
e
s
tate
v
ec
to
r
ty
p
ically
r
ep
r
esen
ts
SOC
an
d
SOH,
th
e
c
o
n
tr
o
l
in
p
u
t
co
r
r
esp
o
n
d
s
to
ap
p
lied
cu
r
r
e
n
t
o
r
ex
citatio
n
,
a
n
d
th
e
m
e
asu
r
em
en
t
v
ec
to
r
in
clu
d
es
ter
m
in
al
v
o
ltag
e
a
n
d
tem
p
er
atu
r
e
r
ea
d
in
g
s
.
T
ab
le
1
.
Def
in
itio
n
s
o
f
v
a
r
iab
l
es
in
UKF
s
tate
-
s
p
ac
e
m
o
d
el
S
y
mb
o
l
D
e
f
i
n
i
t
i
o
n
Ty
p
i
c
a
l
b
a
t
t
e
r
y
c
o
n
t
e
x
t
S
t
a
t
e
v
e
c
t
o
r
a
t
t
i
me
st
e
p
k
S
O
C
,
S
O
H
,
i
n
t
e
r
n
a
l
r
e
s
i
st
a
n
c
e
,
c
a
p
a
c
i
t
y
C
o
n
t
r
o
l
i
n
p
u
t
v
e
c
t
o
r
a
t
t
i
me
st
e
p
k
A
p
p
l
i
e
d
c
u
r
r
e
n
t
(
c
h
a
r
g
e
/
d
i
sc
h
a
r
g
e
)
,
e
x
c
i
t
a
t
i
o
n
M
e
a
su
r
e
me
n
t
v
e
c
t
o
r
a
t
t
i
me
s
t
e
p
k
Te
r
mi
n
a
l
v
o
l
t
a
g
e
,
c
u
r
r
e
n
t
,
t
e
m
p
e
r
a
t
u
r
e
(
,
)
N
o
n
-
l
i
n
e
a
r
st
a
t
e
t
r
a
n
s
i
t
i
o
n
f
u
n
c
t
i
o
n
G
o
v
e
r
n
s
h
o
w
st
a
t
e
s
e
v
o
l
v
e
w
i
t
h
d
y
n
a
mi
c
s
ℎ
(
)
N
o
n
-
l
i
n
e
a
r
mea
su
r
e
me
n
t
f
u
n
c
t
i
o
n
M
a
p
s s
t
a
t
e
s
t
o
me
a
s
u
r
a
b
l
e
o
u
t
p
u
t
s
P
r
o
c
e
ss
n
o
i
s
e
(
z
e
r
o
-
m
e
a
n
,
c
o
v
a
r
i
a
n
c
e
)
U
n
c
e
r
t
a
i
n
t
y
i
n
s
t
a
t
e
e
v
o
l
u
t
i
o
n
M
e
a
su
r
e
me
n
t
n
o
i
s
e
(
z
e
r
o
-
me
a
n
,
c
o
v
a
r
i
a
n
c
e
)
S
e
n
s
o
r
n
o
i
se
a
n
d
d
i
st
u
r
b
a
n
c
e
s
P
r
o
c
e
ss
n
o
i
s
e
c
o
v
a
r
i
a
n
c
e
ma
t
r
i
x
Tu
n
i
n
g
p
a
r
a
m
e
t
e
r
(
sy
s
t
e
m
d
y
n
a
m
i
c
s
u
n
c
e
r
t
a
i
n
t
y
)
M
e
a
su
r
e
me
n
t
n
o
i
s
e
c
o
v
a
r
i
a
n
c
e
ma
t
r
i
x
Tu
n
i
n
g
p
a
r
a
m
e
t
e
r
(
se
n
so
r
a
c
c
u
r
a
c
y
)
Est
i
m
a
t
i
o
n
e
r
r
o
r
c
o
v
a
r
i
a
n
c
e
ma
t
r
i
x
C
o
n
f
i
d
e
n
c
e
i
n
f
i
l
t
e
r
’
s s
t
a
t
e
e
st
i
ma
t
e
I
n
th
e
UKF
f
r
a
m
ewo
r
k
,
th
r
ee
ty
p
es
o
f
co
v
a
r
ian
ce
m
atr
ices
ar
e
u
s
ed
:
p
r
o
ce
s
s
n
o
is
e
co
v
a
r
ian
ce
(
),
m
ea
s
u
r
em
en
t
n
o
is
e
co
v
ar
ian
ce
(
)
,
an
d
esti
m
atio
n
er
r
o
r
co
v
ar
ian
ce
(
)
.
W
h
ile
an
d
ch
ar
ac
ter
iz
e
u
n
ce
r
tain
ties
in
th
e
s
y
s
tem
an
d
m
ea
s
u
r
em
en
ts
,
r
esp
ec
tiv
ely
,
is
u
p
d
ated
iter
ativ
ely
with
in
th
e
f
ilter
to
r
ep
r
esen
t
th
e
ev
o
lv
in
g
co
n
f
id
en
ce
i
n
th
e
esti
m
ated
s
tates.
−
Un
s
ce
n
ted
p
ar
ticle
f
ilter
(
UPF)
Pre
d
ictin
g
th
e
b
atter
y
m
an
a
g
e
m
en
t
s
y
s
tem
'
s
r
em
ain
in
g
u
s
ab
le
life
(
R
UL
)
is
c
r
itical
[
1
4
]
.
T
h
is
p
ap
e
r
d
escr
ib
es
a
r
em
ain
in
g
u
s
ef
u
l
life
(
R
UL
)
p
r
ed
ictio
n
ap
p
r
o
ac
h
f
o
r
lith
iu
m
-
io
n
b
atter
ies
wit
h
a
d
u
al
f
ilter
,
d
ata
-
d
r
iv
en
,
m
o
d
el
-
b
ased
f
u
s
io
n
alg
o
r
ith
m
.
On
e
d
is
tin
g
u
is
h
in
g
a
s
p
ec
t
o
f
th
e
f
r
a
m
ewo
r
k
is
th
e
co
n
cu
r
r
en
t
o
n
lin
e
iter
atio
n
o
f
b
if
u
r
ca
te
d
f
u
s
io
n
alg
o
r
ith
m
f
ilter
s
,
wh
ich
u
p
d
at
e
b
o
th
th
e
b
atter
y
'
s
s
ta
te
o
f
c
ap
ac
ity
d
eg
r
ad
atio
n
an
d
th
e
p
ar
a
m
eter
s
o
f
th
e
L
SS
VM
m
o
d
el.
T
h
e
u
n
s
ce
n
ted
p
ar
ticle
f
ilter
in
g
(
UPF)
is
u
s
ed
in
r
atio
m
ea
s
u
r
em
en
t
eq
u
atio
n
s
to
p
r
o
ce
s
s
f
u
s
io
n
alg
o
r
ith
m
s
,
wh
ich
in
t
u
r
n
s
u
p
p
ly
th
e
L
SS
VM
m
o
d
el
with
v
ir
tu
al
f
u
tu
r
e
m
ea
s
u
r
em
en
ts
.
I
n
UPF
p
r
o
ce
s
s
in
g
,
th
e
u
n
ce
r
tai
n
ty
o
f
t
h
e
p
r
o
g
n
o
s
tic
r
esu
lt
is
ex
p
r
ess
ed
b
y
a
p
r
o
b
a
b
ilit
y
d
en
s
ity
f
u
n
ctio
n
(
PDF)
[
1
5
]
.
−
Sp
h
er
ical
cu
b
atu
r
e
p
ar
ticle
f
ilt
er
(
SC
PF
)
A
t
h
ev
en
in
m
o
d
el
is
d
ev
el
o
p
e
d
an
d
a
lin
ea
r
Kalm
an
f
ilter
(
L
KF)
is
ap
p
lied
to
esti
m
ate
p
a
r
am
eter
s
in
r
ea
l
-
tim
e.
T
h
e
r
esu
lts
s
h
o
w
th
at
th
e
co
n
v
er
g
e
n
ce
o
f
L
KF
o
u
tp
er
f
o
r
m
s
r
h
eo
n
o
m
ic
least
s
q
u
a
r
es
(
R
L
S
)
an
d
R
L
S
with
f
o
r
g
ettin
g
f
ac
t
o
r
(
R
L
S
-
FF
)
[
1
6
]
.
Als
o
,
i
n
s
tead
o
f
a
9
t
h
o
r
d
er
p
o
ly
n
o
m
ial
f
it,
th
e
eq
u
ilib
r
iu
m
p
o
ten
tial
eq
u
atio
n
(
E
PE)
is
u
s
ed
to
d
escr
ib
e
th
e
OC
V
v
er
s
u
s
So
C
r
elatio
n
s
h
ip
.
An
ac
c
u
r
ate
m
o
d
el
o
f
th
e
test
ed
b
atter
y
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
.
1
,
Ma
r
ch
20
2
6
:
257
-
2
7
4
262
an
d
a
f
ast
So
C
esti
m
atio
n
ap
p
r
o
ac
h
ar
e
r
e
q
u
ir
ed
to
i
m
p
r
o
v
e
th
e
esti
m
atio
n
ac
cu
r
ac
y
o
f
So
C
[
1
7
]
.
T
h
e
ex
p
er
im
e
n
ts
wer
e
co
n
d
u
ct
ed
o
n
p
r
is
m
atic
L
iFePO4
b
atter
ies
at
r
o
o
m
tem
p
er
at
u
r
e
u
n
d
er
th
e
New
E
u
r
o
p
ea
n
Dr
iv
in
g
C
y
cle
an
d
Ur
b
an
D
y
n
am
o
m
eter
Dr
iv
in
g
Sch
ed
u
le.
It
h
as
b
e
en
d
is
co
v
er
ed
th
at
W
MI
C
K
F
ex
ce
ed
s
b
o
t
h
th
e
m
u
lti
-
in
n
o
v
atio
n
C
KF
an
d
th
e
s
tan
d
ar
d
cu
b
atu
r
e
Kalm
an
f
ilter
in
p
er
f
o
r
m
an
ce
.
W
MI
C
KF
ac
h
iev
es
a
So
C
esti
m
atio
n
er
r
o
r
o
f
less
th
an
1
%
(
0
.
9
1
%)
,
wh
ich
is
lo
wer
th
a
n
C
KF'
s
1
.
3
0
%
an
d
MI
C
KF's
2
.
7
1
%.
T
h
e
p
r
o
p
o
s
ed
t
ec
h
n
iq
u
e
is
th
o
r
o
u
g
h
ly
test
ed
b
y
d
eter
m
in
in
g
th
e
m
ea
n
ab
s
o
lu
te
er
r
o
r
(
MA
E
)
,
r
o
o
t
m
ea
n
s
q
u
ar
e
er
r
o
r
(
R
MSE
)
,
an
d
co
ef
f
icie
n
t
o
f
d
eter
m
in
atio
n
(R
-
s
q
u
ar
e)
.
Fu
r
th
er
m
o
r
e,
th
e
r
esil
ien
ce
o
f
W
MI
C
KF
i
s
d
em
o
n
s
tr
ated
f
o
r
a
v
ar
iety
o
f
n
o
is
e
d
is
tu
r
b
an
ce
ty
p
es
an
d
in
i
tial
So
C
f
au
lts
[
1
8
]
.
A
4
8
V
s
y
s
tem
o
f
f
er
s
a
lo
w
-
co
s
t
v
eh
icle
elec
tr
if
icatio
n
o
p
tio
n
th
at
ca
n
d
ec
r
ea
s
e
C
O2
em
is
s
io
n
s
b
y
1
5
%
to
2
0
%
wh
ile
p
r
eser
v
in
g
th
e
cu
r
r
e
n
t
au
to
m
o
tiv
e
ar
c
h
itectu
r
e.
C
h
ar
g
in
g
an
d
d
is
ch
ar
g
in
g
cy
cles
h
av
e
s
tr
in
g
e
n
t
ch
ar
g
in
g
an
d
d
is
ch
ar
g
in
g
c
r
iter
ia,
elev
at
io
n
f
o
r
B
MS
in
ter
m
s
o
f
ac
cu
r
a
te
s
tate
esti
m
atio
n
.
T
h
e
f
ir
s
t
co
n
tr
ib
u
tio
n
o
f
th
is
p
ap
er
is
d
o
wn
to
a
m
u
lti
-
s
ca
le
co
-
esti
m
atio
n
ap
p
r
o
ac
h
in
p
ar
am
eter
s
o
f
s
tate
o
f
ch
a
r
g
e
an
d
s
tate
o
f
h
ea
lth
o
f
4
8
V
b
atter
y
s
y
s
tem
.
First,
we
co
n
s
tr
u
ct
a
f
r
a
m
ewo
r
k
f
o
r
m
u
lti
-
s
ca
le
esti
m
ate
b
ased
o
n
m
ain
f
ea
tu
r
es
an
d
p
ar
am
eter
s
o
f
th
e
b
atter
y
.
Seco
n
d
,
we
d
er
iv
e
th
e
in
ter
n
al
r
esis
tan
ce
o
f
th
e
b
atter
y
u
s
in
g
th
e
eq
u
iv
alen
t c
i
r
cu
it m
o
d
el
with
r
ec
u
r
s
iv
e
lea
s
t sq
u
ar
es.
B
atter
y
ch
ar
g
e
a
n
d
ca
p
ac
ity
esti
m
atio
n
ar
e
p
er
f
o
r
m
ed
u
s
in
g
a
d
v
an
ce
d
f
ilter
in
g
tech
n
iq
u
es
lik
e
cu
b
a
tu
r
e
Kalm
an
f
ilter
a
n
d
H
-
i
n
f
in
ity
.
T
h
e
em
b
ed
d
e
d
B
MS
with
h
ig
h
ac
cu
r
ac
y
r
eq
u
ir
em
en
ts
also
b
en
e
f
its
f
r
o
m
r
ed
u
ce
d
co
m
p
u
tatio
n
r
eso
u
r
c
e
d
em
an
d
s
th
r
o
u
g
h
m
u
lti
-
tim
escale
co
-
esti
m
atio
n
ap
p
r
o
ac
h
es.
T
h
e
ev
alu
atio
n
o
f
t
h
e
m
eth
o
d
’
s
ap
p
r
o
ac
h
c
o
n
s
is
ts
o
f
p
er
f
o
r
m
in
g
s
ev
er
al
s
im
u
latio
n
s
b
ased
o
n
s
tan
d
a
r
d
d
r
iv
in
g
cy
cles.
T
h
e
o
u
tco
m
es
ar
e
th
en
an
aly
ze
d
alo
n
g
s
id
e
r
esu
lts
p
r
o
d
u
ce
d
f
r
o
m
o
th
e
r
in
d
u
s
tr
y
ap
p
r
o
ac
h
es.
Ou
tco
m
e
r
esu
lts
th
at
v
alid
ate
th
e
s
u
g
g
esti
o
n
’
s
ap
p
r
o
ac
h
is
b
etter
th
an
all
o
th
er
co
m
p
etin
g
ap
p
r
o
ac
h
es
b
y
a
m
ea
n
ab
s
o
lu
te
er
r
o
r
0
.
6
4
%
in
ca
p
ac
ity
esti
m
atio
n
an
d
0
.
8
8
i
n
SOC
.
4.
I
NT
E
L
L
I
G
E
N
T
M
L
T
E
CH
NIQU
E
S F
O
R
RUL
AN
D
S
O
H
4
.
1
.
Art
if
ici
a
l
inte
llig
ence
(
AI)
Fig
u
r
e
1
illu
s
tr
ates
AI
-
b
ased
t
ec
h
n
iq
u
es
th
at
s
im
u
late
d
e
g
r
a
d
atio
n
tr
e
n
d
s
f
r
o
m
o
b
s
er
v
a
b
le
d
ata
u
s
in
g
m
ac
h
in
e
lear
n
in
g
.
B
y
d
eter
m
in
in
g
f
ailu
r
e
th
r
esh
o
ld
s
a
n
d
ex
tr
ap
o
latin
g
d
e
g
r
ad
atio
n
tr
en
d
s
f
r
o
m
p
ast
p
er
f
o
r
m
an
ce
d
ata,
th
e
y
f
o
r
ec
ast
R
UL
.
T
h
e
ar
ch
itectu
r
e
in
teg
r
ates
m
u
ltip
le
m
ac
h
in
e
l
ea
r
n
in
g
tech
n
iq
u
es
,
in
clu
d
in
g
ar
tific
ial
n
eu
r
al
n
et
wo
r
k
s
(
ANN)
,
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
i
n
es
(
SVM)
,
r
elev
an
c
e
v
ec
to
r
m
ac
h
in
es
(
R
VM
)
,
d
ee
p
n
eu
r
al
n
etwo
r
k
s
(
DNN)
,
an
d
h
y
b
r
id
AI
a
p
p
r
o
a
ch
es
(
ML
co
m
b
in
ed
with
p
h
y
s
ics
-
b
ased
m
o
d
els),
f
ee
d
in
g
in
t
o
a
ce
n
tr
alize
d
p
r
e
d
ictio
n
en
g
in
e
f
o
r
ac
c
u
r
ate
b
att
er
y
d
iag
n
o
s
tics
.
4
.
2
.
M
et
ho
ds
ba
s
ed
o
n phy
s
i
cs
Fo
r
s
y
s
tem
s
s
u
ch
as
b
atter
ies,
p
h
y
s
ics
-
b
ased
ap
p
r
o
ac
h
es
a
r
e
h
elp
f
u
l
b
ec
au
s
e
th
ey
r
ep
licate
t
h
e
p
h
y
s
ical
an
d
ch
em
ical
p
r
o
ce
s
s
es th
at
lead
to
s
y
s
tem
d
eter
io
r
atio
n
.
Ho
wev
er
,
in
ter
n
al
c
o
n
d
itio
n
s
ar
e
d
if
f
icu
lt to
o
b
s
er
v
e
d
ir
ec
tly
.
As a
r
esu
lt,
m
o
d
elin
g
n
o
n
lin
ea
r
p
r
o
ce
s
s
es su
ch
as b
atter
y
d
eg
r
a
d
atio
n
r
e
m
ain
s
a
c
h
allen
g
e.
Fig
u
r
e
1
.
AI
-
b
ased
m
o
d
el
f
o
r
p
r
ed
ictin
g
b
atter
y
r
e
m
ain
in
g
u
s
ef
u
l lif
e
(
R
UL
)
an
d
s
tate
o
f
h
ea
lth
(
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
Ma
ch
in
e
lea
r
n
in
g
-
d
r
iven
p
r
o
g
n
o
s
tics
fo
r
lith
iu
m
-
io
n
b
a
tter
ies:
e
n
h
a
n
cin
g
…
(
B
o
d
a
p
a
ti V
e
n
ka
ta
R
a
j
a
n
n
a
)
263
4
.
3
.
H
y
brid m
et
ho
ds
Als
o
,
co
m
b
in
in
g
d
if
f
er
e
n
t
ap
p
r
o
ac
h
es
s
till
y
ield
s
b
etter
p
e
r
f
o
r
m
a
n
ce
:
f
o
r
e
x
am
p
le,
b
y
b
o
o
s
tin
g
th
e
r
eliab
ilit
y
o
f
th
e
esti
m
ates
u
n
d
er
lo
a
d
;
an
a
d
v
an
ce
d
m
ac
h
in
e
lear
n
in
g
s
y
s
tem
th
at
d
e
p
lo
y
s
s
p
ec
if
ic
h
ea
lth
m
etr
ics
in
co
n
ju
n
ctio
n
with
h
y
b
r
id
en
s
em
b
le
m
o
d
els
m
a
d
e
u
p
o
f
r
an
d
o
m
v
ec
to
r
f
u
n
ctio
n
al
lin
k
s
an
d
ex
tr
e
m
e
lear
n
in
g
m
ac
h
i
n
es
ca
n
ac
h
iev
e
s
u
p
er
io
r
ac
cu
r
ac
y
.
T
h
e
R
UL
m
o
d
el
h
as
its
ad
v
an
tag
es
b
u
t
also
lim
itatio
n
s
.
I
n
p
r
ac
tical
ter
m
s
,
tr
ad
itio
n
al
m
eth
o
d
s
m
er
g
e
d
with
an
aly
tical
o
n
es
in
cr
ea
s
e
th
e
r
eliab
ilit
y
o
f
p
r
e
d
ictio
n
s
u
b
s
tan
tially
as
tech
n
o
lo
g
y
e
v
o
lv
es.
C
o
m
p
lex
ities
in
b
atter
ies
r
eq
u
ir
e
m
o
r
e
s
o
p
h
is
ticated
m
eth
o
d
o
l
o
g
ies
s
u
ch
as
m
ac
h
in
e
lear
n
in
g
a
n
d
AI
a
s
im
p
o
r
tan
t
to
o
ls
f
o
r
t
h
eir
r
e
m
ain
in
g
u
s
ef
u
l
life
esti
m
atio
n
s
an
d
p
r
ed
ictio
n
s
o
f
f
u
tu
r
e
s
tates
with
h
ig
h
ac
cu
r
ac
y
.
T
h
e
f
ir
s
t
s
ec
tio
n
in
tr
o
d
u
ce
s
an
in
v
esti
g
atio
n
o
n
t
y
p
ical
in
t
ellig
en
t
alg
o
r
ith
m
s
s
u
ch
as
ANNs,
SVMs,
R
VM
s
,
an
d
DNNs.
T
h
en
it
s
h
o
ws
r
ec
en
t
d
ev
el
o
p
m
en
ts
i
n
ea
c
h
an
d
r
elativ
e
ef
f
icien
cies
wh
en
b
en
c
h
m
ar
k
ed
a
g
ain
s
t d
if
f
er
en
t sets
o
f
d
ata
p
o
in
ts
.
L
ast
b
u
t n
o
t
least,
th
is
s
ec
tio
n
s
u
m
m
ar
izes lim
itatio
n
s
r
eg
ar
d
in
g
h
ig
h
d
ata
r
eq
u
ir
em
e
n
ts
,
h
ig
h
co
m
p
u
tatio
n
al
r
eso
u
r
ce
s
,
an
d
lo
w
in
te
r
p
r
etab
ilit
y
;
a
t
th
e
s
am
e
tim
e,
it
r
ef
lects
o
n
its
p
o
ten
tial
to
s
ig
n
if
ican
tly
im
p
ac
t
th
e
f
u
tu
r
e
o
f
b
atter
y
m
an
a
g
em
en
t
s
y
s
tem
s
.
T
h
e
s
o
p
h
is
ticated
p
r
o
ce
s
s
in
g
o
f
d
ata
an
d
th
e
cr
ea
tio
n
o
f
p
r
ed
ictio
n
alg
o
r
it
h
m
s
co
n
ce
r
n
i
n
g
t
h
e
r
e
m
ain
in
g
u
s
ef
u
l
life
(
R
UL
)
o
f
lith
iu
m
-
io
n
b
atter
ies
(
L
I
B
s
)
r
elies
g
r
ea
tly
o
n
in
tellig
en
t
s
y
s
tem
s
,
esp
ec
ially
th
o
s
e
th
at
u
tili
ze
ar
tific
ial
in
tellig
en
ce
(
AI
)
.
T
h
e
s
e
m
eth
o
d
s
o
p
tim
ize
ac
cu
r
ac
y
an
d
r
e
liab
ilit
y
o
f
p
r
ed
ictio
n
s
with
d
ata
f
r
o
m
m
u
ltip
l
e
s
o
u
r
ce
s
.
W
e
will d
is
cu
s
s
ex
ec
u
tiv
e
AI
-
b
ased
a
p
p
r
o
ac
h
es f
o
r
R
UL
esti
m
atio
n
as d
ep
icted
in
Fig
u
r
e
2
.
Fig
u
r
e
2
.
Dif
f
e
r
en
t p
a
r
am
eter
s
ev
alu
atio
n
f
o
r
Li
-
i
o
n
b
atter
y
u
s
in
g
AN
N
4
.
4
.
Art
if
ici
a
l
neura
l net
wo
rk
s
(
A
N
N
s
)
L
ith
iu
m
-
io
n
b
atter
ies
ca
n
n
o
t
b
e
ev
alu
ated
with
clo
s
ed
-
f
o
r
m
eq
u
atio
n
s
u
s
in
g
m
ea
s
u
r
ab
le
f
a
cto
r
s
f
r
o
m
o
u
ts
id
e
th
e
s
y
s
tem
.
T
h
er
e
ex
i
s
t
m
an
y
m
ath
em
atica
l
an
d
el
ec
tr
ical
cir
cu
it
m
o
d
els
wh
ich
aim
to
ex
p
lain
th
e
wo
r
k
in
g
p
r
in
cip
les
o
f
lith
iu
m
-
io
n
b
atter
ies.
All
o
f
th
ese
m
o
d
els
h
av
e
b
ee
n
s
h
o
w
n
to
b
e
s
o
m
ewh
at
in
ac
cu
r
ate,
o
v
er
ly
s
o
p
h
is
ticated
,
an
d
r
elia
n
t
o
n
n
u
m
er
o
u
s
o
p
er
atin
g
co
n
d
itio
n
s
.
T
h
is
is
wh
y
r
esear
ch
er
s
h
av
e
b
ee
n
s
tu
d
y
in
g
th
e
im
p
lem
en
tatio
n
o
f
m
ac
h
i
n
e
lear
n
in
g
(
ML
)
tech
n
i
q
u
es
th
at
esti
m
ate
th
e
s
tate
o
f
a
b
atte
r
y
f
r
o
m
ea
s
ily
m
ea
s
u
r
ab
le
p
a
r
am
eter
s
lik
e
d
is
ch
ar
g
e
cu
r
r
en
t,
o
u
tp
u
t
v
o
ltag
e,
s
u
r
f
ac
e
tem
p
er
atu
r
e
s
o
f
th
e
ce
ll
an
d
co
r
r
esp
o
n
d
in
g
en
v
i
r
o
n
m
e
n
t
t
em
p
er
atu
r
es.
Ad
d
itio
n
ally
,
A
I
an
d
ML
s
tr
ateg
ies
h
av
e
b
ee
n
ap
p
lied
in
th
e
m
an
u
f
ac
tu
r
in
g
p
r
o
ce
s
s
es o
f
L
i
-
io
n
b
atter
ies,
p
er
f
o
r
m
ed
p
r
o
c
ess
es r
elate
d
to
r
ec
y
clin
g
to
ass
em
b
le
n
ew
b
atter
y
p
ac
k
s
f
o
r
s
ev
er
al
s
u
b
s
eq
u
e
n
t
u
s
es,
as
we
ll
as
p
r
ed
ictin
g
th
e
m
ass
o
f
L
i
-
io
n
b
atter
ies
an
d
th
eir
r
em
ain
in
g
u
s
ef
u
l
life
(
R
UL
)
.
Ar
tific
ial
n
eu
r
al
n
etwo
r
k
s
(
ANN)
th
at
co
-
esti
m
ate
th
e
p
o
wer
an
d
ch
ar
g
e
s
tatu
s
h
a
v
e
ac
h
iev
ed
s
ig
n
if
ican
t su
cc
ess
.
T
h
ese
m
o
d
els ar
e
c
o
m
p
r
is
ed
o
f
in
ter
c
o
n
n
ec
ted
n
etwo
r
k
s
o
f
n
o
d
es
th
at
s
im
u
late
n
eu
r
o
n
s
in
a
h
u
m
a
n
b
r
ain
an
d
ar
e
co
n
n
e
cted
with
s
ig
n
als
an
alo
g
o
u
s
t
o
b
io
lo
g
ical
s
y
n
ap
s
es.
Neu
r
al
n
etwo
r
k
s
s
ee
k
to
r
ep
licate
a
h
u
m
a
n
b
r
ain
’
s
lear
n
in
g
an
d
r
ea
s
o
n
i
n
g
p
r
o
ce
s
s
es
b
y
m
o
d
if
y
in
g
th
e
p
r
ev
io
u
s
ly
e
s
tab
lis
h
ed
weig
h
ted
co
n
n
ec
tio
n
s
wh
ile
tr
y
in
g
to
f
in
d
r
elatio
n
s
h
ip
s
b
etwe
en
s
y
s
tem
p
ar
am
eter
s
d
u
r
in
g
th
e
lear
n
in
g
p
h
ase.
As
r
ep
o
r
ted
in
[
1
9
]
,
ar
t
if
i
cial
n
eu
r
al
n
etwo
r
k
s
ar
e
tr
ain
ed
o
n
r
ea
l
-
tim
e
d
ata
to
p
r
ed
ict
th
e
s
tates o
f
L
i
-
io
n
b
atter
ies.
4
.
5
.
Su
pp
o
rt
v
ec
t
o
r
ma
chines
(
S
VM
)
Fo
r
ef
f
icien
t
m
o
n
ito
r
in
g
o
f
b
a
tter
y
m
an
a
g
em
en
t
s
y
s
tem
s
,
th
e
SOH
an
d
R
UL
o
f
lith
iu
m
-
io
n
b
atter
ies
(
L
I
B
s
)
n
ee
d
to
b
e
ac
cu
r
ately
ass
ess
ed
.
Du
e
to
th
e
c
o
m
p
lex
in
t
er
n
al
ch
em
ical
ch
an
g
es
an
d
n
o
n
lin
ea
r
d
e
g
r
ad
atio
n
o
f
L
I
B
s
,
d
ir
ec
t
e
v
alu
atio
n
o
f
SOH
an
d
R
UL
is
n
ea
r
im
p
o
s
s
ib
le
[
2
0
]
.
T
h
e
T
W
SVM
ap
p
r
o
ac
h
is
u
s
ed
to
tack
le
th
ese
d
if
f
icu
lties
an
d
esti
m
ate
SOH
an
d
R
UL
.
I
n
o
r
d
e
r
to
asc
er
tain
th
e
m
o
s
t
p
r
o
m
in
en
t d
r
iv
er
s
o
f
d
eg
r
ad
atio
n
in
b
atter
y
p
er
f
o
r
m
an
c
e,
th
e
co
n
s
tan
t
cu
r
r
en
t
ch
ar
g
in
g
tim
e
o
f
a
lith
iu
m
b
atter
y
is
tr
ea
ted
a
s
a
h
ea
lth
in
d
icato
r
(
HI
)
.
Dec
o
m
p
o
s
itio
n
is
p
er
f
o
r
m
ed
u
s
in
g
VM
D
an
d
th
e
im
p
o
r
tan
ce
o
f
r
an
d
o
m
f
o
r
est
f
ea
tu
r
es
is
u
s
ed
to
co
m
p
u
te
th
e
f
ea
tu
r
e
c
o
r
r
elatio
n
s
co
r
es [
2
1
]
.
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
.
1
,
Ma
r
ch
20
2
6
:
257
-
2
7
4
264
Mo
r
eo
v
er
,
th
e
g
l
o
b
al
s
ea
r
ch
in
g
ca
p
ab
ilit
y
o
f
C
OA
is
b
o
o
s
ted
th
r
o
u
g
h
a
p
p
licatio
n
o
f
th
e
d
if
f
er
en
tial
ev
o
lu
tio
n
m
eth
o
d
alo
n
g
with
g
o
o
d
p
o
in
t
s
et
th
eo
r
y
.
S
OH
an
d
R
UL
p
r
ed
ictio
n
m
o
d
els
ar
e
cr
ea
ted
b
y
o
p
tim
izin
g
T
W
SVM
p
ar
am
ete
r
s
u
s
in
g
th
e
im
p
r
o
v
ed
co
n
v
o
lu
tio
n
o
p
tim
izatio
n
alg
o
r
ith
m
(
I
C
OA)
[
2
2
]
.
Fin
ally
,
th
e
p
r
o
p
o
s
ed
m
o
d
els
ar
e
v
alid
ated
u
s
in
g
d
ata
f
r
o
m
NASA
an
d
th
e
C
AL
C
E
lith
iu
m
-
io
n
b
at
ter
ies.
E
x
p
er
im
en
tal
r
esu
lts
s
h
o
w
th
at
th
e
s
u
g
g
ested
m
o
d
els
ac
h
iev
e
a
r
elativ
e
er
r
o
r
in
R
UL
p
r
ed
ictio
n
r
an
g
in
g
f
r
o
m
-
1
.
8
%
to
2
%,
with
R
MSE
an
d
MA
PE
f
o
r
S
OH
an
d
R
UL
p
r
ed
ictio
n
o
f
n
o
m
o
r
e
th
a
n
0
.
0
0
7
an
d
0
.
0
0
8
2
,
r
esp
ec
tiv
ely
.
T
h
e
p
r
o
p
o
s
ed
m
o
d
el
o
u
tp
e
r
f
o
r
m
s
ex
is
tin
g
m
o
d
els
in
ter
m
s
o
f
r
o
b
u
s
tn
ess
an
d
f
it
[
2
3
]
.
Su
p
p
o
r
t
v
ec
to
r
m
ac
h
in
es
(
SVM)
is
p
o
wer
f
u
l
f
o
r
class
if
icatio
n
an
d
r
eg
r
ess
io
n
in
b
atter
y
SOH/R
UL
est
im
atio
n
.
T
h
eir
o
p
tim
izatio
n
p
r
o
b
lem
is
f
o
r
m
u
lated
as sh
o
wn
in
(
3
)
an
d
(
4
)
.
,
,
1
2
‖
‖
2
+
∑
=
1
(
3
)
(
⋅
(
)
+
)
≥
1
−
,
≥
0
(
4
)
T
h
e
(
3
)
an
d
(
4
)
a
n
d
Alg
o
r
ith
m
2
illu
s
tr
ate
th
e
s
tan
d
ar
d
SVM
f
r
am
ewo
r
k
ap
p
lied
in
SOH/R
UL
e
s
tim
atio
n
.
T
h
ese
f
o
r
m
u
latio
n
s
en
ab
le
r
o
b
u
s
t p
r
ed
ictio
n
s
ev
en
u
n
d
er
s
m
all
d
ataset
co
n
d
itio
n
s
.
Alg
o
r
ith
m
2
.
SVM
f
o
r
SOH/R
UL
p
r
ed
ictio
n
1)
I
n
p
u
t tr
ai
n
in
g
d
ata
{
,
}
f
o
r
=
1
,
…
.
2)
Select
k
er
n
el
f
u
n
ctio
n
(
,
)
3)
So
lv
e
q
u
ad
r
atic
o
p
tim
izatio
n
p
r
o
b
lem
→
o
b
tain
(
L
ag
r
a
n
g
e
m
u
ltip
lier
s
)
4)
I
d
en
tify
s
u
p
p
o
r
t
v
ec
to
r
s
(
s
am
p
les with
>
0
)
an
d
co
m
p
u
te
weig
h
t
s
5)
Fo
r
a
n
ew
s
am
p
le
,
p
r
ed
ict:
(
)
=
(
(
,
)
+
)
(
5
)
I
n
th
e
d
u
al
f
o
r
m
u
latio
n
o
f
SVM,
r
ep
r
esen
ts
th
e
L
ag
r
a
n
g
e
m
u
ltip
lier
ass
o
ciate
d
with
th
e
ℎ
tr
ain
in
g
s
am
p
le.
No
n
-
ze
r
o
co
r
r
esp
o
n
d
s
to
s
u
p
p
o
r
t
v
ec
to
r
s
th
at
d
ef
in
e
th
e
d
e
cisi
o
n
b
o
u
n
d
ar
y
.
T
h
e
d
ec
is
io
n
f
u
n
ctio
n
(
)
is
th
en
g
iv
e
n
b
y
(
5
)
as sh
o
wn
in
Alg
o
r
ith
m
2
,
wh
ich
co
m
b
in
es
lab
els
,
an
d
th
e
k
er
n
el
s
im
ilar
ity
f
u
n
ctio
n
.
T
h
e
v
ar
iab
les
u
s
ed
in
th
e
SVM
o
p
tim
izatio
n
p
r
o
b
lem
(
e
q
u
ati
o
n
s
(
3
)
,
(
4
)
,
an
d
(
5
)
an
d
Alg
o
r
i
th
m
2
)
ar
e
s
u
m
m
ar
ized
in
T
ab
le
2
.
B
o
th
an
d
r
ep
r
esen
t
tr
ain
in
g
s
a
m
p
les,
wh
er
e
(
)
an
d
(
)
d
en
o
te
t
h
eir
n
o
n
lin
ea
r
tr
a
n
s
f
o
r
m
atio
n
s
in
t
o
a
h
ig
h
er
-
d
im
en
s
io
n
al
f
ea
tu
r
e
s
p
ac
e.
T
h
e
k
er
n
el
f
u
n
ctio
n
(
,
)
co
m
p
u
tes
th
eir
s
im
ilar
ity
in
th
is
s
p
ac
e
with
o
u
t e
x
p
licitly
ev
al
u
atin
g
(
⋅
)
.
T
ab
le
2
.
Def
in
itio
n
s
o
f
v
a
r
iab
l
es
in
SVM
m
o
d
el
S
y
mb
o
l
D
e
f
i
n
i
t
i
o
n
C
o
n
t
e
x
t
i
n
b
a
t
t
e
r
y
S
O
H
/
R
U
L
e
s
t
i
mat
i
o
n
I
n
p
u
t
f
e
a
t
u
r
e
v
e
c
t
o
r
o
f
t
h
e
ℎ
t
r
a
i
n
i
n
g
sa
mp
l
e
M
e
a
su
r
e
d
b
a
t
t
e
r
y
d
a
t
a
(
v
o
l
t
a
g
e
,
c
u
r
r
e
n
t
,
t
e
mp
e
r
a
t
u
r
e
,
a
n
d
c
y
c
l
e
c
o
u
n
t
)
I
n
p
u
t
f
e
a
t
u
r
e
v
e
c
t
o
r
o
f
t
h
e
ℎ
t
r
a
i
n
i
n
g
sa
mp
l
e
U
sed
w
h
e
n
c
o
m
p
u
t
i
n
g
p
a
i
r
w
i
s
e
si
mi
l
a
r
i
t
i
e
s i
n
k
e
r
n
e
l
sp
a
c
e
O
u
t
p
u
t
l
a
b
e
l
o
f
t
h
e
ℎ
samp
l
e
Ta
r
g
e
t
v
a
r
i
a
b
l
e
(
e
.
g
.
,
S
O
H
c
l
a
s
s,
R
U
L
v
a
l
u
e
)
W
e
i
g
h
t
v
e
c
t
o
r
(
mo
d
e
l
p
a
r
a
me
t
e
r
s)
D
e
f
i
n
e
s t
h
e
se
p
a
r
a
t
i
n
g
h
y
p
e
r
p
l
a
n
e
i
n
f
e
a
t
u
r
e
sp
a
c
e
B
i
a
s
t
e
r
m
S
h
i
f
t
s
t
h
e
d
e
c
i
si
o
n
b
o
u
n
d
a
r
y
f
r
o
m t
h
e
o
r
i
g
i
n
S
l
a
c
k
v
a
r
i
a
b
l
e
f
o
r
t
h
e
ℎ
sam
p
l
e
A
l
l
o
w
s
mar
g
i
n
v
i
o
l
a
t
i
o
n
s
/
mi
s
c
l
a
ssi
f
i
c
a
t
i
o
n
t
o
l
e
r
a
n
c
e
R
e
g
u
l
a
r
i
z
a
t
i
o
n
(
p
e
n
a
l
t
y
)
p
a
r
a
me
t
e
r
B
a
l
a
n
c
e
s
mar
g
i
n
m
a
x
i
mi
z
a
t
i
o
n
a
n
d
m
i
scl
a
ssi
f
i
c
a
t
i
o
n
(
)
N
o
n
l
i
n
e
a
r
f
e
a
t
u
r
e
m
a
p
p
i
n
g
o
f
i
n
p
u
t
Tr
a
n
sf
o
r
ms e
a
c
h
r
a
w
i
n
p
u
t
i
n
t
o
a
h
i
g
h
e
r
-
d
i
m
e
n
s
i
o
n
a
l
sp
a
c
e
(
)
N
o
n
l
i
n
e
a
r
f
e
a
t
u
r
e
m
a
p
p
i
n
g
o
f
i
n
p
u
t
U
sed
t
o
g
e
t
h
e
r
w
i
t
h
(
)
w
h
e
n
c
o
m
p
u
t
i
n
g
k
e
r
n
e
l
s
(
,
)
K
e
r
n
e
l
f
u
n
c
t
i
o
n
d
e
f
i
n
e
d
a
s
(
,
)
=
(
)
⋅
(
)
C
o
m
p
u
t
e
s si
mi
l
a
r
i
t
y
i
n
t
r
a
n
sf
o
r
me
d
s
p
a
c
e
w
i
t
h
o
u
t
e
x
p
l
i
c
i
t
ma
p
p
i
n
g
N
u
mb
e
r
o
f
t
r
a
i
n
i
n
g
sam
p
l
e
s
S
i
z
e
o
f
t
h
e
a
v
a
i
l
a
b
l
e
b
a
t
t
e
r
y
d
a
t
a
s
e
t
La
g
r
a
n
g
e
mu
l
t
i
p
l
i
e
r
f
o
r
t
h
e
ℎ
t
r
a
i
n
i
n
g
s
a
mp
l
e
(
n
o
n
-
z
e
r
o
o
n
l
y
f
o
r
s
u
p
p
o
r
t
v
e
c
t
o
r
s)
W
e
i
g
h
t
a
ssi
g
n
e
d
t
o
s
u
p
p
o
r
t
v
e
c
t
o
r
i
n
d
e
c
i
si
o
n
b
o
u
n
d
a
r
y
c
a
l
c
u
l
a
t
i
o
n
4
.
6
.
Rele
v
a
nce
v
ec
t
o
r
ma
chines
(
RVM)
Alth
o
u
g
h
it
d
o
es
n
o
t
n
ec
ess
itate
a
th
o
r
o
u
g
h
u
n
d
e
r
s
tan
d
in
g
o
f
th
e
b
atter
y
'
s
co
m
p
lex
elec
tr
o
ch
em
ical
p
r
o
ce
s
s
es,
th
e
d
ata
-
d
r
iv
e
n
ap
p
r
o
ac
h
is
u
n
s
u
cc
ess
f
u
l
f
o
r
s
tate
o
f
ch
ar
g
e
(
SOC
)
f
lu
ctu
atio
n
s
in
d
u
ce
d
b
y
f
ast
ch
an
g
es
in
b
atter
y
cu
r
r
e
n
t.
C
u
r
r
en
tly
,
th
e
m
ajo
r
ity
o
f
d
at
a
-
d
r
iv
en
f
ilter
in
g
r
esear
c
h
f
o
c
u
s
es
o
n
n
eu
r
al
n
etwo
r
k
-
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
Ma
ch
in
e
lea
r
n
in
g
-
d
r
iven
p
r
o
g
n
o
s
tics
fo
r
lith
iu
m
-
io
n
b
a
tter
ies:
e
n
h
a
n
cin
g
…
(
B
o
d
a
p
a
ti V
e
n
ka
ta
R
a
j
a
n
n
a
)
265
b
ased
alg
o
r
ith
m
s
,
with
litt
le
atten
tio
n
p
aid
to
th
e
r
ele
v
an
ce
v
ec
to
r
m
ac
h
in
e
(
R
VM
)
alg
o
r
i
th
m
,
wh
ich
h
as
th
e
ad
v
an
tag
e
o
f
b
ein
g
a
b
le
to
m
a
n
ag
e
s
m
all
s
am
p
le
s
izes,
h
ig
h
s
p
ar
s
ity
,
an
d
p
r
o
b
a
b
ilis
tic
d
is
tr
ib
u
tio
n
s
.
Mo
tiv
ated
b
y
th
is
ass
u
m
p
tio
n
,
th
e
cu
r
r
e
n
t
p
ap
e
r
p
r
esen
ts
a
n
o
v
el
f
u
s
io
n
s
tr
ateg
y
,
th
e
C
o
u
lo
m
b
_
A
KF_
OI
R
VM
(
C
F
R
)
alg
o
r
ith
m
,
wh
ich
i
n
tr
icate
ly
co
m
b
in
es th
e
o
p
tim
ized
i
n
cr
em
en
tal
R
VM
(
OI
R
V
M)
with
th
e
C
o
u
lo
m
b
co
u
n
tin
g
tech
n
iq
u
e
[
2
4
]
.
Acc
o
r
d
in
g
to
t
h
e
test
r
esu
lts
,
th
e
C
FR
tech
n
iq
u
e
h
as
a
r
em
ar
k
ab
le
a
b
ilit
y
to
g
en
er
alize
ac
r
o
s
s
a
wid
e
r
an
g
e
o
f
o
p
er
atin
g
cy
cles
an
d
tem
p
er
atu
r
es,
d
em
o
n
s
tr
at
in
g
g
r
ea
t e
s
tim
atio
n
ac
cu
r
ac
y
an
d
d
u
r
ab
ilit
y
.
T
h
e
ex
p
er
im
en
tal
a
n
d
p
u
b
lic
d
atasets
h
ad
m
ea
n
ab
s
o
lu
te
er
r
o
r
,
m
ax
im
u
m
a
b
s
o
lu
te
er
r
o
r
,
d
eter
m
in
atio
n
co
e
f
f
icien
t
,
an
d
av
er
a
g
e
r
o
o
t m
ea
n
s
q
u
ar
e
er
r
o
r
o
f
0
.
8
5
%,
0
.
2
5
%,
0
.
2
%,
an
d
0
.
2
%,
r
esp
ec
tiv
ely
.
4
.
7
.
Dee
p
neura
l net
wo
rk
s
(
D
N
N
)
To
u
s
e
lith
iu
m
-
io
n
b
atter
ies
s
u
cc
ess
f
u
lly
an
d
s
af
ely
,
th
e
s
ta
te
o
f
h
ea
lth
(
SOH)
an
d
r
em
ai
n
in
g
u
s
ab
le
life
(
R
UL
)
m
u
s
t
b
e
ac
c
u
r
ately
ev
alu
ated
.
B
ec
au
s
e
o
f
th
eir
g
o
o
d
n
o
n
lin
ea
r
f
itti
n
g
,
ad
a
p
tab
ilit
y
,
an
d
s
elf
-
lear
n
in
g
ca
p
ab
ilit
ies,
n
eu
r
al
n
etwo
r
k
s
a
r
e
co
m
m
o
n
ly
em
p
lo
y
ed
to
esti
m
ate
th
e
SOH
an
d
R
UL
o
f
lith
iu
m
-
io
n
b
atter
ies
.
C
o
n
v
o
lu
tio
n
al
n
e
u
r
al
n
etwo
r
k
s
(
C
NN)
ar
e
ef
f
icien
t
at
ex
tr
ac
tin
g
k
ey
f
ea
tu
r
es,
wh
er
ea
s
r
ec
u
r
r
en
t
n
eu
r
al
n
etwo
r
k
s
(
R
NN)
ar
e
ex
ce
llen
t
at
r
ec
o
r
d
in
g
tim
e
in
f
o
r
m
atio
n
in
s
er
ies
d
ata;
h
o
wev
er
,
b
ec
a
u
s
e
C
NN
lack
s
th
e
ab
ilit
y
to
lear
n
tim
e
ass
o
ciatio
n
s
,
tr
ad
itio
n
al
s
er
ies
m
o
d
el
s
th
at
co
m
b
in
e
C
NN
an
d
R
NN
m
ay
r
esu
lt
in
in
f
o
r
m
atio
n
leak
a
g
e.
I
n
th
is
p
ap
er
,
we
p
r
o
v
id
e
a
u
n
iq
u
e
p
a
r
allel
ar
ch
itectu
r
e
th
at
co
m
b
i
n
es
lo
n
g
s
h
o
r
t
-
ter
m
m
em
o
r
y
(
L
STM
)
an
d
tem
p
o
r
a
l c
o
n
v
o
l
u
tio
n
al
n
etwo
r
k
s
.
W
h
en
co
m
p
a
r
ed
to
s
eq
u
en
ti
al
C
NN
-
R
NN
tech
n
iq
u
es,
th
is
s
y
n
er
g
is
tic
ar
ch
itectu
r
e
c
o
n
s
id
er
ab
ly
im
p
r
o
v
es
t
h
e
ac
cu
r
ac
y
o
f
SOH
an
d
R
UL
esti
m
atio
n
f
o
r
lit
h
iu
m
-
io
n
b
atter
ies
b
y
allo
win
g
th
e
s
im
u
ltan
eo
u
s
ex
tr
ac
tio
n
o
f
s
p
atial
d
eg
r
a
d
at
io
n
f
ea
tu
r
es
v
ia
T
C
N
an
d
d
y
n
am
ic
tem
p
o
r
al
d
ep
e
n
d
en
cies
v
ia
L
STM
.
R
UL
esti
m
atio
n
co
m
p
ar
is
o
n
,
we
ev
alu
ate
th
e
s
er
ies
an
d
p
ar
allel
m
o
d
els
o
n
p
u
b
licly
av
ailab
l
e
b
atter
y
d
ata
f
r
o
m
Ox
f
o
r
d
Un
iv
er
s
ity
an
d
th
e
Natio
n
al
Aer
o
n
au
tics
an
d
Sp
ac
e
Ad
m
in
is
tr
atio
n
(
NASA)
.
I
n
d
a
tasets
1
an
d
2
,
SOH
esti
m
ati
o
n
p
r
ec
is
io
n
im
p
r
o
v
e
d
to
tally
b
y
m
o
r
e
t
h
an
2
9
%
an
d
3
7
%
r
esp
ec
tiv
ely
.
Par
all
el
m
o
d
els,
with
th
e
ass
u
m
p
tio
n
th
at
an
ac
ce
p
tab
le
in
cr
em
en
t
in
tim
e
co
n
s
u
m
p
tio
n
ca
n
b
e
allo
wed
,
m
ay
b
e
as
ef
f
ec
tiv
e
as
s
er
ies
m
o
d
els f
o
r
R
UL
esti
m
atio
n
[
2
5
]
.
E
v
en
th
o
u
g
h
DNNs
ca
n
lear
n
co
m
p
lex
n
o
n
lin
ea
r
r
elatio
n
s
h
ip
s
,
th
ey
o
f
ten
f
ac
e
p
r
o
b
le
m
s
wh
ile
p
r
o
ce
s
s
in
g
tem
p
o
r
al
s
eq
u
en
ce
s
d
u
e
to
is
s
u
es
o
f
v
an
is
h
in
g
g
r
ad
ien
ts
.
T
o
g
et
ar
o
u
n
d
th
is
li
m
itatio
n
,
lo
n
g
s
h
o
r
t
-
ter
m
m
em
o
r
y
u
n
its
o
f
th
e
r
ec
u
r
r
en
t
n
eu
r
al
n
etwo
r
k
s
wer
e
u
s
ed
in
th
is
wo
r
k
.
th
e
lo
n
g
s
h
o
r
t
-
ter
m
m
em
o
r
y
n
etwo
r
k
is
a
p
p
r
o
p
r
iate
f
o
r
s
tate
o
f
h
ea
lth
an
d
r
em
ai
n
in
g
u
s
ef
u
l
life
esti
m
atio
n
o
f
th
e
b
atter
i
es,
co
n
s
id
er
in
g
th
eir
v
ar
io
u
s
tem
p
o
r
al
r
elatio
n
s
h
ip
s
in
th
e
d
eter
io
r
atio
n
p
r
o
ce
s
s
es.
4
.
8
.
L
o
ng
s
ho
rt
-
t
er
m
m
emo
ry
(
L
ST
M
)
net
wo
rk
s
As
in
s
ec
tio
n
f
o
u
r
,
we
f
o
cu
s
o
n
d
ee
p
n
eu
r
al
n
etwo
r
k
s
.
Sev
e
n
ex
ce
l
at
m
o
d
ellin
g
co
m
p
lex
n
o
n
lin
ea
r
r
elatio
n
s
h
ip
s
b
u
t
o
f
ten
f
in
d
it
d
if
f
icu
lt
to
m
o
d
el
s
eq
u
en
ce
s
d
u
e
to
th
e
p
r
o
b
lem
c
alled
v
a
n
is
h
in
g
g
r
ad
ien
ts
,
wh
ic
h
im
p
air
s
th
eir
a
b
ilit
y
to
lea
r
n
l
o
n
g
-
ter
m
d
ep
e
n
d
en
cies.
T
o
o
v
e
r
co
m
e
th
is
p
r
o
b
lem
,
r
esear
ch
er
s
d
ev
elo
p
e
d
R
NNs;
h
o
wev
er
,
th
ese
s
tan
d
ar
d
m
o
d
e
ls
s
u
f
f
er
f
r
o
m
u
n
s
tab
le
g
r
ad
ien
t
is
s
u
es
d
u
r
in
g
lea
r
n
in
g
.
L
o
n
g
s
h
o
r
t
-
ter
m
m
e
m
o
r
y
n
etwo
r
k
s
ar
e
a
n
ex
ten
s
io
n
o
f
R
NNs
wh
er
e
m
em
o
r
y
ce
lls
an
d
g
atin
g
m
ec
h
an
is
m
s
allo
w
th
e
n
etwo
r
k
t
o
ch
o
o
s
e
wh
eth
er
o
r
n
o
t
to
k
ee
p
in
f
o
r
m
a
tio
n
f
r
o
m
p
r
e
v
io
u
s
s
tep
s
.
T
h
is
m
ak
es
L
STM
s
p
ar
ticu
lar
ly
we
ll
-
s
u
ited
f
o
r
b
atter
y
s
tate
o
f
h
ea
lth
(
SOH)
an
d
r
e
m
ain
in
g
u
s
ef
u
l
life
(
R
UL
)
p
r
ed
ictio
n
,
wh
e
r
e
b
o
th
s
h
o
r
t
-
ter
m
f
lu
ctu
atio
n
s
(
e.
g
.
,
ch
ar
g
e/d
is
ch
ar
g
e
cy
cles)
an
d
lo
n
g
-
ter
m
d
eg
r
a
d
atio
n
tr
e
n
d
s
m
u
s
t
b
e
m
o
d
eled
s
im
u
ltan
eo
u
s
ly
.
T
h
e
g
o
v
er
n
i
n
g
eq
u
atio
n
s
o
f
t
h
e
L
STM
ce
ll
ar
e
p
r
esen
ted
in
(
6
)
-
(
1
4
)
,
f
o
llo
w
ed
b
y
th
e
alg
o
r
ith
m
ic
im
p
lem
e
n
tatio
n
as
s
h
o
wn
in
Alg
o
r
ith
m
3
a
n
d
d
etailed
v
ar
ia
b
le
d
ef
in
itio
n
s
as sh
o
wn
in
T
a
b
le
3
.
=
(
[
ℎ
−
1
,
]
+
)
(
6
)
=
(
[
ℎ
−
1
,
]
+
)
(
7
)
C
t
̃
=
ta
n
h
(
[
ℎ
−
1
,
]
+
)
(
8
)
=
⊙
−
1
+
⊙
C
t
̃
(
9
)
=
(
[
ℎ
−
1
,
]
+
)
(
1
0
)
ℎ
=
⊙
ta
n
h
(
)
(
1
1
)
̂
=
(
ℎ
;
,
)
(
1
2
)
̂
=
ℎ
+
(
1
3
)
̂
=
(
ℎ
+
)
(
1
4
)
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
.
1
,
Ma
r
ch
20
2
6
:
257
-
2
7
4
266
Alg
o
r
ith
m
3
.
L
STM
p
r
o
ce
d
u
r
e
f
o
r
b
atter
y
SOH/R
UL
p
r
ed
ictio
n
1
.
I
n
p
u
t tim
e
-
s
er
ies tr
ain
in
g
d
ata
{
,
}
f
o
r
=
1
,
…,
T
2
.
I
n
itialize
L
STM
p
ar
am
eter
s
:
weig
h
ts
{
,
,
,
,
},
b
iases
{
,
,
,
,
},
h
id
d
en
s
tate
ℎ
0
000
,
an
d
ce
ll st
ate
0
3
.
Fo
r
ea
ch
tim
e
s
tep
=
1
,
…,
T
:
a.
C
o
m
p
u
te
f
o
r
g
et
g
ate:
=
(
[
ℎ
−
1
,
]
+
)
b
.
C
o
m
p
u
te
in
p
u
t g
ate:
=
(
[
ℎ
−
1
,
]
+
)
c.
C
o
m
p
u
te
ca
n
d
id
ate
ce
ll st
ate:
C
t
̃
=
ta
n
h
(
[
ℎ
−
1
,
]
+
)
d
.
Up
d
ate
ce
ll st
ate:
=
⊙
−
1
+
⊙
C
t
̃
e.
C
o
m
p
u
te
o
u
tp
u
t g
ate:
=
(
[
ℎ
−
1
,
]
+
)
f
.
Up
d
ate
h
id
d
e
n
s
tate:
ℎ
=
⊙
ta
n
h
(
)
4
.
Af
ter
f
in
al
tim
e
s
tep
,
u
s
e
h
i
d
d
en
s
tate
ℎ
f
o
r
p
r
e
d
ictio
n
̂
=
(
ℎ
;
,
)
Fo
r
r
eg
r
ess
io
n
(
SOH/R
UL
)
:
̂
=
ℎ
+
Fo
r
class
if
icatio
n
:
̂
=
(
ℎ
+
)
5
.
T
r
ain
t
h
e
n
etwo
r
k
b
y
m
in
im
izin
g
lo
s
s
f
u
n
ctio
n
L
(
,
̂
y
)
b
etwe
en
p
r
ed
icted
o
u
tp
u
t
̂
an
d
g
r
o
u
n
d
tr
u
th
–
Me
an
s
q
u
ar
e
d
er
r
o
r
(
MSE
)
f
o
r
r
eg
r
ess
io
n
–
C
r
o
s
s
-
e
n
tr
o
p
y
l
o
s
s
f
o
r
class
if
icatio
n
Up
d
ate
weig
h
ts
an
d
b
iases
u
s
in
g
b
ac
k
p
r
o
p
a
g
atio
n
t
h
r
o
u
g
h
tim
e
(
B
PTT
)
with
o
p
tim
izer
(
e.
g
.
,
Ad
a
m
,
SGD)
6
.
Fo
r
in
f
e
r
en
ce
,
f
ee
d
n
ew
in
p
u
t seq
u
en
ce
{
}
in
to
tr
ain
ed
L
STM
to
p
r
ed
ict
SOH
o
r
R
UL
I
n
s
u
m
m
a
r
y
,
L
STM
n
etwo
r
k
s
o
f
f
er
s
ig
n
if
ican
t
ad
v
a
n
tag
es
o
v
er
c
o
n
v
e
n
tio
n
al
ANN
an
d
DNN
ar
ch
itectu
r
es
wh
en
a
p
p
lied
to
b
atter
y
p
r
o
g
n
o
s
tics
.
C
ellu
lar
s
tate
an
d
co
n
t
r
o
l
p
r
o
ce
s
s
es
in
f
o
r
m
th
e
u
s
e
o
f
L
o
n
g
Sh
o
r
t
-
T
er
m
Me
m
o
r
y
n
etwo
r
k
s
in
s
p
o
ttin
g
f
lu
ctu
atio
n
s
o
v
er
s
m
all
tim
escales
co
u
p
led
with
lo
n
g
er
-
s
ca
le
p
atter
n
s
in
d
icativ
e
o
f
eq
u
ip
m
en
t
d
eter
i
o
r
atio
n
,
wh
ich
c
o
n
s
titu
te
th
e
m
ain
b
asis
f
o
r
p
r
ec
is
e
SOH
an
d
R
UL
esti
m
atio
n
s
.
Un
lik
e
s
tatic
m
o
d
els th
at
r
ely
o
n
h
an
d
-
d
esig
n
ed
f
ea
t
u
r
es,
L
STM
n
etwo
r
k
s
in
h
er
en
tly
lear
n
s
eq
u
en
tial
p
atter
n
s
f
r
o
m
th
e
d
ata
with
o
u
t
ex
p
lici
t
p
r
ep
r
o
ce
s
s
in
g
,
f
u
r
t
h
er
im
p
r
o
v
in
g
r
o
b
u
s
tn
ess
to
p
er
tu
r
b
at
io
n
s
an
d
n
o
n
lin
ea
r
d
y
n
a
m
i
c
s
.
T
h
i
s
i
s
w
h
a
t
e
n
a
b
l
es
g
e
n
e
r
a
l
i
z
a
ti
o
n
u
n
d
e
r
v
a
r
i
a
n
t
c
y
c
l
i
n
g
c
o
n
d
i
t
i
o
n
s
,
m
a
k
i
n
g
t
h
e
m
s
u
i
t
a
b
le
f
o
r
p
r
a
c
t
i
c
al
a
p
p
l
i
c
a
t
i
o
n
s
o
f
b
a
t
te
r
y
m
a
n
a
g
em
e
n
t
s
y
s
t
e
m
s
.
H
o
w
e
v
e
r
,
c
h
a
l
l
en
g
e
s
s
u
c
h
as
h
i
g
h
c
o
m
p
u
t
at
i
o
n
a
l
c
o
s
t
a
n
d
t
r
a
i
n
i
n
g
tim
e
r
em
ain
,
m
o
ti
v
atin
g
th
e
e
x
p
lo
r
atio
n
o
f
lig
h
tweig
h
t L
ST
M
v
ar
ian
ts
an
d
h
y
b
r
i
d
ap
p
r
o
ac
h
es in
f
u
tu
r
e
wo
r
k
.
T
ab
le
3
.
Def
in
itio
n
s
o
f
v
a
r
iab
l
es
in
L
STM
m
o
d
el
S
y
mb
o
l
D
e
f
i
n
i
t
i
o
n
C
o
n
t
e
x
t
i
n
b
a
t
t
e
r
y
S
O
H
/
R
U
L
e
s
t
i
mat
i
o
n
I
n
p
u
t
v
e
c
t
o
r
a
t
t
i
me
s
t
e
p
B
a
t
t
e
r
y
m
e
a
s
u
r
e
m
e
n
t
d
a
t
a
a
t
t
i
me
(
v
o
l
t
a
g
e
,
c
u
r
r
e
n
t
,
t
e
m
p
e
r
a
t
u
r
e
,
a
n
d
c
y
c
l
e
s)
ℎ
−
1
H
i
d
d
e
n
s
t
a
t
e
f
r
o
m
p
r
e
v
i
o
u
s
t
i
me
st
e
p
−
1
C
a
p
t
u
r
e
s
t
e
m
p
o
r
a
l
d
e
p
e
n
d
e
n
c
i
e
s fr
o
m
p
a
s
t
b
a
t
t
e
r
y
st
a
t
e
s
ℎ
H
i
d
d
e
n
s
t
a
t
e
a
t
c
u
r
r
e
n
t
t
i
me
st
e
p
O
u
t
p
u
t
f
e
a
t
u
r
e
r
e
p
r
e
se
n
t
a
t
i
o
n
f
o
r
p
r
e
d
i
c
t
i
o
n
(
S
O
H
o
r
R
U
L)
−
1
C
e
l
l
s
t
a
t
e
f
r
o
m
p
r
e
v
i
o
u
s t
i
me
st
e
p
−
1
Lo
n
g
-
t
e
r
m m
e
m
o
r
y
c
a
r
r
y
i
n
g
d
e
g
r
a
d
a
t
i
o
n
h
i
s
t
o
r
y
U
p
d
a
t
e
d
c
e
l
l
st
a
t
e
a
t
t
i
me
st
e
p
S
t
o
r
e
s
l
o
n
g
-
t
e
r
m
d
e
g
r
a
d
a
t
i
o
n
f
e
a
t
u
r
e
s
C
t
̃
C
a
n
d
i
d
a
t
e
c
e
l
l
s
t
a
t
e
(
p
r
o
p
o
se
d
u
p
d
a
t
e
b
e
f
o
r
e
g
a
t
i
n
g
)
R
e
p
r
e
se
n
t
s
n
e
w
d
e
g
r
a
d
a
t
i
o
n
i
n
f
o
r
m
a
t
i
o
n
t
o
b
e
i
n
t
e
g
r
a
t
e
d
i
n
t
o
memo
r
y
F
o
r
g
e
t
g
a
t
e
a
c
t
i
v
a
t
i
o
n
v
e
c
t
o
r
C
o
n
t
r
o
l
s
w
h
a
t
f
r
a
c
t
i
o
n
o
f
p
a
s
t
c
e
l
l
s
t
a
t
e
i
s
r
e
t
a
i
n
e
d
I
n
p
u
t
g
a
t
e
a
c
t
i
v
a
t
i
o
n
v
e
c
t
o
r
C
o
n
t
r
o
l
s
h
o
w
m
u
c
h
n
e
w
i
n
p
u
t
i
n
f
o
r
mat
i
o
n
e
n
t
e
r
s
t
h
e
c
e
l
l
O
u
t
p
u
t
g
a
t
e
a
c
t
i
v
a
t
i
o
n
v
e
c
t
o
r
C
o
n
t
r
o
l
s
h
o
w
m
u
c
h
o
f
t
h
e
c
e
l
l
s
t
a
t
e
i
s
e
x
p
o
se
d
t
o
h
i
d
d
e
n
st
a
t
e
̂
P
r
e
d
i
c
t
e
d
o
u
t
p
u
t
a
t
t
i
me
st
e
p
,
c
o
m
p
u
t
e
d
f
r
o
m
h
i
d
d
e
n
st
a
t
e
ℎ
Est
i
m
a
t
e
d
S
O
H
(
%)
o
r
R
U
L
(
r
e
ma
i
n
i
n
g
c
y
c
l
e
s)
G
r
o
u
n
d
-
t
r
u
t
h
l
a
b
e
l
a
t
t
i
me
s
t
e
p
A
c
t
u
a
l
S
O
H
o
r
R
U
L
f
r
o
m
d
a
t
a
s
e
t
,
,
,
W
e
i
g
h
t
m
a
t
r
i
c
e
s
f
o
r
f
o
r
g
e
t
,
i
n
p
u
t
,
o
u
t
p
u
t
,
a
n
d
c
e
l
l
g
a
t
e
s
Le
a
r
n
e
d
p
a
r
a
m
e
t
e
r
s ma
p
p
i
n
g
[
ℎ
−
1
,
]
t
o
g
a
t
e
a
c
t
i
v
a
t
i
o
n
s
W
e
i
g
h
t
m
a
t
r
i
x
o
f
f
i
n
a
l
o
u
t
p
u
t
l
a
y
e
r
M
a
p
s
h
i
d
d
e
n
s
t
a
t
e
ℎ
t
o
p
r
e
d
i
c
t
i
o
n
̂
,
,
,
B
i
a
s
v
e
c
t
o
r
s
f
o
r
f
o
r
g
e
t
,
i
n
p
u
t
,
o
u
t
p
u
t
,
a
n
d
c
e
l
l
g
a
t
e
s
Le
a
r
n
e
d
p
a
r
a
m
e
t
e
r
s t
h
a
t
s
h
i
f
t
a
c
t
i
v
a
t
i
o
n
s
B
i
a
s
v
e
c
t
o
r
o
f
f
i
n
a
l
o
u
t
p
u
t
l
a
y
e
r
S
h
i
f
t
s
f
i
n
a
l
p
r
e
d
i
c
t
i
o
n
o
u
t
p
u
t
(
⋅
)
S
i
g
m
o
i
d
a
c
t
i
v
a
t
i
o
n
f
u
n
c
t
i
o
n
N
o
r
mal
i
z
e
s
g
a
t
e
o
u
t
p
u
t
s
b
e
t
w
e
e
n
0
a
n
d
1
t
a
n
h
(
⋅
)
H
y
p
e
r
b
o
l
i
c
t
a
n
g
e
n
t
a
c
t
i
v
a
t
i
o
n
f
u
n
c
t
i
o
n
S
c
a
l
e
s
v
a
l
u
e
s
b
e
t
w
e
e
n
–
1
a
n
d
1
f
o
r
n
o
n
l
i
n
e
a
r
i
t
y
⊙
El
e
m
e
n
t
-
w
i
se
m
u
l
t
i
p
l
i
c
a
t
i
o
n
C
o
m
b
i
n
e
s
g
a
t
e
a
c
t
i
v
a
t
i
o
n
s
w
i
t
h
st
a
t
e
s
5.
DATAS
E
T
S AN
D
F
E
A
T
UR
E
E
NG
I
NE
E
RI
NG
Stan
d
ar
d
ized
d
atasets
p
lay
a
c
r
itical
r
o
le
in
en
a
b
lin
g
ML
-
b
a
s
ed
p
r
o
g
n
o
s
tics
an
d
h
ea
lth
m
a
n
ag
em
en
t
o
f
lith
iu
m
-
io
n
b
atter
ies.
Sev
er
al
wid
ely
u
s
ed
d
atasets
in
clu
d
e
NASA,
C
AL
C
E
,
Ox
f
o
r
d
,
an
d
MI
T
r
ep
o
s
ito
r
ies,
wh
ich
p
r
o
v
id
e
ex
p
er
im
en
tal
d
ata
ac
r
o
s
s
d
if
f
er
en
t c
h
e
m
is
tr
ies,
ch
ar
g
in
g
p
r
o
to
c
o
ls
,
an
d
d
eg
r
ad
atio
n
co
n
d
itio
n
s
.
L
ith
iu
m
-
io
n
a
n
d
lith
iu
m
co
b
a
lt
o
x
id
e
b
atter
ies
ar
e
ju
s
t
s
o
m
e
o
f
th
e
m
an
y
b
atter
ies
with
p
u
b
licly
ac
ce
s
s
ib
le
Evaluation Warning : The document was created with Spire.PDF for Python.