I
nte
rna
t
io
na
l J
o
urna
l o
f
Appl
ied P
o
w
er
E
ng
ineering
(
I
J
AP
E
)
Vo
l.
4
,
No
.
1
,
A
p
r
il
201
5
,
p
p
.
1
~
12
I
SS
N:
2252
-
8792
1
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
jo
u
r
n
a
l.c
o
m/o
n
lin
e/in
d
ex
.
p
h
p
/I
J
APE
An
Ada
pt
iv
e
RT
RL
Ba
sed
Neu
ro
co
ntroller
for
Da
mping
Pow
er
Sy
ste
m
O
s
cilla
tions
K
.
C.
Sindhu
T
ha
m
pa
t
t
y
*
,
P
.
C.
Reg
hu
Ra
j
**
*
D
e
p
a
rt
m
e
n
t
o
f
El
e
c
tri
c
a
l
a
n
d
El
e
c
tro
n
ics
En
g
in
e
e
rin
g
,
A
m
rit
a
V
iswa
V
id
y
a
p
e
e
th
a
m
,
Et
ti
m
a
d
a
i,
Co
im
b
a
to
re
,
Am
rit
a
n
a
g
a
r
(P
.
O),
T
a
m
il
n
a
d
u
,
I
n
d
ia
.
*
*
De
p
a
rt
m
e
n
t
o
f
Co
m
p
u
ter S
c
ien
c
e
a
n
d
E
n
g
in
e
e
rin
g
,
G
o
v
t.
En
g
in
e
e
rin
g
Co
ll
e
g
e
,
S
re
e
k
rish
n
a
p
u
ra
m
,
P
a
lak
k
a
d
,
Ke
ra
la,
In
d
ia
.
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
an
2
,
2
0
1
5
R
ev
i
s
ed
Feb
1
0
,
2
0
1
5
A
cc
ep
ted
Mar
2
3
,
2
0
1
5
T
h
e
m
a
in
o
b
jec
ti
v
e
o
f
th
is
p
a
p
e
r
is
to
p
re
se
n
t
t
h
e
d
e
sig
n
o
f
a
n
a
d
a
p
ti
v
e
n
e
u
ro
-
c
o
n
tr
o
ll
e
r
f
o
r
se
ries
c
o
n
n
e
c
ted
F
A
CT
S
d
e
v
ice
s
li
k
e
T
h
y
risto
r
Co
n
tr
o
ll
e
d
S
e
ries
Ca
p
a
c
it
o
r
(T
CS
C)
a
n
d
T
h
y
risto
r
c
o
n
tro
ll
e
d
P
o
w
e
r
A
n
g
le
Re
g
u
lato
r
(T
CP
A
R).
T
h
is
c
o
n
tr
o
l
sc
h
e
m
e
is
su
it
a
b
le
f
o
r
n
o
n
-
li
n
e
a
r
s
y
ste
m
c
o
n
tro
l,
i
n
w
h
ich
th
e
e
x
a
c
t
li
n
e
a
rise
d
m
a
th
e
m
a
ti
c
a
l
m
o
d
e
l
o
f
th
e
sy
st
e
m
is
n
o
t
re
q
u
ire
d
.
T
h
e
p
r
o
p
o
se
d
c
o
n
tr
o
ll
e
r
d
e
sig
n
is
b
a
se
d
o
n
Re
a
l
T
i
m
e
Re
c
u
rre
n
t
L
e
a
rn
in
g
(RT
R
L
)
a
lg
o
r
it
h
m
in
w
h
ich
th
e
Ne
u
ra
l
Ne
t
w
o
rk
(NN
)
is
train
e
d
i
n
re
a
l
ti
m
e
.
T
h
is
c
o
n
tro
l
sc
h
e
m
e
re
q
u
ires
tw
o
se
ts
o
f
n
e
u
ra
l
n
e
tw
o
rk
s.
T
h
e
fi
rst
se
t
is
a
n
e
u
ro
-
id
e
n
ti
fi
e
r
a
n
d
th
e
se
c
o
n
d
se
t
i
s
a
n
e
u
ro
-
c
o
n
tro
ll
e
r
w
h
ich
g
e
n
e
ra
tes
th
e
r
e
q
u
ired
c
o
n
tr
o
l
sig
n
a
ls
f
o
r
t
h
e
th
y
rist
o
rs.
P
e
rf
o
rm
a
n
c
e
o
f
th
e
s
y
ste
m
is
a
n
a
ly
s
e
d
w
it
h
th
e
p
ro
p
o
se
d
c
o
n
tr
o
ll
e
r
u
si
n
g
sta
n
d
a
rd
sim
u
latio
n
e
n
v
iro
n
m
e
n
ts
li
k
e
M
ATLA
B/S
IM
UL
INK
a
n
d
it
h
a
s
b
e
e
n
o
b
se
rv
e
d
th
a
t
th
e
c
o
n
tr
o
ll
e
ris
ro
b
u
st
a
n
d
t
h
e
re
sp
o
n
se
is
v
e
r
y
fa
st.
P
e
rf
o
rm
a
n
c
e
o
f
th
e
s
y
ste
m
w
it
h
p
ro
p
o
se
d
c
o
n
tro
ll
e
r
is
c
o
m
p
a
re
d
w
it
h
c
o
n
v
e
n
ti
o
n
a
l
P
I
c
o
n
tr
o
ll
e
rs
a
n
d
GA
b
a
se
d
P
I
c
o
n
tro
ll
e
rs.
P
e
rf
o
rm
a
c
e
o
f
th
e
p
ro
p
o
se
d
c
o
n
tr
o
ll
e
r
is
e
x
trem
e
l
y
g
o
o
d
.
K
ey
w
o
r
d
:
T
h
y
r
i
s
to
r
C
o
n
tr
o
lled
Ser
ies
C
ap
ac
ito
r
(
T
C
SC
)
T
h
y
r
is
to
r
co
n
tr
o
lled
P
o
w
er
A
n
g
le
R
eg
u
lato
r
(
T
C
P
A
R
)
R
ea
l T
i
m
e
R
ec
u
r
r
en
t L
ea
r
n
i
n
g
A
l
g
o
r
ith
m
(
R
T
R
L
)
R
ec
u
r
r
en
t N
e
u
r
al
Net
w
o
r
k
(
R
NN)
D
y
n
a
m
ic
Neu
r
al
Net
w
o
r
k
(
DNN)
Co
p
y
rig
h
t
©
2
0
1
4
In
stit
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
.
Al
l
rig
h
ts
re
se
rv
e
d
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
K.
C
.
Si
n
d
h
u
T
h
a
m
p
att
y
Dep
ar
t
m
en
t o
f
E
lectr
ical
an
d
E
lectr
o
n
ics E
n
g
i
n
ee
r
in
g
,
Am
r
ita
Vi
s
w
a
Vid
y
ap
ee
t
h
a
m
,
E
tti
m
ad
ai,
C
o
i
m
b
ato
r
e,
Am
r
ita
n
a
g
ar
(
P
.
O)
,
T
am
iln
ad
u
,
I
n
d
ia
-
641112
.
E
m
ail
:
k
c_
s
in
d
h
u
@
y
a
h
o
o
.
co
.
in
1.
I
NT
RO
D
UCT
I
O
N
In
lar
g
e
in
ter
co
n
n
ec
ted
p
o
w
er
s
y
s
te
m
s
,
lo
w
f
r
eq
u
e
n
c
y
elec
tr
o
m
ec
h
a
n
ical
o
s
cilla
tio
n
s
w
i
th
f
r
eq
u
en
c
y
0
.
2
Hz
to
2
Hz
h
av
e
b
ee
n
o
b
s
er
v
ed
f
r
eq
u
e
n
tl
y
,
w
h
ich
w
i
ll
r
ed
u
ce
th
e
e
f
fi
cie
n
c
y
of
t
h
e
s
y
s
te
m
co
n
s
id
er
ab
l
y
.
C
o
n
v
en
t
io
n
all
y
,
t
h
ese
o
s
c
illati
o
n
s
ar
e
d
a
m
p
ed
u
s
i
n
g
P
o
w
er
S
y
s
te
m
Stab
ilizer
s
(
P
SS
)
.
In
o
r
d
er
to
d
am
p
th
e
s
e
elec
tr
o
m
ec
h
a
n
ical
o
s
ci
llatio
n
s
in
p
o
w
er
s
y
s
te
m
s
,
s
u
p
p
le
m
e
n
tar
y
co
n
tr
o
l
ac
tio
n
,
p
o
w
er
o
s
cillatio
n
d
a
m
p
i
n
g
(
P
OD)
can
be
ap
p
lied
to
s
o
m
e
of
th
e
F
AC
T
S d
ev
ices.
Mo
s
t
o
f
t
h
e
co
n
tr
o
ller
s
d
esi
g
n
ed
f
o
r
d
a
m
p
in
g
p
o
w
er
s
y
s
te
m
o
s
cilla
tio
n
s
[
1
]
.
[
3
]
a
r
e
b
ased
o
n
a
lin
ea
r
is
ed
m
o
d
el,
th
e
p
er
f
o
r
m
an
ce
o
f
w
h
ic
h
w
il
l d
eter
io
r
ate
f
o
r
w
id
e
v
ar
y
in
g
o
p
er
atin
g
co
n
d
itio
n
s
in
p
r
esen
ce
o
f
lar
g
e
d
is
t
u
r
b
an
ce
.
I
n
o
r
d
er
to
o
v
er
co
m
e
t
h
i
s
,
in
tel
lig
e
n
t
a
d
ap
t
iv
e
co
n
tr
o
ller
s
ar
e
d
esig
n
ed
w
h
ich
d
o
es
n
o
t
r
eq
u
ir
e
a
p
er
f
ec
t
m
at
h
e
m
atic
al
m
o
d
el
o
f
t
h
e
s
y
s
te
m
[
4
,
5
]
.
T
h
e
v
ar
iatio
n
i
n
p
lan
t
p
a
r
a
m
eter
s
a
n
d
p
lan
t
s
tr
u
ct
u
r
es
ca
n
b
e
ef
f
ec
ti
v
el
y
u
p
d
ated
in
an
in
te
lli
g
en
t
co
n
t
r
o
l
s
tr
ateg
y
s
o
th
a
t
th
e
s
y
s
te
m
is
m
o
r
e
r
o
b
u
s
t.
A
n
eu
r
al
n
et
w
o
r
k
b
ased
co
n
tr
o
ll
er
is
s
u
c
h
an
i
n
tell
ig
e
n
t c
o
n
tr
o
ller
w
h
ic
h
ca
n
o
p
er
ate
v
er
y
f
a
s
t.
Mo
s
t
o
f
t
h
e
co
m
m
o
n
l
y
a
v
aila
b
le
n
eu
r
al
n
et
w
o
r
k
s
h
a
v
e
d
r
a
w
b
ac
k
s
s
u
c
h
as
lar
g
e
tr
ai
n
i
n
g
ti
m
e,
lar
g
e
n
u
m
b
er
o
f
n
eu
r
o
n
s
a
n
d
m
o
r
e
h
id
d
en
la
y
er
s
r
eq
u
ir
ed
to
d
ea
l
w
ith
co
m
p
lex
p
r
o
b
le
m
s
[
6
,
7
]
.
T
h
e
tr
ain
in
g
tech
n
iq
u
es
a
v
ailab
le
f
o
r
t
h
ese
n
eu
r
al
n
et
w
o
r
k
[
8
]
.
.
[
1
0
]
ar
e
u
n
ab
le
to
r
etain
th
e
i
n
f
o
r
m
atio
n
ab
o
u
t
th
e
i
n
fin
ite
p
ast
w
h
ich
i
s
ess
e
n
tial
f
o
r
r
ea
l
-
ti
m
e
ap
p
licatio
n
s
.
D
y
n
a
m
ic
R
ec
u
r
r
en
t
Ne
u
r
o
n
(
DR
N)
w
it
h
a
f
ee
d
b
ac
k
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8792
IJ
A
P
E
Vo
l.
4
,
No
.
1
,
A
p
r
il
2
0
1
4
:
1
–
12
2
co
n
n
ec
ti
o
n
f
r
o
m
t
h
e
o
u
tp
u
t
to
th
e
in
p
u
t
is
t
h
e
m
o
s
t
s
u
i
tab
le
ar
ch
itectu
r
e
f
o
r
r
ea
l
-
ti
m
e
lear
n
i
n
g
.
DR
N
r
ep
r
esen
ts
t
h
e
i
n
ter
n
a
l o
r
h
id
d
en
s
ta
tes,
i
n
a
p
o
ten
t
iall
y
d
is
tr
i
b
u
ted
f
as
h
io
n
,
w
h
ich
lead
s
to
ca
p
ab
ilit
ies th
a
t a
r
e
s
i
m
ilar
to
th
o
s
e
o
f
a
n
o
b
s
er
v
er
in
m
o
d
er
n
co
n
tr
o
l t
h
eo
r
y
.
Dif
f
er
en
t
tr
ai
n
i
n
g
m
et
h
o
d
s
ar
e
av
ailab
le
f
o
r
r
ec
u
r
r
en
t
n
eu
r
o
n
s
[
1
1
]
.
[
1
3
]
.
R
ea
l
tim
e
R
ec
u
r
r
en
t
L
ea
r
n
i
n
g
(
R
T
R
L
)
al
g
o
r
ith
m
[
1
4
]
is
o
n
e
o
f
th
e
m
o
s
t
s
u
ita
b
le
alg
o
r
ith
m
f
o
r
tr
ai
n
in
g
t
h
e
R
ec
u
r
r
en
t
Neu
r
al
Net
w
o
r
k
(
R
NN)
f
o
r
r
ea
l
-
ti
m
e
a
p
p
licatio
n
.
In
all
th
er
e
p
o
r
ted
w
o
r
k
s
,
a
f
u
ll
y
co
n
n
ec
ted
R
NN
is
s
eld
o
m
u
s
ed
f
o
r
t
h
e
d
a
m
p
i
n
g
of
p
o
wer
s
y
s
te
m
o
s
cillatio
n
s
.
T
h
e
p
r
o
p
o
s
ed
n
eu
r
o
-
co
n
tr
o
ller
ar
ch
itect
u
r
e
is
s
u
itab
le
f
o
r
Mu
ltip
le
-
I
n
p
u
t
Mu
ltip
le
-
Ou
tp
u
t
(
MI
MO
)
n
o
n
-
lin
ea
r
s
y
s
te
m
s
w
h
ic
h
ca
n
b
e
tr
ai
n
ed
b
y
R
T
R
L
alg
o
r
it
h
m
.
I
t
is
a
n
o
p
ti
m
al
alg
o
r
ith
m
w
h
ic
h
m
i
n
i
m
is
e
s
t
h
e
i
n
s
tan
ta
n
eo
u
s
s
q
u
ar
ed
er
r
o
r
at
th
e
o
u
tp
u
t
o
f
R
NN
f
o
r
ev
er
y
d
is
cr
ete
ti
m
e,
w
h
ile
th
e
n
et
w
o
r
k
i
s
r
u
n
n
i
n
g
.
Nu
m
b
er
o
f
n
e
u
r
o
n
s
i
n
t
h
e
o
u
tp
u
t
la
y
er
o
f
R
NN
is
e
q
u
al
to
t
h
e
n
u
m
b
er
of
s
tate
s
of
th
e
s
y
s
te
m
a
n
d
t
h
e
n
u
m
b
er
of
n
e
u
r
o
n
s
in
t
h
e
co
n
t
r
o
ller
n
et
w
o
r
k
m
u
s
t
b
esa
m
e
as
th
e
n
u
m
b
er
co
n
tr
o
l i
n
p
u
t
s
.
T
h
is
p
ap
er
ex
a
m
i
n
es
t
h
e
i
m
p
r
o
v
e
m
e
n
t
in
d
a
m
p
i
n
g
o
f
p
o
w
e
r
s
y
s
te
m
o
s
cillatio
n
s
w
it
h
R
T
R
L
b
ased
n
eu
r
o
co
n
tr
o
ller
f
o
r
s
er
ie
s
co
n
n
ec
ted
F
AC
T
S
d
ev
ices
li
k
e
T
C
SC
an
d
T
C
P
A
R
.
A
s
y
s
te
m
atic
p
r
o
ce
d
u
r
e
f
o
r
m
o
d
eli
n
g
a
n
d
s
i
m
u
latio
n
o
f
a
s
i
n
g
le
-
m
ac
h
i
n
e
i
n
fin
ite
-
bus
(
SMI
B
)
p
o
w
er
s
y
s
te
m
in
s
talle
d
w
i
th
T
C
SC
an
d
T
C
P
A
R
is
d
ev
elo
p
ed
.
T
h
e
d
ev
elo
p
ed
p
o
w
er
s
y
s
te
m
m
o
d
el
is
s
i
m
u
lated
u
s
in
g
M
A
T
L
A
B
/SIM
U
L
I
NK
f
o
r
d
if
f
er
e
n
t o
p
er
atin
g
co
n
d
itio
n
s
an
d
its
p
er
f
o
r
m
a
n
ce
v
er
i
fied
.
T
h
e
p
ap
e
r
is
o
r
g
an
is
ed
as
f
o
ll
o
w
s
.
Sectio
n
2
b
r
iefly
e
x
p
lai
n
s
th
e
m
ath
e
m
atica
l
m
o
d
elli
n
g
o
f
p
o
w
er
s
y
s
te
m
w
i
th
T
C
SC
a
n
d
T
C
PAR
.
T
h
e
d
etailed
d
esig
n
o
f
th
e
p
r
o
p
o
s
ed
a
d
ap
tiv
e
n
eu
r
o
-
co
n
tr
o
ller
is
ex
p
lain
ed
in
s
ec
tio
n
3
f
o
llo
w
ed
b
y
t
h
e
d
i
s
cu
s
s
io
n
o
n
s
i
m
u
latio
n
r
es
u
lt
s
in
s
ec
tio
n
4
.
Sectio
n
5
g
i
v
es t
h
e
co
n
clu
s
io
n
.
2.
P
O
WE
R
SY
ST
E
M
M
O
DE
L
I
n
th
i
s
s
t
u
d
y
,
SMI
B
s
y
s
te
m
in
s
tal
led
w
it
h
T
C
SC
a
n
d
T
C
P
AR
is
i
n
v
e
s
ti
g
ated
.
T
h
e
s
y
n
ch
r
o
n
o
u
s
g
en
er
ato
r
is
d
eliv
er
i
n
g
p
o
w
er
to
th
e
in
fin
ite
-
bus
th
r
o
u
g
h
s
er
ies
co
m
p
en
s
ated
tr
a
n
s
m
is
s
i
o
n
lin
e
as
s
h
o
w
n
in
Fig
u
r
e
.1
in
w
h
ic
h
th
e
li
n
e
is
co
n
n
ec
ted
w
it
h
T
C
SC
an
d
T
C
P
AR
.
In
Fig
u
r
e
.
1
,
Vs
an
d
Vr
ar
e
th
e
g
en
er
ato
r
ter
m
i
n
al
an
d
i
n
fin
ite
b
u
s
v
o
ltag
e
r
esp
ec
ti
v
el
y
.
I
n
th
i
s
wo
r
k
,
th
e
m
ec
h
a
n
ical
s
y
s
te
m
,
elec
tr
ical
s
y
s
te
m
,
tr
an
s
m
is
s
io
n
li
n
es
a
n
d
F
A
C
T
S
d
ev
ices
ar
e
m
o
d
eled
s
ep
ar
atel
y
an
d
ar
e
th
en
i
n
ter
co
n
n
ec
ted
to
f
o
r
m
th
e
co
m
p
lete
s
y
s
te
m
[
1
4
]
.
Fig
u
r
e
1
.
On
e
L
in
e
Dia
g
r
a
m
o
f
T
h
e
S
y
s
te
m
2
.
1
.
M
o
dellin
g
o
f
T
CSC
a
nd
T
CP
AR
Gen
er
all
y
,
th
e
s
er
ies
co
n
n
ec
t
ed
FAC
T
S
d
ev
ices
ar
e
u
s
ed
t
o
i
m
p
r
o
v
e
t
h
e
p
o
w
er
fl
o
w
th
r
o
u
g
h
th
e
tr
an
s
m
is
s
io
n
li
n
es.
D
u
e
to
th
e
d
ev
elo
p
m
en
t
i
n
p
o
w
er
elec
tr
o
n
ics,
th
e
co
n
tr
o
l
o
f
p
o
w
er
fl
o
w
ca
n
b
e
d
o
n
e
m
u
c
h
f
aster
.
Als
o
t
h
e
e
n
h
a
n
ce
m
en
t
o
f
p
o
w
er
s
y
s
te
m
s
tab
ilit
y
ca
n
b
e
ac
h
ie
v
ed
w
it
h
t
h
ese
f
a
s
t
ac
tin
g
s
w
itch
e
s
.
I
n
th
is
w
o
r
k
,
t
w
o
s
er
ies
co
n
n
ec
t
ed
FA
C
T
S
d
ev
ices,
T
C
SC
an
d
T
C
P
A
R
ar
e
co
n
tr
o
lled
b
y
co
n
tr
o
llin
g
th
e
fi
r
i
n
g
an
g
le
s
o
f
th
e
s
w
itc
h
es.
A
Q
u
ad
r
atu
r
e
B
o
o
s
ter
T
r
an
s
f
o
r
m
er
(
QB
T
)
,
in
j
ec
tin
g
v
o
lta
g
e
i
n
9
0
⸰
w
i
th
p
h
a
s
e
v
o
ltag
e
is
co
n
s
id
er
ed
as
th
e
Po
w
er
An
g
le
R
e
g
u
la
to
r
.
Mo
d
el
lin
g
o
f
T
C
SC
an
d
T
C
P
A
R
ar
e
ex
p
lain
ed
in
th
e
f
o
llo
w
in
g
s
e
s
s
io
n
s
.
(
1
)
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
A
P
E
I
SS
N:
2252
-
8792
A
n
A
d
a
p
tive
R
TRL B
a
s
ed
N
eu
r
o
co
n
t
r
o
ller
fo
r
Da
mp
in
g
P
o
w
er S
ystem…
(
K
.
C
.
S
in
d
h
u
Th
a
mp
a
tty
)
3
(
2
)
w
h
er
e,
W
h
en
th
e
t
h
y
r
i
s
to
r
s
ar
e
tu
r
n
e
d
o
f
f
,
th
e
cu
r
r
en
t
t
h
r
o
u
g
h
t
h
e
in
d
u
cto
r
is
ze
r
o
.
I
n
th
is
co
n
d
itio
n
,
th
e
T
C
SC
can
be
m
o
d
eled
as
a
s
i
m
p
le
s
er
ie
s
ca
p
ac
ito
r
cir
cu
it.
T
h
e
co
r
r
esp
o
n
d
in
g
v
o
ltag
e
eq
u
atio
n
f
o
r
p
h
a
s
e
a
ca
n
b
e
w
r
itte
n
as:
(
3
)
C
o
m
b
i
n
in
g
eq
u
at
io
n
s
1
,
2
an
d
3
,
th
e
s
tate
eq
u
atio
n
o
f
T
C
SC
f
o
r
p
h
ase
a
ca
n
b
e
w
r
i
tten
a
s
:
(
4
)
(
5
)
w
h
er
e
Si
m
i
lar
s
tate
s
p
ac
e
eq
u
atio
n
s
ca
n
b
e
w
r
itte
n
f
o
r
o
th
er
p
h
ase
s
also
.
I
n
o
r
d
er
to
o
p
er
ate
th
e
T
C
SC
in
th
e
ca
p
ac
itiv
e
r
e
g
io
n
,
t
h
e
fi
r
in
g
an
g
le
α
m
u
s
t b
e
v
ar
ied
f
r
o
m
150
⸰
to
1
7
5
⸰
.
2
.
1
.
2
.
M
o
dellin
g
of
P
ha
s
e
An
g
le
Reg
ula
t
o
r
We
ad
o
p
t
a
q
u
ad
r
atu
r
e
v
o
lta
g
e
in
j
ec
tio
n
f
o
r
r
eg
u
lati
n
g
t
h
e
p
h
ase
a
n
g
le
[
1
5
]
.
T
h
is
can
be
ac
h
ie
v
ed
u
s
i
n
g
QB
T
.
T
h
e
in
j
ec
ted
v
o
lta
g
e
is
g
iv
e
n
as
:
(
6
)
(
7
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8792
IJ
A
P
E
Vo
l.
4
,
No
.
1
,
A
p
r
il
2
0
1
4
:
1
–
12
4
(
8
)
(
9
)
(
1
0
)
3.
R
T
R
L
B
ASE
D
AD
AP
T
I
VE
CO
NT
RO
L
L
E
R
T
h
e
p
r
o
p
o
s
ed
n
eu
r
o
-
co
n
tr
o
ll
er
s
h
o
w
n
in
Fig
u
r
e
.
2
co
n
s
i
s
ts
o
f
a
n
e
u
r
o
-
id
en
ti
fi
er
a
n
d
a
n
eu
r
o
-
co
n
tr
o
ller
,
w
h
ich
ca
n
b
e
r
ea
l
ized
u
s
in
g
a
f
u
ll
y
co
n
n
ec
ted
Dyn
a
mic
R
ec
u
r
r
en
t
N
eu
r
a
l
N
etw
o
r
k
(
D
R
NN
)
ar
ch
itect
u
r
e
[
1
8
]
w
i
th
all
th
e
s
y
s
te
m
o
u
tp
u
ts
f
ed
b
ac
k
to
t
h
e
in
p
u
t
t
h
r
o
u
g
h
a
d
ela
y
.
T
h
e
n
e
u
r
o
-
id
en
t
i
fi
er
tr
ac
k
s
th
e
d
y
n
a
m
ic
p
r
o
p
er
ties
o
f
th
e
s
y
s
te
m
an
d
th
e
n
eu
r
o
-
co
n
tr
o
ller
p
r
o
v
id
es
th
e
n
ec
ess
ar
y
c
o
n
tr
o
l
s
ig
n
al
to
th
e
p
lan
t.
A
s
et
o
f
f
ee
d
b
ac
k
w
ei
g
h
ts
co
n
tr
o
ls
t
h
e
a
m
o
u
n
t o
f
f
ee
d
b
ac
k
to
ea
ch
p
r
o
ce
s
s
in
g
n
e
u
r
o
n
.
Fig
u
r
e
2
.
R
T
R
L
-
B
ased
C
o
n
tr
o
ller
w
it
h
Ne
u
r
o
-
I
d
en
ti
fier
an
d
Neu
r
o
-
C
o
n
tr
o
ller
3
.
1
.
Rec
urre
nt
Neura
l
Net
wo
rk
(
RNN)
Arc
hite
ct
ure
R
ec
u
r
r
en
t
Ne
u
r
al
Net
w
o
r
k
s
i
s
a
d
y
n
a
m
ic
s
y
s
te
m
s
w
h
ic
h
is
ca
p
ab
le
o
f
s
to
r
in
g
te
m
p
o
r
al
in
f
o
r
m
atio
n
s
[
1
3
]
.
T
h
e
o
u
tp
u
ts
o
f
R
NN
ar
e
n
o
t
o
n
l
y
a
f
u
n
ct
io
n
o
f
th
e
cu
r
r
en
t
i
n
p
u
t
s
,
b
u
t
a
ls
o
a
f
u
n
ctio
n
o
f
p
r
ev
io
u
s
i
n
p
u
t
s
an
d
o
u
tp
u
t
s
an
d
it
is
f
o
r
m
u
lat
ed
f
o
r
th
e
s
tate
s
p
ac
e
r
ea
lis
at
i
o
n
.
T
h
e
b
lo
ck
d
iag
r
a
m
s
h
o
w
n
in
Fi
g
u
r
e
3
s
h
o
w
s
th
e
s
ta
te
s
p
ac
e
m
o
d
el
o
f
a
r
ec
u
r
r
en
t
n
eu
r
o
n
in
w
h
ic
h
,
th
e
s
t
ates
o
f
t
h
e
s
y
s
te
m
ar
e
u
p
d
ated
w
it
h
t
h
e
ex
ter
n
al
in
p
u
t
s
an
d
th
e
ac
ti
v
atio
n
f
r
o
m
th
e
p
r
ev
io
u
s
f
o
r
w
ar
d
p
r
o
p
ag
atio
n
.
I
t
h
as
t
w
o
la
y
er
s
o
f
n
eu
r
o
n
s
-
a
n
o
n
l
in
ea
r
h
id
d
en
la
y
er
a
n
d
a
li
n
ea
r
o
u
tp
u
t
la
y
er
as
s
h
o
w
n
i
n
Fi
g
u
r
e
3
.
T
h
e
o
u
tp
u
ts
f
r
o
m
t
h
e
h
id
d
en
l
a
y
er
x
(n
+
1
)
ar
e
f
ed
b
ac
k
to
th
e
in
p
u
t
t
h
r
o
u
g
h
a
s
et
o
f
u
n
it
d
ela
y
s
.
T
h
e
o
u
t
p
u
t
o
f
th
e
li
n
ea
r
o
u
tp
u
t
la
y
er
y
(
n
+
1
)
is
ta
k
e
n
th
r
o
u
g
h
u
n
it d
ela
y
s
w
h
ic
h
p
r
o
d
u
ce
th
e
o
u
tp
u
t
y
(
n
)
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
A
P
E
I
SS
N:
2252
-
8792
A
n
A
d
a
p
tive
R
TRL B
a
s
ed
N
eu
r
o
co
n
t
r
o
ller
fo
r
Da
mp
in
g
P
o
w
er S
ystem…
(
K
.
C
.
S
in
d
h
u
Th
a
mp
a
tty
)
5
Fig
u
r
e
3
.
B
lo
ck
d
iag
r
a
m
r
ep
r
esen
tat
io
n
o
f
a
s
ta
te
s
p
ac
e
m
o
d
el
o
f
r
ec
u
r
r
en
t n
e
u
r
o
n
T
h
e
in
p
u
t
o
f
R
NN
co
n
s
is
t
s
o
f
a
co
n
ca
ten
atio
n
o
f
f
ee
d
b
ac
k
n
o
d
es
an
d
th
e
co
n
tr
o
l
n
o
d
es.
E
a
ch
n
e
u
r
o
n
in
t
h
e
p
r
o
ce
s
s
i
n
g
la
y
er
is
ac
t
iv
ated
b
y
a
n
o
n
-
lin
ea
r
ac
tiv
a
tio
n
f
u
n
ct
io
n
w
h
er
ea
s
th
e
o
u
tp
u
t
n
eu
r
o
n
i
s
ac
ti
v
ated
b
y
li
n
ea
r
ac
ti
v
atio
n
f
u
n
ctio
n
.
T
h
e
n
u
m
b
er
o
f
u
n
it
d
ela
y
s
(
)
u
s
ed
to
f
ee
d
t
h
e
o
u
tp
u
t
o
f
th
e
h
id
d
en
la
y
e
r
(
p
r
o
ce
s
s
in
g
la
y
er
)
to
t
h
e
i
n
p
u
t
la
y
er
d
eter
m
in
e
s
t
h
e
o
r
d
er
o
f
th
e
m
o
d
el.
Fo
r
a
f
u
ll
y
co
n
n
ec
t
ed
r
ec
u
r
r
en
t
n
eu
r
al
n
et
w
o
r
k
,
th
e
n
u
m
b
er
o
f
u
n
it
d
ela
y
s
co
n
n
ec
ti
n
g
b
et
w
ee
n
o
u
t
p
u
t
an
d
in
p
u
t
w
ill
b
e
eq
u
al
to
th
e
n
u
m
b
er
o
f
s
tates
o
f
t
h
e
s
y
s
te
m
.
T
h
e
ar
ch
itec
t
u
r
e
o
f
th
e
p
r
o
p
o
s
ed
co
n
tr
o
ller
u
s
i
n
g
R
NN
is
e
x
p
lai
n
ed
i
n
d
etail
in
t
h
e
n
ex
t
s
ec
tio
n
.
3
.
2
.
P
r
o
po
s
ed
Ada
ptiv
e
Neuro
-
Co
ntr
o
ller
Du
e
to
t
h
e
n
o
n
li
n
ea
r
,
ti
m
e
v
ar
y
in
g
n
a
tu
r
e
o
f
t
h
e
p
o
w
er
s
y
s
te
m
,
it
is
es
s
e
n
tial
f
o
r
a
co
n
tr
o
ller
to
ch
an
g
e
its
o
w
n
b
eh
a
v
io
u
r
ac
c
o
r
d
in
g
to
th
e
ch
an
g
es
i
n
th
e
s
y
s
te
m
.
T
h
e
co
n
tr
o
ller
p
r
o
p
o
s
e
d
in
th
is
s
ec
tio
n
is
s
u
c
h
an
ad
ap
tiv
e
co
n
tr
o
ller
b
ased
o
n
R
T
R
L
alg
o
r
ith
m
.
R
T
R
L
-
b
ased
co
n
tr
o
ller
in
cl
u
d
es
t
w
o
s
u
b
-
n
et
w
o
r
k
s
.
T
h
e
fi
r
s
t
o
n
e
is
R
NN
w
h
ic
h
ac
ts
as
a
n
eu
r
o
-
id
en
ti
fier
th
at
tr
ac
k
th
e
d
y
n
a
m
ic
ac
tiv
ities
of
th
e
s
y
s
te
m
.
T
h
e
s
ec
o
n
d
n
et
w
o
r
k
ac
t
s
as
a
n
e
u
r
o
-
co
n
tr
o
ller
to
p
r
o
v
id
e
p
r
o
p
er
co
n
tr
o
l
s
ig
n
al
to
t
h
e
s
y
s
te
m
[
1
4
]
.
I
n
ea
ch
o
p
er
atio
n
s
tep
,
th
e
co
n
tr
o
ller
g
e
n
er
ates a
co
n
tr
o
l in
p
u
t u
t
h
at
ca
u
s
e
s
th
e
p
lan
t to
p
r
o
d
u
ce
th
e
d
esire
d
o
u
tp
u
t.
3
.
2
.
1
.
Neuro
-
identifi
er
T
h
e
o
b
j
ec
tiv
e
o
f
th
is
s
u
b
n
et
wo
r
k
is
to
id
en
tify
a
p
r
o
p
er
m
o
d
el
o
f
th
e
s
y
s
te
m
to
b
e
co
n
tr
o
lled
.
T
h
e
m
ai
n
s
y
s
te
m
to
b
e
id
en
ti
fi
ed
b
y
th
e
co
n
tr
o
ller
ca
n
b
e
r
ep
r
esen
ted
as:
(
1
1
)
w
h
er
e
T
h
e
r
ec
u
r
r
en
t
n
eu
r
al
n
et
w
o
r
k
as
n
eu
r
o
-
id
en
ti
fier
is
tr
ain
ed
u
s
i
n
g
R
T
R
L
alg
o
r
it
h
m
b
y
a
s
s
u
m
in
g
th
e
n
o
n
li
n
ea
r
f
u
n
ctio
n
F
,
w
h
ic
h
r
ep
r
esen
t
t
h
e
d
y
n
a
m
ic
r
elat
io
n
s
h
ip
o
f
t
h
e
s
y
s
te
m
.
Fo
r
t
h
e
n
eu
r
o
-
id
en
t
i
fi
er
,
t
h
e
in
p
u
t
v
ec
to
r
is
:
W
h
er
e,
p
: th
e
to
tal
n
u
m
b
er
o
f
o
u
tp
u
ts
f
ed
b
ac
k
to
th
e
in
p
u
t sid
e
an
d
m
: th
e
to
tal
n
u
m
b
er
o
f
co
n
tr
o
ls
o
f
th
e
s
y
s
te
m
.
T
h
e
n
et
w
o
r
k
is
tr
ai
n
ed
u
s
i
n
g
a
k
n
o
w
n
d
esire
d
tr
aj
ec
to
r
y
.
T
h
e
p
r
ed
icted
s
y
s
te
m
o
u
tp
u
t
at
(
)
in
ter
v
a
l
is
r
ep
r
esen
ted
as
̂
(n
+
1
)
.
E
v
er
y
d
is
cr
ete
i
n
ter
v
al,
weig
h
ts
o
f
th
e
n
eu
r
o
-
id
en
ti
fier
ar
e
u
p
d
ated
u
s
in
g
R
T
R
L
al
g
o
r
ith
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8792
IJ
A
P
E
Vo
l.
4
,
No
.
1
,
A
p
r
il
2
0
1
4
:
1
–
12
6
3
.
2
.
2
.
P
er
f
o
r
m
a
nce
I
n
dex
o
f
neuro
-
identilier
T
h
e
P
erfo
r
ma
n
ce
I
n
d
ex
(
PI)
of
th
e
n
e
u
r
o
-
id
e
n
ti
fi
er
s
h
o
w
s
t
h
e
ef
fi
cien
c
y
of
t
h
e
id
en
ti
fi
er
in
tr
ac
k
i
n
g
th
e
d
esire
d
tr
aj
ec
to
r
y
an
d
it is
ca
lcu
lated
u
s
i
n
g
E
q
n
.
1
2
.
(
1
2
)
w
h
er
e
(
n
)
d
en
o
tes th
e
d
esire
d
tar
g
et
v
al
u
e
o
f
t
h
e
o
u
tp
u
t a
t
tim
e
in
ter
v
al.
T
h
e
o
p
tim
al
v
al
u
e
o
f
P
I
af
ter
th
e
co
n
v
er
g
e
n
ce
i
n
ea
ch
s
a
m
p
lin
g
in
ter
v
al
i
s
p
lo
tted
w
i
th
r
esp
ec
t
to
ti
m
e
a
s
s
h
o
w
n
i
n
Fig
u
r
e
.
4
.
P
er
f
o
r
m
an
ce
o
f
th
i
s
n
e
u
r
o
-
id
en
ti
fi
er
is
test
ed
f
o
r
v
ar
io
u
s
o
p
er
atin
g
co
n
d
itio
n
s
b
y
v
ar
y
i
n
g
th
e
r
ea
l p
o
w
er
f
r
o
m
0
.
2
p
u
to
1
.
2
p
u
.
I
t h
as b
ee
n
o
b
s
er
v
ed
th
at
t
h
e
p
r
o
p
o
s
ed
n
eu
r
o
-
id
en
ti
fi
er
p
er
f
o
r
m
s
w
ell
i
n
all
t
h
ese
co
n
d
itio
n
s
.
Fig
u
r
e
4
.
P
er
f
o
r
m
a
n
ce
I
n
d
ex
o
f
Neu
r
o
-
I
d
en
ti
fi
er
3
.
2
.
3
.
Neuro
-
co
ntr
o
ller
T
h
is
is
t
h
e
s
ec
o
n
d
s
u
b
-
n
et
w
o
r
k
o
f
t
h
e
p
r
o
p
o
s
ed
R
T
R
L
-
b
ase
d
co
n
tr
o
ller
.
Fo
r
th
e
n
e
u
r
o
-
co
n
tr
o
ller
,
th
e
in
p
u
t
v
ec
to
r
is
th
e
s
y
s
te
m
s
tat
es
w
h
ic
h
ca
n
b
e
r
ep
r
esen
ted
a
s
:
w
h
er
e
x
r
ep
r
esen
t
t
h
e
u
p
d
ate
d
s
y
s
te
m
s
tate
s
a
n
d
q
r
ep
r
esen
t
t
h
e
n
u
m
b
er
o
f
s
tate
s
o
f
t
h
e
s
y
s
te
m
.
T
h
ese
s
tates
ar
e
f
ed
b
ac
k
f
r
o
m
t
h
e
o
u
tp
u
t
to
th
e
i
n
p
u
t
s
id
e.
W
h
en
t
h
e
d
esire
d
o
u
tp
u
t
i
s
g
iv
e
n
at
a
n
y
ti
m
e
in
ter
v
a
l,
s
a
y
(
)
in
s
ta
n
t,
t
h
e
n
t
h
e
r
eq
u
ir
ed
co
n
tr
o
l
i
n
p
u
t
is
g
e
n
er
ated
by
t
h
e
co
n
tr
o
ller
w
ith
th
e
av
ailab
le
k
n
o
w
led
g
e
of
th
e
s
ta
tes
in
t
h
e
ti
m
e
i
n
ter
v
al.
T
h
e
o
u
tp
u
t
o
f
th
e
n
e
u
r
o
-
co
n
tr
o
ller
i
s
r
ep
r
esen
ted
as
a
f
u
n
ctio
n
o
f
th
e
s
y
s
te
m
s
tate
s
an
d
th
e
b
ias as:
(
1
3
)
w
h
er
e
x
(
n
)
i
s
th
e
s
tate
v
ec
to
r
in
in
s
tan
t.
3
.
2
.
4
.
P
er
f
o
rm
a
nce
I
nd
e
x
o
f
Neuro
-
Co
ntr
o
ller
T
h
e
ef
f
ec
ti
v
en
e
s
s
o
f
t
h
e
n
e
u
r
o
-
co
n
tr
o
ller
is
q
u
a
n
titi
v
el
y
m
ea
s
u
r
ed
u
s
i
n
g
a
p
er
f
o
r
m
an
ce
i
n
d
ex
w
h
ic
h
is
ca
lcu
la
ted
as:
(
1
4
)
Fig
u
r
e
5
.
s
h
o
w
s
th
e
v
ar
iatio
n
o
f
P
I
o
f
t
h
e
n
e
u
r
o
-
co
n
tr
o
ll
er
w
i
th
r
esp
ec
t
t
o
t
i
m
e.
Fo
r
th
e
n
e
u
r
o
-
co
n
tr
o
ller
,
th
e
d
esire
d
o
u
tp
u
t c
an
n
o
t b
e
d
e
fin
ed
ex
p
lic
itl
y
.
T
h
er
ef
o
r
e,
th
e
n
eu
r
o
-
co
n
tr
o
ller
h
as to
b
e
tr
ain
ed
b
y
d
r
iv
in
g
t
h
e
n
e
u
r
o
-
id
e
n
ti
fi
er
a
n
d
th
e
er
r
o
r
b
et
w
ee
n
t
h
e
d
es
ir
ed
an
d
p
r
ed
icted
s
y
s
te
m
o
u
tp
u
t
w
ill
b
e
b
ac
k
p
r
o
p
ag
ated
b
y
th
e
n
e
u
r
o
-
id
en
ti
fi
er
an
d
th
e
s
tate
s
o
f
th
e
s
y
s
te
m
ar
e
m
o
d
i
fi
ed
ac
co
r
d
in
g
to
th
is
er
r
o
r
.
T
h
e
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
A
P
E
I
SS
N:
2252
-
8792
A
n
A
d
a
p
tive
R
TRL B
a
s
ed
N
eu
r
o
co
n
t
r
o
ller
fo
r
Da
mp
in
g
P
o
w
er S
ystem…
(
K
.
C
.
S
in
d
h
u
Th
a
mp
a
tty
)
7
r
eq
u
ir
ed
co
n
tr
o
l
to
th
e
s
y
s
te
m
is
g
e
n
er
ated
b
ased
o
n
th
is
m
o
d
i
fi
ed
s
tates.
T
h
e
p
ar
a
m
eter
s
o
f
th
e
n
e
u
r
o
-
id
en
ti
fi
er
a
n
d
co
n
tr
o
ller
w
i
ll
b
e
ad
j
u
s
ted
in
e
v
er
y
s
a
m
p
lin
g
p
er
io
d
.
T
h
is
allo
w
s
th
e
co
n
tr
o
ller
to
tr
ac
k
th
e
d
y
n
a
m
ic
v
ar
iatio
n
s
o
f
t
h
e
p
o
w
er
s
y
s
te
m
a
n
d
p
r
o
v
id
e
th
e
b
est
co
n
tr
o
l
ac
tio
n
.
E
ac
h
co
n
tr
o
ller
n
eu
r
o
n
is
co
n
n
ec
ted
to
all
th
e
s
tates i
n
t
h
e
in
p
u
t la
y
er
a
n
d
g
i
v
en
a
n
ex
ter
n
al
b
ias.
Fig
u
r
e
5
.
P
er
f
o
r
m
a
n
ce
I
n
d
ex
o
f
Neu
r
o
-
C
o
n
tr
o
ller
T
h
e
s
y
s
te
m
s
tate
s
at
(
)
in
s
ta
n
t c
an
b
e
ca
lcu
lated
as:
(
1
5
)
w
h
er
e,
T
h
e
d
esire
d
o
u
tp
u
t a
t
(
)
in
s
ta
n
t i
s
g
i
v
en
b
y
:
(
1
6
)
w
h
er
e
C
is
t
h
e
o
u
tp
u
t
m
atr
ix
o
f
th
e
s
y
s
te
m
.
T
h
e
co
n
tr
o
l
in
p
u
t
g
e
n
er
ated
b
y
t
h
e
co
n
tr
o
ller
in
(
)
in
s
ta
n
t
is
g
i
v
e
n
b
y
:
(
1
7
)
a.
A
t
ti
m
e
s
tep
,
o
u
tp
u
ts
ar
e
s
a
m
p
le
d
as
y
(
n
)
.
b.
(
n
)
,
(
n
)
.
.
.
etc.
,
ar
e
u
s
ed
to
f
o
r
m
i
n
p
u
t
v
ec
to
r
o
f
n
eu
r
o
co
n
tr
o
ller
at
ti
m
e
s
tep
.
A
t
th
e
s
a
m
e
ti
m
e,
t
h
e
w
ei
g
h
ts
o
f
th
e
n
e
u
r
o
-
id
en
ti
fier
ar
e
u
p
d
ated
u
s
i
n
g
R
T
R
L
al
g
o
r
ith
m
to
m
i
n
i
m
is
e
th
e
er
r
o
r
b
et
w
ee
n
th
e
d
esire
d
o
u
tp
u
t
an
d
th
e
p
r
ed
icted
o
u
tp
u
t
̂
at
in
s
t
an
t.
c.
(
n
)
,
(
n
)
........
u
(
n
)
ar
e
u
s
ed
to
f
o
r
m
an
i
n
p
u
t
v
ec
to
r
f
o
r
th
e
n
e
u
r
o
-
id
en
ti
fier
an
d
th
e
te
m
p
o
r
al
o
u
tp
u
t
̂
(n
+
1
)
is
ca
lcu
lated
.
d.
T
h
e
w
ei
g
h
ts
o
f
t
h
e
n
e
u
r
o
co
n
tr
o
ller
ar
e
u
p
d
ated
to
m
in
i
m
is
e
th
e
er
r
o
r
af
ter
e
v
er
y
e
x
ec
u
ti
o
n
o
f
t
h
e
alg
o
r
ith
m
.
e.
T
h
e
s
y
s
te
m
s
tate
s
ar
e
u
p
d
ated
u
s
i
n
g
E
q
n
.
1
5
f.
T
h
e
o
u
tp
u
t
o
f
th
e
n
e
u
r
o
co
n
tr
o
ller
u
(
n
)
i
s
ca
lc
u
lated
a
g
ai
n
w
it
h
th
e
late
s
t
u
p
d
ated
s
tate
s
as
t
h
e
i
n
p
u
t
v
ec
to
r
an
d
th
e
n
e
w
w
ei
g
h
t
s
ca
lcu
lated
in
t
h
e
p
r
ev
io
u
s
s
tep
.
g.
T
h
e
co
n
tr
o
l
s
ig
n
al
u
(
n
)
is
ap
p
lied
to
th
e
p
lan
t,
a
n
d
to
th
e
n
e
u
r
o
-
id
en
t
i
fi
er
ag
a
in
to
ca
lc
u
lat
e
̂
(
n
+
1
)
f
o
r
(
)
ti
m
e
s
tep
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8792
IJ
A
P
E
Vo
l.
4
,
No
.
1
,
A
p
r
il
2
0
1
4
:
1
–
12
8
3
.
3
.
Rea
l
T
i
m
e
Rec
urre
nt
L
ea
rn
ing
(
RT
RL
)
Alg
o
rit
h
m
T
h
is
is
a
f
o
r
w
ar
d
g
r
ad
ien
t
alg
o
r
ith
m
,
w
h
ic
h
m
a
k
es
u
s
e
o
f
a
m
atr
ix
o
f
p
ar
tial
d
er
iv
ati
v
es
o
f
th
e
n
et
w
o
r
k
s
tate
v
al
u
es
w
it
h
r
esp
ec
t
to
ev
er
y
w
ei
g
h
t.
T
h
e
alg
o
r
ith
m
atte
m
p
t
s
to
m
i
n
i
m
iz
e
th
e
in
s
ta
n
ta
n
eo
u
s
s
q
u
ar
ed
er
r
o
r
at
th
e
o
u
tp
u
t
o
f
th
e
n
eu
r
o
n
.
T
h
e
m
ai
n
d
i
f
fi
c
u
lt
y
r
elate
d
to
t
h
e
r
ec
u
r
s
iv
e
tr
ain
i
n
g
o
f
r
ec
u
r
r
e
n
t
n
et
w
o
r
k
ar
i
s
es
f
r
o
m
th
e
f
ac
t
th
at
t
h
e
o
u
tp
u
t
o
f
t
h
e
n
et
w
o
r
k
a
n
d
its
p
ar
tial
d
er
i
v
ati
v
es
w
it
h
r
e
s
p
ec
t
to
th
e
w
ei
g
h
ts
d
ep
en
d
o
n
t
h
e
i
n
p
u
ts
.
I
n
th
is
m
et
h
o
d
,
th
e
p
ar
tial
d
er
i
v
ati
v
es
o
f
ea
ch
n
o
d
e
w
it
h
r
esp
ec
t
to
ea
ch
w
ei
g
h
t
ar
e
co
m
p
u
ted
at
ev
er
y
iter
atio
n
[
1
4
]
.
I
f
q
is
th
e
n
u
m
b
er
o
f
s
t
ates o
f
t
h
e
s
y
s
te
m
,
p
t
h
e
n
u
m
b
er
o
f
o
u
tp
u
t
s
an
d
m
th
e
n
u
m
b
er
o
f
co
n
tr
o
l
in
p
u
ts
to
th
e
s
y
s
te
m
,
th
e
n
t
h
e
d
y
n
a
m
ic
s
y
s
te
m
ca
n
b
e
r
ep
r
esen
te
d
b
y
t
h
e
n
o
n
-
li
n
ea
r
d
if
f
er
e
n
ce
eq
u
atio
n
s
g
i
v
e
n
as :
(
1
8
)
(
1
9
)
(
2
0
)
(
2
1
)
(
2
2
)
a.
I
n
E
q
n
.
(
1
8
)
,
th
e
to
tal
w
e
ig
h
t
s
ar
e
s
p
lit
in
to
tw
o
,
n
a
m
el
y
,
an
d
.
T
h
e
m
atr
ix
r
ep
r
esen
ts
th
e
s
y
n
ap
tic
w
eig
h
t
s
ass
o
ciate
d
w
it
h
t
h
e
q
n
eu
r
o
n
s
in
t
h
e
h
id
d
en
la
y
er
th
at
ar
e
f
ed
b
ac
k
t
o
th
e
in
p
u
t
la
y
er
.
T
h
e
m
atr
ix
r
ep
r
esen
ts
th
e
s
y
n
ap
tic
w
eig
h
t
s
ass
o
ciat
ed
w
i
th
t
h
is
h
id
d
en
la
y
er
,
w
h
ich
ar
e
co
n
n
ec
ted
to
th
e
i
n
p
u
t
s
o
u
r
ce
s
in
cl
u
d
in
g
t
h
e
b
ias.
T
h
u
s
,
th
e
b
ias
ter
m
s
o
f
t
h
e
h
id
d
en
n
e
u
r
o
n
s
ar
e
in
cl
u
d
ed
in
.
b.
T
h
e
m
atr
i
x
C
r
ep
r
esen
ts
t
h
e
s
y
n
ap
tic
w
eig
h
t
s
o
f
p
o
u
tp
u
t
n
eu
r
o
n
s
co
n
n
ec
ted
to
th
e
h
id
d
en
l
a
y
er
.
c.
T
h
e
n
eu
r
o
n
s
i
n
t
h
e
h
id
d
en
la
y
er
ar
e
w
i
th
h
y
p
er
b
o
lic
tan
g
e
n
t
n
o
n
li
n
ea
r
ac
ti
v
atio
n
f
u
n
ctio
n
,
g
iv
e
n
b
y
:
(
2
3
)
o
r
a
lo
g
is
tic
f
u
n
c
tio
n
:
(
2
4
)
Fo
r
ev
er
y
i
n
s
ta
n
t,
t
h
e
s
tates
ar
e
o
b
s
er
v
ed
by
n
e
u
r
o
id
en
ti
fi
er
an
d
th
e
r
eq
u
ir
ed
co
n
tr
o
l
s
ig
n
al
is
g
en
er
ated
u
s
i
n
g
R
T
R
L
al
g
o
r
it
h
m
[
1
4
]
,
[
1
8
]
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
A
P
E
I
SS
N:
2252
-
8792
A
n
A
d
a
p
tive
R
TRL B
a
s
ed
N
eu
r
o
co
n
t
r
o
ller
fo
r
Da
mp
in
g
P
o
w
er S
ystem…
(
K
.
C
.
S
in
d
h
u
Th
a
mp
a
tty
)
9
4.
SI
M
UL
AT
I
O
N
R
E
S
UL
T
A
m
at
h
e
m
atica
l
m
o
d
el
o
f
th
e
s
y
s
te
m
s
h
o
w
n
i
n
Fi
g
u
r
e
.
1
is
d
ev
elo
p
ed
.
State
s
p
ac
e
m
o
d
el
o
f
ex
cite
r
,
g
en
er
ato
r
,
tr
an
s
m
i
s
s
io
n
lin
e
,
m
ec
h
a
n
ical
s
y
s
te
m
a
n
d
s
e
r
ies
co
n
n
ec
ted
F
AC
T
S
d
ev
ices
ar
e
d
ev
elo
p
ed
s
ep
ar
atel
y
a
n
d
t
h
en
in
ter
co
n
n
ec
ted
to
g
eth
er
to
f
o
r
m
f
u
ll
s
y
s
te
m
.
T
h
e
r
eq
u
ir
ed
co
m
p
en
s
a
tio
n
i
s
p
r
o
v
id
ed
b
y
T
C
SC
an
d
T
C
P
A
R
.
I
n
t
h
i
s
an
al
y
s
i
s
w
e
ca
lc
u
lated
an
o
p
ti
m
al
lev
el
o
f
co
m
p
e
n
s
atio
n
p
r
o
v
id
ed
b
y
T
C
SC
a
n
d
T
C
P
A
R
an
d
b
y
k
ee
p
in
g
th
e
T
C
P
AR
co
m
p
en
s
atio
n
co
n
s
tan
t
,
o
n
l
y
T
C
SC
is
co
n
tr
o
lled
.
T
h
er
e
ar
e
16
s
tates
in
th
e
g
iv
e
n
s
y
s
te
m
w
h
ic
h
is
s
i
m
u
lated
at
d
i
f
f
er
e
n
t
o
p
er
atin
g
c
o
n
d
itio
n
s
I
n
a
n
al
y
s
is
,
th
e
m
ac
h
in
e
d
a
m
p
i
n
g
is
n
o
t
in
cl
u
d
ed
s
o
t
h
at
all
t
h
e
d
a
m
p
i
n
g
ef
f
ec
t
s
ar
e
d
u
e
to
t
h
e
F
A
C
T
S
d
ev
ices
co
n
n
ec
ted
in
s
er
ie
s
.
T
h
e
an
al
y
s
is
h
a
s
b
ee
n
d
o
n
e
by
g
i
v
i
n
g
a
s
tep
ch
an
g
e
in
t
h
e
m
ec
h
a
n
ical
p
o
w
er
in
p
u
t
to
t
h
e
s
y
s
te
m
.
P
er
f
o
r
m
an
ce
o
f
t
h
e
s
y
s
te
m
w
it
h
t
h
e
p
r
o
p
o
s
ed
co
n
tr
o
ller
is
co
m
p
ar
ed
w
it
h
th
o
s
e
o
f
t
h
e
co
n
v
en
t
io
n
al
P
I
co
n
tr
o
ller
an
d
GA
b
ased
P
I
co
n
tr
o
ller
w
h
ich
i
s
ex
p
lai
n
ed
i
n
d
etail
in
t
h
e
f
o
llo
w
in
g
s
ec
tio
n
s
.
4
.
1
.
GA
B
a
s
ed
P
I
Co
ntr
o
ller
A
co
n
v
e
n
tio
n
a
l
P
I
co
n
tr
o
ller
is
w
id
el
y
u
s
ed
i
n
p
o
w
er
s
y
s
t
e
m
ap
p
licatio
n
s
b
ec
a
u
s
e
it
i
s
s
i
m
p
le
i
n
s
tr
u
ct
u
r
e,
lo
w
co
s
t
an
d
h
ig
h
l
y
r
eliab
le.
Ho
w
ev
er
,
i
m
p
r
o
p
er
P
I
p
ar
am
eter
tu
n
in
g
ca
n
lea
d
to
s
lo
w
r
ec
o
v
er
y
,
p
o
o
r
r
o
b
u
s
tn
ess
a
n
d
in
w
o
r
s
t
ca
s
e
th
e
f
ail
u
r
e
o
f
th
e
s
y
s
te
m
o
p
er
atio
n
.
Gen
etic
Alg
o
r
ith
m
(
G
A
)
i
s
an
o
p
tim
is
atio
n
al
g
o
r
ith
m
w
h
ich
ca
n
b
e
u
s
ed
f
o
r
fi
n
d
i
n
g
t
h
e
o
p
tim
a
l
p
ar
a
m
eter
s
f
o
r
P
I
co
n
tr
o
ller
s
o
t
h
at
t
h
e
p
ar
am
eter
s
w
il
l
al
w
a
y
s
r
es
u
lt
in
g
lo
b
al
m
i
n
i
m
u
m
r
at
h
er
t
h
a
n
i
n
lo
ca
l
m
in
i
m
u
m
co
n
d
itio
n
s
.
T
h
is
s
ec
tio
n
g
iv
e
s
a
co
m
p
ar
is
o
n
o
f
t
h
e
p
er
f
o
r
m
an
ce
o
f
th
e
s
y
s
te
m
w
i
th
c
o
n
v
e
n
tio
n
al
co
n
tr
o
ller
a
n
d
G
A
b
a
s
ed
co
n
tr
o
ller
.
P
ar
am
ete
r
s
u
s
ed
in
G
A
ar
e
g
i
v
en
in
T
ab
le
1
.
T
ab
le
1
.
P
ar
am
eter
s
U
s
ed
f
o
r
T
u
n
in
g
G
A
T
h
e
o
p
tim
i
s
ed
v
al
u
e
o
f
t
h
e
t
u
n
in
g
p
ar
a
m
eter
f
o
r
P
I
co
n
tr
o
ller
u
s
ed
is
g
i
v
e
n
in
T
ab
le
2
w
h
er
e
K
is
t
h
e
g
ain
,
an
d
ar
e
th
e
ti
m
e
co
n
s
ta
n
ts
o
f
t
h
e
P
I
co
n
tr
o
ller
.
T
h
e
p
er
f
o
r
m
an
ce
o
f
th
e
s
y
s
te
m
w
i
t
h
G
A
b
ased
P
I
co
n
tr
o
ller
is
co
m
p
ar
ed
w
it
h
th
e
co
n
v
en
tio
n
al
P
I
co
n
tr
o
ller
as
s
h
o
w
n
i
n
Fi
g
u
r
e
8
.
I
t
tak
es
a
l
m
o
s
t
3
s
ec
o
n
d
s
to
r
ea
ch
s
tead
y
s
tate
i
n
ca
s
e
o
f
c
o
n
v
e
n
tio
n
al
co
n
tr
o
ller
,
w
h
er
ea
s
,
w
ith
t
h
e
G
A
b
ased
co
n
tr
o
lle
r
,
v
ar
iatio
n
r
ed
u
ce
s
to
ze
r
o
b
ef
o
r
e
2
s
ec
o
n
d
s
.
Si
m
u
latio
n
s
ar
e
ca
r
r
ied
o
u
t
f
o
r
d
if
f
er
en
t
o
p
er
atin
g
co
n
d
itio
n
s
a
n
d
m
o
s
t
of
t
h
e
ca
s
es
th
e
p
er
f
o
r
m
a
n
ce
of
t
h
e
s
y
s
te
m
w
it
h
G
A
b
ased
P
I
co
n
tr
o
ller
is
f
ar
b
etter
th
a
n
th
e
co
n
v
e
n
tio
n
al
co
n
tr
o
ller
.
T
ab
le
2
.
GA
T
u
n
ed
P
I
P
ar
am
e
ter
s
4
.
2
.
RT
RL
B
a
s
ed
ANN
Co
n
t
ro
ller
Deta
iled
s
i
m
u
latio
n
o
f
th
e
test
s
y
s
te
m
w
it
h
R
T
R
L
-
b
a
s
ed
co
n
tr
o
ller
is
e
x
ec
u
ted
u
s
i
n
g
M
AT
L
A
B
f
o
r
th
e
s
a
m
e
s
tep
c
h
an
g
e
in
m
ec
h
an
ical
i
n
p
u
t
to
t
h
e
g
e
n
er
ato
r
.
T
h
e
Po
w
er
An
g
le
R
e
g
u
lato
r
is
s
et
f
o
r
a
co
n
s
tan
t
in
j
ec
ted
v
o
ltag
e
a
n
d
th
e
T
C
S
C
is
co
n
tr
o
lled
s
m
o
o
th
l
y
f
o
r
th
er
eq
u
ir
ed
o
p
er
atin
g
co
n
d
itio
n
s
.
Si
m
u
latio
n
s
ar
e
ca
r
r
ied
o
u
tf
o
r
w
id
eo
p
er
atin
g
co
n
d
itio
n
s
w
h
er
e,
th
e
r
ea
l
p
o
w
er
P
is
v
ar
ied
f
r
o
m
0
.
2
p
u
to
1
.
2
p
u
an
d
th
e
th
y
r
i
s
to
r
fi
r
in
g
a
n
g
le
o
f
T
C
S
C
is
v
ar
ied
f
r
o
m
1
5
0
⸰
to
1
7
5
⸰
(
C
ap
ac
itiv
e
m
o
d
e
of
o
p
er
atio
n
of
T
C
SC
)
.
T
h
e
lear
n
in
g
r
ate
p
ar
a
m
eter
η
w
as
tak
en
as0
.
4
2
f
o
r
tr
ain
i
n
g
t
h
e
n
eu
r
al
n
et
w
o
r
k
w
h
ich
p
r
o
v
id
e
f
ast
co
n
v
er
g
e
n
ce
.
T
h
e
m
a
x
i
m
u
m
to
ler
an
ce
le
v
el
s
et
f
o
r
th
e
er
r
o
r
is
0
.
0
0
5
.
I
n
itiall
y
,
t
h
e
s
y
n
ap
tic
w
ei
g
h
ts
as
s
o
ciate
d
w
it
h
ea
ch
n
eu
r
o
n
ar
e
g
e
n
er
ated
r
an
d
o
m
l
y
f
o
r
ea
ch
s
a
m
p
le.
T
h
e
n
o
n
lin
ea
r
ac
ti
v
atio
n
f
u
n
c
tio
n
φ
(
E
q
n
.
23)
an
d
th
e
lin
e
ar
ac
tiv
atio
n
f
u
n
ct
io
n
γ
ar
e
s
e
lecte
d
f
o
r
th
er
e
cu
r
r
en
t
n
e
u
r
al
n
et
w
o
r
k
a
n
d
th
e
co
n
tr
o
ller
n
eu
r
o
n
s
r
esp
ec
tiv
el
y
.
T
h
e
in
itial
s
tates
s
u
p
p
lied
to
th
e
n
eu
r
o
n
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8792
IJ
A
P
E
Vo
l.
4
,
No
.
1
,
A
p
r
il
2
0
1
4
:
1
–
12
10
ar
e
s
et
to
a
s
m
al
l
v
al
u
e.
I
n
t
h
is
s
t
u
d
y
,
th
e
d
e
s
ir
ed
o
u
tp
u
t
s
u
p
p
lied
to
th
e
n
eu
r
al
n
et
w
o
r
k
is
s
et
to
b
e
ze
r
o
d
u
r
in
g
tr
ai
n
in
g
.
T
h
e
co
n
tr
o
l
s
i
g
n
al
g
e
n
e
r
ated
b
y
th
e
p
r
o
p
o
s
e
d
A
NN
co
n
tr
o
ller
i
s
g
iv
e
n
to
t
h
e
s
y
s
te
m
,
w
h
ich
is
th
e
r
eq
u
ir
ed
tr
ig
g
er
in
g
p
u
l
s
e
f
o
r
th
e
th
y
r
is
to
r
s
in
T
C
SC
.
T
h
er
e
s
p
o
n
s
es
ar
e
s
i
m
u
lated
f
o
r
an
in
itial
c
h
an
g
e
of
40%
in
cr
ea
s
e
in
th
e
s
h
af
t
to
r
q
u
e.
T
h
e
co
n
tr
o
l
s
ig
n
al
g
e
n
er
at
ed
b
y
th
e
A
NN
co
n
tr
o
ller
is
s
h
o
w
n
i
n
Fig
u
r
e
6
.
T
h
is
is
th
e
r
eq
u
ir
ed
d
ev
iatio
n
o
f
th
y
r
i
s
to
r
fi
r
i
n
g
a
n
g
le
f
o
r
T
C
S
C
f
r
o
m
t
h
e
in
it
ial
o
p
er
atin
g
p
o
in
t,
in
o
r
d
er
to
d
am
p
th
e
p
o
w
er
s
y
s
te
m
o
s
cill
atio
n
s
.
T
h
e
s
q
u
ar
ed
er
r
o
r
p
lo
t
o
f
ea
c
h
s
a
m
p
le
i
s
s
h
o
w
n
i
n
Fig
u
r
e
7
.
w
h
er
e
th
e
er
r
o
r
to
le
r
an
ce
is
s
et
to
a
v
al
u
e
0
.
0
0
5
.
Fig
u
r
e
6
.
E
x
p
an
d
ed
v
ie
w
o
f
t
h
e
o
u
tp
u
t o
f
t
h
e
A
NN
co
n
tr
o
ller
Fig
u
r
e
7
.
Sq
u
ar
ed
E
r
r
o
r
Fig
u
r
es
9
.
To
1
5
s
h
o
w
s
t
h
e
p
er
f
o
r
m
an
ce
o
f
th
e
s
y
s
te
m
w
it
h
d
if
f
er
en
t
o
p
er
atin
g
co
n
d
itio
n
s
.
Fi
g
.
9
s
h
o
w
s
t
h
e
d
ev
ia
tio
n
o
f
s
lip
w
it
h
G
A
b
ased
co
n
tr
o
ller
an
d
A
N
N
co
n
tr
o
ller
.
T
h
e
m
a
x
i
m
u
m
d
e
v
iatio
n
i
s
0
.
0
0
1
5
p
u
w
it
h
A
NN
co
n
tr
o
lle
r
w
h
er
ea
s
0
.
0
0
3
8
p
u
w
it
h
G
A
b
ased
co
n
tr
o
ller
.
De
v
ia
tio
n
o
f
s
lip
r
ed
u
ce
d
to
le
s
s
th
an
t
h
e
to
ler
an
ce
v
al
u
e
w
i
th
in
o
n
e
s
ec
o
n
d
in
th
e
ca
s
e
o
f
ANN
co
n
tr
o
ller
w
h
er
ea
s
it
t
ak
es
m
o
r
e
t
h
an
5
s
ec
o
n
d
s
f
o
r
G
A
b
ased
P
I
co
n
tr
o
ller
.
Fig
u
r
e
1
0
s
h
o
w
s
th
e
d
ev
iatio
n
o
f
g
en
er
ato
r
r
o
to
r
an
g
le
δ
an
d
s
lip
w
it
h
th
e
p
r
o
p
o
s
ed
A
N
N
co
n
tr
o
ller
.
I
n
th
i
s
ca
s
e
t
h
e
r
ea
l
p
o
w
er
is
s
et
to
0
.
8
p
u
an
d
v
o
ltag
e
=
1
p
u
.
I
t
h
as
b
ee
n
o
b
s
er
v
ed
f
r
o
m
th
i
s
fi
g
u
r
e
th
a
t
δ
d
ev
iatio
n
r
ed
u
ce
s
to
ze
r
o
b
ef
o
r
e
0
.
1
s
ec
o
n
d
an
d
s
lip
r
ed
u
ce
s
to
less
th
a
n
to
ler
an
ce
lev
el
b
ef
o
r
e
1
.
5
s
ec
o
n
d
s
.
T
h
e
d
ev
iatio
n
o
f
to
r
q
u
e
w
i
th
o
u
t
a
n
d
w
it
h
p
r
o
p
o
s
ed
ANN
co
n
tr
o
ller
i
s
s
h
o
w
n
in
Fi
g
u
r
e
1
1
.
W
ith
ANN
co
n
tr
o
ller
,
th
e
o
s
cil
latio
n
s
d
a
m
p
ed
o
u
t
b
ef
o
r
e
0
.
7
s
ec
o
n
d
s
w
h
er
ea
s
w
it
h
o
u
t
co
n
tr
o
ller
,
it
ta
k
es
m
o
r
e
th
a
n
2
s
ec
o
n
d
s
.
T
h
e
m
ax
i
m
u
m
a
m
p
lit
u
d
e
o
f
t
h
e
o
s
cil
latio
n
is
m
o
r
e
w
i
th
A
N
N
co
n
tr
o
l
ler
b
u
t
it
ex
is
ts
o
n
l
y
f
o
r
fi
r
s
t
f
e
w
c
y
cle
s
.
Si
m
ilar
l
y
in
Fi
g
u
r
e
12,
p
er
f
o
r
m
a
n
ce
of
th
e
s
y
s
te
m
is
co
m
p
ar
ed
w
it
h
G
A
b
ased
P
I
co
n
tr
o
ller
an
d
A
NN
b
ased
co
n
t
r
o
ller
f
o
r
P
=
0
.
8
p
u
an
d
it h
a
s
b
ee
n
o
b
s
er
v
ed
t
h
at
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
s
y
s
te
m
w
it
h
A
N
N
co
n
tr
o
ller
is
f
ar
b
e
tter
th
a
n
th
e
G
A
b
a
s
ed
co
n
tr
o
l
ler
.
A
p
er
f
o
r
m
a
n
ce
co
m
p
ar
is
o
n
is
m
ad
e
w
it
h
ANN
co
n
tr
o
ller
an
d
co
n
v
e
n
tio
n
al
co
n
tr
o
ller
as
s
h
o
w
n
in
Fi
g
u
r
e
1
3
.
T
h
e
to
r
q
u
e
o
s
cillatio
n
s
p
er
s
is
ts
f
o
r
a
lo
n
g
d
u
r
atio
n
w
i
th
co
n
v
e
n
tio
n
al
c
o
n
tr
o
ller
w
h
er
ea
s
w
it
h
ANN
b
ased
co
n
tr
o
ller
,
th
e
d
am
p
i
n
g
is
v
er
y
f
a
s
t
ev
e
n
t
h
o
u
g
h
t
h
e
m
ag
n
it
u
d
e
o
f
o
s
ci
llat
io
n
is
lar
g
e
d
u
r
i
n
g
fi
r
s
t
f
e
w
c
y
cles.
I
t
h
as
b
ee
n
o
b
s
er
v
ed
th
at
th
e
p
er
f
o
r
m
an
ce
o
f
th
i
s
is
f
ar
b
etter
th
an
w
it
h
t
h
e
co
n
v
e
n
tio
n
al
co
n
tr
o
ller
.
Fig
u
r
e
8
.
Var
iatio
n
of
s
lip
w
it
h
co
n
v
en
tio
n
al
co
n
tr
o
ller
an
d
GA
b
ased
co
n
tr
o
ller
Fig
u
r
e
9
.
Var
iatio
n
o
f
s
lip
w
it
h
a)
GA
b
ased
co
n
tr
o
ller
an
d
b
)
w
ith
ANN
c
o
n
tr
o
ller
P
=
1
pu
Evaluation Warning : The document was created with Spire.PDF for Python.