I
nte
rna
t
io
na
l J
o
urna
l o
f
P
o
w
er
E
lect
ro
nics
a
nd
Driv
e
Sy
s
t
e
m
(
I
J
P
E
DS
)
Vo
l.
9
,
No
.
4
,
Dec
em
b
er
2
0
1
8
,
p
p
.
1
4
8
6
~
1
5
0
2
I
SS
N:
2
0
8
8
-
8
6
9
4
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
p
ed
s
.
v9
.
i
4
.
pp
1
4
8
6
-
1
5
0
2
1486
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
co
r
e.
co
m/jo
u
r
n
a
ls
/in
d
ex
.
p
h
p
/
I
JP
E
DS
New
Versio
n
o
f
A
da
ptive Speed
O
b
serv
er
b
as
ed
o
n
N
eura
l
Netw
o
rk
f
o
r S
PI
M
Ng
o
c
T
hu
y
P
ha
m
1
,
Diep
P
hu
Ng
uy
en
2
,
K
hu
o
ng
H
uu
N
g
uy
en
3
,
Nho
Va
n Ng
uy
en
4
1,
2
De
p
ar
t
m
e
n
t
o
f
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
T
e
c
h
n
o
l
o
g
y
,
In
d
u
strial
Un
i
v
e
rsit
y
o
f
Ho
Ch
i
M
in
h
Cit
y
,
V
iet
n
a
m
3
De
p
ar
t
m
e
n
t
o
f
T
E
a
n
d
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
,
Ho
Ch
i
M
in
h
Cit
y
Un
iv
e
rsit
y
o
f
T
r
a
n
sp
o
rt
,
V
iet
n
a
m
4
De
p
ar
t
m
e
n
t
o
f
El
e
c
tri
c
a
l
a
n
d
El
e
c
tro
n
ics
En
g
i
n
e
e
rin
g
,
Un
iv
e
rsity
o
f
T
e
c
h
n
o
l
o
g
y
Ho
Ch
i
M
in
h
Cit
y
,
V
iet
n
a
m
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Ma
y
21
,
2
0
1
8
R
ev
i
s
ed
A
u
g
30
,
2
0
1
8
A
cc
ep
ted
Sep
13
,
2
0
1
8
T
h
is
p
a
p
e
r
p
re
se
n
ts
a
n
o
v
e
l
S
tato
r
Cu
rre
n
t
b
a
se
d
M
o
d
e
l
Re
f
e
r
e
n
c
e
A
d
a
p
ti
v
e
S
y
st
e
m
(S
C_
M
RA
S
)
sp
e
e
d
o
b
se
rv
e
r
f
o
r
h
ig
h
-
p
e
rf
o
rm
a
n
c
e
S
ix
P
h
a
se
s
In
d
u
c
ti
o
n
M
o
to
r
(S
P
I
M
)
d
r
iv
e
s
u
sin
g
li
n
e
a
r
n
e
u
ra
l
n
e
tw
o
rk
.
T
h
e
a
rti
c
le
a
im
is
in
ten
d
e
d
to
im
p
ro
v
e
p
e
rf
o
rm
a
n
c
e
o
f
a
n
S
C
_
M
RA
S
o
b
se
rv
e
r,
w
h
ich
w
e
r
e
p
re
se
n
ted
in
t
h
e
li
tera
tu
re
.
I
n
t
h
is
p
ro
p
o
se
d
sc
h
e
m
e
,
th
e
m
e
a
s
u
re
d
sta
to
r
c
u
rre
n
t
c
o
m
p
o
n
e
n
ts
a
re
u
se
d
a
s
t
h
e
re
f
e
re
n
c
e
m
o
d
e
l
o
f
th
e
M
RAS
o
b
se
rv
e
r
to
a
v
o
id
th
e
u
se
o
f
a
p
u
re
in
teg
ra
to
r
a
n
d
re
d
u
c
e
th
e
in
f
lu
e
n
c
e
o
f
m
o
to
r
p
a
ra
m
e
ter
v
a
riatio
n
.
T
h
e
a
d
a
p
ti
v
e
m
o
d
e
l
u
se
s
a
t
w
o
-
la
y
e
r
Ne
u
ra
l
Ne
t
w
o
rk
(NN
)
to
e
stim
a
te
th
e
sta
to
r
c
u
rre
n
t,
w
h
ich
h
a
s
b
e
e
n
train
e
d
o
n
li
n
e
b
y
m
e
a
n
s
o
f
a
Lea
st
S
q
u
a
re
s
(
L
S
)
a
l
g
o
rit
h
m
in
ste
a
d
o
f
u
se
s
a
n
o
n
li
n
e
a
r
Ba
c
k
P
r
o
p
a
g
a
ti
o
n
Ne
tw
o
rk
(BP
N)
a
lg
o
rit
h
m
to
re
d
u
c
e
th
e
c
o
m
p
lex
it
y
a
n
d
c
o
m
p
u
tatio
n
a
l
b
u
r
d
e
n
,
it
a
lso
h
e
l
p
t
o
im
p
ro
v
e
so
m
e
d
isa
d
v
a
n
tag
e
s
c
a
u
se
b
y
th
e
in
h
e
re
n
t
n
o
n
li
n
e
a
rit
y
o
f
t
h
e
B
P
N
a
lg
o
rit
h
m
a
s
lo
c
a
l
m
i
n
im
a
,
t
w
o
h
e
u
risti
c
a
ll
y
c
h
o
se
n
p
a
ra
m
e
ters
,
in
it
ializa
ti
o
n
,
a
n
d
c
o
n
v
e
rg
e
n
c
e
p
ro
b
lem
s,
p
a
ra
ly
sis
o
f
th
e
n
e
u
ra
l
n
e
t
w
o
rk
.
T
h
e
a
d
a
p
ti
v
e
m
o
d
e
l
o
f
th
e
p
ro
p
o
se
d
s
c
h
e
m
e
is
e
m
p
lo
y
e
d
in
p
re
d
ictio
n
m
o
d
e
,
n
o
t
i
n
sim
u
latio
n
m
o
d
e
a
s is
u
su
a
ll
y
th
e
c
a
s
e
in
th
e
li
tera
tu
re
,
t
h
is
m
a
d
e
th
e
p
ro
p
o
se
d
o
b
se
rv
e
r
o
p
e
ra
te
b
e
tt
e
r
a
c
c
u
ra
c
y
a
n
d
sta
b
il
it
y
.
In
t
h
e
p
r
o
p
o
se
d
o
b
se
r
v
e
r,
sta
to
r
a
n
d
ro
to
r
re
sista
n
c
e
v
a
lu
e
s
a
re
e
sti
m
a
ted
o
n
li
n
e
,
t
h
e
se
v
a
lu
e
s
th
e
re
a
f
ter
we
re
u
p
d
a
ted
f
o
r
th
e
c
u
rre
n
t
o
b
se
rv
e
r
a
n
d
ro
to
r
f
lu
x
id
e
n
ti
f
ier
to
e
n
h
a
n
c
e
th
e
a
c
c
u
ra
c
y
,
ro
b
u
stn
e
ss
a
n
d
in
se
n
siti
v
it
y
to
p
a
ra
m
e
ters
v
a
riatio
n
f
o
r
th
e
p
ro
p
o
se
d
o
b
se
rv
e
r.
T
h
e
p
ro
p
o
se
d
L
S
S
C_
M
RA
S
o
b
se
rv
e
r
h
a
s
b
e
e
n
v
e
ri
f
ied
th
o
u
g
h
t
t
h
e
sim
u
latio
n
a
n
d
c
o
m
p
a
re
d
w
it
h
th
e
B
P
N
M
RA
S
o
b
se
rv
e
r.
T
h
e
si
m
u
latio
n
re
su
l
ts
h
a
v
e
p
ro
v
e
n
th
a
t
th
e
sp
e
e
d
is
e
stim
a
t
e
d
a
c
o
n
se
q
u
e
n
t
q
u
ick
e
r
c
o
n
v
e
rg
e
n
c
e
,
d
o
n
o
t
n
e
e
d
th
e
e
stim
a
t
e
d
sp
e
e
d
f
il
ter,
lo
w
e
r
e
stim
a
ti
o
n
e
rro
rs b
o
th
in
tran
sie
n
t
a
n
d
ste
a
d
y
sta
te o
p
e
ra
ti
o
n
,
b
e
tt
e
r
b
e
h
a
v
io
r
in
lo
w
a
n
d
z
e
ro
sp
e
e
d
o
p
e
ra
ti
o
n
.
K
ey
w
o
r
d
:
Neu
r
al
n
et
w
o
r
k
Sen
s
o
r
les
s
v
ec
to
r
co
n
tr
o
l
Six
p
h
a
s
e
in
d
u
ctio
n
m
o
to
r
d
r
iv
e
MR
A
S o
b
s
er
v
er
Stato
r
cu
r
r
en
t b
ased
MR
AS
Co
p
y
rig
h
t
©
201
8
In
s
t
it
u
te o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
rin
g
a
n
d
S
c
ien
c
e
.
All
rig
h
ts
re
se
rv
e
d
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Ng
o
c
T
h
u
y
P
h
a
m
,
Dep
ar
t
m
en
t
o
f
E
lectr
ical
E
n
g
i
n
ee
r
in
g
T
ec
h
n
o
lo
g
y
,
I
n
d
u
s
tr
ial
U
n
i
v
er
s
it
y
o
f
Ho
C
h
i M
in
h
C
it
y
,
1
2
Ng
u
y
e
n
Va
n
B
ao
s
tr
ee
t,
w
a
r
d
4
,
H
o
C
h
i M
in
h
C
it
y
,
V
ietn
a
m
.
E
m
ail: N
g
o
cp
h
a
m
1
0
2
0
@
g
m
ai
l.c
o
m
1.
I
NT
RO
D
UCT
I
O
N
I
n
r
ec
en
t
d
ec
ad
es,
th
e
s
en
s
o
r
less
v
ec
to
r
co
n
tr
o
lled
SP
I
M
d
r
iv
es
ar
e
b
ein
g
v
ig
o
r
o
u
s
l
y
d
e
v
elo
p
ed
f
o
r
h
ig
h
p
er
f
o
r
m
a
n
ce
i
n
d
u
s
tr
ial
d
r
iv
e
s
y
s
te
m
s
.
T
h
e
m
a
in
ad
v
a
n
tag
e
s
o
f
t
h
e
s
e
n
s
o
r
les
s
SP
I
M
d
r
iv
es
ar
e
h
i
g
h
e
r
to
r
q
u
e
d
en
s
it
y
,
g
r
ea
ter
e
f
f
icie
n
c
y
,
r
ed
u
ce
d
to
r
q
u
e
p
u
ls
atio
n
s
,
f
a
u
lt
to
ler
an
ce
,
a
n
d
r
ed
u
ct
i
o
n
i
n
t
h
e
r
eq
u
ir
ed
r
atin
g
p
er
i
n
v
er
ter
le
g
[
1
]
an
d
its
r
eliab
le
w
o
r
k
in
g
c
h
ar
ac
t
er
is
tics
a
n
d
h
i
g
h
f
a
ilu
r
e
to
ler
an
ce
.
O
n
th
e
o
t
h
er
h
an
d
,
th
e
s
en
s
o
r
less
d
r
iv
e
h
av
e
h
i
g
h
r
eliab
ilit
y
an
d
m
ec
h
an
ica
l
r
o
b
u
s
tn
e
s
s
,
s
a
v
e
c
o
s
t.
T
h
e
d
if
f
er
en
t
tech
n
iq
u
es
f
o
r
th
e
s
p
ee
d
s
en
s
o
r
less
co
n
tr
o
l o
f
i
n
d
u
ctio
n
m
o
t
o
r
s
h
a
v
e
b
ee
n
p
r
o
p
o
s
ed
.
T
h
ey
u
s
u
all
y
ar
e
d
i
v
id
ed
in
to
t
w
o
ca
te
g
o
r
ies,
th
e
f
u
n
d
a
m
en
tal
m
o
d
el
b
ased
o
b
s
er
v
er
s
an
d
an
is
o
tr
o
p
ies
m
o
d
el
b
ased
o
b
s
er
v
er
s
.
Mo
d
el
-
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
P
o
w
E
lec
&
Dr
i
S
y
s
t
I
SS
N:
2
0
8
8
-
8
694
N
ew V
er
s
io
n
o
f A
d
a
p
tive
S
p
ee
d
Ob
s
erver B
a
s
ed
o
n
N
eu
r
a
l
N
etw
o
r
k
fo
r
S
P
I
M (
N
g
o
c
Th
u
y
P
h
a
m)
1487
b
ased
esti
m
atio
n
s
tr
ate
g
ies
i
n
clu
d
e
o
p
en
-
lo
o
p
o
b
s
er
v
er
s
[
2
]
,
s
lid
in
g
-
m
o
d
e
o
b
s
er
v
er
s
[
3
]
,
E
x
te
n
d
ed
Kal
m
an
Fil
ter
[
4
]
,
B
ac
k
s
tep
p
in
g
[
5
]
,
m
o
d
el
r
e
f
er
en
ce
ad
ap
ti
v
e
s
y
s
t
e
m
s
(
M
R
A
S)
[
6
]
an
d
ar
tific
i
al
in
tell
ig
e
n
ce
(
A
I
)
[
7
]
.
R
ec
en
t
r
esear
ch
also
u
s
ed
p
r
ed
ictiv
e
cu
r
r
en
t
co
n
tr
o
l
f
o
r
s
en
s
o
r
les
s
I
M
d
r
iv
es
[
8
]
.
Sen
s
o
r
less
d
r
iv
es
h
av
e
b
ee
n
s
u
cc
e
s
s
f
u
ll
y
ap
p
lied
in
m
ed
iu
m
a
n
d
h
i
g
h
s
p
ee
d
r
eg
io
n
s
[
9
]
,
[
1
0
]
,
b
u
t
lo
w
an
d
ze
r
o
s
p
ee
d
o
p
er
ati
o
n
is
s
till
a
lar
g
e
ch
al
len
g
e.
I
n
o
r
d
er
to
o
v
er
co
m
e
t
h
ese
p
r
o
b
le
m
s
a
h
ig
h
f
r
eq
u
e
n
c
y
v
o
lta
g
e
o
r
cu
r
r
en
t
ca
r
r
ier
w
er
e
in
j
ec
ted
,
n
ee
d
ed
to
ex
cite
th
e
s
alien
c
y
it
s
el
f
[
1
1
]
.
T
h
is
m
eth
o
d
w
o
r
k
s
w
e
ll
at
lo
w
a
n
d
n
ea
r
ze
r
o
s
p
ee
d
r
eg
io
n
.
Ho
w
e
v
er
,
t
h
eir
m
aj
o
r
d
is
ad
v
an
tag
e
s
ar
e
co
m
p
u
ta
tio
n
al
co
m
p
lex
it
y
,
t
h
e
n
ee
d
o
f
ex
ter
n
al
h
ar
d
w
ar
e
f
o
r
s
i
g
n
a
l
in
j
ec
tio
n
an
d
t
h
e
ad
v
er
s
e
e
f
f
ec
t o
f
in
j
ec
tin
g
s
i
g
n
al
o
n
t
h
e
m
a
ch
in
e
p
er
f
o
r
m
an
ce
.
D
u
e
to
its
s
i
m
p
lic
it
y
a
n
d
ea
s
e
o
f
i
m
p
le
m
e
n
tatio
n
t
h
e
m
o
d
el
b
ased
m
et
h
o
d
s
an
d
esp
ec
iall
y
MR
A
S
b
a
s
ed
m
et
h
o
d
s
ar
e,
u
n
til
n
o
w
,
t
h
e
m
o
s
t
w
id
el
y
u
s
ed
.
T
h
e
m
ai
n
p
r
o
b
lem
s
a
s
s
o
ciate
d
w
it
h
t
h
e
lo
w
s
p
ee
d
o
p
er
atio
n
o
f
m
o
d
el
-
b
ased
s
en
s
o
r
less
d
r
iv
e
s
ar
e
r
elate
d
to
m
ac
h
i
n
e
p
ar
a
m
eter
s
en
s
iti
v
it
y
,
s
tato
r
v
o
lta
g
e
an
d
cu
r
r
en
t
ac
q
u
is
itio
n
,
a
n
d
f
l
u
x
p
u
r
e
i
n
teg
r
atio
n
p
r
o
b
lem
s
[
12
]
-
[
15
]
.
Nu
m
er
o
u
s
M
R
A
S
h
av
e
b
ee
n
p
r
o
p
o
s
ed
.
Am
o
n
g
th
e
m
,
t
h
e
r
o
to
r
f
lu
x
MR
AS
f
ir
s
t
in
tr
o
d
u
ce
d
b
y
Sc
h
a
u
d
e
r
[
16
]
,
Flu
x
B
ac
k
s
tep
p
in
g
Ob
s
er
v
er
[
1
7
]
,
b
o
th
s
u
f
f
er
f
r
o
m
DC
d
r
if
t
p
r
o
b
lem
s
ass
o
ciate
d
w
it
h
p
u
r
e
in
teg
r
ati
o
n
an
d
s
e
n
s
it
iv
i
t
y
to
s
ta
to
r
r
esis
ta
n
ce
v
ar
iat
io
n
,
esp
ec
iall
y
in
t
h
e
lo
w
s
p
ee
d
r
eg
io
n
.
I
n
o
r
d
er
to
im
p
r
o
v
e
th
e
p
er
f
o
r
m
a
n
ce
o
f
o
b
s
er
v
er
o
v
er
co
m
e
th
e
e
f
f
ec
t
b
y
s
en
s
it
iv
i
t
y
to
s
tato
r
r
esis
ta
n
ce
v
ar
iatio
n
,
o
n
l
in
e
ad
ap
tatio
n
o
f
th
e
s
tato
r
r
esis
tan
c
e
[
1
8
]
,
th
e
p
u
r
e
in
teg
r
atio
n
p
r
o
b
lem
s
,
E
x
ten
d
ed
Kal
m
a
n
Fil
ter
(
E
K
F),
a
m
o
d
if
ied
to
r
q
u
e
b
ased
o
n
MR
AS
s
ch
e
m
e
s
h
a
v
e
p
r
o
p
o
s
ed
in
[
19
],
[
20
]
,
r
esp
ec
tiv
el
y
.
A
lt
h
o
u
g
h
[
19
],
[
20
]
h
av
e
s
h
o
w
n
t
h
at
th
e
s
e
ap
p
r
o
ac
h
es
s
ig
n
i
f
ican
tl
y
i
m
p
r
o
v
e
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
R
F
-
MR
A
S
at
lo
w
s
p
ee
d
,
th
ese
s
ch
e
m
e
r
e
m
a
in
t
h
e
e
f
f
ec
ted
b
y
th
e
s
en
s
iti
v
it
y
to
p
ar
a
m
e
ter
v
ar
iatio
n
s
.
An
im
p
r
o
v
ed
r
o
to
r
f
l
u
x
esti
m
ati
o
n
to
eli
m
i
n
ate
t
h
e
p
u
r
e
in
te
g
r
atio
n
p
r
o
b
le
m
s
an
d
th
e
ef
f
ec
t
o
f
s
en
s
iti
v
it
y
to
p
ar
am
eter
v
ar
iatio
n
s
f
o
r
a
T
o
r
q
u
e
MR
AS
is
p
r
o
p
o
s
ed
in
[
21
]
.
Si
m
u
latio
n
a
n
d
ex
p
er
i
m
e
n
ta
l
r
esu
lt
s
ar
e
s
h
o
w
n
th
e
s
e
n
s
o
r
les
s
co
n
tr
o
l
d
r
iv
e
o
p
er
atin
g
at
lo
w
a
n
d
ze
r
o
s
p
ee
d
s
,
w
it
h
b
o
th
m
o
to
r
in
g
a
n
d
r
eg
en
er
ativ
e
o
p
er
atio
n
s
co
n
s
id
er
ed
.
T
h
e
p
er
f
o
r
m
an
ce
o
f
th
e
o
b
s
er
v
er
i
n
lo
w
s
p
ee
d
r
eg
en
er
ati
n
g
r
e
g
io
n
a
n
d
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
tr
an
s
ie
n
t
a
n
d
s
tead
y
s
ta
te
w
e
r
e
s
ig
n
if
ican
t
l
y
i
m
p
r
o
v
ed
at
v
er
y
lo
w
a
n
d
ze
r
o
r
o
to
r
s
p
ee
d
s
.
An
al
y
s
i
s
o
f
t
h
e
ef
f
ec
t
o
f
p
ar
a
m
eter
v
ar
iat
io
n
o
n
th
e
s
c
h
e
m
e
p
e
r
f
o
r
m
a
n
ce
h
as
s
h
o
w
n
i
m
p
r
o
v
ed
r
o
b
u
s
t
n
es
s
ag
ai
n
s
t
s
tato
r
an
d
r
o
to
r
r
esis
tan
ce
v
ar
iatio
n
o
v
er
a
w
id
er
r
an
g
e
o
f
lo
ad
to
r
q
u
es
co
m
p
ar
ed
to
r
esu
lt
s
p
r
ev
io
u
s
l
y
p
u
b
li
s
h
ed
f
o
r
th
e
co
n
v
e
n
tio
n
al
s
c
h
e
m
e.
Ho
w
ev
er
,
th
e
est
i
m
a
ted
er
r
o
r
in
cr
ea
s
e
at
v
er
y
lo
w
(
3
.
1
4
r
ad
/s
)
an
d
ze
r
o
s
p
ee
d
r
an
g
e
i
s
r
ec
o
r
d
e
d
.
T
h
e
p
er
f
o
r
m
an
ce
o
f
th
e
s
p
ee
d
esti
m
atio
n
in
lo
w
s
p
ee
d
r
eg
en
er
atin
g
r
eg
io
n
an
d
th
e
p
er
f
o
r
m
a
n
ce
o
f
th
e
tr
an
s
ien
t a
n
d
s
tead
y
s
tate
i
s
n
o
t r
ea
ll
y
s
atis
f
ied
.
An
o
th
er
ap
p
r
o
ac
h
,
th
e
s
tato
r
cu
r
r
en
t
M
R
A
S
s
ch
e
m
e
h
as
b
e
en
i
n
tr
o
d
u
ce
d
i
n
[
22
]
-
[
23
]
.
[
23
]
p
r
esen
ts
a
s
tato
r
c
u
r
r
en
t
b
ased
MR
AS
s
p
ee
d
o
b
s
er
v
er
u
s
in
g
NN,
w
h
ic
h
i
s
a
n
e
v
o
lu
t
io
n
o
f
[
22
]
.
I
n
t
h
is
p
r
o
p
o
s
ed
s
ch
e
m
e,
to
a
v
o
id
t
h
e
e
f
f
ec
t
o
f
a
p
u
r
e
in
teg
r
ato
r
a
n
d
r
ed
u
ce
i
n
f
lu
e
n
ce
o
f
m
o
to
r
p
ar
am
eter
v
ar
iatio
n
s
,
t
h
e
m
ea
s
u
r
ed
s
ta
to
r
cu
r
r
en
t
co
m
p
o
n
en
t
s
ar
e
u
s
ed
a
s
th
e
r
ef
er
e
n
ce
m
o
d
el.
T
h
e
ad
ap
tiv
e
m
o
d
el
o
f
t
h
e
p
r
o
p
o
s
ed
o
b
s
er
v
er
in
[
23
]
u
s
e
s
a
t
w
o
-
la
y
er
NN
w
i
th
a
B
P
N
al
g
o
r
ith
m
to
esti
m
ate
t
h
e
r
o
to
r
s
p
ee
d
,
an
o
f
f
-
li
n
e
tr
ai
n
ed
m
u
ltil
a
y
er
f
ee
d
-
f
o
r
w
ar
d
n
e
u
r
a
l n
et
w
o
r
k
i
s
p
r
o
p
o
s
ed
as a
r
o
to
r
f
lu
x
o
b
s
er
v
er
.
T
h
e
s
i
m
u
la
ti
o
n
an
d
ex
p
er
i
m
e
n
tal
r
esu
lt
s
h
a
v
e
p
r
o
v
en
th
at
th
e
s
i
g
n
i
f
ica
n
tl
y
i
m
p
r
o
v
e
m
en
t
o
p
er
atio
n
p
er
f
o
r
m
a
n
ce
i
n
lo
w
an
d
ze
r
o
s
p
ee
d
r
an
g
es,
th
e
lo
w
e
s
t
s
p
ee
d
li
m
it
2
5
r
p
m
(
2
.
6
r
ad
/s
)
w
as
r
ep
o
r
ted
.
T
h
e
r
esu
lt
s
i
n
[
23
]
als
o
d
e
m
o
n
s
tr
at
e
th
at
t
h
e
p
r
o
p
o
s
ed
o
b
s
er
v
er
ca
n
h
a
n
d
le
t
h
e
p
ar
am
eter
v
ar
iatio
n
p
r
o
b
lem
w
it
h
a
g
o
o
d
lev
el
o
f
r
o
b
u
s
t
n
ess
,
s
e
n
s
o
r
les
s
p
er
f
o
r
m
a
n
c
e
w
it
h
a
5
0
%
v
ar
iat
io
n
i
n
r
e
s
is
t
an
ce
s
at
lo
w
s
p
ee
d
,
2
5
%
lo
ad
.
Alth
o
u
g
h
[
23
]
ca
n
o
v
er
co
m
e
th
e
m
ai
n
p
r
o
b
le
m
s
ass
o
ciate
d
w
it
h
th
e
lo
w
ze
r
o
an
d
s
p
ee
d
o
p
er
atio
n
,
h
o
w
e
v
er
,
d
u
e
to
[
23
]
th
e
u
s
e
o
f
th
e
n
o
n
li
n
ea
r
B
P
N
alg
o
r
ith
m
to
tr
ain
i
n
g
a
n
eu
r
al
n
et
w
o
r
k
ca
u
s
es
s
o
m
e
p
r
o
b
lem
as
lo
ca
l
m
in
i
m
a,
p
ar
al
y
s
is
o
f
th
e
n
e
u
r
al
n
et
w
o
r
k
,
n
ee
d
o
f
t
w
o
h
eu
r
i
s
ticall
y
ch
o
s
en
p
ar
a
m
eter
s
,
i
n
itia
lizati
o
n
p
r
o
b
lem
s
,
a
n
d
co
n
v
er
g
e
n
ce
p
r
o
b
le
m
s
.
T
h
ese
m
ak
e
th
e
p
er
f
o
r
m
a
n
ce
o
f
o
b
s
er
v
er
in
[
23
]
is
n
o
t
r
ea
ll
y
as
e
x
p
e
cted
.
T
h
e
s
p
ee
d
esti
m
atio
n
e
r
r
o
r
an
d
o
s
cillatio
n
p
h
en
o
m
e
n
o
n
at
lo
w
a
n
d
ze
r
o
in
cr
ea
s
e.
Ot
h
er
s
id
e,
t
h
e
ad
a
p
tiv
e
m
o
d
el
i
n
[
23
]
is
u
s
ed
i
n
s
i
m
u
lat
io
n
m
o
d
e,
w
h
ic
h
m
ea
n
s
t
h
at
its
o
u
tp
u
ts
a
r
e
f
ed
b
ac
k
r
ec
u
r
s
i
v
el
y
,
t
h
is
al
s
o
m
a
k
e
r
ed
u
ce
t
h
e
ac
c
u
r
ac
y
an
d
s
tab
ili
t
y
o
f
t
h
e
r
esp
o
n
s
es
o
f
o
b
s
er
v
er
.
Fin
a
ll
y
,
th
e
u
s
e
o
f
t
w
o
n
e
u
r
al
n
et
wo
r
k
s
:
th
e
f
ir
s
t
is
o
n
li
n
e
tr
ain
e
d
f
o
r
s
tato
r
cu
r
r
en
t
esti
m
atio
n
an
d
th
e
s
ec
o
n
d
is
o
f
f
-
lin
e
tr
ai
n
ed
f
o
r
r
o
to
r
f
lu
x
esti
m
atio
n
m
ad
e
in
cr
ea
s
e
t
h
e
co
m
p
lex
it
y
an
d
co
m
p
u
tatio
n
al
b
u
r
d
en
r
eq
u
ir
e
h
i
g
h
ab
o
u
t
h
ar
d
w
ar
e
a
n
d
ti
m
e
h
a
n
d
le
t
h
e
d
ata.
T
h
is
i
m
p
o
s
e
a
lar
g
e
d
is
ad
v
an
ta
g
e
o
f
M
R
A
S [
23
].
T
h
is
p
ap
er
p
r
o
p
o
s
es
a
n
o
v
el
SC
_
MR
AS
s
c
h
e
m
e.
I
n
th
e
p
r
o
p
o
s
e
d
L
S_
SC
_
M
R
A
S
o
b
s
er
v
er
,
th
e
r
ef
er
en
ce
m
o
d
el
u
s
es
t
h
e
s
tat
o
r
cu
r
r
en
t
co
m
p
o
n
e
n
ts
to
f
r
ee
o
f
p
u
r
e
i
n
te
g
r
atio
n
p
r
o
b
le
m
s
an
d
i
n
s
e
n
s
itiv
e
to
m
o
to
r
p
ar
a
m
eter
v
ar
ia
tio
n
s
.
T
h
e
n
e
w
p
o
in
t
s
i
n
t
h
i
s
S
C
_
M
R
A
S
s
c
h
e
m
e
ar
e,
f
ir
s
t:
A
d
a
p
tiv
e
Mo
d
el
u
s
es
a
t
w
o
la
y
er
lin
ea
r
n
eu
r
al
n
et
w
o
r
k
,
w
h
ic
h
i
s
tr
ain
ed
o
n
l
in
e
b
y
a
lin
ea
r
L
S
al
g
o
r
ith
m
,
th
is
al
g
o
r
ith
m
r
eq
u
ir
e
s
th
e
less
co
m
p
u
tatio
n
ef
f
o
r
t
an
d
o
v
er
co
m
e
s
o
m
e
d
r
a
w
b
ac
k
s
,
w
h
ic
h
ca
u
s
e
b
y
its
in
h
er
en
t
n
o
n
li
n
ea
r
it
y
a
s
i
n
liter
atu
r
e
p
u
b
li
s
h
ed
b
e
f
o
r
e
[
2
3
]
.
T
h
is
s
i
g
n
if
ica
n
tl
y
i
m
p
r
o
v
es
t
h
e
p
er
f
o
r
m
an
ce
o
f
t
h
e
p
r
o
p
o
s
ed
o
b
s
er
v
er
.
S
ec
o
n
d
: t
h
e
ad
ap
tiv
e
m
o
d
el
b
ased
o
n
N
N
i
s
i
m
p
le
m
en
ted
in
th
e
p
r
ed
ictio
n
m
o
d
e.
T
h
is
i
m
p
r
o
v
e
m
en
t
e
n
s
u
r
e
s
th
e
p
r
o
p
o
s
ed
o
b
s
er
v
er
o
p
er
at
e
b
etter
ac
cu
r
ac
y
an
d
s
tab
ilit
y
.
T
h
ir
d
:
th
e
r
o
to
r
f
lu
x
,
w
h
ic
h
is
n
ee
d
ed
f
o
r
th
e
s
tato
r
cu
r
r
en
t
es
ti
m
a
tio
n
o
f
t
h
e
ad
ap
tiv
e
m
o
d
el,
is
id
e
n
ti
f
ier
b
y
t
h
e
Vo
lta
g
e
Mo
d
el
(
VM
)
w
it
h
t
h
e
s
tato
r
r
esis
ta
n
ce
v
al
u
e
is
est
i
m
a
ted
o
n
li
n
e
to
en
h
a
n
ce
th
e
p
er
f
o
r
m
a
n
ce
o
f
th
e
p
r
o
p
o
s
ed
L
S_
SC
_
MR
A
S
o
b
s
er
v
er
,
in
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
694
I
n
t J
P
o
w
E
lec
&
Dr
i
S
y
s
t
,
Vo
l.
9
,
No
.
4
,
Dec
em
b
er
2
0
1
8
:
1486
–
1
5
0
2
1488
ad
d
itio
n
,
u
s
i
n
g
VM
w
i
ll
av
o
i
d
th
e
in
s
tab
ilit
y
i
n
t
h
e
r
eg
e
n
er
atin
g
m
o
d
e.
I
n
th
i
s
p
r
o
p
o
s
ed
,
r
o
to
r
r
esis
tan
ce
v
alu
e
a
ls
o
h
a
s
b
ee
n
e
s
ti
m
ated
b
ased
o
n
its
v
ar
iatio
n
p
r
o
p
o
r
tio
n
al
to
th
a
t
o
n
e
o
f
s
tato
r
r
esis
ta
n
ce
,
th
e
n
t
h
e
esti
m
ated
r
esis
ta
n
ce
v
al
u
es
w
er
e
u
p
d
ate
d
f
o
r
th
e
cu
r
r
en
t
o
b
s
er
v
er
to
esti
m
ate
t
h
e
cu
r
r
en
t
ex
ac
tl
y
m
o
r
e.
Fin
all
y
,
t
h
e
m
o
d
if
ied
E
u
ler
in
teg
r
atio
n
h
as
b
ee
n
u
s
ed
in
t
h
e
ad
ap
tiv
e
m
o
d
el
to
s
o
lv
e
th
e
i
n
s
tab
ili
t
y
p
r
o
b
le
m
s
d
u
e
to
t
h
e
d
is
cr
etiza
t
io
n
o
f
t
h
e
r
o
to
r
eq
u
atio
n
s
o
f
t
h
e
m
a
ch
in
e
en
h
a
n
ce
t
h
e
p
er
f
o
r
m
an
ce
o
f
o
b
s
er
v
er
.
T
h
e
th
eo
r
etica
l
an
al
y
s
i
s
is
v
alid
at
ed
b
y
s
i
m
u
latio
n
te
s
ts
o
f
t
h
e
s
en
s
o
r
les
s
SP
I
M
d
r
iv
e
s
y
s
te
m
u
n
d
er
d
if
f
er
e
n
t
o
p
er
atin
g
co
n
d
itio
n
s
.
Si
m
u
lati
o
n
r
esu
l
ts
ar
e
g
iv
e
n
to
co
m
p
ar
e
th
e
p
er
f
o
r
m
a
n
ce
o
f
th
e
p
r
o
p
o
s
ed
o
b
s
er
v
er
w
it
h
r
ec
en
t
p
r
o
p
o
s
ed
o
b
s
er
v
er
[
1
7
],
[
2
0
]
-
[
2
3
]
.
T
h
e
co
m
p
ar
is
o
n
d
ata
h
av
e
p
r
o
v
en
th
a
t
th
e
p
r
o
p
o
s
ed
L
S_
SC
_
MR
AS
o
b
s
er
v
er
ar
e
q
u
ick
er
co
n
v
er
g
e
n
ce
in
s
p
ee
d
esti
m
atio
n
,
b
ette
r
d
y
n
a
m
ic
p
er
f
o
r
m
a
n
ce
s
;
lo
w
er
esti
m
atio
n
er
r
o
r
s
b
o
th
in
tr
an
s
ien
t
a
n
d
s
tead
y
-
s
t
ate
o
p
er
atio
n
.
T
h
e
ter
m
s
o
f
ac
cu
r
ac
y
o
f
t
h
e
L
S_
SC
_
M
R
AS
o
b
s
er
v
er
s
is
q
u
it
e
h
ig
h
a
n
d
it
is
r
o
b
u
s
t
n
es
s
a
g
ain
s
t
m
o
to
r
p
ar
a
m
eter
v
ar
iat
io
n
s
.
Fro
m
th
e
co
m
p
ar
is
o
n
d
ata
h
av
e
p
r
o
v
en
t
h
at
t
h
e
p
r
o
p
o
s
ed
NNVM
_
SC
_
M
R
A
S
o
b
s
er
v
er
is
m
u
c
h
b
etter
s
o
lu
tio
n
t
h
o
s
e
k
n
o
w
n
f
r
o
m
t
h
e
lit
er
atu
r
e
[
1
7
]
,
[
2
0
]
,
[2
1
]
,
[
2
3
]
esp
ec
iall
y
,
at
lo
w
a
n
d
ze
r
o
s
p
ee
d
r
an
g
e.
T
h
e
p
ap
e
r
is
o
r
g
an
ized
in
to
f
i
v
e
s
ec
tio
n
s
.
I
n
Sect
io
n
2
,
th
e
b
asic
th
eo
r
y
o
f
t
h
e
m
o
d
el
o
f
t
h
e
SP
I
M
an
d
th
e
SP
I
M
d
r
iv
e
ar
e
p
r
esen
ted
.
Sectio
n
3
i
n
tr
o
d
u
ce
s
t
h
e
p
r
o
p
o
s
ed
L
S_
SC
MR
AS
o
b
s
er
v
er
.
Si
m
u
latio
n
an
d
d
is
cu
s
s
ar
e
p
r
esen
ted
in
S
ec
tio
n
4
.
Fin
all
y
,
th
e
co
n
clu
d
i
n
g
i
s
p
r
o
v
id
ed
in
Sectio
n
5
.
2.
M
O
DE
L
VE
CT
O
R
CO
N
T
R
O
L
O
F
SPIM
DRIVE
S
2
.
1
.
M
o
del v
ec
t
o
r
c
o
ntr
o
l o
f
SPIM
driv
es
T
h
e
s
y
s
te
m
u
n
d
er
s
t
u
d
y
co
n
s
i
s
ts
o
f
a
n
SP
I
M
f
ed
b
y
a
s
ix
-
p
h
ase
V
SI
(
v
o
lta
g
e
So
u
r
ce
I
n
v
er
ter
)
an
d
a
DC
l
in
k
.
A
d
etailed
s
ch
e
m
e
o
f
t
h
e
d
r
iv
e
i
s
p
r
o
v
id
ed
i
n
Fig
u
r
e
1
.
T
h
is
SP
I
M
is
a
co
n
ti
n
u
o
u
s
s
y
s
te
m
th
a
t
ca
n
b
e
d
escr
ib
ed
b
y
a
s
et
o
f
d
i
f
f
e
r
en
tial
eq
u
atio
n
s
.
T
h
e
m
o
d
el
o
f
th
e
s
y
s
te
m
ca
n
b
e
s
i
m
p
li
f
i
ed
b
y
m
ea
n
s
o
f
t
h
e
v
ec
to
r
s
p
ac
e
d
ec
o
m
p
o
s
itio
n
(
VSD)
.
B
y
ap
p
l
y
i
n
g
t
h
is
tec
h
n
iq
u
e,
t
h
e
o
r
ig
in
al
s
i
x
-
d
i
m
e
n
s
io
n
al
s
p
ac
e
o
f
t
h
e
m
ac
h
in
e
i
s
tr
an
s
f
o
r
m
ed
i
n
to
th
r
ee
t
w
o
-
d
i
m
e
n
s
io
n
al
o
r
th
o
g
o
n
al
s
u
b
s
p
ac
es
i
n
th
e
s
tatio
n
ar
y
r
ef
er
en
ce
f
a
m
e
(
D
-
Q)
,
(
x
-
y
)
a
n
d
(
zl
-
z2
)
.
T
h
is
tr
an
s
f
o
r
m
atio
n
is
o
b
tain
ed
b
y
m
e
an
s
o
f
6
x
6
tr
an
s
f
o
r
m
atio
n
m
a
tr
ix
e
q
u
atio
n
(
1
)
.
(
1
)
I
n
t
h
at,
a
n
a
m
p
lit
u
d
e
i
n
v
ar
ian
t
cr
iter
io
n
w
a
s
u
s
ed
.
Fro
m
th
e
m
o
to
r
m
o
d
el
o
b
tai
n
ed
b
y
u
s
i
n
g
t
h
e
VSD
ap
p
r
o
ac
h
,
th
e
f
o
llo
w
i
n
g
co
n
cl
u
s
io
n
s
s
h
o
u
ld
b
e
e
m
p
h
asized
:
1.
T
h
e
elec
tr
o
m
ec
h
an
ica
l
en
er
g
y
co
n
v
er
s
io
n
v
ar
iab
les
ar
e
m
ap
p
ed
to
th
e
(
D
-
Q
)
s
u
b
s
p
ac
e.
T
h
e
n
o
n
-
elec
tr
o
m
ec
h
a
n
ical
e
n
er
g
y
c
o
n
v
er
s
io
n
v
ar
iab
les ca
n
b
e
f
o
u
n
d
in
o
th
er
s
u
b
s
p
ac
es.
2
.
T
h
e
cu
r
r
en
t
co
m
p
o
n
en
t
s
i
n
th
e
(
x
-
y
)
s
u
b
s
p
ac
e
d
o
n
o
t
co
n
tr
ib
u
te
to
th
e
air
g
ap
fl
u
x
s
o
th
e
y
s
h
o
u
ld
b
e
co
n
tr
o
lled
to
b
e
as sm
a
ll a
s
p
o
s
s
ib
le.
3
.
T
h
e
v
o
ltag
e
v
ec
to
r
s
i
n
t
h
e
(
zl
-
z2
)
ar
e
ze
r
o
d
u
e
to
th
e
s
ep
ar
ated
n
eu
tr
als co
n
f
i
g
u
r
atio
n
o
f
t
h
e
m
ac
h
i
n
e.
A
V
SI
h
a
s
a
d
is
cr
ete
n
a
tu
r
e,
ac
tu
all
y
,
it
h
as
a
to
tal
n
u
m
b
er
o
f
6
2
6
4
d
if
f
er
e
n
t
s
w
i
tch
i
n
g
s
ta
tes
d
ef
in
ed
b
y
s
i
x
s
w
itc
h
in
g
f
u
n
ctio
n
s
co
r
r
esp
o
n
d
in
g
to
th
e
s
ix
in
v
er
ter
leg
s
[
Sa,
S
x
,
Sb
,
S
y
,
Sc,
Sz]
,
w
h
er
e
Si
є
{0
,
1
}.
On
t
h
e
o
t
h
er
h
an
d
,
a
t
r
an
s
f
o
r
m
atio
n
m
atr
ix
m
u
s
t
b
e
u
s
ed
to
r
ep
r
esen
t
t
h
e
s
tatio
n
ar
y
r
ef
er
e
n
ce
f
a
m
e
(
D
-
Q
)
in
t
h
e
d
y
n
a
m
ic
r
ef
er
en
c
e
(
d
-
q
)
.
T
h
is
m
atr
ix
i
s
g
i
v
e
n
:
(
2
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
P
o
w
E
lec
&
Dr
i
S
y
s
t
I
SS
N:
2
0
8
8
-
8
694
N
ew V
er
s
io
n
o
f A
d
a
p
tive
S
p
ee
d
Ob
s
erver B
a
s
ed
o
n
N
eu
r
a
l
N
etw
o
r
k
fo
r
S
P
I
M (
N
g
o
c
Th
u
y
P
h
a
m)
1489
Fig
u
r
e
1
.
A
g
e
n
er
al
s
c
h
e
m
e
o
f
an
SP
I
M
d
r
iv
e
Fig
u
r
e
2
.
S
w
i
tch
i
n
g
s
tate
s
in
(
D
-
Q)
an
d
(
x
-
y
)
s
u
b
s
p
ac
es f
o
r
a
SP
VSI
T
h
e
d
if
f
er
en
t
s
w
itc
h
i
n
g
s
tate
s
an
d
th
e
v
o
ltag
e
o
f
t
h
e
DC
l
in
k
d
ef
in
e
t
h
e
p
h
a
s
e
v
o
lta
g
es
w
h
i
ch
ca
n
i
n
tu
r
n
b
e
m
ap
p
ed
to
th
e
(
D
-
Q
)
-
(x
-
y
)
s
p
ac
e
ac
co
r
d
in
g
to
th
e
V
ec
to
r
s
p
ac
e
d
ec
o
m
p
o
s
itio
n
VS
D
ap
p
r
o
ac
h
.
2
.
2
.
M
o
del o
f
SPI
M
I
n
t
h
is
p
ar
t
a
s
ix
p
h
a
s
e
i
n
d
u
c
tio
n
m
o
to
r
,
w
h
ic
h
co
n
tai
n
s
t
w
o
s
ets
o
f
th
r
ee
p
h
ase
w
i
n
d
i
n
g
s
p
atiall
y
s
h
i
f
ed
b
y
3
0
elec
tr
ical
d
eg
r
ee
s
w
it
h
i
s
o
lated
n
e
u
tr
al
p
o
in
t
s
(
as
d
ep
icted
in
F
ig
u
r
e
1
)
,
is
m
o
d
eled
.
Stato
r
an
d
r
o
to
r
v
o
ltag
e
eq
u
atio
n
f
o
r
th
is
m
o
d
el
i
s
as
f
o
llo
w
s
:
(
3
)
w
h
er
e:
r
esp
ec
ti
v
el
y
,
[
V]
,
[
I
]
,
[
R
]
,
[
L
]
an
d
[
M]
ar
e
v
o
lta
g
e,
cu
r
r
en
t,
r
esi
s
tan
t,
s
el
f
a
n
d
m
u
tu
a
l
in
d
u
cta
n
ce
v
ec
to
r
s
.
A
s
t
h
e
s
e
eq
u
atio
n
s
i
m
p
lies
,
t
h
e
elec
tr
o
m
ec
h
a
n
ical
co
n
v
er
s
io
n
,
o
n
l
y
ta
k
es
p
lace
i
n
t
h
e
D
-
Q
s
u
b
s
p
ac
e
an
d
th
e
o
th
er
s
u
b
s
p
ac
es j
u
s
t p
r
o
d
u
ce
lo
s
s
es.
So
th
e
to
r
q
u
e
eq
u
atio
n
ca
n
b
e
w
r
it
ten
as
f
o
llo
w
s
:
(
4
)
(5
)
w
h
er
e:
r
esp
ec
ti
v
el
y
,
J
i,
ω
r
,
B
i,
T
m
,
T
L
,
P
ar
e
i
n
er
tia
co
ef
f
icien
t,
an
g
u
lar
s
p
ee
d
,
f
ict
io
n
f
ac
to
r
,
t
h
e
elec
tr
o
m
ag
n
etic
to
r
q
u
e
t
h
at
g
en
er
ated
b
y
t
h
e
m
o
to
r
,
lo
ad
to
r
q
u
e,
n
u
m
b
er
o
f
p
o
les
an
d
s
tato
r
flu
x
li
n
k
ag
e
a
t
th
e
r
elate
d
s
u
b
s
p
ac
e.
3.
L
S NN
_
SC_
M
RAS
SPE
E
D
O
B
SE
RVER
3
.
1
.
P
I
_
SC_
M
RAS
o
bs
er
v
er
I
n
t
h
e
cla
s
s
ical
r
o
to
r
f
l
u
x
MR
A
S
s
p
ee
d
o
b
s
er
v
er
,
t
h
e
r
ef
er
en
ce
m
o
d
el,
u
s
u
all
y
e
x
p
r
ess
ed
as
a
Vo
ltag
e
Mo
d
el
(
VM
)
,
r
ep
r
ese
n
ts
t
h
e
s
tato
r
eq
u
atio
n
an
d
ca
n
b
e
w
r
itten
a
s
f
o
llo
w
i
n
g
:
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
694
I
n
t J
P
o
w
E
lec
&
Dr
i
S
y
s
t
,
Vo
l.
9
,
No
.
4
,
Dec
em
b
er
2
0
1
8
:
1486
–
1
5
0
2
1490
Ѱ
̂
=
(
−
−
σ
x
)
Ѱ
̂
=
(
−
−
σ
x
)
(
6
)
w
h
er
e:
r
s
:
s
tato
r
r
esis
ta
n
ce
s
,
x
s
=
x
m
+
x
s
σ
;
x
r
=
x
m
+
x
r
σ
;
x
m
:
r
esp
ec
tiv
el
y
s
tato
r
,
r
o
to
r
r
ea
ctan
ce
s
an
d
m
ag
n
etiz
in
g
,
x
σ
s
,
x
σ
r
:
s
tato
r
an
d
r
o
to
r
leak
ag
e
r
ea
ctan
ce
s
,
p
=d
/d
t;
T
n
=
1
/2
πf
s
n
,
σ
=
1
-
x
m
2
/
x
s
x
r
,
f
s
n
:
n
o
m
i
n
al
f
r
eq
u
en
c
y
.
T
h
e
ad
ap
t
iv
e
m
o
d
el,
u
s
u
all
y
r
ep
r
esen
te
d
b
y
th
e
C
u
r
r
en
t
Mo
d
el
(
C
M)
,
d
escr
ib
es
th
e
r
o
to
r
eq
u
atio
n
w
h
er
e
th
e
r
o
to
r
f
lu
x
co
m
p
o
n
e
n
t
s
ar
e
ex
p
r
ess
ed
in
ter
m
s
o
f
s
tato
r
cu
r
r
en
t
co
m
p
o
n
en
t
s
an
d
t
h
e
r
o
to
r
s
p
ee
d
.
d
Ѱ
̂
=
[
(
−
Ѱ
)
−
ω
̂
Ѱ
]
d
Ѱ
̂
=
[
(
−
Ѱ
)
+
ω
̂
Ѱ
]
(
7
)
L
o
o
k
i
n
g
at
t
h
e
f
o
r
m
u
la
(
6
)
,
i
t
is
ea
s
y
to
f
i
n
d
th
e
p
r
esen
ce
o
f
r
s
an
d
r
o
to
r
f
lu
x
,
T
h
ese
m
ak
e
t
h
e
tr
ad
itio
n
al
R
F_
MR
AS
o
b
s
er
v
er
s
u
f
f
er
ed
b
y
p
u
r
e
i
n
teg
r
at
io
n
p
r
o
b
lem
s
,
w
h
ich
b
ein
g
ab
le
t
o
ca
u
s
e
d
c
d
r
if
t
an
d
in
itial
co
n
d
itio
n
p
r
o
b
le
m
s
,
a
n
d
in
s
e
n
s
iti
v
e
to
m
o
to
r
p
ar
a
m
eter
v
ar
iatio
n
s
.
I
n
o
r
d
er
to
o
v
er
co
m
e
t
h
ese
p
r
o
b
lem
s
a
n
o
th
er
ap
p
r
o
ac
h
,
th
e
s
tato
r
cu
r
r
e
n
t
MR
AS
s
ch
e
m
e
h
a
s
b
ee
n
p
r
o
p
o
s
ed
,
th
e
s
tato
r
cu
r
r
e
n
t
co
m
p
o
n
e
n
t
s
is
u
s
ed
as
a
r
ef
er
en
ce
m
o
d
el.
T
h
e
s
tato
r
cu
r
r
e
n
t
esti
m
ato
r
is
ad
j
u
s
tab
le
m
o
d
el.
T
h
e
esti
m
ated
s
tato
r
cu
r
r
en
t
co
m
p
o
n
e
n
ts
ar
e
co
m
p
ar
ed
w
i
th
t
h
eir
m
ea
s
u
r
ed
v
alu
e
s
,
an
d
th
e
s
ig
n
al
e
is
is
u
s
ed
i
n
th
e
ad
ap
tatio
n
m
ec
h
an
is
m
(
9
)
to
g
en
er
ate
th
e
r
o
to
r
s
p
ee
d
.
I
n
t
h
is
o
b
s
er
v
er
,
t
h
e
m
ath
e
m
at
ical
m
o
d
el
o
f
t
h
e
s
tato
r
cu
r
r
en
t
o
b
s
er
v
er
ca
n
b
e
ca
lc
u
lated
f
r
o
m
t
h
e
co
m
b
i
n
ed
v
o
lt
ag
e
an
d
cu
r
r
e
n
t
m
o
d
els
a
n
d
is
d
escr
ib
ed
b
y
t
h
e
f
o
llo
w
in
g
eq
u
at
io
n
:
d
̂
=
1
[
u
−
−
(
−
Ѱ
−
ω
Ѱ
)
]
d
̂
=
1
[
u
−
−
(
−
Ѱ
+
ω
Ѱ
)
]
(
8
)
T
h
e
ad
j
u
s
tab
le
m
o
d
el
(
8
)
r
eq
u
ir
es
in
f
o
r
m
atio
n
ab
o
u
t
th
e
r
o
t
o
r
f
lu
x
.
T
h
is
i
s
ca
lcu
la
ted
o
n
th
e
b
asis
o
f
v
o
ltag
e
m
o
d
el
(
VM
)
(
6
)
o
r
cu
r
r
en
t
m
o
d
el
(
C
M)
(
7
)
.
I
n
t
h
e
P
I
_
SC
_
MRAS
o
b
s
er
v
er
,
t
h
e
u
s
ed
ad
ap
tatio
n
alg
o
r
ith
m
is
b
ased
o
n
t
h
e
er
r
o
r
b
et
w
ee
n
es
ti
m
ated
an
d
m
ea
s
u
r
ed
s
tato
r
cu
r
r
en
t
d
ev
elo
p
ed
in
[
24
]
(
b
asin
g
o
n
th
e
m
in
i
m
iza
tio
n
o
f
t
h
e
L
y
ap
u
n
o
v
f
u
n
ctio
n
)
ω
̂
=
K
p
(
e
Ѱ
−
e
Ѱ
)
+
K
I
∫
(
e
Ѱ
−
e
Ѱ
)
dt
(
9
)
w
h
er
e
e
isD
=
i
sD
−
i
esD
,
e
isQ
=
i
sQ
−
i
esQ
is
t
h
e
er
r
o
r
b
et
w
ee
n
esti
m
ated
a
n
d
m
ea
s
u
r
ed
s
tato
r
cu
r
r
en
t.
T
h
e
o
b
tain
ed
r
o
to
r
s
p
ee
d
v
alu
e
is
u
s
ed
in
t
h
e
s
tato
r
cu
r
r
en
t
es
ti
m
ato
r
as
ch
a
n
g
ea
b
le
p
ar
a
m
et
er
,
as
s
h
o
w
n
i
n
th
e
Fig
u
r
e
3
.
S
P
I
M
A
d
a
p
ti
v
e
M
o
d
e
l
v
sD
C
M
-
+
ε
ω
v
sQ
R
e
f
e
r
e
n
c
e
M
o
d
e
l
i
sQ
D
VM
R
o
t
o
r
f
l
u
x
i
d
e
n
t
i
f
i
e
r
-
+
-
+
P
I
c
o
n
tr
o
l
e
r
i
sD
Q
i
sD
i
sQ
ε
I
S
D
ε
I
S
Q
A
d
a
p
ta
ti
o
n
M
e
c
h
a
n
i
s
m
Fig
u
r
e
3.
P
I
_
SC
_
MR
A
S sp
e
ed
o
b
s
er
v
er
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
P
o
w
E
lec
&
Dr
i
S
y
s
t
I
SS
N:
2
0
8
8
-
8
694
N
ew V
er
s
io
n
o
f A
d
a
p
tive
S
p
ee
d
Ob
s
erver B
a
s
ed
o
n
N
eu
r
a
l
N
etw
o
r
k
fo
r
S
P
I
M (
N
g
o
c
Th
u
y
P
h
a
m)
1491
3
.
2
.
S NN
_
SC_
M
RAS
o
bs
er
v
er
3
.
2
.
1
Str
uct
ure
o
f
t
he
L
S_
SC_
M
RAS
O
bs
er
v
er
I
n
t
h
is
s
c
h
e
m
e,
t
h
e
m
ea
s
u
r
ed
s
tato
r
cu
r
r
en
t
co
m
p
o
n
e
n
t
s
ar
e
also
u
s
ed
a
s
t
h
e
r
e
f
er
en
ce
m
o
d
el
o
f
t
h
e
MR
A
S
o
b
s
er
v
er
to
av
o
id
t
h
e
u
s
e
o
f
a
p
u
r
e
in
te
g
r
ato
r
an
d
r
ed
u
ce
in
f
l
u
en
ce
o
f
m
o
to
r
p
ar
am
eter
v
ar
iat
io
n
as
i
n
[
22
]
-
[
23
].
T
h
e
ad
ap
tiv
e
m
o
d
el
is
a
t
w
o
-
la
y
er
li
n
ea
r
NN
to
esti
m
ate
th
e
s
tato
r
cu
r
r
en
t
h
a
s
b
ee
n
tr
ain
ed
o
n
lin
e
b
y
m
ea
n
s
o
f
a
least
-
s
q
u
ar
es
alg
o
r
ith
m
.
T
h
is
ad
ap
tiv
e
m
o
d
el
is
d
escr
ib
ed
b
y
th
e
co
m
b
in
ed
v
o
ltag
e
-
an
d
cu
r
r
en
t
m
o
d
el
s
i
n
th
e
s
tato
r
r
e
f
er
en
ce
f
r
a
m
e
(
8
)
.
E
q
u
atio
n
(
8
)
,
T
h
en
t
h
e
y
b
ee
n
d
i
v
id
ed
b
y
T
n
,
b
e
r
e
w
r
it
ten
i
n
th
e
f
o
llo
w
i
n
g
as:
̇
=
AX
+
Bu
(
1
0
)
w
h
er
e
̇
=
[
d
i
sD
dt
d
i
sQ
dt
]
;
A
=
[
−
(
1
+
x
m
2
x
r
2
)
r
s
x
s
σ
−
(
1
+
x
m
2
x
r
2
)
r
s
x
s
σ
]
;
B
=
[
1
x
s
σ
0
0
1
x
s
σ
1
x
s
σ
r
r
x
m
x
r
2
1
x
s
σ
x
m
ω
r
x
r
−
1
x
s
σ
x
m
ω
r
x
r
1
x
s
σ
r
r
x
m
x
r
2
]
;
X
=
[
i
sD
i
sQ
]
;
u
=
[
v
sD
v
sQ
Ψ
̂
rD
Ψ
̂
rQ
]
I
ts
co
r
r
esp
o
n
d
in
g
d
is
cr
ete
m
o
d
el
is
,
th
er
ef
o
r
e,
g
i
v
en
b
y
:
X
̂
(
k
)
=
e
[
A
]
T
s
X
(
k
−
1
)
+
(
e
AT
s
−
I
)
A
−
1
Bu
(
k
−
1
)
(
1
1
)
e
A
T
s
:
is
g
en
er
all
y
co
m
p
u
ted
b
y
tr
u
n
ca
ti
n
g
i
ts
p
o
w
er
s
er
ie
s
ex
p
an
s
io
n
,
i.e
.
,
e
AT
s
=
I
+
A
T
s
1
!
+
A
2
T
s
2
!
+
⋯
+
A
n
T
s
n
!
(
1
2
)
I
f
n
=1
,
t
h
e
s
i
m
p
le
f
o
r
w
ar
d
E
u
ler
m
eth
o
d
is
o
b
tain
ed
,
w
h
i
ch
g
iv
e
s
t
h
e
f
o
llo
w
in
g
f
i
n
ite
d
if
f
er
e
n
ce
eq
u
at
io
n
[
1
5
]
-
[
1
7
]
.
i
̂
sD
(
k
)
=
w
1
i
̂
sD
(
k
−
1
)
+
w
2
v
sD
(
k
−
1
)
+
w
3
Ψ
̂
rD
(
k
−
1
)
+
w
4
Ψ
̂
rQ
(
k
−
1
)
i
̂
sQ
(
k
)
=
w
1
i
̂
sQ
(
k
−
1
)
+
w
2
v
sQ
(
k
−
1
)
+
w
3
Ψ
̂
rQ
(
k
−
1
)
−
w
4
Ψ
̂
rD
(
k
−
1
)
(
1
3
)
w
h
er
e
m
ar
k
s
t
h
e
v
ar
iab
les
e
s
ti
m
ated
w
ith
t
h
e
ad
ap
tiv
e
m
o
d
el
a
n
d
is
th
e
c
u
r
r
en
t
ti
m
e
s
a
m
p
le.
A
n
e
u
r
al
n
et
w
o
r
k
c
a
n
r
ep
r
o
d
u
ce
th
ese
eq
u
atio
n
s
,
w
h
er
e
ar
e
th
e
w
ei
g
h
t
s
o
f
th
e
n
e
u
r
al
n
et
w
o
r
k
s
d
ef
i
n
ed
as:
2
s
s
s
m
1
s
s
r
r
T
r
T
L
w
=
1
-
-
;
σ
L
σ
L
L
T
s
2
s
T
w
=
;
σL
^
r
s
m
s
m
34
s
r
s
r
T
L
T
L
ω
w
=
;
w
=
σ
L
T
σ
L
L
(
1
4
)
w
h
er
e:
i
̂
s
(
k
)
th
e
cu
r
r
e
n
t
v
ar
iab
les
e
s
ti
m
ated
w
it
h
t
h
e
ad
ap
tiv
e
m
o
d
el
an
d
k
is
th
e
cu
r
r
en
t
ti
m
e
s
a
m
p
le,
T
s
is
th
e
s
a
m
p
lin
g
ti
m
e
f
o
r
th
e
s
tat
o
r
cu
r
r
en
t
o
b
s
er
v
er
.
T
h
e
A
NN
h
as,
th
u
s
,
f
o
u
r
in
p
u
ts
a
n
d
t
w
o
o
u
tp
u
ts
[
22
]
–
[
23
]
.
I
n
t
h
e
A
NN,
th
e
w
ei
g
h
ts
w
1
,
w
2
a
n
d
w
3
ar
e
k
ep
t
co
n
s
tan
t
t
o
th
eir
v
al
u
es
co
m
p
u
ted
o
f
f
lin
e
w
h
ile
o
n
l
y
w
4
i
s
ad
o
p
ted
o
n
lin
e.
T
h
ese
eq
u
atio
n
s
ar
e
th
e
s
a
m
e
as t
h
o
s
e
o
b
tain
ed
in
[
23
]
.
I
n
th
e
s
c
h
e
m
e
i
s
p
r
esen
ted
in
[
23
]
,
th
e
ad
ap
tiv
e
m
o
d
el
is
c
h
ar
ac
ter
iz
ed
b
y
t
h
e
f
ee
d
b
ac
k
o
f
d
ela
y
ed
esti
m
ated
s
tato
r
cu
r
r
e
n
t
co
m
p
o
n
en
ts
to
t
h
e
i
n
p
u
t
o
f
th
e
n
e
u
r
al
n
e
t
w
o
r
k
,
w
h
ich
m
ea
n
s
t
h
at
th
e
ad
ap
t
iv
e
m
o
d
el
e
m
p
lo
y
ed
is
i
n
s
i
m
u
latio
n
m
o
d
e.
Mo
r
eo
v
er
,
th
e
ad
ap
tiv
e
m
o
d
el
is
tu
n
ed
o
n
li
n
e
(
tr
ain
in
g
)
b
y
m
ea
n
s
o
f
a
B
P
N
alg
o
r
ith
m
,
h
o
w
ev
er
,
n
o
n
li
n
ea
r
in
its
n
at
u
r
e
w
it
h
th
e
co
n
s
eq
u
e
n
t
d
r
a
w
b
ac
k
s
(
l
o
ca
l
m
i
n
i
m
a,
h
eu
r
i
s
tics
i
n
t
h
e
ch
o
ice
o
f
t
h
e
n
et
w
o
r
k
p
ar
a
m
eter
s
,
p
ar
al
y
s
i
s
,
co
n
v
er
g
e
n
ce
p
r
o
b
le
m
s
)
.
I
n
t
h
is
L
S_
SC
_
M
R
A
S
o
b
s
er
v
er
p
r
o
p
o
s
ed
,
th
e
ad
ap
tiv
e
m
o
d
el
b
ased
o
n
t
h
e
A
D
AL
I
NE
h
as
b
ee
n
i
m
p
r
o
v
ed
,
A
li
n
ea
r
least
-
s
q
u
ar
e
alg
o
r
ith
m
,
w
h
ic
h
is
m
o
r
e
s
u
i
tab
le
th
an
a
n
o
n
lin
ea
r
o
n
e,
lik
e
th
e
B
P
N,
is
u
s
ed
to
r
ed
u
ce
th
e
co
m
p
u
tati
o
n
e
f
f
o
r
t
an
d
o
v
er
co
m
e
s
o
m
e
d
r
a
wb
ac
k
s
,
w
h
ich
ca
u
s
e
b
y
i
ts
i
n
h
er
en
t
n
o
n
li
n
ea
r
it
y
.
Fu
r
t
h
er
m
o
r
e,
t
h
e
e
m
p
lo
y
m
en
t
o
f
t
h
e
ad
ap
tiv
e
m
o
d
el
i
n
p
r
ed
i
ctio
n
m
o
d
e
lead
s
to
a
q
u
ic
k
er
co
n
v
er
g
e
n
ce
o
f
t
h
e
alg
o
r
ith
m
,
a
h
i
g
h
er
b
an
d
w
id
th
o
f
th
e
s
p
ee
d
co
n
tr
o
l
lo
o
p
,
a
b
etter
b
e
h
av
io
r
at
ze
r
o
-
s
p
ee
d
,
lo
w
er
s
p
ee
d
esti
m
atio
n
er
r
o
r
s
b
o
th
in
tr
an
s
i
en
t a
n
d
s
tead
y
s
tate
co
n
d
it
io
n
s
.
An
i
n
te
g
r
atio
n
m
et
h
o
d
m
o
r
e
ef
f
icie
n
t
t
h
a
n
t
h
at
u
s
ed
i
n
(
1
5
)
is
th
e
s
o
ca
lled
m
o
d
i
f
ied
E
u
ler
in
te
g
r
atio
n
,
w
h
ic
h
al
s
o
ta
k
es
in
to
co
n
s
id
er
atio
n
th
e
v
al
u
e
s
o
f
t
h
e
v
ar
iab
les
i
n
t
w
o
p
r
ev
io
u
s
ti
m
e
s
tep
s
[
25
]
.
Fro
m
(
9
)
,
th
e
f
o
llo
w
i
n
g
d
is
cr
e
te
ti
m
e
eq
u
at
io
n
s
ca
n
b
e
o
b
tain
ed
,
as
s
h
o
w
n
i
n
(
1
5
)
.
A
ls
o
,
i
n
t
h
is
ca
s
e,
a
n
e
u
r
al
n
et
w
o
r
k
ca
n
r
ep
r
o
d
u
ce
th
ese
eq
u
atio
n
s
,
w
h
er
e
an
d
ar
e
th
e
weig
h
ts
o
f
t
h
e
n
e
u
r
al
n
et
w
o
r
k
s
d
ef
in
ed
a
s
(
1
6
)
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
694
I
n
t J
P
o
w
E
lec
&
Dr
i
S
y
s
t
,
Vo
l.
9
,
No
.
4
,
Dec
em
b
er
2
0
1
8
:
1486
–
1
5
0
2
1492
^
^
^
^
^
s
D
(
k
)
1
s
D
(
k
-
1
)
2
s
D
(
k
-
1
)
3
4
5
s
D
(
k
-
2
)
6
s
D
(
k
-
2
)
7
8
r
D
(
k
-
1
)
r
Q
(
k
-
1
)
r
D
(
k
-
2
)
r
Q
(
k
-
2
)
^
^
^
s
Q
(
k
)
1
s
Q
(
k
-
1
)
2
s
Q
(
k
-
1
)
3
4
5
s
Q
(
k
-
2
)
6
s
Q
(
k
-
2
)
r
Q
(
k
-
1
)
r
D(
k
-
1
)
i
=
w
i
+
w
u
+
w
ψ
+
w
ψ
w
i
-
w
u
-
w
ψ
-
w
ψ
i
=
w
i
+
w
u
+
w
ψ
-
w
ψ
+
w
i
-
w
u
^^
78
r
Q
(
k
-
2
)
r
D
(
k
-
2
)
-
w
ψ
+
w
ψ
(
1
5
)
22
ss
m
m
m
m
1
2
3
4
r
5
s
r
s
r
s
r
s
r
r
s
s
r
s
r
mm
6
7
8
r
s
r
s
r
r
s
3
T
R
3
T
R
3
T
L
3
T
L
3
T
L
T
L
3T
w
=
1
-
-
;
w
=
;
w
=
;
w
=
ω
;
w
=
+
;
2
σ
L
2
σ
L
L
T
2
σ
L
2
σ
L
L
T
2
σ
L
L
2
σ
L
2
σ
L
L
T
T
L
T
L
T
w
=
;
w
=
;
w
=
ω
2
σ
L
2
σ
L
L
T
2
σ
L
L
(
1
6
)
R
ea
r
r
an
g
in
g
(
1
5
)
,
th
e
m
atr
i
x
e
q
u
atio
n
i
s
o
b
tain
ed
i
n
p
r
ed
icti
o
n
m
o
d
e;
s
ee
(
1
7
)
.
T
h
is
m
atr
i
x
eq
u
at
io
n
ca
n
b
e
s
o
lv
ed
b
y
an
y
least
s
q
u
ar
e
tech
n
iq
u
e.
^^
mm
r
Q
(
k
-
1
)
r
Q
(
k
-
2
)
r
s
r
s
r
(
k
-
1
)
^^
mm
r
D
(
k
-
1
)
r
D
(
k
-
2
)
r
s
r
s
3
T
L
T
L
ψψ
2
σ
L
L
2
σ
L
L
ω
3
T
L
T
L
ψψ
2
σ
L
L
2
σ
L
L
^
^
^
s
Q
(
k
)
1
s
Q
(
k
-
1
)
2
s
Q
(
k
-
1
)
3
5
s
Q
(
k
-
2
)
6
s
Q
(
k
-
2
)
7
r
Q
(
k
-
1
)
r
Q
(
k
-
2
)
^
^
^
s
Q
(
k
)
1
s
Q
(
k
-
1
)
2
s
Q
(
k
-
1
)
3
5
s
Q
(
k
-
2
)
6
s
Q
(
k
-
2
)
7
r
Q
(
k
-
1
)
r
Q
(
k
-
2
)
i
-
w
i
-
w
u
-
w
ψ
-
w
i
+
w
u
+
w
ψ
i
-
w
i
-
w
u
-
w
ψ
-
w
i
+
w
u
+
w
ψ
(
1
7
)
Ma
tr
ix
eq
u
atio
n
(
1
7
)
ca
n
b
e
w
r
itten
:
Ax
≈
b
.
T
h
is
is
a
cla
s
s
i
ca
l
m
atr
i
x
eq
u
atio
n
o
f
t
h
e
t
y
p
e,
w
h
er
e
A
is
ca
lled
a
“
d
ata
m
atr
ix
”,
b
i
s
ca
lled
an
“
o
b
s
er
v
atio
n
v
ec
to
r
,
”
an
d
ω
r
is
th
e
s
ca
lar
u
n
k
n
o
w
n
.
I
n
t
h
is
ap
p
licatio
n
a
class
ic
al
L
S
alg
o
r
ith
m
i
n
a
r
ec
u
r
s
iv
e
f
o
r
m
h
as
b
ee
n
e
m
p
lo
y
ed
;
T
h
is
alg
o
r
ith
m
is
d
es
cr
ib
ed
in
d
etail
in
[
25
]
,
[
26
]
.
Fig
u
r
e
4
s
h
o
w
s
t
h
e
b
lo
ck
d
iag
r
a
m
o
f
t
h
e
L
S_
S
C
_
MR
A
S
s
p
ee
d
o
b
s
er
v
er
.
I
n
l
iter
atu
r
e
t
h
er
e
ex
i
s
t
th
r
ee
L
ea
s
t
-
Sq
u
ar
es
tec
h
n
iq
u
e
s
,
i.e
.
th
e
Or
d
in
ar
y
L
ea
s
t
-
Sq
u
ar
es
(
OL
S),
th
e
T
o
tal
L
ea
s
t
-
S
q
u
ar
es
(
T
L
S)
an
d
th
e
Data
L
ea
s
t
-
Sq
u
ar
es
(
D
L
S)
w
h
ic
h
ar
is
e
w
h
en
er
r
o
r
s
ar
e
r
esp
ec
tiv
el
y
p
r
esen
t
o
n
l
y
in
b
o
r
b
o
th
in
A
a
n
d
in
b
o
r
o
n
l
y
in
A
.
T
h
e
L
S
tech
n
i
q
u
e
s
o
l
v
es
f
o
r
t
h
i
s
p
r
o
b
le
m
b
y
ca
lcu
la
tin
g
th
e
v
a
lu
e
o
f
ω
r
wh
ich
m
i
n
i
m
is
e
s
t
h
e
s
u
m
o
f
s
q
u
ar
es
o
f
th
e
d
is
ta
n
ce
s
a
m
o
n
g
th
e
ele
m
e
n
ts
(
ai
,
bi
)
,
w
i
th
i
=
1
,
.
.
.
,
m
,
an
d
th
e
lin
e
it
s
el
f
.
OL
S
m
i
n
i
m
is
e
s
th
e
s
u
m
o
f
s
q
u
ar
es
o
f
th
e
d
is
tan
ce
s
in
t
h
e
b
d
ir
e
ctio
n
(
er
r
o
r
o
n
l
y
in
t
h
e
o
b
s
er
v
atio
n
v
ec
to
r
)
.
T
L
S
m
i
n
i
m
is
e
s
th
e
s
u
m
o
f
s
q
u
a
r
es
in
t
h
e
d
ir
ec
tio
n
o
r
th
o
g
o
n
al
to
th
e
lin
e
(
f
o
r
th
i
s
r
ea
s
o
n
T
L
S
is
also
ca
lled
o
r
th
o
g
o
n
al
r
eg
r
es
s
io
n
)
w
h
ile
DL
S
m
i
n
i
m
is
e
s
th
e
s
u
m
o
f
s
q
u
ar
es
in
th
e
A
d
ir
ec
tio
n
(
er
r
o
r
s
o
n
l
y
in
t
h
e
d
ata
m
atr
i
x
)
.
I
n
th
i
s
p
ap
er
,
au
th
o
r
s
f
o
cu
s
o
n
an
al
y
s
in
g
O
L
S a
l
g
o
r
ith
m
.
^^
mm
r
Q
(
k
-
1
)
r
Q
(
k
-
2
)
r
s
r
s
1
^^
2
mm
r
D
(
k
-
1
)
r
D
(
k
-
2
)
r
s
r
s
3
T
L
T
L
ψψ
2
σ
L
L
2
σ
L
L
A
A
A
3
T
L
T
L
ψψ
2
σ
L
L
2
σ
L
L
^
^
^
s
Q
(
k
)
1
s
Q
(
k
-
1
)
2
s
Q
(
k
-
1
)
3
5
s
Q
(
k
-
2
)
6
s
Q
(
k
-
2
)
7
r
Q
(
k
-
1
)
r
Q
(
k
-
2
)
1
^
^
^
2
s
Q
(
k
)
1
s
Q
(
k
-
1
)
2
s
Q
(
k
-
1
)
3
5
s
Q
(
k
-
2
)
6
s
Q
(
k
-
2
)
7
r
Q
(
k
-
1
)
r
Q
(
k
-
2
)
i
-
w
i
-
w
u
-
w
ψ
-
w
i
+
w
u
+
w
ψ
b
b
b
i
-
w
i
-
w
u
-
w
ψ
-
w
i
+
w
u
+
w
ψ
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
P
o
w
E
lec
&
Dr
i
S
y
s
t
I
SS
N:
2
0
8
8
-
8
694
N
ew V
er
s
io
n
o
f A
d
a
p
tive
S
p
ee
d
Ob
s
erver B
a
s
ed
o
n
N
eu
r
a
l
N
etw
o
r
k
fo
r
S
P
I
M (
N
g
o
c
Th
u
y
P
h
a
m)
1493
(
R
e
f
e
r
e
n
c
e
M
o
d
e
l
)
S
P
I
M
A
N
N
Cur
re
nt
o
bs
e
rv
e
r
(
A
da
pt
i
ve
M
ode
l
)
L
ea
r
n
i
n
g
A
l
g
o
r
i
th
m
i
s
(
V
o
l
ta
g
e
M
o
d
el
)
LS
u
s
VM
Rs
,
R
r
E
s
t
i
m
a
t
or
Fig
u
r
e
4
.
L
S_
SC
_
MR
AS sp
ee
d
o
b
s
er
v
er
3
.
2
.
2
.
Ro
t
o
r
Sp
ee
d E
s
t
i
m
a
t
io
n Alg
o
rit
h
m
Ax
~
b
is
t
h
e
l
in
ea
r
r
e
g
r
ess
io
n
p
r
o
b
lem
u
n
d
er
h
a
n
d
.
A
ll
L
S
p
r
o
b
lem
s
h
a
v
e
b
ee
n
g
en
er
alize
d
b
y
u
s
in
g
a
p
ar
a
m
eter
ized
f
o
r
m
u
latio
n
(
g
en
er
al
ized
L
S)
o
f
an
e
r
r
o
r
f
u
n
c
tio
n
w
h
o
s
e
m
in
i
m
izatio
n
y
ield
s
th
e
co
r
r
esp
o
n
d
in
g
s
o
l
u
tio
n
.
T
h
is
e
r
r
o
r
is
g
iv
en
b
y
:
(
X
)
=
(
Ax
−
b
)
T
(
Ax
−
b
)
1
−
+
(
1
8
)
w
h
er
e
T
r
e
p
r
esen
ts
th
e
tr
an
s
p
o
s
e
an
d
is
eq
u
al
to
0
.
5
f
o
r
T
L
S,
1
f
o
r
DL
S
an
d
0
f
o
r
OL
S.
U
s
in
g
O
L
S
alg
o
r
ith
m
,
t
h
is
er
r
o
r
is
g
i
v
en
b
y
:
O
L
S
=
(
Ax
−
b
)
T
(
Ax
−
b
)
(
1
9
)
w
h
er
e:
(
Ax
−
b
)
=
ε
=
[
ε
i
s
D
ε
i
s
Q
]
=
[
i
sD
(
k
)
−
i
̂
sD
(
k
)
i
sQ
(
k
)
−
i
̂
sQ
(
k
)
]
(
2
0
)
T
h
is
er
r
o
r
ca
n
b
e
m
in
i
m
ized
w
it
h
a
g
r
ad
ie
n
t d
escen
t
m
et
h
o
d
:
(
+
1
)
=
(
)
−
(
)
(
)
(
2
1
)
w
h
er
e:
(
)
=
(
)
(
)
−
(
)
(
2
2
)
w
h
er
e
is
t
h
e
lear
n
i
n
g
r
ate,
a(
k
)
is
th
e
r
o
w
o
f
A
f
ed
at
in
s
ta
n
t
k
,
an
d
b
(
k
)
is
th
e
co
r
r
esp
o
n
d
in
g
o
b
s
er
v
atio
n
.
3
.
2
.
3
.
Ro
t
o
r
F
lux
E
s
t
i
m
a
t
io
n a
nd
S
t
a
t
o
r
Resis
t
a
nce
O
nli
ne
E
s
t
i
m
a
t
io
n Alg
o
rit
h
m
Fro
m
(
1
5
)
ca
n
b
ee
n
s
ee
n
t
h
at,
th
e
ad
ap
tiv
e
m
o
d
el
g
e
n
er
ates
s
tato
r
cu
r
r
en
t
esti
m
atio
n
v
al
u
e
s
b
ased
o
n
r
o
to
r
s
p
ee
d
in
f
o
r
m
atio
n
,
s
tato
r
v
o
ltag
es
a
n
d
r
o
to
r
f
lu
x
.
I
n
th
is
p
r
o
p
o
s
ed
o
b
s
er
v
er
,
th
e
r
o
to
r
f
lu
x
v
al
u
e
s
w
er
e
g
en
er
ate
d
b
ase
o
n
eq
u
atio
n
(
6
)
(
VM
)
.
Ѱ
̂
=
(
−
−
σ
x
)
Ѱ
̂
=
(
−
−
σ
x
)
(
2
3
)
T
h
ese
r
o
to
r
f
lu
x
co
m
p
o
n
e
n
ts
ar
e
id
en
tif
ied
f
r
o
m
th
e
co
n
tr
o
l
v
o
ltag
e
a
n
d
th
e
m
ea
s
u
r
ed
s
tat
o
r
cu
r
r
en
t.
Usi
n
g
t
h
i
s
VM
to
id
en
ti
f
y
r
o
to
r
f
lu
x
w
il
l
o
v
er
co
m
e
th
e
in
s
tab
il
it
y
p
r
o
b
lem
in
th
e
r
e
g
en
er
ati
n
g
m
o
d
e
o
f
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
694
I
n
t J
P
o
w
E
lec
&
Dr
i
S
y
s
t
,
Vo
l.
9
,
No
.
4
,
Dec
em
b
er
2
0
1
8
:
1486
–
1
5
0
2
1494
o
p
er
atio
n
,
w
h
at
ap
p
ea
r
u
s
i
n
g
C
M
r
o
to
r
f
l
u
x
id
en
ti
f
ie
r
.
Fro
m
(
6
)
,
(
1
5
)
an
d
(
1
6
)
,
is
ea
s
y
s
ee
th
at,
th
e
r
esis
tan
ce
p
ar
am
eter
s
n
ec
e
s
s
ar
y
f
o
r
esti
m
ati
n
g
t
h
e
s
p
ee
d
.
Ho
w
ev
er
,
d
u
r
in
g
m
o
to
r
o
p
er
atio
n
,
th
e
s
e
p
ar
a
m
eter
s
w
ill
ch
an
g
e
w
it
h
th
e
i
n
cr
ea
s
e
o
f
te
m
p
er
at
u
r
e,
esp
ec
iall
y
,
at
lo
w
s
p
ee
d
.
T
h
er
ef
o
r
e
to
th
e
p
er
f
o
r
m
a
n
ce
i
m
p
r
o
v
e
m
en
t
o
f
th
e
o
b
s
er
v
er
,
esp
ec
iall
y
at
l
o
w
s
p
ee
d
,
th
e
r
esi
s
tan
ce
s
o
n
li
n
e
id
en
ti
f
icatio
n
is
n
ec
es
s
ar
y
.
I
n
th
e
p
r
o
p
o
s
ed
L
S
b
ased
ad
ap
tiv
e
s
p
ee
d
o
b
s
er
v
er
th
e
o
n
li
n
e
r
s
esti
m
atio
n
m
eth
o
d
o
lo
g
ies
p
r
o
p
o
s
ed
in
[
27
]
h
av
e
b
ee
n
u
s
ed
,
s
u
m
m
ar
ized
in
th
e
f
o
llo
w
i
n
g
.
I
n
p
ar
ticu
lar
R
s
is
est
i
m
a
ted
o
n
th
e
b
asis
o
f
t
h
e
i
sD,
i
sD
m
ea
s
u
r
ed
an
d
^^
s
D
s
Q
i
,
i
esti
m
ated
s
tato
r
cu
r
r
en
t c
o
m
p
o
n
e
n
ts
,
b
y
m
ea
n
s
o
f
t
h
e
f
o
llo
w
in
g
u
p
d
ate
la
w:
^
^
^
^
^
s
s
D
s
D
s
Q
s
Q
s
D
s
Q
dR
=
-
(
(
i
-
i
)
i
+
(
i
-
i
)
i
)
dt
(
2
4
)
w
h
er
e
is
a
p
r
o
p
er
ly
ch
o
s
e
n
p
o
s
iti
v
e
co
n
s
ta
n
t.
w
h
er
e
λ
is
a
p
r
o
p
er
ly
ch
o
s
en
p
o
s
iti
v
e
co
n
s
ta
n
t.
I
n
t
h
is
ca
s
e,
b
ec
au
s
e
it c
an
n
o
t a
p
p
lied
th
e
s
a
m
e
es
ti
m
atio
n
s
ch
e
m
e
to
r
o
to
r
r
esis
tan
ce
esti
m
at
io
n
i
n
s
en
s
o
r
less
d
r
iv
es,
R
r
h
as b
ee
n
e
s
ti
m
ated
b
ased
o
n
its
v
ar
iatio
n
p
r
o
p
o
r
tio
n
al
to
th
at
o
n
e
o
f
th
e
R
s
o
n
t
h
e
b
asis
o
f
t
h
e
f
o
llo
w
i
n
g
la
w
:
^^
rs
r
R
=
K
R
(
2
5
)
w
h
er
e
Kr
is
th
e
r
atio
o
f
t
h
e
r
at
ed
v
alu
es o
f
th
e
s
tato
r
an
d
r
o
to
r
r
esis
tan
ce
s
.
T
h
e
esti
m
ated
r
e
s
is
tan
ce
v
alu
e
s
w
er
e
u
p
d
ate
f
o
r
th
e
cu
r
r
en
t o
b
s
er
v
er
to
es
ti
m
ate
t
h
e
c
u
r
r
en
t
ex
ac
tl
y
m
o
r
e.
4.
SI
M
UL
I
NK
AND
D
I
SCU
SS
I
O
N
I
n
o
r
d
er
to
v
er
if
y
an
d
e
v
al
u
at
e
th
e
p
er
f
o
r
m
an
ce
o
f
t
h
e
S
C
_
MR
A
S
u
s
in
g
N
N
o
b
s
er
v
er
a
s
en
s
o
r
less
v
ec
to
r
co
n
tr
o
l
o
f
SP
I
M
d
r
iv
e
s
y
s
te
m
,
as
s
h
o
w
n
i
n
Fi
g
u
r
e
5
h
as
b
ee
n
s
i
m
u
lated
at
d
if
f
er
en
t
s
p
ee
d
r
an
g
es
th
r
o
u
g
h
Ma
tlab
s
i
m
u
latio
n
s
o
f
t
w
ar
e,
s
p
ec
iall
y
s
u
r
v
e
y
ed
at
lo
w
s
p
ee
d
r
an
g
e.
T
ests
i
n
t
h
is
s
ec
tio
n
ar
e
co
n
d
u
cted
b
ased
o
n
r
ec
o
m
m
e
n
d
ed
b
en
ch
m
ar
k
te
s
ts
[
17
],
[2
0
],
[
21
]
,
[
2
3
]
.
SP
I
M
p
ar
am
eter
s
:
1
HP
,
2
2
0
V,
5
0
Hz,
4
p
o
le
,
1
4
5
0
r
p
m
.
R
s
=
1
0
.
1
,
R
r
=
9
.
8
5
4
6
,
L
s
=
0
.
8
3
3
4
5
7
H,
L
r
=
0
.
8
3
0
8
1
1
H,
m
=
0
.
7
8
3
1
0
6
H,
J
i
=
0
.
0
0
8
8
k
g
.
m
2
.
R
s
i
s
n
o
m
i
n
al
v
alu
e
o
f
s
ta
to
r
r
esis
tan
ce
.
3
_
P
h
a
se
A
C
S
u
p
p
l
y
-
+
-
+
-
+
PI
PI
PI
T
2
T
2
-
1
R
ect
i
f
i
er
b
r
i
d
g
e
6
P
h
a
s
e
I
n
v
e
r
t
e
r
D
C
l
i
n
k
i
s
d
C
u
r
r
en
t
C
a
l
c
u
l
a
t
o
r
i
s
d
C
u
r
r
e
n
t
C
o
n
t
r
o
l
l
e
r
i
s
q
C
u
r
r
en
t
C
o
n
t
r
o
l
l
er
S
p
e
e
d
C
o
n
t
r
o
ll
er
L
e
a
r
n
i
n
g
A
l
g
o
r
i
t
h
m
LS
N
e
u
r
a
l
n
e
t
w
o
r
k
A
d
a
p
t
i
v
e
M
o
d
e
l
VM
S
P
I
M
T
6
-
1
T
6
i
sd
i
sq
u
sd
u
sd
u
sD
i
sD
i
sQ
i
sq
i
sd
u
sQ
u
sD
u
sQ
i
sD
i
sQ
i
s
u
s
i
s
θ
s
ω
*
*
*
*
ψ
r
LS
_
SC
_
MR
A
S
O
b
s
e
r
v
e
r
Rs
,
R
r
E
s
t
i
m
a
t
o
r
R
o
t
o
r
f
l
u
x
I
d
e
n
t
i
f
i
e
r
Fig
u
r
e
5
.
Sen
s
o
r
les
s
v
ec
to
r
co
n
tr
o
l o
f
SP
I
M
d
r
iv
e
u
s
i
n
g
L
S_
SC
_
MR
AS o
b
s
er
v
er
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
P
o
w
E
lec
&
Dr
i
S
y
s
t
I
SS
N:
2
0
8
8
-
8
694
N
ew V
er
s
io
n
o
f A
d
a
p
tive
S
p
ee
d
Ob
s
erver B
a
s
ed
o
n
N
eu
r
a
l
N
etw
o
r
k
fo
r
S
P
I
M (
N
g
o
c
Th
u
y
P
h
a
m)
1495
4
.
1
.
Dy
na
m
ic
P
er
f
o
r
m
a
nce
I
n
th
i
s
f
ir
s
t
p
ar
t,
th
e
d
y
n
a
m
ic
p
er
f
o
r
m
a
n
ce
o
f
t
h
e
d
r
iv
e
an
d
o
b
s
er
v
er
h
av
e
b
ee
n
v
er
i
f
ied
b
y
th
e
te
s
ts
ar
e
co
n
d
u
cted
b
ased
o
n
r
ec
o
m
m
e
n
d
ed
B
en
ch
m
ar
k
tes
ts
in
[
1
7
]
.
T
est
1
p
r
esen
ts
r
a
p
id
tr
an
s
itio
n
s
a
n
d
o
p
er
atin
g
ar
ea
s
at
lar
g
e
a
n
d
ze
r
o
s
p
ee
d
s
,
T
est 2
co
n
s
is
ts
o
f
lo
w
a
n
d
v
er
y
lo
w
s
p
ee
d
o
p
er
atio
n
an
d
r
ev
er
s
al.
I
n
T
est
1
,
th
e
r
ef
er
en
ce
s
p
ee
d
ar
e
i
m
p
o
s
ed
f
r
o
m
ze
r
o
i
n
cr
ea
s
ed
to
1
5
5
r
a
d
/s
at
0
.
2
s
an
d
co
n
s
ta
n
t
u
p
to
1
,
5
s
.
T
h
en
,
it
is
r
ed
u
ce
d
t
o
ze
r
o
at
1
,
7
s
,
th
e
r
ef
er
en
ce
s
p
ee
d
is
th
e
n
m
a
in
tai
n
ed
at
s
t
an
d
s
till
u
n
til
2
,
5
s
.
T
h
en
,
b
et
w
ee
n
2
,
5
s
an
d
4
,
8
s
,
a
q
u
asi
-
s
y
m
m
etr
ic
v
e
lo
cit
y
p
r
o
f
i
le
is
i
m
p
o
s
ed
i
n
th
e
o
p
p
o
s
ite
r
o
tatio
n
al
d
ir
ec
tio
n
,
d
ef
i
n
i
n
g
a
s
ec
o
n
d
ar
ea
o
f
cr
itica
l o
p
er
atin
g
at
a
r
at
e
o
f
-
1
5
5
r
ad
/s
b
et
w
ee
n
4
s
a
n
d
4
,
8
s
.
Fi
n
all
y
,
t
h
e
ma
ch
in
e
s
p
ee
d
is
a
g
ai
n
i
n
cr
ea
s
ed
to
1
5
5
r
ad
/s
,
to
b
e
th
er
e
m
ai
n
tai
n
ed
u
p
to
6
s
.
R
ated
lo
ad
is
ap
p
lied
at
0
.
8
s
an
d
r
ej
ec
ted
at
1
.
2
s
.
T
h
e
ap
p
licatio
n
a
n
d
r
e
m
o
v
al
o
f
a
to
r
q
u
e
lo
ad
at
th
e
m
o
m
e
n
ts
0
,
8
s
an
d
1
,
2
s
w
ill
also
ass
es
s
t
h
e
i
m
p
ac
t
o
f
t
h
i
s
t
y
p
e
o
f
d
is
t
u
r
b
an
ce
.
T
h
en
,
5
0
%
lo
ad
is
ap
p
lied
an
d
r
ej
ec
ted
at
3
,
2
5
s
an
d
5
,
5
s
,
r
esp
ec
tiv
el
y
.
T
h
e
r
esu
l
ts
i
n
F
ig
u
r
e
6
s
h
o
w
t
h
e
s
p
ee
d
r
esp
o
n
s
es,
t
h
e
s
p
ee
d
esti
m
at
io
n
er
r
o
r
,
th
e
s
tato
r
cu
r
r
en
t
an
d
to
r
q
u
e
d
u
r
in
g
te
s
t
1
.
Fro
m
t
h
ese
s
i
m
u
la
tio
n
r
es
u
lt
s
s
h
o
w
t
h
at
al
th
o
u
g
h
s
u
r
v
e
y
ed
w
it
h
lar
g
er
s
p
ee
d
r
an
g
e
co
m
p
ar
e
to
i
n
[
1
7
]
,
th
e
esti
m
atio
n
p
er
f
o
r
m
a
n
ce
o
f
L
S_
S
C
_
MR
A
S
o
b
s
er
v
er
is
v
e
r
y
g
o
o
d
at
h
ig
h
,
lo
w
an
d
ze
r
o
s
p
ee
d
an
d
r
ev
er
s
al.
T
h
e
s
p
ee
d
esti
m
atio
n
er
r
o
r
is
t
h
e
esti
m
ated
s
p
ee
d
p
er
f
ec
tl
y
f
o
llo
w
s
th
e
m
ac
h
i
n
e
s
p
ee
d
w
it
h
g
o
o
d
b
eh
av
io
r
in
t
er
m
s
o
f
tr
ac
k
i
n
g
an
d
d
is
tu
r
b
a
n
ce
r
ej
ec
tio
n
.
B
o
th
th
e
p
r
o
p
o
s
ed
s
ch
e
m
e
i
n
[
1
7
]
an
d
L
S_
S
C
_
MR
AS
s
c
h
e
m
e,
t
h
e
s
p
ee
d
r
ev
er
s
al
is
ac
co
m
p
lis
h
ed
in
les
s
th
a
n
1
s
an
d
th
at
t
h
e
to
r
q
u
e
r
esp
o
n
s
e
is
in
s
ta
n
ta
n
eo
u
s
.
T
h
e
s
p
ee
d
er
r
o
r
is
m
a
x
i
m
u
m
at
ze
r
o
cr
o
s
s
i
n
g
a
n
d
d
u
r
i
n
g
th
e
s
p
ee
d
tr
an
s
i
e
n
t
a
n
d
it
is
ab
o
u
t
as
m
u
c
h
as
0
.
1
2
r
ad
/s
in
L
S_
SC
_
MR
A
S o
b
s
er
v
er
,
0
.
5
r
ad
/s
in
th
e
p
r
o
p
o
s
ed
o
b
s
er
v
er
[
1
7
]
.
(
a)
(
b
)
Fig
u
r
e
6
.
T
h
e
p
e
r
f
o
r
m
a
n
ce
o
f
L
S S
C
_
MR
AS o
b
s
ev
er
d
u
r
i
n
g
h
i
g
h
s
p
ee
d
r
ev
er
al
(
a)
Sp
ee
d
r
esp
o
n
s
es a
n
d
er
r
o
r
,
r
esp
ec
tiv
el
y
; (
b
)
Stato
r
cu
r
r
en
t p
h
ase
A
an
d
to
r
q
u
e
r
esp
o
n
e
Fu
r
t
h
er
m
o
r
e,
a
s
p
ee
d
r
ev
er
s
al
f
r
o
m
1
5
r
ad
/s
to
-
8
r
ad
/s
a
n
d
8
r
ad
/s
,
at
5
0
%lo
ad
h
as
b
ee
n
p
er
f
o
r
m
ed
f
o
r
test
in
g
t
h
e
d
y
n
a
m
ic
p
er
f
o
r
m
an
ce
s
o
f
th
e
d
r
iv
e
u
s
i
n
g
L
S
-
S
C
_
MR
AS
at
lo
w
s
p
ee
d
.
Fig
u
r
e
7
s
h
o
w
s
th
e
s
p
ee
d
an
d
to
r
q
u
e
r
esp
o
n
s
e
o
f
L
S_
S
C
_
MR
AS
d
r
iv
e
is
v
er
y
q
u
ick
a
n
d
th
at
a
f
ter
f
e
w
o
s
cill
atio
n
s
it
co
n
v
er
g
e
s
to
th
e
r
ef
er
en
ce
v
al
u
e.
W
e
o
b
s
er
v
e
v
er
y
s
m
all
o
s
cillat
io
n
s
d
u
r
in
g
tr
a
n
s
ie
n
t
s
an
d
v
er
y
lo
w
o
r
ien
tatio
n
er
r
o
r
s
in
th
e
t
w
o
cr
itical
ar
ea
s
o
f
t
h
e
p
r
o
p
o
s
ed
o
b
s
er
v
er
.
0
1
2
3
4
5
-
2
0
0
-
1
0
0
0
100
200
T
i
m
e
(s
)
S
p
e
e
d
(ra
d
/
s
)
R
e
f
e
r
e
n
c
e
M
e
a
s
u
r
e
d
E
s
t
i
m
a
t
e
d
0
2
4
6
-1
-0.5
0
0.5
1
T
i
m
e
(s
)
E
rro
r
(ra
d
/
s
)
0
1
2
3
4
5
-5
0
5
T
i
m
e
(s
)
S
t
a
t
o
r
c
u
rre
n
t
i
s
a
(A
)
0
1
2
3
4
5
-1
0
0
10
T
i
m
e
(s
)
T
o
rq
u
e
(N
.
m
)
L
oa
d
t
or
qu
e
E
l
e
c
t
r
om
a
gn
e
t
i
c
t
or
qu
e
0
.
6
0
.
8
1
1
.
2
1
.
4
150
160
Evaluation Warning : The document was created with Spire.PDF for Python.