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
.
2
,
J
u
n
e
2
0
1
8
,
p
p
.
738
~
7
4
3
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
2
.
pp
7
3
8
-
743
738
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
I
m
ple
m
e
ntatio
n o
f
NN
Co
ntrolled
DVR
for Enha
nci
ng
t
he
Po
w
er Q
ua
lity
b
y
Mitig
a
ting H
a
r
mo
nics
P
.A
bi
ra
m
i
1
,
Me
r
i
n L
i
z
bet
h G
eor
ge
2
De
p
a
rte
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
,
S
a
th
y
a
b
a
m
a
In
stit
u
te
o
f
S
c
ien
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
,
In
d
ia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Dec
2
2
,
2
0
1
7
R
ev
i
s
ed
J
an
1
1
,
2
0
1
8
A
cc
ep
ted
J
an
3
1
,
2
0
1
8
No
w
a
d
a
y
s
t
h
er
e
is
a
w
id
es
p
r
ea
d
u
s
e
o
f
s
e
m
ico
n
d
u
cto
r
d
ev
ices,
w
h
ic
h
ar
e
m
o
s
tl
y
i
m
p
le
m
e
n
te
d
as
th
e
p
o
w
er
s
w
itc
h
es
f
o
r
co
n
v
er
ter
s
an
d
in
v
er
ter
s
.
T
h
ese
co
n
v
er
t
er
s
an
d
in
v
er
ter
s
p
la
y
a
v
i
t
al
r
o
le
in
p
o
w
er
s
y
s
te
m
s
b
o
th
i
n
tr
a
n
s
m
is
s
io
n
a
n
d
d
is
tr
ib
u
t
io
n
s
y
s
t
e
m
s
.
T
h
is
p
r
o
v
id
es
a
w
a
y
f
o
r
th
e
i
n
tr
o
d
u
ctio
n
o
f
h
ar
m
o
n
ic
s
in
t
h
e
p
o
w
er
s
y
s
te
m
w
h
ic
h
lead
s
to
p
o
o
r
p
o
w
er
q
u
alit
y
.
T
o
o
v
er
co
m
e
th
i
s
m
an
y
s
o
lu
tio
n
s
h
av
e
b
ee
n
s
u
g
g
ested
b
y
t
h
e
r
esear
ch
co
m
m
u
n
it
y
b
u
t
ea
ch
s
o
lu
t
io
n
h
o
ld
s
its
o
w
n
m
er
its
a
n
d
d
em
er
its
.
O
f
all
t
h
ese
s
u
g
g
ested
s
o
lu
tio
n
s
,
th
e
D
y
n
a
m
ic
Vo
lta
g
e
R
esto
r
er
is
o
n
e
o
f
th
e
m
o
s
t
co
s
t
ef
f
ec
tiv
e
s
y
s
te
m
s
f
o
r
v
ar
io
u
s
p
o
w
er
q
u
alit
y
i
s
s
u
es.
I
n
th
is
p
ap
er
th
e
DVR
i
s
co
n
s
id
er
ed
f
o
r
en
h
an
c
in
g
t
h
e
p
o
w
er
q
u
alit
y
b
y
r
ed
u
cin
g
th
e
h
ar
m
o
n
ics
g
en
er
ated
b
ec
au
s
e
o
f
s
en
s
iti
v
e
lo
ad
s
.
Her
e
th
e
p
o
w
er
q
u
alit
y
i
s
en
h
a
n
ce
d
b
y
co
n
tr
o
llin
g
t
h
e
DV
R
u
s
i
n
g
Ne
u
r
al
Net
w
o
r
k
C
o
n
tr
o
ller
w
h
ic
h
is
tr
ain
ed
b
y
L
e
v
e
n
b
er
g
Ma
r
q
u
ar
d
t
alg
o
r
ith
m
.
I
n
th
is
p
ap
er
th
e
T
HD
an
al
y
s
i
s
o
f
th
e
v
o
lta
g
e
q
u
a
n
tit
y
i
s
a
n
al
y
s
ed
b
y
in
tr
o
d
u
ci
n
g
a
n
u
n
b
ala
n
ce
d
th
r
ee
p
h
ase
f
a
u
lt
i
n
th
e
s
y
s
t
e
m
.
T
h
e
s
i
m
u
lat
io
n
is
d
o
n
e
b
y
u
s
i
n
g
MA
T
L
A
B
/
S
i
m
u
li
n
k
.
Fro
m
th
e
r
esu
lts
,
it
is
v
er
if
ied
t
h
at
t
h
e
h
ar
m
o
n
ic
s
ar
e
r
ed
u
ce
d
b
y
t
h
e
NN
co
n
tr
o
lled
DV
R
u
n
i
t.
A
ls
o
t
h
e
s
i
m
u
latio
n
r
esu
lt
s
ar
e
v
er
i
f
ied
w
it
h
t
h
e
h
ar
d
w
ar
e
r
esu
lt
s
.
K
ey
w
o
r
d
:
DVR
Har
m
o
n
ic
s
Neu
r
al
Net
w
o
r
k
C
o
n
tr
o
ller
P
o
w
er
Q
u
alit
y
T
HD
Co
p
y
rig
h
t
©
2
0
1
8
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
:
P
.
A
b
ir
a
m
i
,
Dep
ar
te
m
en
t o
f
E
lectr
ical
an
d
E
lectr
o
n
ics E
n
g
i
n
ee
r
in
g
,
Sath
y
ab
a
m
a
I
n
s
tit
u
te
o
f
Scie
n
ce
an
d
T
ec
h
n
o
lo
g
y
,
I
n
d
ia.
E
m
ail:
ab
ir
a
m
ir
a
m
k
u
m
ar
8
0
@
g
m
a
il.c
o
m
1.
I
NT
RO
D
UCT
I
O
N
T
h
er
e
o
cc
u
r
s
a
p
o
w
er
q
u
a
lit
y
p
r
o
b
lem
i
n
a
p
o
w
er
s
y
s
te
m
w
h
ic
h
p
o
s
s
es
s
a
n
o
n
-
s
ta
n
d
ar
d
v
o
ltag
e,
cu
r
r
en
t
an
d
f
r
eq
u
e
n
c
y
.
T
h
is
r
esu
lts
in
t
h
e
ec
o
n
o
m
ic
c
r
is
is
at
co
n
s
u
m
er
’
s
p
r
e
m
i
s
e
s
an
d
also
to
th
e
m
an
u
f
ac
t
u
r
er
s
o
f
elec
tr
ical
eq
u
ip
m
e
n
t
.
I
n
d
u
s
tr
ial
a
u
to
m
at
io
n
in
c
lu
d
es
m
aj
o
r
ap
p
licatio
n
s
o
f
p
o
w
er
elec
tr
o
n
ic
s
y
s
te
m
s
[
1
]
.
T
h
ese
p
o
w
er
elec
tr
o
n
ic
s
y
s
te
m
s
ar
e
s
e
n
s
i
tiv
e
to
d
is
tu
r
b
an
ce
s
a
n
d
b
ec
o
m
e
les
s
to
ler
an
t
to
p
o
w
er
q
u
alit
y
p
r
o
b
le
m
s
s
u
c
h
as
v
o
ltag
e
s
a
g
s
,
s
w
el
ls
an
d
h
ar
m
o
n
ics.
I
n
th
is
p
ap
er
Dy
n
a
m
i
c
v
o
ltag
e
r
esto
r
er
tech
n
iq
u
e
i
s
u
s
ed
to
m
iti
g
ate
th
e
h
ar
m
o
n
ics
p
r
o
d
u
ce
d
d
u
e
t
o
th
e
s
e
n
s
i
tiv
e
lo
ad
.
D
y
n
a
m
i
c
v
o
ltag
e
r
esto
r
er
s
(
DVRs
)
ar
e
n
o
w
b
ec
o
m
i
n
g
m
o
r
e
e
s
tab
lis
h
ed
i
n
i
n
d
u
s
tr
y
to
r
ed
u
ce
th
e
i
m
p
ac
t
o
f
v
o
lta
g
e
s
a
g
s
to
s
e
n
s
iti
v
e
lo
ad
s
.
Ho
w
ev
er
,
DV
R
s
s
p
e
nd
m
o
s
t
o
f
th
e
ir
ti
m
e
in
s
ta
n
d
b
y
m
o
d
e,
s
in
ce
v
o
lta
g
e
s
a
g
s
o
cc
u
r
v
er
y
i
n
f
r
eq
u
en
t
l
y
,
an
d
h
e
n
ce
th
eir
u
ti
lizatio
n
is
lo
w
[
2
]
.
So
,
th
e
DVR
ca
n
b
e
u
tili
s
ed
to
co
m
p
e
n
s
ate
f
o
r
th
e
r
ed
u
c
tio
n
o
f
h
ar
m
o
n
ics
i
n
th
e
p
o
w
er
s
y
s
t
e
m
.
T
h
u
s
t
h
e
p
o
w
er
q
u
alit
y
o
f
th
e
s
y
s
te
m
ca
n
b
e
i
m
p
r
o
v
ed
to
ac
h
iev
e
t
h
e
s
atis
f
ac
tio
n
o
f
t
h
e
co
n
s
u
m
er
s
.
D
y
n
a
m
ic
Vo
lta
g
e
R
e
s
to
r
er
(
DVR)
is
a
m
o
s
t
co
s
t
e
f
f
ec
tiv
e
a
n
d
ef
f
icie
n
t
ap
p
r
o
ac
h
to
im
p
r
o
v
e
th
e
v
o
ltag
e
q
u
alit
y
at
lo
ad
s
id
e.
DVR
i
s
a
p
o
w
er
elec
tr
o
n
ic
co
n
v
er
ter
b
ased
Dis
tr
ib
u
ted
-
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:
2088
-
8
694
I
mp
leme
n
ta
tio
n
o
f NN
C
o
n
tr
o
lled
DV
R
fo
r
E
n
h
a
n
cin
g
th
e
P
o
w
er…
(
P
.
A
b
ir
a
mi
)
739
Static
S
y
n
ch
r
o
n
o
u
s
Ser
ies
C
o
m
p
e
n
s
ato
r
(
DSSS
C
)
,
d
esi
g
n
e
d
to
p
r
o
tect
th
e
s
en
s
iti
v
e
lo
ad
[
3
]
.
I
t
is
co
n
n
ec
ted
in
s
er
ie
s
w
it
h
d
i
s
tr
ib
u
tio
n
n
et
w
o
r
k
,
w
h
ic
h
m
ai
n
tai
n
s
th
e
s
e
n
s
it
iv
e
lo
ad
v
o
ltag
e
w
h
ich
ca
n
b
e
ad
j
u
s
ted
f
o
r
it
s
p
h
ase,
Ma
g
n
it
u
d
e
an
d
f
r
eq
u
en
c
y
[
4
]
.
Her
e
th
e
v
o
ltag
e
is
in
j
ec
ted
b
y
th
e
D
VR
t
h
r
o
u
g
h
a
n
in
j
ec
tio
n
tr
an
s
f
o
r
m
er
a
n
d
an
L
C
f
il
ter
an
d
h
e
n
ce
t
h
e
D
VR
u
n
it
is
c
o
n
n
ec
ted
i
n
s
er
ies
w
i
th
t
h
e
s
en
s
it
iv
e
lo
ad
.
T
h
u
s
DVR
is
co
n
s
id
er
ed
as
a
v
o
lta
g
e
r
esto
r
er
w
h
ic
h
h
elp
s
in
e
n
h
an
ci
n
g
t
h
e
p
o
w
er
q
u
alit
y
o
f
th
e
s
y
s
te
m
[
5
–
8
]
.
T
h
u
s
a
r
eq
u
ir
ed
v
o
ltag
e
w
it
h
d
esire
d
m
a
g
n
i
tu
d
e
an
d
p
h
ase
an
g
le
i
s
in
j
ec
ted
in
to
th
e
s
y
s
t
e
m
to
i
m
p
r
o
v
e
th
e
p
o
w
er
q
u
alit
y
.
A
ls
o
it
h
elp
s
i
n
r
esto
r
in
g
t
h
e
b
alan
ce
d
a
n
d
s
i
n
u
s
o
id
al
v
o
lta
g
e
o
f
t
h
e
s
y
s
te
m
u
n
d
er
t
h
e
d
is
to
r
ted
co
n
d
itio
n
s
[
9
]
.
2.
NE
URA
L
N
E
T
WO
RK
I
n
o
u
r
d
ay
to
d
ay
li
f
e
w
e
ar
e
s
ea
r
ch
in
g
f
o
r
ad
v
an
ce
d
tech
n
o
lo
f
ies
f
o
r
i
m
p
r
o
v
in
g
o
u
r
ac
tiv
itie
s
.
T
o
ac
h
iev
e
th
i
s
m
an
y
r
esear
c
h
er
s
h
av
e
co
m
p
u
ted
s
o
f
t c
o
m
p
u
ti
n
g
tec
h
n
iq
u
es li
k
e
Fu
zz
y
,
A
NF
I
S,
A
N
N
co
n
tr
o
lled
s
y
s
te
m
s
etc.
,
Of
all
th
e
s
e
tech
n
iq
u
es,
in
t
h
is
p
ap
er
NN
co
n
tr
o
ller
is
co
n
s
id
er
ed
f
o
r
co
n
tr
o
llin
g
t
h
e
DVR
to
s
u
p
p
l
y
co
m
p
en
s
ated
v
o
ltag
e
t
o
th
e
s
en
s
iti
v
e
lo
ad
s
.
Ne
u
r
al
n
et
w
o
r
k
s
i
n
cl
u
d
es
t
h
e
tr
ain
i
n
g
o
f
n
e
u
r
o
n
s
to
atta
in
a
s
p
ec
if
ic
c
h
ar
ater
is
tic
s
.
A
n
e
u
r
o
n
is
n
o
t
h
in
g
b
u
t
d
er
iv
ed
f
r
o
m
th
e
s
t
u
d
ies
o
f
h
u
m
a
n
b
r
ai
n
n
e
u
r
o
n
s
.
A
n
e
u
r
o
n
s
tr
u
ct
u
r
e
co
n
s
i
s
ts
o
f
in
p
u
t
(
d
en
tr
ites
)
a
n
d
an
o
u
tp
u
t
(
Ax
o
n
)
to
co
m
m
u
n
ica
te
w
it
h
th
e
n
eig
h
b
o
u
r
n
e
u
r
o
n
s
.
B
y
th
e
i
n
f
lu
e
n
ce
o
f
t
h
ese
d
e
n
tr
it
es
an
d
a
x
o
n
b
a
s
ic
s
tr
u
ct
u
r
e
f
o
r
Neu
r
al
n
et
w
o
r
k
is
d
ev
el
o
p
ed
[
1
0
]
.
B
ef
o
r
e
co
n
n
ec
ti
n
g
t
h
e
NN
co
n
tr
o
ller
in
th
e
s
y
s
te
m
,
th
e
n
eu
r
o
n
s
s
h
o
u
ld
b
e
tr
ai
n
ed
ef
f
ec
ti
v
el
y
b
y
u
s
i
n
g
s
o
m
e
b
asic a
lg
o
r
ith
m
s
.
Neu
r
al
n
et
w
o
r
k
s
ca
n
b
e
tr
ain
ed
in
th
eir
p
latf
o
r
m
to
p
r
o
v
id
e
b
etter
p
er
f
o
r
m
an
ce
.
T
h
e
y
ca
n
b
e
u
tili
s
ed
to
ap
p
r
o
x
i
m
ate
s
o
m
e
o
f
t
h
e
s
m
o
o
th
n
o
n
-
li
n
ea
r
f
u
n
ctio
n
s
a
n
d
h
en
ce
t
h
e
y
ar
e
n
a
m
ed
as
u
n
i
v
er
s
al
ap
p
r
o
x
im
a
to
r
s
.
Fo
r
b
u
ild
in
g
a
n
ef
f
icie
n
t
n
e
u
r
al
n
et
s
,
it
h
as
t
o
b
e
lear
n
ed
an
d
tr
ain
ed
p
r
o
p
er
l
y
b
y
th
e
ex
p
er
ts
.
Neu
r
al
n
et
w
o
r
k
tr
ai
n
i
n
g
ca
n
b
e
p
er
f
o
r
m
ed
ei
th
er
i
n
o
f
f
li
n
e
o
r
o
n
lin
e
m
o
d
e
[
1
1
-
1
2
]
.
A
v
ar
iet
y
o
f
al
g
o
r
it
h
m
s
ar
e
ex
is
ti
n
g
to
tr
ain
a
n
eu
r
al
n
et
w
o
r
k
co
n
tr
o
ller
.
Her
e
L
e
v
en
b
er
g
m
ar
q
u
ar
d
t
al
g
o
r
ith
m
is
co
n
s
id
er
ed
f
o
r
tr
ain
i
n
g
t
h
e
n
eu
r
al
n
ets.
T
h
e
L
M
m
et
h
o
d
is
also
k
n
o
w
n
a
s
Da
m
p
ed
L
ea
s
t
Sq
u
ar
es
Me
th
o
d
s
(
DL
S)
to
s
o
lv
e
n
o
n
-
li
n
ea
r
least
s
q
u
ar
es
p
r
o
b
le
m
s
.
T
h
i
s
is
th
e
m
o
s
t
w
id
el
y
u
s
ed
o
p
ti
m
izatio
n
al
g
o
r
ith
m
f
o
r
t
h
e
n
o
n
-
li
n
ea
r
least sq
u
ar
es p
r
o
b
le
m
s
[
1
3
]
.
I
t p
o
s
s
ess
t
h
e
ad
v
an
tag
e
s
o
f
b
o
th
g
r
ad
ien
t
-
d
esce
n
t a
n
d
Gau
s
s
-
Ne
w
to
n
m
eth
o
d
s
.
Fig
u
r
e
1
r
ep
r
esen
t
s
t
h
e
Ne
u
r
o
n
s
tr
u
ctu
r
e
o
f
t
h
e
le
v
n
eb
er
g
m
ar
q
u
ar
d
t
alg
o
r
it
h
m
f
o
r
tr
ain
i
n
g
th
e
n
eu
r
o
n
s
i
n
t
h
e
n
eu
r
al
n
et
w
o
r
k
co
n
tr
o
ller
.
Her
e
t
h
e
s
tr
u
c
tu
r
e
co
n
s
is
ts
o
f
t
w
o
i
n
p
u
t
la
y
er
s
(
R
ef
er
e
n
ce
I
n
p
u
t
an
d
A
ct
u
al
I
n
p
u
t)
an
d
ten
h
id
d
en
lay
er
s
an
d
o
n
e
o
u
tp
u
t
la
y
er
(
Du
t
y
R
atio
)
.
T
h
e
Neu
r
o
n
s
in
th
is
al
g
o
r
ith
m
ar
e
tr
ain
ed
b
ased
o
n
t
h
e
p
er
f
o
r
m
an
ce
o
f
th
e
s
y
s
te
m
a
n
d
t
h
e
t
r
ain
ed
u
n
it
is
co
n
n
ec
ted
w
it
h
th
e
DVR
u
n
it
to
i
m
p
r
o
v
e
its
p
er
f
o
r
m
a
n
ce
to
r
ed
u
ce
th
e
h
ar
m
o
n
ic
s
to
o
b
tain
th
e
s
tab
ilit
y
o
f
v
o
lta
g
e
in
t
h
e
s
y
s
te
m
w
h
ich
i
n
tu
r
n
en
h
a
n
ce
s
t
h
e
p
o
w
er
q
u
al
it
y
o
f
t
h
e
s
y
s
te
m
[
1
4
]
.
A
s
s
aid
b
ef
o
r
e
th
e
h
ar
m
o
n
ics
ar
e
i
n
j
ec
ted
in
to
t
h
e
s
y
s
te
m
b
ea
ca
u
s
e
o
f
n
o
n
-
lin
ea
r
lo
ad
s
.
T
h
u
s
th
e
s
y
s
te
m
i
s
co
n
tr
o
lled
b
y
a
tr
ai
n
ed
NN
co
n
tr
o
ller
to
im
p
r
o
v
e
th
e
q
u
al
it
y
o
f
th
e
s
y
s
te
m
b
y
m
i
tig
a
tin
g
t
h
e
h
ar
m
o
n
ics p
r
o
p
ag
ated
in
th
e
d
is
tr
ib
u
tio
n
n
et
w
o
r
k
s
[
1
5
-
1
7
]
.
Fig
u
r
e
2
g
i
v
es
th
e
in
f
er
en
ce
s
ab
o
u
t
th
e
p
er
f
o
r
m
an
ce
c
u
r
v
e
o
f
th
e
tr
ain
ed
Ne
u
r
al
Net
w
o
r
k
co
n
tr
o
ller
.
T
h
e
b
est
v
alid
atio
n
p
er
f
o
r
m
a
n
ce
is
o
b
tai
n
ed
at
ep
o
ch
1
a
n
d
th
e
n
eu
r
o
n
s
ar
e
tr
ain
ed
to
t
h
i
s
b
est
f
it
l
in
e.
T
h
ese
n
eu
r
o
n
s
ar
e
tr
ain
ed
b
y
L
ev
e
n
b
er
g
Ma
r
q
u
ar
d
t
A
lg
o
r
ith
m
.
Af
ter
th
at
t
h
e
tr
ain
ed
n
eu
r
o
n
s
ar
e
u
s
ed
to
f
o
r
m
a
n
eu
r
al
n
e
t
w
o
r
k
s
tr
u
ctu
r
e
w
it
h
in
p
u
t,
h
id
d
en
an
d
o
u
tp
u
t
la
y
er
s
an
d
th
ese
s
tr
u
ctu
r
e
is
u
s
e
d
as
a
c
o
n
tr
o
ller
to
co
n
tr
o
l
th
e
DVR
u
n
it
f
o
r
its
b
etter
p
er
f
o
r
m
a
n
ce
.
Fig
u
r
e
3
r
ep
r
esen
ts
t
h
e
Neu
r
al
Net
w
o
r
k
b
lo
ck
o
f
lev
en
b
er
g
Ma
r
q
u
ar
d
t
alg
o
r
ith
m
.
Her
e
th
er
e
ar
e
t
w
o
in
p
u
t
la
y
er
s
w
h
ic
h
r
ep
r
esen
ts
th
e
r
e
f
er
en
ce
i
n
p
u
t
an
d
ac
t
u
al
i
n
p
u
t
,
ten
h
id
d
en
la
y
er
s
an
d
o
n
e
o
u
t
p
u
t
la
y
er
w
h
ic
h
r
ep
r
esen
t
s
th
e
d
u
t
y
r
atio
o
f
th
e
s
y
s
te
m
f
o
r
g
en
er
atin
g
t
h
e
p
u
ls
e
s
f
o
r
th
e
s
w
i
tch
e
s
in
t
h
e
co
n
v
er
t
er
s
.
Fig
u
r
e
1
.
Neu
r
o
n
Stru
c
tu
r
e
o
f
L
e
v
en
b
er
g
Ma
r
q
u
ar
d
t A
l
g
o
r
ith
m
Fig
u
r
e
2
.
P
er
f
o
r
m
a
n
ce
C
u
r
v
e
f
o
r
T
r
ain
ed
Neu
r
o
n
s
o
f
L
e
v
e
n
b
er
g
Ma
r
q
u
ar
d
t A
lg
o
r
it
h
m
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
.
2
,
J
u
n
e
2
0
1
8
:
7
3
8
–
743
740
Fig
u
r
e
3
.
Neu
r
al
Net
w
o
r
k
B
lo
ck
f
o
r
L
e
v
e
n
b
er
g
Ma
r
q
u
ar
d
t
A
l
g
o
r
ith
m
3.
SI
M
UL
AT
I
O
N
R
E
S
UL
T
S
AND
DIS
CUSS
I
O
NS
Fig
u
r
e
4
r
ep
r
esen
ts
t
h
e
s
i
m
u
l
atio
n
d
iag
r
a
m
f
o
r
Ne
u
r
al
Net
w
o
r
k
co
n
tr
o
lled
DVR
u
n
it
f
o
r
r
ed
u
cin
g
h
ar
m
o
n
ics
d
u
e
to
s
en
s
iti
v
e
l
o
ad
s
.
I
t
co
n
s
is
ts
o
f
DV
R
u
n
it
w
h
ic
h
is
co
n
tr
o
lled
b
y
t
h
e
Neu
r
al
Net
w
o
r
k
C
o
n
tr
o
ller
.
T
h
e
o
u
tp
u
t
o
f
DV
R
is
f
ed
in
to
t
h
e
tr
an
s
m
i
s
s
io
n
li
n
e
i.e
.
,
a
co
u
n
ter
v
o
ltag
e
i
s
f
ed
to
th
e
li
n
e
to
co
m
p
e
n
s
ate
f
o
r
th
e
v
o
ltag
e
d
r
o
p
in
th
e
lin
e
d
u
e
to
th
e
g
en
er
atio
n
o
f
h
ar
m
o
n
ics
b
ec
au
s
e
o
f
s
en
s
it
iv
e
lo
ad
s
.
T
h
e
lin
e
v
o
lta
g
e
is
d
is
tr
ac
ted
b
y
g
i
v
in
g
a
u
n
b
ala
n
ce
d
th
r
ee
p
h
ase
f
a
u
lt o
n
t
h
e
lo
ad
s
id
e
o
f
th
e
s
y
s
t
e
m
.
Fig
u
r
e
5
r
ep
r
esen
t
s
t
h
e
h
ar
m
o
n
ic
v
o
ltag
e
o
f
t
h
e
s
y
s
te
m
w
h
ic
h
is
cr
ea
ted
b
y
i
n
tr
o
d
u
ci
n
g
a
t
h
r
ee
p
h
ase
f
au
lt
o
n
th
e
lo
ad
s
id
e.
Her
e
th
e
th
r
ee
p
h
a
s
e
v
o
ltag
e
s
ar
e
h
a
v
in
g
d
if
f
er
en
t
f
r
eq
u
en
cie
s
a
n
d
a
m
p
lit
u
d
e.
Fig
u
r
e
6
r
ep
r
esen
ts
t
h
e
o
u
tp
u
t
v
o
ltag
e
o
f
t
h
e
s
y
s
te
m
af
ter
t
h
e
h
ar
m
o
n
ics
h
a
v
e
b
ee
n
r
ed
u
ce
d
b
y
t
h
e
i
n
tr
o
d
u
ctio
n
o
f
DVR
u
n
it.
Her
e
all
t
h
e
t
h
r
ee
p
h
ase
v
o
lta
g
es
h
av
e
s
a
m
e
m
ag
n
it
u
d
e
an
d
a
m
p
lit
u
d
e.
Fro
m
t
h
e
d
iag
r
a
m
i
t is c
lear
th
at
t
h
e
e
f
f
ec
ti
v
e
r
e
m
o
v
al
o
f
h
ar
m
o
n
ics
is
o
b
tai
n
ed
i
n
t
h
is
s
y
s
te
m
.
Fi
g
u
r
e
7
,
8
,
9
r
ep
r
esen
ts
th
e
FF
T
an
al
y
s
is
o
f
t
h
e
v
o
ltag
e
w
a
v
ef
o
r
m
b
ef
o
r
e
m
iti
g
ati
n
g
t
h
e
h
ar
m
o
n
ics
f
o
r
all
t
h
e
th
r
ee
p
h
ase
s
(
A
,
B
,
C
)
.
I
n
P
h
a
s
e
A
t
h
e
T
o
tal
Har
m
o
n
icD
is
to
r
tio
n
i
s
m
ea
s
u
r
ed
as
3
.
8
0
%.
I
n
P
h
ase
B
its
v
al
u
e
i
s
0
.
2
6
%
an
d
i
n
C
it
i
s
o
b
s
er
v
ed
as
0
.
3
3
%.
Fig
u
r
e
1
0
,
1
1
,
1
2
r
ep
r
esen
t
s
t
h
e
F
FT
an
al
y
s
i
s
o
f
v
o
ltag
e
q
u
an
t
it
y
f
o
r
t
h
r
ee
p
h
a
s
e
s
(
A
,
B
,
C
)
w
h
o
s
e
T
HD
v
alu
e
h
av
e
b
ee
n
i
m
p
r
o
v
ed
.
Fo
r
p
h
ase
A
th
e
T
o
tal
Har
m
o
n
ic
Di
s
to
r
tio
n
is
0
.
0
9
%.
Fo
r
p
h
ase
B
it
is
o
b
s
er
v
ed
as 0
.
1
3
% a
n
d
f
o
r
p
h
ase
C
it is
o
b
tai
n
ed
as 0
.
1
7
%
Fig
u
r
e
4
.
Si
m
u
latio
n
Diag
r
a
m
f
o
r
Neu
r
al
Net
w
o
r
k
C
o
n
tr
o
lle
d
DVR
Un
it
f
o
r
Har
m
o
n
ics R
ed
u
ctio
n
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:
2088
-
8
694
I
mp
leme
n
ta
tio
n
o
f NN
C
o
n
tr
o
lled
DV
R
fo
r
E
n
h
a
n
cin
g
th
e
P
o
w
er…
(
P
.
A
b
ir
a
mi
)
741
Fig
u
r
e
5
.
Har
m
o
n
ic
Vo
lta
g
e
D
u
e
to
T
h
r
ee
P
h
ase
Fau
lt
C
r
ea
ted
o
n
T
h
e
L
o
ad
Sid
e
.
Fig
u
r
e
6
.
Ou
tp
u
t V
o
lta
g
e
o
f
T
h
e
S
y
s
te
m
Af
ter
th
e
R
e
m
o
v
al
o
f
Har
m
o
n
ic
s
.
Fig
u
r
e
7
.
FF
T
A
n
al
y
s
i
s
f
o
r
P
h
ase
A
B
e
f
o
r
e
Mitig
at
in
g
Har
m
o
n
ics
Fig
u
r
e
8
.
FF
T
A
n
al
y
s
i
s
f
o
r
P
h
ase
B
B
ef
o
r
e
Mitig
at
in
g
Har
m
o
n
ics
Fig
u
r
e
9
.
FF
T
A
n
al
y
s
i
s
f
o
r
P
h
ase
C
B
ef
o
r
e
Mitig
at
in
g
Har
m
o
n
ics
Fig
u
r
e
10
.
FF
T
A
n
al
y
s
i
s
f
o
r
P
h
ase
A
Af
ter
th
e
R
e
m
o
v
al
o
f
Har
m
o
n
ics
Fig
u
r
e
11
.
FF
T
A
n
al
y
s
i
s
f
o
r
P
h
ase
B
A
f
ter
t
h
e
R
e
m
o
v
a
l
o
f
Har
m
o
n
ics
Fig
u
r
e
12
.
FF
T
A
n
al
y
s
i
s
f
o
r
P
h
ase
C
A
f
ter
t
h
e
R
e
m
o
v
al
o
f
Har
m
o
n
ic
s
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
.
2
,
J
u
n
e
2
0
1
8
:
7
3
8
–
743
742
T
ab
le
1
g
iv
es
d
etails
ab
o
u
t
th
e
T
HD
v
al
u
e
b
ef
o
r
e
r
ed
u
c
in
g
t
h
e
h
ar
m
o
n
ics
f
o
r
d
if
f
er
en
t
p
h
ase
s
o
f
th
e
s
y
s
te
m
.
T
ab
le
2
g
iv
es
d
etails
ab
o
u
t
th
e
T
HD
v
al
u
e
af
ter
r
ed
u
ci
n
g
t
h
e
h
ar
m
o
n
ics
f
o
r
d
if
f
er
e
n
t
p
h
a
s
es
o
f
th
e
s
y
s
te
m
.
T
ab
le
-
1
T
HD
B
ef
o
r
e
R
ed
u
cin
g
Har
m
o
n
ics
T
ab
le
-
2
T
HD
af
ter
r
ed
u
cin
g
h
ar
m
o
n
ics
4.
H
ARDWA
R
E
RE
SUL
T
S A
ND
DIS
C
USS
I
O
NS
Fig
u
r
e
1
3
r
ep
r
esen
ts
t
h
e
h
ar
d
w
ar
e
d
iag
r
a
m
o
f
DVR
u
n
i
t
f
o
r
r
ed
u
ctio
n
o
f
h
ar
m
o
n
ics
.
I
t
co
n
s
is
ts
o
f
MO
SF
E
T
Dr
iv
er
cir
cu
it
w
h
ic
h
ca
r
r
ies
th
e
MO
S
FET
s
w
itc
h
es
f
o
r
th
e
co
n
v
er
ter
s
.
A
N
N
b
ased
co
n
tr
o
ller
f
o
r
d
r
iv
in
g
th
e
DV
R
.
A
i
n
s
u
latio
n
t
r
an
s
f
o
r
m
er
s
f
o
r
p
r
o
tectin
g
th
e
co
n
tr
o
l
cir
cu
itr
y
f
r
o
m
t
h
e
p
o
w
er
cir
cu
itr
y
a
n
d
a
s
tep
d
o
w
n
tr
a
n
s
f
o
r
m
er
f
o
r
r
ed
u
cin
g
th
e
s
u
p
p
l
y
v
o
ltag
e
2
3
0
V
to
1
2
0
V.
I
t
also
c
o
n
s
is
t
s
o
f
a
Vo
ltag
e
So
u
r
ce
I
n
v
er
ter
w
h
ich
ac
ts
a
s
a
m
ain
co
m
p
o
n
e
n
t
o
f
t
h
e
DV
R
u
n
it.
I
t
co
n
s
i
s
ts
o
f
a
d
r
iv
er
cir
cu
it
b
o
ar
d
f
o
r
d
r
iv
in
g
t
h
e
MO
SF
E
T
s
w
itch
e
s
.
T
h
is
h
ar
d
w
ar
e
m
o
d
u
le
is
d
esi
g
n
ed
b
ased
o
n
th
e
s
i
m
u
la
tio
n
r
es
u
l
ts
o
b
tain
ed
f
r
o
m
th
e
s
o
f
t
w
ar
e.
F
ig
u
r
e
1
4
r
ep
r
esen
t
s
t
h
e
v
o
ltag
e
w
av
e
f
o
r
m
o
f
t
h
e
p
o
w
e
r
s
y
s
te
m
b
e
f
o
r
e
th
e
r
ed
u
c
tio
n
o
f
h
ar
m
o
n
ics
.
Fi
g
u
r
e
1
5
g
iv
es
t
h
e
o
u
tp
u
t
v
o
lta
g
e
o
f
t
h
e
h
ar
d
war
e
u
n
it
a
f
ter
th
e
r
e
m
o
v
al
o
f
h
ar
m
o
n
ics
f
r
o
m
t
h
e
s
y
s
te
m
.
Fig
u
r
e
13
.
Har
d
w
ar
e
o
f
DV
R
Un
it
f
o
r
R
ed
u
ctio
n
o
f
Har
m
o
n
ics
Fig
u
r
e
14
.
Vo
ltag
e
W
av
ef
o
r
m
o
f
th
e
Har
d
w
ar
e
Un
it
B
ef
o
r
e
C
o
m
p
en
s
a
tio
n
f
o
r
Har
m
o
n
ic
s
Fig
u
r
e
14
.
Vo
ltag
e
W
av
ef
o
r
m
o
f
th
e
Har
d
w
ar
e
Un
it
B
ef
o
r
e
C
o
m
p
en
s
a
tio
n
f
o
r
Har
m
o
n
ic
s
.
Fig
u
r
e
15
.
Ou
tp
u
t V
o
lta
g
e
W
av
ef
o
r
m
o
f
t
h
e
Har
d
w
ar
e
Un
i
t
A
f
ter
C
o
m
p
en
s
atio
n
f
o
r
Har
m
o
n
ic
s
5.
CO
NCLU
SI
O
N
D
y
n
a
m
ic
Vo
lta
g
e
r
esto
r
er
is
an
ef
f
icie
n
t
s
y
s
te
m
w
h
ic
h
is
a
p
p
lied
to
co
m
p
en
s
a
te
f
o
r
p
o
w
er
q
u
alit
y
is
s
u
es
li
k
e
Vo
ltag
e
Sa
g
/S
w
el
l
An
d
h
ar
m
o
n
ic
s
th
at
o
cc
u
r
s
d
u
e
to
ex
ter
n
al
d
is
tu
r
b
an
ce
s
.
A
s
t
h
e
ex
ter
n
al
d
is
tu
r
b
an
ce
s
ar
e
d
u
e
to
n
atu
r
al
ca
la
m
ities
,
m
o
s
t
o
f
th
e
ti
m
e
t
h
e
DVR
is
u
n
d
er
Static
co
n
d
it
io
n
w
h
ile
it
is
u
s
ed
f
o
r
co
m
p
en
s
ati
n
g
th
e
v
o
lta
g
e
s
ag
/
s
w
ell
o
f
t
h
e
s
y
s
te
m
.
So
,
it
ca
n
b
e
e
f
f
icie
n
tl
y
u
s
ed
f
o
r
h
ar
m
o
n
ics
co
m
p
e
n
s
at
io
n
d
u
e
to
s
e
n
s
it
iv
e
lo
ad
s
.
I
n
t
h
is
p
ap
er
th
e
h
a
r
m
o
n
ics
o
f
t
h
e
p
o
w
er
s
y
s
te
m
d
u
e
to
s
en
s
iti
v
e
lo
ad
s
ar
e
r
ec
tif
ied
b
y
u
s
i
n
g
a
NN
co
n
tr
o
lled
DVR
u
n
it.
T
h
e
L
e
v
en
b
er
g
m
ar
q
u
ar
d
t
al
g
o
r
ith
m
f
o
r
tr
ain
i
g
t
h
e
N
N
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:
2088
-
8
694
I
mp
leme
n
ta
tio
n
o
f NN
C
o
n
tr
o
lled
DV
R
fo
r
E
n
h
a
n
cin
g
th
e
P
o
w
er…
(
P
.
A
b
ir
a
mi
)
743
co
n
tr
o
ller
is
ex
p
lai
n
e
d
h
er
e.
A
n
o
n
-
li
n
ea
r
th
r
ee
p
h
a
s
e
f
a
u
l
t
is
in
tr
o
d
u
ce
d
in
to
th
e
s
y
s
te
m
to
d
ea
l
ab
o
u
t
th
e
h
ar
m
o
n
ics
g
e
n
er
atio
n
an
d
c
o
m
p
e
n
s
atio
n
f
o
r
t
h
e
h
ar
m
o
n
ic
s
d
u
e
to
s
e
n
s
iti
v
e
lo
ad
s
b
y
DVR.
T
h
e
T
HD
a
m
n
a
l
y
s
is
f
o
r
b
ef
o
r
e
an
d
af
ter
co
m
p
en
s
atio
n
f
o
r
h
ar
m
o
n
ic
s
is
a
n
l
y
s
ed
.
Fro
m
t
h
e
s
i
m
u
lati
o
n
r
es
u
l
t
s
a
n
d
FF
T
an
al
y
s
is
it
i
s
o
b
v
io
u
s
t
h
at
th
e
h
ar
m
o
n
ics
o
f
t
h
e
s
y
s
te
m
h
as
b
ee
n
r
ed
u
ce
d
.
T
h
e
h
ar
d
w
ar
e
i
s
also
m
ad
e
f
o
r
th
e
s
y
s
te
m
to
p
er
f
o
r
m
th
e
h
ar
m
o
n
ic
a
n
al
y
s
i
s
.
T
h
u
s
th
e
T
HD
a
n
d
th
e
d
is
to
r
tio
n
i
n
t
h
e
lo
ad
v
o
ltag
e
ar
e
r
ed
u
ce
d
w
it
h
th
e
a
p
p
licatio
n
o
f
NN
c
o
n
tr
o
lled
DVR.
Fro
m
t
h
e
ab
o
v
e
r
esu
lts
i
t
is
clea
r
th
a
t
t
h
e
p
r
o
p
o
s
ed
s
y
s
te
m
p
r
o
v
id
es b
etter
m
i
tig
at
io
n
o
f
h
ar
m
o
n
ics d
u
e
to
s
en
s
iti
v
e
lo
ad
s
.
RE
F
E
R
E
NC
E
S
[1
]
S
a
n
d
e
sh
Ja
in
,
P
r
o
f
.
S
h
iv
e
n
d
ra
S
i
n
g
h
T
h
a
k
u
r,
P
r
o
f
.
S
.
P
.
P
h
u
lam
b
rik
a
r
,
“
F
u
z
z
y
Co
n
tro
ll
e
r
Ba
se
d
D
V
R
T
o
M
it
ig
a
te
P
o
w
e
r
Qu
a
li
ty
A
n
d
Re
d
u
c
e
T
h
e
Ha
r
m
o
n
ics
Disto
rti
o
n
Of
S
e
n
si
ti
v
e
L
o
a
d
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
A
d
v
a
n
c
e
d
Res
e
a
rc
h
in
El
e
c
trica
l
,
El
e
c
tro
n
ics
a
n
d
In
str
u
me
n
ta
t
io
n
E
n
g
i
n
e
e
rin
g
V
o
l.
1
,
Iss
u
e
5
,
No
v
e
m
b
e
r
2
0
1
2
[2
]
M
ich
a
e
l
Jo
h
n
Ne
wm
a
n
,
M
e
m
b
e
r
,
IEE
E,
Do
n
a
ld
G
ra
h
a
m
e
Ho
lme
s,
S
e
n
io
r
M
e
m
b
e
r,
IEE
E,
Jo
h
n
G
o
d
sk
Nie
lse
n
,
M
e
m
b
e
r,
IEE
E,
a
n
d
F
re
d
e
Blaa
b
j
e
rg
,
F
e
ll
o
w
,
IEE
E
“
A
D
y
n
a
m
ic V
o
lt
a
g
e
Re
sto
re
r
(DV
R)
W
it
h
S
e
le
c
ti
v
e
Ha
r
m
o
n
ic
Co
m
p
e
n
sa
ti
o
n
a
t
M
e
d
iu
m
V
o
l
tag
e
L
e
v
e
l”,
IEE
E
tra
n
sa
c
ti
o
n
s
o
n
in
d
u
stry
a
p
p
li
c
a
ti
o
n
s
,
v
o
l.
4
1
,
n
o
.
6
,
n
o
v
e
m
b
e
r/d
e
c
e
m
b
e
r
2
0
0
5
[3
]
S
a
th
ish
Ba
b
u
P
a
n
d
u
a
n
d
Ka
m
a
ra
j
Na
g
a
p
p
a
n
“
A
No
v
e
l
M
u
lt
io
b
jec
ti
v
e
Co
n
tro
l
o
f
DV
R
to
E
n
h
a
n
c
e
P
o
w
e
r
Qu
a
li
t
y
o
f
S
e
n
siti
v
e
L
o
a
d
”
Hin
d
a
w
i
P
u
b
li
sh
in
g
Co
r
p
o
ra
ti
o
n
S
c
ien
ti
f
ic W
o
rld
Jo
u
rn
a
l
Vo
lu
m
e
2
0
1
5
.
[4
]
K.S
u
re
n
d
a
r,
M
.
Ra
m
y
a
,
M
.
M
u
ru
g
a
n
a
n
d
a
m
,
“
Co
n
tro
l
o
f
Re
d
u
c
e
d
Ra
ti
n
g
Dy
n
a
m
i
c
V
o
lt
a
g
e
Re
sto
r
e
r
W
it
h
En
e
rg
y
S
to
ra
g
e
S
y
ste
m
u
sin
g
F
u
z
z
y
L
o
g
ic”
In
ter
n
a
ti
o
n
a
l
J
o
u
r
n
a
l
o
f
In
n
o
v
a
t
ive
Res
e
a
rc
h
in
S
c
ien
c
e
,
En
g
iee
rin
g
a
n
d
T
e
c
h
n
o
l
o
g
y
,
Vo
l.
4
,
S
p
e
c
ial
Iss
u
e
6
,
M
a
y
2
0
1
5
.
[5
]
W
a
li
d
F
ra
n
g
ieh
,
M
a
g
e
d
B.
Na
jj
a
r
“
Ac
ti
v
e
c
o
n
tro
l
f
o
r
p
o
w
e
r
q
u
a
li
ty
i
m
p
ro
v
e
m
e
n
t
in
h
y
b
rid
p
o
w
e
r
s
y
ste
m
s”
IEE
E
X
p
l
o
re
T
e
c
h
n
o
l
o
g
ica
l
Ju
n
e
2
0
1
5
.
[6
]
Krisc
h
o
n
m
e
Bh
u
m
k
it
ti
p
ich
a
n
d
Na
d
a
ra
jah
M
it
h
u
lan
a
n
th
a
n
“
P
e
rf
o
rm
a
n
c
e
En
h
a
n
c
e
m
e
n
t
o
f
DV
R
f
o
r
M
it
ig
a
ti
n
g
V
o
l
tag
e
S
a
g
/S
we
ll
u
sin
g
V
e
c
to
r
Co
n
tr
o
l
S
trate
g
y
”
El
se
v
ier E
n
e
rg
y
P
ro
c
e
d
ia 9
(
2
0
1
1
)
3
6
6
–
3
7
9
.
[7
]
M
.
Na
b
ip
o
u
r
n
,
M
.
Ra
z
a
z
,
S
.
G
H.S
e
if
o
ss
a
d
a
t,
S
.
S
.
M
o
rtaz
a
v
i
“
A
n
o
v
e
l
a
d
a
p
ti
v
e
f
u
z
z
y
m
e
m
b
e
r
sh
ip
f
u
n
c
ti
o
n
t
u
n
in
g
a
lg
o
rit
h
m
f
o
r
ro
b
u
st
c
o
n
tro
l
o
f
a
P
V
-
b
a
se
d
Dy
n
a
m
ic
V
o
lt
a
g
e
Re
sto
re
r
(DV
R)”
El
se
v
ier
2
0
1
6
En
g
in
e
e
rin
g
A
p
p
li
c
a
ti
o
n
so
f
A
rti
f
icia
l
In
telli
g
e
n
c
e
5
3
(2
0
1
6
)1
5
5
–
1
7
5
.
[8
]
C.
K.
S
u
n
d
a
ra
b
a
lan
⇑
,
K.
S
e
lv
i
“
Co
m
p
e
n
sa
ti
o
n
o
f
v
o
lt
a
g
e
d
istu
rb
a
n
c
e
s
u
sin
g
P
EM
F
C
su
p
p
o
rted
D
y
n
a
m
i
c
V
o
lt
a
g
e
Re
sto
re
r”
El
se
v
ier
El
e
c
tri
c
a
l
P
o
w
e
r
a
n
d
E
n
e
rg
y
S
y
ste
m
s 7
1
(2
0
1
5
)
7
7
–
9
2
.
[9
]
K.
Ch
a
n
d
ra
se
k
a
ra
n
,
V
.
K.
Ra
m
a
c
h
a
n
d
a
ra
m
u
rth
y
“
A
n
im
p
ro
v
e
d
Dy
n
a
m
i
c
V
o
lt
a
g
e
Re
sto
re
r
f
o
r
p
o
w
e
r
q
u
a
li
ty
im
p
ro
v
e
m
e
n
t
”
El
se
v
ier 2
0
1
6
El
e
c
trica
l
Po
we
r
a
n
d
E
n
e
rg
y
S
y
ste
ms
8
2
(
2
0
1
6
)
3
5
4
–
3
6
2
.
[1
0
]
Jü
rg
e
n
S
c
h
m
id
h
u
b
e
r
“
De
e
p
lea
rn
i
n
g
in
n
e
u
ra
l
n
e
tw
o
rk
s: A
n
o
v
e
rv
i
e
w”
El
se
v
ier 2
0
1
5
.
[1
1
]
L
e
ste
r
S
.
H.
Ng
ia,
Jo
n
a
s
S
jö
b
e
rg
,
“
Eff
icie
n
t
T
ra
in
in
g
o
f
Ne
u
ra
l
Ne
ts
f
o
r
No
n
li
n
e
a
r
A
d
a
p
ti
v
e
F
i
lt
e
rin
g
Us
in
g
a
Re
c
u
rsiv
e
Lev
e
n
b
e
rg
–
M
a
rq
u
a
rd
t
A
l
g
o
rit
h
m
”
IEE
E
T
ra
n
s
a
c
ti
o
n
s O
n
S
i
g
n
a
l
Pr
o
c
e
ss
in
g
,
Vo
l.
4
8
,
No
.
7
,
J
u
ly
2
0
0
0
.
[1
2
]
G
.
Lera
a
n
d
M
.
P
in
z
o
las
“
Ne
ig
h
b
o
r
h
o
o
d
Ba
se
d
L
e
v
e
n
b
e
rg
–
M
a
rq
u
a
rd
t
A
lg
o
rit
h
m
f
o
r
Ne
u
ra
l
Ne
tw
o
rk
T
ra
in
in
g
”
IEE
E
T
ra
n
sa
c
ti
o
n
s On
Ne
u
ra
l
Ne
tw
o
rk
s,
V
o
l.
1
3
,
No
.
5
,
S
e
p
tem
b
e
r
2
0
0
2
.
[1
3
]
Kit
Ya
n
Ch
a
n
,
T
h
a
ra
m
S
.
Dill
o
n
,
Ja
ip
a
l
S
in
g
h
,
a
n
d
El
iza
b
e
th
Ch
a
n
g
“
Ne
u
ra
l
-
N
e
t
w
o
rk
-
Ba
se
d
M
o
d
e
ls
f
o
r
S
h
o
rt
-
T
e
r
m
T
r
a
ff
ic
F
lo
w
F
o
re
c
a
stin
g
Us
in
g
a
H
y
b
rid
Ex
p
o
n
e
n
ti
a
l
S
m
o
o
t
h
in
g
a
n
d
L
e
v
e
n
b
e
rg
–
M
a
rq
u
a
r
d
t
A
lg
o
rit
h
m
”
IEE
E
T
ra
n
sa
c
ti
o
n
s O
n
In
telli
g
e
n
t
T
ra
n
s
p
o
rt
a
ti
o
n
S
y
ste
ms
,
Vo
l.
1
3
,
No
.
2
,
Ju
n
e
2
0
1
2
[1
4
]
R.
Ba
las
u
b
ra
m
a
n
ian
,
S
.
P
a
lan
i,
“
S
im
u
latio
n
A
n
d
P
e
rf
o
rm
a
n
c
e
Ev
a
lu
a
ti
o
n
Of
S
h
u
n
t
Hy
b
rid
p
o
w
e
r
F
il
ter
F
o
r
P
o
w
e
r
Qu
a
li
ty
I
m
p
ro
v
e
m
e
n
t
Us
in
g
P
Q
T
h
e
o
r
y
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
El
e
c
trica
l
a
n
d
Co
m
p
u
ter
En
g
in
e
e
rin
g
”
Vo
l.
6
,
No
.
6
De
c
e
m
b
e
r
2
0
1
6
p
p
2
6
0
3
-
2
6
0
9
.
[1
5
]
M
.
Ja
w
a
d
G
h
o
rb
a
n
i
*
,
H.
M
o
k
h
ta
ri
*
*
“
I
m
p
a
c
t
o
f
Ha
r
m
o
n
ics
o
n
P
o
w
e
r
Qu
a
li
t
y
a
n
d
L
o
ss
e
s
in
P
o
we
r
Distrib
u
ti
o
n
S
y
st
e
m
s”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
E
lec
trica
l
a
n
d
C
o
mp
u
ter
E
n
g
i
n
e
e
rin
g
,
Vo
l.
5
,
No
.
1
,
F
e
b
ru
a
ry
2
0
1
5
,
p
p
.
1
6
6
~
1
7
4
[1
6
]
Ja
ru
p
u
la
S
o
m
lal,
V
e
n
u
G
o
p
a
la
Ra
o
.
M
a
n
n
a
m
,
Na
rsi
m
h
a
Ra
o
.
V
u
t
lap
a
ll
i
“
P
o
w
e
r
Qu
a
li
ty
I
m
p
ro
v
e
m
e
n
t
in
Distrib
u
ti
o
n
S
y
ste
m
u
sin
g
A
NN
Ba
se
d
S
h
u
n
t
A
c
ti
v
e
P
o
w
e
r
F
il
ter
”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
Po
we
r
El
e
c
tro
n
ics
a
n
d
Dr
ive
S
y
ste
m
V
o
l.
5
,
N
o
.
4
,
A
p
ril
2
0
1
5
,
p
p
.
5
6
8
~
5
7
5
[1
7
]
G
.
R
a
m
y
a
,
V
.
G
a
n
a
p
a
th
y
,
P
.
S
u
re
sh
“
P
o
w
e
r
Qu
a
li
t
y
I
m
p
ro
v
e
m
e
n
t
Us
in
g
M
u
lt
i
-
lev
e
l
In
v
e
rter
b
a
se
d
DV
R
a
n
d
DST
AT
COM
Us
in
g
Ne
u
ro
-
f
u
z
z
y
Co
n
tro
ll
e
r
”
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
P
o
we
r
El
e
c
tro
n
ics
a
n
d
Dr
ive
S
y
ste
m
(
IJ
PE
DS
)
V
o
l.
8
,
No
.
1
,
M
a
rc
h
2
0
1
7
,
p
p
.
3
1
6
~
3
2
4
Evaluation Warning : The document was created with Spire.PDF for Python.