I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
pu
t
er
E
ng
ineering
(
I
J
E
CE
)
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
25
,
p
p
.
2599
~
2
6
1
5
I
SS
N:
2088
-
8
7
0
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijece.
v
15
i
3
.
pp
2
5
9
9
-
2
6
1
5
2599
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m
H
y
brid op
timiza
t
io
n t
uned
deep
ne
ura
l net
wo
rk
-
ba
s
ed wind
po
wer genera
tion
sy
stem f
o
r
perm
a
nent
ma
g
net
sy
n
chro
no
us
g
enerato
r control
P
ra
s
ha
nt
K
um
a
r
S.
Chin
a
ma
lli,
M
un
g
a
m
uri Sa
s
ik
a
la
D
e
p
a
r
t
me
n
t
o
f
El
e
c
t
r
i
c
a
l
a
n
d
El
e
c
t
r
o
n
i
c
s
En
g
i
n
e
e
r
i
n
g
,
S
h
a
r
n
b
a
sv
a
U
n
i
v
e
r
si
t
y
,
K
a
l
a
b
u
r
a
g
i
,
I
n
d
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Oct
1
9
,
2
0
2
4
R
ev
is
ed
Dec
2
1
,
2
0
2
4
Acc
ep
ted
J
an
1
6
,
2
0
2
5
Wi
n
d
e
n
e
rg
y
,
a
c
o
st
-
e
ffe
c
ti
v
e
re
n
e
wa
b
le
so
u
rc
e
,
h
a
s
se
e
n
su
b
sta
n
ti
a
l
g
ro
wt
h
.
p
e
rm
a
n
e
n
t
m
a
g
n
e
t
sy
n
c
h
ro
n
o
u
s
g
e
n
e
ra
to
r
(P
M
S
G
)
e
q
u
ip
p
e
d
win
d
tu
r
b
in
e
s
d
e
m
o
n
stra
te
su
p
e
ri
o
r
p
e
rfo
rm
a
n
c
e
in
v
a
riab
le
-
sp
e
e
d
a
p
p
li
c
a
ti
o
n
s.
Ho
we
v
e
r,
th
e
re
re
m
a
in
s
a
n
o
tab
le
re
se
a
rc
h
g
a
p
in
o
p
ti
m
izin
g
th
e
o
v
e
r
a
ll
sy
ste
m
e
fficie
n
c
y
fo
r
su
c
h
win
d
e
n
e
rg
y
s
y
ste
m
s.
Th
e
re
fo
re
,
th
is
re
se
a
rc
h
p
re
se
n
ts
to
d
e
v
e
lo
p
a
d
e
e
p
lea
rn
i
n
g
-
b
a
se
d
o
p
ti
m
iza
ti
o
n
tec
h
n
i
q
u
e
t
h
a
t
imp
ro
v
e
s
t
h
e
e
fficie
n
c
y
o
f
P
M
S
G
-
b
a
se
d
win
d
e
n
e
rg
y
sy
ste
m
s
b
y
m
in
imiz
in
g
o
v
e
ra
l
l
sy
ste
m
lo
ss
e
s
a
n
d
m
a
x
imiz
in
g
e
n
e
rg
y
o
u
t
p
u
t.
C
o
re
l
o
ss
a
n
d
r
o
t
o
r
s
p
e
e
d
d
a
ta
we
re
fe
d
in
to
a
d
e
e
p
n
e
u
ra
l
n
e
two
rk
fo
r
v
a
rio
u
s o
p
e
ra
ti
n
g
c
o
n
d
i
ti
o
n
s
ra
n
g
in
g
fro
m
5
0
to
1
0
0
0
rp
m
,
t
o
d
e
term
in
e
o
p
ti
m
a
l
sy
ste
m
p
a
ra
m
e
ters
.
Th
is
wo
r
k
in
tro
d
u
c
e
s
a
h
y
b
rid
ly
re
b
i
rd
-
b
a
se
d
c
o
a
ti
o
p
ti
m
iza
ti
o
n
a
lg
o
r
it
h
m
(
LB
-
COA
)
to
o
p
ti
m
ize
t
h
e
d
e
e
p
n
e
u
ra
l
n
e
two
rk
s
(DN
N)
c
las
sifier,
c
o
m
b
i
n
in
g
tw
o
a
d
v
a
n
c
e
d
o
p
ti
m
iza
ti
o
n
tec
h
n
i
q
u
e
s
t
o
imp
ro
v
e
m
o
d
e
l
p
e
r
fo
rm
a
n
c
e
.
S
imu
latio
n
re
su
lt
s
v
a
li
d
a
te t
h
a
t
t
h
e
p
ro
p
o
se
d
o
p
t
imiz
a
ti
o
n
stra
teg
y
e
fficie
n
tl
y
b
o
o
sts
th
e
sy
ste
m
'
s d
y
n
a
m
ic p
e
rf
o
rm
a
n
c
e
a
n
d
o
v
e
ra
l
l
p
o
we
r
e
ffici
e
n
c
y
.
K
ey
w
o
r
d
s
:
C
o
ati
o
p
tim
izatio
n
Dee
p
n
eu
r
al
n
etwo
r
k
o
p
tim
izatio
n
L
o
s
s
m
in
im
izatio
n
L
y
r
eb
ir
d
o
p
tim
izatio
n
Per
m
an
en
t m
ag
n
et
s
y
n
ch
r
o
n
o
u
s
g
en
er
ato
r
W
in
d
p
o
wer
g
e
n
er
atio
n
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Pra
s
h
an
t K
u
m
ar
S.
C
h
in
am
alli
Dep
ar
tm
en
t o
f
E
lectr
ical
an
d
E
lectr
o
n
ics E
n
g
in
ee
r
i
n
g
,
Sh
a
r
n
b
asv
a
Un
iv
er
s
ity
Kala
b
u
r
ag
i,
I
n
d
ia
E
m
ail:
p
r
ash
an
tv
n
ec
@
g
m
ail.
c
o
m
1.
I
NT
RO
D
UCT
I
O
N
T
h
e
escalatin
g
g
lo
b
al
d
em
a
n
d
f
o
r
s
u
s
tain
ab
le
en
er
g
y
a
lter
n
ativ
es
to
f
o
s
s
il
f
u
els
h
as
d
r
iv
en
s
ig
n
if
ican
t
in
ter
est
in
win
d
en
er
g
y
tec
h
n
o
lo
g
y
[
1
]
.
I
ts
co
m
p
ellin
g
attr
ib
u
tes,
s
u
ch
as
co
s
t
-
ef
f
ec
tiv
en
ess
,
ea
s
e
o
f
in
s
tallatio
n
,
an
d
m
in
im
al
m
ain
ten
an
ce
r
eq
u
ir
em
en
ts
,
h
a
v
e
co
n
tr
ib
u
te
d
to
its
r
ap
id
g
r
o
wth
in
r
ec
en
t
y
ea
r
s
[
2
]
,
[
3
]
.
Per
m
a
n
en
t
m
a
g
n
et
s
y
n
ch
r
o
n
o
u
s
g
e
n
er
ato
r
s
(
PMSG)
b
ased
win
d
e
n
er
g
y
co
n
v
er
s
io
n
s
y
s
tem
s
(
W
E
C
S
)
h
av
e
g
ain
ed
s
u
b
s
tan
tial
p
o
p
u
l
ar
ity
b
ec
au
s
e
o
f
th
eir
co
m
p
ac
t
s
ize,
h
ig
h
o
u
tp
u
t
p
o
wer
,
an
d
h
ig
h
er
to
r
q
u
e
-
to
-
in
er
tia
r
atio
.
U
n
lik
e
d
o
u
b
l
y
-
f
ed
in
d
u
ctio
n
g
en
er
at
o
r
(
DFI
G)
s
y
s
tem
s
,
PMSGs
f
ea
tu
r
e
a
s
im
p
ler
i
n
ter
n
al
s
tr
u
ctu
r
e
with
o
u
t th
e
n
ee
d
f
o
r
co
m
p
lex
g
ea
r
b
o
x
es o
r
b
r
u
s
h
es
.
T
h
ese
ad
v
an
ta
g
es h
av
e
p
o
s
itio
n
ed
PMSG
-
b
ased
W
E
C
S
as
a
f
o
ca
l p
o
in
t
o
f
r
esear
ch
an
d
d
ev
elo
p
m
en
t
in
th
e
win
d
en
er
g
y
s
ec
to
r
[
4
]
.
T
h
e
f
lu
ctu
atin
g
a
n
d
e
r
r
atic
n
atu
r
e
o
f
win
d
s
p
ee
d
p
o
s
es
a
s
u
b
s
tan
tial
ch
allen
g
e
to
m
ax
im
izin
g
th
e
o
u
tp
u
t
o
f
h
ig
h
-
p
o
wer
win
d
g
en
er
ato
r
s
.
C
o
n
s
eq
u
en
tly
,
th
ese
s
y
s
tem
s
o
f
ten
o
p
e
r
ate
b
elo
w
th
eir
f
u
ll
p
o
ten
tial,
h
in
d
e
r
in
g
th
e
o
v
er
all
ef
f
ec
tiv
en
ess
o
f
win
d
en
er
g
y
u
tili
za
tio
n
an
d
d
im
in
is
h
in
g
th
e
an
ticip
ated
a
id
s
o
f
lar
g
e
-
s
ca
le
win
d
p
o
wer
g
en
er
ati
o
n
.
T
h
e
in
ter
m
itten
t
n
atu
r
e
o
f
win
d
s
p
ee
d
p
o
s
es
s
ig
n
if
ican
t
ch
allen
g
es
f
o
r
g
r
id
in
teg
r
atio
n
o
f
lar
g
e
-
s
ca
le
win
d
f
ar
m
s
.
Flu
ctu
atio
n
s
in
win
d
en
e
r
g
y
o
u
tp
u
t
ca
n
i
n
tr
o
d
u
ce
in
s
tab
ilit
y
in
to
th
e
p
o
wer
s
y
s
tem
.
W
in
d
tu
r
b
in
es,
co
m
p
lex
m
ac
h
in
es
in
f
lu
en
ce
d
b
y
v
ar
i
o
u
s
m
eteo
r
o
lo
g
ical
f
ac
to
r
s
,
f
u
r
th
er
co
m
p
licate
th
e
in
teg
r
a
tio
n
p
r
o
ce
s
s
.
W
in
d
en
er
g
y
c
o
n
v
er
s
io
n
is
af
f
ec
ted
b
y
v
ar
io
u
s
f
ac
to
r
s
,
s
u
ch
as
win
d
s
p
ee
d
,
b
lad
e
p
itc
h
an
g
l
e,
an
d
r
o
t
o
r
s
p
ee
d
.
Mo
r
eo
v
er
,
th
e
in
ter
m
itten
t
n
atu
r
e
o
f
win
d
an
d
en
e
r
g
y
lo
s
s
es
with
in
th
e
W
E
C
S
p
o
s
e
s
ig
n
if
ican
t
ch
allen
g
es
to
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
25
:
2
5
9
9
-
2
6
1
5
2600
s
y
s
tem
s
tab
ili
ty
[
5
]
.
T
h
er
ef
o
r
e,
W
E
C
S
is
a
co
m
p
lex
p
r
o
ce
s
s
ca
teg
o
r
ized
b
y
s
u
b
s
tan
tial
f
lu
ctu
atio
n
s
an
d
co
n
ce
r
n
s
.
T
o
a
d
d
r
ess
th
ese
ch
allen
g
es,
r
esear
ch
er
s
h
av
e
ex
p
lo
r
ed
n
u
m
er
o
u
s
co
n
tr
o
l stra
teg
ies an
d
th
eo
r
ies f
o
r
PMSG sy
s
tem
s
o
v
er
th
e
p
ast f
ew
d
ec
ad
es
[
6
]
.
B
ec
au
s
e
o
f
th
eir
u
n
co
m
p
licated
d
esig
n
,
r
eliab
ilit
y
,
an
d
s
tr
aig
h
tf
o
r
war
d
im
p
lem
en
tatio
n
,
p
r
o
p
o
r
tio
n
al
in
teg
r
al
(
PI)
co
n
tr
o
ller
s
r
em
ai
n
a
p
o
p
u
lar
ch
o
ice
f
o
r
en
h
a
n
cin
g
th
e
s
tab
ilit
y
o
f
b
o
th
th
e
g
r
id
s
id
e
co
n
v
er
ter
(
GSC
)
an
d
r
o
to
r
s
id
e
co
n
v
er
ter
(
R
SC
)
in
PMSG
-
b
ased
w
in
d
tu
r
b
i
n
es.
R
ef
er
en
ce
em
p
lo
y
s
a
p
r
o
p
o
r
tio
n
al
in
teg
r
al
(
PI)
co
n
tr
o
ller
to
en
h
an
ce
PMSG
win
d
tu
r
b
in
e
p
er
f
o
r
m
an
ce
u
n
d
er
b
o
th
s
tead
y
an
d
f
lu
ctu
atin
g
win
d
co
n
d
itio
n
s
[
7
]
.
Nev
er
t
h
eless
,
th
e
lin
ea
r
n
atu
r
e
o
f
PI
co
n
tr
o
l lim
its
its
ad
ap
tab
ilit
y
to
th
e
n
o
n
lin
ea
r
d
y
n
am
ics o
f
th
e
PMSG
s
y
s
tem
,
as
well
as
its
r
e
s
ilien
ce
to
v
ar
iatio
n
in
win
d
an
d
tu
r
b
in
e
co
n
d
itio
n
s
.
T
o
ad
d
r
ess
p
o
wer
q
u
ality
ch
allen
g
es
in
win
d
t
u
r
b
in
e
s
y
s
tem
s
,
v
ar
io
u
s
n
o
n
l
in
ea
r
co
n
tr
o
l
s
tr
ateg
ies
h
a
v
e
em
er
g
e
d
.
No
tab
le
ex
am
p
les in
clu
d
e
f
u
zz
y
lo
g
ic
co
n
tr
o
ller
(
FLC)
,
b
ac
k
s
tep
p
in
g
co
n
tr
o
l,
an
d
d
ir
ec
t p
o
we
r
co
n
tr
o
l (
DPC
)
[
8
]
,
[
9
]
.
Fu
zz
y
co
n
tr
o
l,
d
esp
ite
its
p
o
ten
tial
f
o
r
h
ig
h
ac
cu
r
ac
y
,
is
p
r
o
b
lem
atic
to
im
p
lem
en
t
d
u
e
to
its
r
elian
ce
o
n
s
u
b
s
tan
tial
ex
p
er
t
in
p
u
t
an
d
s
u
b
jectiv
e
h
u
m
an
i
n
ter
p
r
etatio
n
,
o
f
te
n
lead
in
g
to
d
elay
e
d
r
es
p
o
n
s
es.
W
h
ile
DPC
o
f
f
er
s
s
u
p
er
i
o
r
tr
an
s
ien
t
p
er
f
o
r
m
an
ce
,
it
r
e
q
u
ir
es
h
ig
h
s
witch
in
g
f
r
eq
u
e
n
cies
to
m
itig
ate
to
r
q
u
e
a
n
d
c
u
r
r
e
n
t
r
ip
p
le,
as r
ep
o
r
ted
in
[
1
0
]
.
Me
ta
-
h
eu
r
is
tic
alg
o
r
ith
m
s
h
av
e
b
ee
n
s
u
cc
ess
f
u
lly
em
p
lo
y
ed
to
ad
d
r
ess
in
tr
icate
e
n
g
in
ee
r
in
g
ch
allen
g
es
[
1
1
]
.
Fo
r
in
s
tan
ce
,
g
en
etic
alg
o
r
ith
m
(
GA)
h
as
b
ee
n
u
tili
ze
d
to
o
p
tim
ize
PI
co
n
tr
o
ller
p
ar
a
m
eter
s
f
o
r
PMSG
d
u
r
in
g
g
r
i
d
d
is
tu
r
b
an
ce
s
[
1
2
]
.
Mo
r
e
o
v
er
,
p
ar
ticl
e
s
war
m
o
p
tim
izatio
n
(
PS
O)
h
as
b
ee
n
ap
p
lied
to
PMSG
s
y
s
tem
s
o
p
er
atin
g
u
n
d
er
v
ar
y
in
g
win
d
s
p
ee
d
s
,
wh
ile
f
ir
ef
ly
alg
o
r
ith
m
(
FA)
h
as
b
e
en
u
s
ed
to
en
h
an
ce
m
ax
im
u
m
p
o
wer
p
o
in
t
tr
ac
k
in
g
(
MPPT)
p
er
f
o
r
m
a
n
ce
th
r
o
u
g
h
p
itch
an
g
le
co
n
tr
o
l
o
f
P
MSG
[
1
3
]
,
[
1
4
]
.
Fo
r
in
s
tan
ce
,
co
m
b
i
n
ed
a
n
a
d
ap
tiv
e
an
t
co
l
o
n
y
o
p
tim
izatio
n
(
A
AC
O)
alg
o
r
ith
m
with
a
g
e
n
er
al
r
eg
r
ess
io
n
n
eu
r
al
n
etwo
r
k
(
GR
NN)
to
im
p
r
o
v
e
MPPT
[
1
5
]
.
Ad
d
itio
n
ally
,
th
e
wh
ale
o
p
tim
izatio
n
alg
o
r
ith
m
h
as
b
ee
n
ap
p
lied
to
o
p
tim
ize
m
ac
h
in
e
-
s
id
e
co
n
v
e
r
ter
p
ar
am
eter
s
f
o
r
MPPT
[
1
6
]
.
C
o
llectiv
ely
,
th
ese
s
tu
d
ies
d
em
o
n
s
tr
ate
th
e
p
o
ten
tial
f
o
r
o
p
tim
izatio
n
alg
o
r
ith
m
s
in
im
p
r
o
v
in
g
b
o
th
MPPT
an
d
lo
w
v
o
ltag
e
r
id
e
th
r
o
u
g
h
(
L
VR
T
)
ca
p
ab
ilit
ies
o
f
PMSG
-
d
r
iv
en
win
d
en
er
g
y
s
y
s
tem
s
b
y
tu
n
i
n
g
eith
er
th
e
m
ac
h
in
e
-
s
id
e
co
n
v
er
ter
(
MSC
)
o
r
g
r
id
s
id
e
co
n
v
e
r
ter
(
GSC
)
p
ar
am
eter
s
[
1
7
]
.
Po
p
u
latio
n
-
b
ased
m
etah
eu
r
is
tic
alg
o
r
ith
m
s
,
s
u
ch
as
t
h
e
ly
r
eb
ir
d
o
p
tim
izatio
n
alg
o
r
ith
m
(
L
OA
)
,
o
f
f
er
s
ev
e
r
al
ad
v
an
ta
g
es
o
v
er
tr
ad
itio
n
al
o
p
tim
izatio
n
t
ec
h
n
iq
u
es.
L
OA’
s
ab
ilit
y
to
b
alan
ce
e
x
p
lo
r
ati
o
n
an
d
ex
p
lo
itatio
n
en
a
b
les
it
to
f
in
d
o
p
tim
al
s
o
lu
tio
n
s
in
a
r
elativ
ely
s
h
o
r
t
tim
e,
m
ak
in
g
it
s
u
itab
le
f
o
r
PMSG
-
b
ased
win
d
p
o
wer
g
e
n
er
at
io
n
.
T
h
e
s
u
b
s
eq
u
en
t
s
ec
tio
n
h
ig
h
lig
h
ts
th
e
k
ey
co
n
tr
ib
u
tio
n
s
:
a.
Pro
p
o
s
es a
d
ee
p
n
e
u
r
al
n
etwo
r
k
s
(
DNN)
m
o
d
el
th
at
h
el
p
s
to
m
in
im
ize
th
e
d
is
cr
ep
an
c
y
b
et
wee
n
ac
tu
al
an
d
p
r
ed
icted
d
ata
u
s
in
g
er
r
o
r
m
e
tr
ics.
T
h
is
m
o
d
el
aim
s
to
en
h
an
ce
th
e
ac
cu
r
ac
y
o
f
p
r
ed
icti
o
n
s
b
y
r
ed
u
cin
g
th
e
er
r
o
r
b
etwe
en
th
e
o
b
s
er
v
e
d
an
d
esti
m
ated
v
al
u
es.
b.
I
n
tr
o
d
u
ce
d
a
h
y
b
r
id
ly
r
eb
ir
d
-
c
o
ati
o
p
tim
izatio
n
alg
o
r
ith
m
(
L
B
-
C
OA)
to
o
p
tim
ize
DNN
cla
s
s
if
ier
weig
h
ts
,
th
er
eb
y
im
p
r
o
v
in
g
th
e
p
e
r
f
o
r
m
an
ce
,
p
r
ed
ictio
n
ac
cu
r
ac
y
,
a
n
d
en
e
r
g
y
ef
f
icien
c
y
o
f
win
d
p
o
wer
g
e
n
er
atio
n
u
s
in
g
PMSGs
.
c.
T
h
e
n
o
v
el
in
teg
r
atio
n
o
f
d
ee
p
lear
n
in
g
with
h
y
b
r
id
o
p
tim
iz
atio
n
tech
n
iq
u
es
a
d
d
r
ess
es
k
e
y
ch
allen
g
es
in
PMSG
-
b
ased
W
E
C
S.
d.
C
o
m
p
ar
ativ
e
an
aly
s
is
r
ev
ea
ls
th
at
th
e
s
u
g
g
ested
m
o
d
el
s
u
r
p
ass
es
tr
ad
itio
n
al
co
u
n
te
r
p
ar
ts
in
g
en
er
atin
g
o
u
tp
u
t
p
o
wer
,
m
i
n
im
izin
g
lo
s
s
es,
an
d
en
h
a
n
cin
g
e
f
f
icien
cy
.
T
h
e
s
tr
u
ctu
r
e
o
f
th
is
p
ap
e
r
is
as
f
o
llo
ws:
s
ec
tio
n
2
p
r
esen
ts
a
th
o
r
o
u
g
h
r
ev
iew
o
f
ex
is
tin
g
liter
atu
r
e.
Sectio
n
3
p
r
esen
ts
a
d
etaile
d
m
o
d
el
o
f
th
e
PMSG
-
b
ased
win
d
tu
r
b
in
e
s
y
s
tem
.
T
h
e
L
B
-
C
OA
tech
n
iq
u
e
f
o
r
p
ar
am
eter
tu
n
i
n
g
is
elab
o
r
ated
in
s
ec
tio
n
4
.
T
h
e
p
er
f
o
r
m
an
ce
e
v
alu
atio
n
an
d
r
esu
lt
s
ar
e
d
is
cu
s
s
ed
in
s
ec
tio
n
5
,
an
d
th
e
p
a
p
er
co
n
cl
u
d
es with
a
s
u
m
m
ar
y
o
f
k
ey
f
i
n
d
in
g
s
in
s
ec
tio
n
6
.
2.
L
I
T
E
R
AT
U
RE
R
E
VI
E
W
2
.
1
.
Rela
t
ed
wo
rk
s
A
n
o
v
el
co
n
tr
o
l
s
tr
ateg
y
co
m
b
in
in
g
f
u
zz
y
f
ield
-
o
r
ie
n
ted
co
n
tr
o
l
(
FF
OC
)
with
d
ir
ec
t
p
o
w
er
co
n
tr
o
l
(
DPC
)
f
o
r
v
a
r
iab
le
s
p
ee
d
win
d
en
er
g
y
c
o
n
v
e
r
s
io
n
s
y
s
tem
s
(
VSW
E
C
S)
was
p
r
o
p
o
s
ed
i
n
[
1
8
]
.
T
h
is
m
eth
o
d
aim
s
to
im
p
r
o
v
e
s
y
s
tem
p
er
f
o
r
m
an
ce
b
y
e
n
h
an
cin
g
en
er
g
y
ex
tr
ac
tio
n
an
d
m
in
im
izin
g
lo
s
s
es
th
r
o
u
g
h
th
e
o
p
tim
izatio
n
o
f
tu
r
b
in
e
o
p
er
atio
n
p
a
r
am
eter
s
.
Stab
ilit
y
,
r
o
b
u
s
tn
ess
,
p
er
f
o
r
m
a
n
ce
,
a
n
d
b
alan
ce
d
cu
r
r
en
t
in
jectio
n
ar
e
all
en
h
a
n
ce
d
b
y
th
e
co
m
b
in
e
d
co
n
t
r
o
l
alg
o
r
ith
m
.
An
in
n
o
v
ativ
e
o
p
tim
iza
tio
n
tech
n
iq
u
e
f
o
r
PMSG
-
b
ased
win
d
tu
r
b
in
es
w
as
p
r
esen
ted
in
s
tu
d
y
[
1
9
]
.
T
h
eir
ap
p
r
o
ac
h
p
r
io
r
itized
m
a
x
i
m
izin
g
W
E
G
wh
ile
co
n
cu
r
r
en
tly
d
ec
r
ea
s
in
g
t
h
e
o
v
er
all
s
y
s
tem
lo
s
s
es.
T
o
ac
h
iev
e
o
p
tim
al
win
d
p
o
wer
o
u
tp
u
t,
th
e
p
r
o
p
o
s
ed
m
eth
o
d
f
o
c
u
s
ed
o
n
o
p
tim
izi
n
g
to
r
q
u
e
p
e
r
am
p
er
e
(
T
PA)
,
as
well
as
q
u
ad
r
atu
r
e
cu
r
r
en
t.
T
h
e
o
b
jectiv
e
f
u
n
ctio
n
,
wh
ich
is
im
p
ac
ted
b
y
b
o
th
d
-
q
ax
is
cu
r
r
en
t,
is
co
r
e
lo
s
s
m
in
im
izatio
n
.
T
h
e
p
er
f
o
r
m
an
ce
o
f
th
e
s
u
g
g
ested
aq
u
ila
with
Af
r
ican
v
u
ltu
r
e
o
p
tim
izatio
n
(
E
A
-
A
VO)
is
ev
alu
ated
b
y
co
m
p
ar
i
n
g
its
o
u
tp
u
t
p
o
wer
,
lo
s
s
es,
ef
f
icac
y
,
an
d
co
n
v
er
g
en
ce
s
p
ee
d
with
estab
lis
h
e
d
ap
p
r
o
ac
h
es.
A
p
io
n
ee
r
in
g
co
n
tr
o
l
s
tr
ateg
y
to
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
Hyb
r
id
o
p
timiz
a
tio
n
tu
n
ed
d
e
ep
n
eu
r
a
l n
etw
o
r
k
-
b
a
s
ed
w
in
d
…
(
P
r
a
s
h
a
n
t Ku
ma
r
S
.
C
h
in
a
ma
lli
)
2601
o
p
tim
ize
in
ter
io
r
p
er
m
a
n
en
t
-
m
ag
n
et
s
y
n
ch
r
o
n
o
u
s
g
en
er
ato
r
(
I
PMSG)
-
b
ased
s
y
s
tem
s
wa
s
in
tr
o
d
u
ce
d
in
[
1
9
]
.
T
h
eir
m
eth
o
d
em
p
l
o
y
ed
p
o
ly
n
o
m
ial
p
ar
a
m
eter
s
to
cu
r
tail
I
PMSG
lo
s
s
e
s
an
d
am
p
lify
W
E
C
S.
T
h
is
s
tu
d
y
p
r
esen
ts
an
in
n
o
v
ativ
e
h
y
b
r
i
d
o
p
tim
izatio
n
tech
n
i
q
u
e
n
am
ed
cr
o
s
s
o
v
er
ass
is
ted
wh
ale
o
p
ti
m
izatio
n
alg
o
r
ith
m
(
C
W
OA)
,
wh
ich
ef
f
ec
tiv
ely
d
eter
m
in
es
o
p
tim
al
c
o
ef
f
ic
ien
ts
f
o
r
m
ax
im
izin
g
p
o
wer
g
en
er
atio
n
.
T
h
is
f
r
am
ewo
r
k
o
p
tim
izes
tip
s
p
e
ed
r
atio
,
co
n
s
id
er
ed
a
cr
itical
p
ar
am
eter
f
o
r
win
d
e
n
er
g
y
co
n
v
er
s
io
n
.
I
t
also
ac
co
u
n
ts
f
o
r
th
e
n
o
n
lin
ea
r
b
e
h
av
io
r
o
f
th
e
I
PMSG
d
u
e
to
m
ag
n
etic
s
atu
r
atio
n
.
T
o
ass
ess
th
e
ef
f
icac
y
o
f
th
e
C
W
O
A
alg
o
r
ith
m
,
it
is
co
m
p
ar
ed
ag
ain
s
t
estab
lis
h
ed
tech
n
iq
u
es
lik
e
W
OA,
f
ir
ef
ly
alg
o
r
ith
m
(
FF
)
,
ar
tific
ial
b
ee
co
lo
n
y
(
AB
C
)
,
an
d
g
e
n
eti
c
alg
o
r
ith
m
(
GA)
.
A
n
o
v
el
s
lid
in
g
m
o
d
e
co
n
tr
o
l
(
SMC
)
s
tr
ateg
y
f
o
r
PMSG
in
win
d
en
er
g
y
c
o
n
v
er
s
io
n
s
y
s
tem
s
,
with
a
f
o
cu
s
o
n
m
a
x
im
izin
g
p
o
wer
o
u
tp
u
t,
was
in
tr
o
d
u
ce
d
in
s
tu
d
y
[
2
0
]
.
SMC
's
r
o
b
u
s
tn
ess
in
h
an
d
lin
g
n
o
n
lin
ea
r
elec
tr
ical
s
y
s
tem
s
h
as
m
ad
e
i
t
a
p
o
p
u
lar
ch
o
ice
in
th
is
d
o
m
ain
.
T
h
is
r
esear
ch
em
p
lo
y
s
a
n
o
n
lin
ea
r
PMSG
m
o
d
el
to
im
p
lem
en
t
SMC
co
n
tr
o
l.
T
h
e
p
r
im
ar
y
g
o
al
is
t
o
r
eg
u
late
s
tato
r
PQ
p
o
wer
,
as
well
as
v
o
ltag
e
f
r
eq
u
e
n
cy
,
f
o
r
o
p
tim
al
g
r
i
d
in
teg
r
atio
n
.
T
h
e
r
esu
lts
d
em
o
n
s
tr
ate
en
h
an
ce
d
s
y
s
tem
r
o
b
u
s
tn
ess
.
A
co
m
p
r
eh
e
n
s
iv
e
m
o
d
el
o
f
a
v
a
r
iab
le
-
s
p
ee
d
win
d
tu
r
b
in
e
eq
u
ip
p
ed
with
a
PMSG
was
p
r
es
en
ted
in
s
tu
d
y
[
2
1
]
.
T
h
eir
co
n
tr
o
l
s
tr
ateg
y
aim
ed
to
o
p
tim
ize
win
d
p
o
wer
ca
p
tu
r
e
b
y
em
p
l
o
y
in
g
f
ield
-
o
r
ien
ted
co
n
tr
o
l
(
FOC
)
an
d
an
id
ea
l
s
p
ee
d
s
etp
o
in
t
d
eter
m
in
ed
b
y
win
d
co
n
d
itio
n
s
.
A
co
m
p
ar
ativ
e
an
aly
s
is
o
f
PS
O
an
d
its
v
ar
ian
ts
was
co
n
d
u
cte
d
to
o
p
tim
ize
PI
c
o
n
t
r
o
ller
g
ai
n
s
at
c
o
n
v
e
r
g
en
ce
.
T
h
e
r
esu
lts
in
d
icate
d
th
at
th
e
PS
O
-
b
ased
co
n
tr
o
ller
ex
h
ib
ited
i
n
f
er
io
r
p
e
r
f
o
r
m
an
c
e
m
etr
ics
ac
r
o
s
s
v
a
r
io
u
s
e
r
r
o
r
cr
iter
ia
wh
e
n
c
o
m
p
ar
e
d
to
th
e
ex
p
licitly
d
ef
in
ed
co
n
tr
o
ller
.
A
n
o
v
el
s
en
s
o
r
le
s
s
tech
n
iq
u
e
ca
p
ab
le
o
f
o
p
e
r
atin
g
PMSGs
ac
r
o
s
s
a
wid
e
s
p
ee
d
r
an
g
e
was
d
ev
elo
p
e
d
in
[
2
2
]
.
T
h
is
tech
n
i
q
u
e
lev
er
ag
es th
e
cu
r
r
en
t c
o
n
t
r
o
ller
’
s
o
u
tp
u
t w
ith
in
th
e
s
p
ee
d
co
n
tr
o
l lo
o
p
.
T
h
is
s
tu
d
y
in
tr
o
d
u
ce
d
an
o
p
p
o
s
i
tio
n
p
ar
ticle
s
war
m
o
p
tim
izatio
n
-
s
u
p
p
o
r
t
v
ec
to
r
r
e
g
r
es
s
io
n
(
OPSO
-
SV
R
)
alg
o
r
ith
m
f
o
r
win
d
s
p
ee
d
p
r
ed
ictio
n
u
s
in
g
h
is
to
r
ical
o
f
f
li
n
e
d
ata.
T
h
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
ad
d
r
ess
es
th
e
ch
allen
g
e
o
f
s
u
p
p
o
r
t
v
ec
to
r
r
eg
r
ess
io
n
(
SVR
)
p
ar
a
m
eter
o
p
tim
izatio
n
ac
r
o
s
s
a
b
r
o
ad
win
d
s
p
ee
d
s
p
ec
tr
u
m
.
E
m
p
lo
y
in
g
OPSO
y
ield
s
f
aste
r
an
d
m
o
r
e
ac
c
u
r
ate
o
p
tim
al
S
VR
p
ar
am
eter
s
.
T
h
e
d
e
v
elo
p
e
d
alg
o
r
ith
m
e
x
ce
ls
at
r
ap
id
an
d
p
r
ec
is
e
win
d
s
p
ee
d
esti
m
atio
n
th
r
o
u
g
h
s
wif
t
o
p
tim
izatio
n
a
n
d
tu
n
in
g
o
f
SVR
p
ar
am
eter
s
.
I
t
ac
cu
r
ately
tr
ac
k
s
r
ea
l
win
d
s
p
ee
d
v
alu
es
an
d
is
s
o
v
er
eig
n
o
f
g
e
n
er
ato
r
t
u
r
b
in
e
c
o
n
s
tan
ts
o
r
to
r
q
u
e
m
ea
s
u
r
em
en
ts
.
T
h
u
s
,
th
e
ex
p
er
im
en
tal
f
in
d
in
g
s
c
o
n
f
ir
m
o
u
ts
tan
d
in
g
p
er
f
o
r
m
an
ce
in
p
r
ed
ictin
g
win
d
s
p
ee
d
an
d
r
o
t
o
r
s
p
ee
d
.
An
t
lio
n
o
p
tim
izer
(
AL
O)
wa
s
p
r
o
p
o
s
ed
t
o
tu
n
e
t
h
e
p
ar
a
m
e
ter
s
o
f
a
co
n
v
en
tio
n
al
PI
co
n
tr
o
ller
f
o
r
a
win
d
en
er
g
y
s
y
s
tem
with
a
PMSG
[
2
3
]
.
T
h
eir
aim
was
MPPT
wh
ile
en
h
an
cin
g
f
au
lt
r
id
e
-
th
r
o
u
g
h
p
er
f
o
r
m
an
ce
.
T
h
e
p
r
o
p
o
s
ed
m
eth
o
d
ef
f
ec
tiv
ely
en
h
an
ce
d
lo
w
v
o
ltag
e
r
id
e
th
r
o
u
g
h
(
L
VR
T
)
p
er
f
o
r
m
an
c
e
wh
ile
m
ax
im
izin
g
p
o
wer
ex
t
r
ac
tio
n
.
T
h
e
f
in
d
in
g
s
d
em
o
n
s
tr
ate
a
s
u
b
s
tan
tial
en
h
an
ce
m
e
n
t
in
o
v
er
all
s
y
s
tem
d
y
n
am
ics
wh
en
co
n
tr
asted
with
th
e
tr
a
d
itio
n
al
PI
c
o
n
tr
o
ller
.
An
ad
v
an
ce
d
MPPT
s
tr
ateg
y
co
u
p
led
with
p
itch
an
g
le
co
n
tr
o
l
was
in
tr
o
d
u
ce
d
in
[
2
4
]
.
T
h
is
r
esear
ch
f
o
cu
s
es
o
n
two
p
r
im
ar
y
o
b
jectiv
es.
T
h
e
f
ir
s
t
is
to
co
o
r
d
in
ate
th
e
g
e
n
er
ato
r
an
d
GSC
to
f
o
llo
w
th
e
o
p
tim
al
win
d
s
p
ee
d
s
etp
o
in
t
d
eter
m
in
ed
b
y
th
e
MPPT
tech
n
iq
u
e.
T
h
e
s
ec
o
n
d
is
to
m
itig
ate
th
e
ch
atter
in
g
is
s
u
e
in
h
er
en
t
i
n
s
tan
d
ar
d
SMC
b
y
in
tr
o
d
u
cin
g
a
n
o
v
e
l
s
m
o
o
th
co
n
tin
u
o
u
s
SMC
s
tr
ateg
y
.
Fu
r
t
h
er
m
o
r
e,
L
y
ap
u
n
o
v
s
tab
ilit
y
an
aly
s
is
v
alid
ates
th
e
p
r
o
p
o
s
ed
s
lid
in
g
m
o
d
e
c
o
n
tr
o
ller
.
Simu
latio
n
r
esu
lts
u
n
d
er
s
co
r
e
th
e
c
o
n
tr
o
ller
's
ef
f
ec
tiv
en
ess
in
ac
h
ie
v
in
g
p
r
ec
is
e,
s
tab
le
o
p
er
atio
n
with
m
in
im
al
o
u
t
p
u
t
cu
r
r
en
t
r
ip
p
le.
2
.
2
.
Rev
iew
T
ab
le
1
p
r
o
v
id
es
an
o
v
er
v
iew
o
f
s
ev
er
al
c
o
n
tr
o
llin
g
m
eth
o
d
s
f
o
r
PMSG
-
b
ased
W
E
C
S.
I
n
itially
,
a
DNN
clas
s
if
ier
,
wh
ich
ca
n
h
an
d
le
u
n
s
tr
u
ctu
r
e
d
an
d
u
n
lab
eled
d
ata,
was
r
ec
o
m
m
en
d
ed
in
[
2
5
]
.
Ho
wev
er
,
o
v
er
f
itti
n
g
r
e
m
ain
s
a
s
ig
n
if
ican
t
ch
allen
g
e
en
co
u
n
ter
ed
i
n
th
is
ap
p
r
o
ac
h
.
Fu
r
th
er
m
o
r
e,
th
e
AI
-
AVO
m
eth
o
d
,
d
esig
n
ed
to
p
r
ev
e
n
t
p
r
e
m
atu
r
e
co
n
v
e
r
g
en
ce
an
d
ac
h
iev
e
o
p
tim
al
s
o
lu
tio
n
s
,
f
ac
es
ch
alle
n
g
es
s
u
ch
as
s
lo
w
co
n
v
er
g
en
ce
r
ates,
h
ig
h
c
o
m
p
u
tatio
n
al
d
em
a
n
d
s
,
an
d
s
en
s
itiv
ity
to
p
ar
am
eter
s
ettin
g
s
[
1
9
]
.
L
i
k
ewise,
th
e
C
W
O
A
s
tr
ateg
y
,
p
r
esen
ted
to
d
ec
r
ea
s
e
p
ar
am
eter
o
p
tim
izat
io
n
tim
e,
s
till
r
eq
u
ir
es
s
ig
n
if
i
ca
n
t
co
m
p
u
tatio
n
al
r
eso
u
r
ce
s
to
d
eter
m
in
e
o
p
tim
al
p
ar
am
eter
s
[
2
0
]
.
Ad
d
itio
n
a
lly
,
SMC
co
n
tr
o
l,
p
r
o
p
o
s
ed
f
o
r
en
h
a
n
ce
d
s
y
s
tem
p
er
f
o
r
m
an
ce
an
d
h
ig
h
r
o
b
u
s
tn
ess
,
en
co
u
n
ter
s
th
e
ch
alle
n
g
e
o
f
t
h
e
ch
atter
in
g
p
h
en
o
m
en
o
n
[
2
1
]
,
[
2
5
]
.
Similar
ly
,
th
e
PI+
PS
O
s
ch
em
e,
in
tr
o
d
u
c
ed
,
is
ea
s
y
to
im
p
le
m
en
t
an
d
co
n
v
er
g
es
f
aster
b
u
t
is
co
m
p
u
tatio
n
ally
ex
p
en
s
iv
e
[
2
2
]
.
I
n
ad
d
itio
n
,
t
h
e
OPSO
-
SV
R
m
o
d
el
s
u
g
g
ested
,
wh
ich
o
f
f
er
s
b
etter
p
er
f
o
r
m
an
ce
an
d
f
aster
co
n
v
er
g
en
ce
,
is
co
m
p
u
tatio
n
al
ly
in
ten
s
iv
e
an
d
r
eq
u
ir
es
ca
r
e
f
u
l
p
ar
am
eter
t
u
n
in
g
f
o
r
o
p
tim
al
r
esu
lts
[
2
3
]
.
T
h
e
AL
O
-
PI
s
tr
ateg
y
,
d
e
v
elo
p
ed
,
o
f
f
er
s
s
u
p
e
r
io
r
p
er
f
o
r
m
an
ce
a
n
d
elim
in
ates
th
e
n
ee
d
f
o
r
g
r
ad
ien
t
in
f
o
r
m
atio
n
;
h
o
wev
er
,
a
m
aj
o
r
d
r
aw
b
ac
k
i
s
its
s
lo
w
co
n
v
er
g
en
ce
a
n
d
is
co
m
p
u
tatio
n
ally
e
x
p
en
s
iv
e
[
2
4
]
.
T
h
er
ef
o
r
e,
th
is
r
esear
ch
en
d
ea
v
o
r
s
to
r
eso
lv
e
th
e
p
r
ev
io
u
s
ly
o
u
tlin
ed
is
s
u
es
an
d
o
p
tim
ize
W
E
C
S
p
er
f
o
r
m
a
n
ce
th
r
o
u
g
h
PMSG in
teg
r
atio
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
25
:
2
5
9
9
-
2
6
1
5
2602
T
ab
le
1
.
Featu
r
es a
n
d
ch
allen
g
es o
f
PMSG
-
b
ased
W
E
C
S u
s
i
n
g
v
ar
i
o
u
s
co
n
tr
o
llin
g
s
tr
ateg
i
es
A
u
t
h
o
r
[
c
i
t
a
t
i
o
n
]
M
e
t
h
o
d
o
l
o
g
y
u
t
i
l
i
z
e
d
F
e
a
t
u
r
e
s
C
h
a
l
l
e
n
g
e
s
S
a
l
i
m
e
e
t
a
l
.
[
1
8
]
F
F
O
C
−
P
o
w
e
r
q
u
a
l
i
t
y
w
i
t
h
l
o
w
h
a
r
m
o
n
i
c
d
i
s
t
o
r
t
i
o
n
−
S
i
mp
l
e
i
m
p
l
e
me
n
t
a
t
i
o
n
−
F
a
c
e
c
h
a
l
l
e
n
g
e
s
e
s
p
e
c
i
a
l
l
y
i
n
sy
s
t
e
m
s
w
i
t
h
f
a
st
e
r
d
y
n
a
mi
c
s
o
r
u
n
d
e
r
e
x
t
r
e
m
e
l
o
a
d
c
o
n
d
i
t
i
o
n
s
C
h
i
n
a
ma
l
l
i
a
n
d
S
a
si
k
a
l
a
[
1
9
]
AI
-
AVO
a
p
p
r
o
a
c
h
−
P
r
e
v
e
n
t
s
p
r
e
ma
t
u
r
e
c
o
n
v
e
r
g
e
n
c
e
a
n
d
a
c
h
i
e
v
e
s
o
p
t
i
m
a
l
so
l
u
t
i
o
n
s
−
H
o
w
e
v
e
r
,
t
h
e
y
a
l
so
f
a
c
e
c
h
a
l
l
e
n
g
e
s
i
n
t
e
r
ms
o
f
sl
o
w
c
o
n
v
e
r
g
e
n
c
e
r
a
t
e
s,
h
i
g
h
c
o
mp
u
t
a
t
i
o
n
a
l
d
e
m
a
n
d
s
a
n
d
s
e
n
si
t
i
v
e
p
a
r
a
me
t
e
r
s
e
t
t
i
n
g
s
C
h
i
n
a
ma
l
l
i
a
n
d
S
a
si
k
a
l
a
[
2
0
]
C
W
O
A
t
e
c
h
n
i
q
u
e
−
D
e
c
r
e
a
s
e
p
a
r
a
m
e
t
e
r
o
p
t
i
m
i
z
a
t
i
o
n
t
i
m
e
−
D
e
t
e
r
m
i
n
i
n
g
o
p
t
i
ma
l
p
a
r
a
m
e
t
e
r
s c
a
n
b
e
c
o
m
p
u
t
a
t
i
o
n
a
l
l
y
d
e
m
a
n
d
i
n
g
La
a
b
i
d
i
n
e
e
t
a
l
.
[
2
1
]
S
M
C
c
o
n
t
r
o
l
−
O
f
f
e
r
s h
i
g
h
l
e
v
e
l
s
o
f
r
o
b
u
s
t
n
e
ss f
o
r
e
n
h
a
n
c
e
d
sy
s
t
e
m
p
e
r
f
o
r
ma
n
c
e
−
A
si
g
n
i
f
i
c
a
n
t
c
h
a
l
l
e
n
g
e
a
s
so
c
i
a
t
e
d
w
i
t
h
S
M
C
i
s t
h
e
c
h
a
t
t
e
r
i
n
g
p
h
e
n
o
me
n
o
n
H
a
n
n
a
c
h
i
e
t
a
l
.
[
2
2
]
P
I
+
P
S
O
−
Ea
sy
t
o
i
mp
l
e
me
n
t
−
C
o
n
v
e
r
g
e
s fas
t
e
r
−
C
o
m
p
u
t
a
t
i
o
n
a
l
l
y
e
x
p
e
n
s
i
v
e
A
b
o
-
K
h
a
l
i
l
e
t
a
l
.
[
2
3
]
O
P
S
O
-
S
V
R
mo
d
e
l
−
B
e
t
t
e
r
p
e
r
f
o
r
ma
n
c
e
−
C
o
n
v
e
r
g
e
s
q
u
i
c
k
l
y
−
H
o
w
e
v
e
r
,
i
t
i
s
c
o
mp
u
t
a
t
i
o
n
a
l
l
y
i
n
t
e
n
s
i
v
e
a
n
d
r
e
q
u
i
r
e
s
c
a
r
e
f
u
l
p
a
r
a
m
e
t
e
r
t
u
n
i
n
g
f
o
r
o
p
t
i
m
a
l
p
e
r
f
o
r
m
a
n
c
e
.
H
a
r
i
d
y
e
t
a
l
.
[
2
4
]
A
LO
-
P
I
st
r
a
t
e
g
y
−
O
f
f
e
r
s su
p
e
r
i
o
r
p
e
r
f
o
r
m
a
n
c
e
−
El
i
m
i
n
a
t
e
s
t
h
e
n
e
e
d
f
o
r
g
r
a
d
i
e
n
t
i
n
f
o
r
mat
i
o
n
.
−
C
o
n
v
e
r
g
e
s s
l
o
w
l
y
−
C
o
m
p
u
t
a
t
i
o
n
a
l
l
y
e
x
p
e
n
s
i
v
e
M
a
j
o
u
t
e
t
a
l
.
[
2
5
]
P
S
M
C
c
o
n
t
r
o
l
−
O
f
f
e
r
s h
i
g
h
l
e
v
e
l
s
o
f
r
o
b
u
s
t
n
e
ss f
o
r
e
n
h
a
n
c
e
d
sy
s
t
e
m
p
e
r
f
o
r
ma
n
c
e
−
A
si
g
n
i
f
i
c
a
n
t
c
h
a
l
l
e
n
g
e
a
s
so
c
i
a
t
e
d
w
i
t
h
S
M
C
i
s t
h
e
c
h
a
t
t
e
r
i
n
g
p
h
e
n
o
me
n
o
n
3.
SYST
E
M
M
O
D
E
L
O
F
A
P
M
SG
-
B
A
SE
D
WI
ND
T
UR
B
I
NE
G
E
NE
R
AT
I
O
N
SYS
T
E
M
B
y
em
p
lo
y
in
g
a
p
er
m
a
n
en
t
m
ag
n
et
to
tr
an
s
f
o
r
m
th
e
m
e
ch
an
ical
en
er
g
y
g
en
e
r
ated
b
y
th
e
win
d
tu
r
b
in
e
in
to
AC
elec
tr
icity
,
th
e
I
PMSG
ac
ts
as
a
g
en
er
ato
r
.
T
h
e
g
en
e
r
ato
r
tr
a
n
s
f
o
r
m
s
th
e
m
ec
h
an
ical
r
o
tatio
n
in
to
elec
tr
ical
e
n
er
g
y
b
y
u
s
e
o
f
m
a
g
n
etic
f
ield
s
.
A
g
en
er
at
o
r
'
s
elec
tr
ic
o
u
tp
u
t
m
ay
r
is
e
with
an
i
n
cr
ea
s
e
i
n
lo
ad
cu
r
r
en
t
o
r
r
o
to
r
s
p
ee
d
.
An
in
s
u
lated
-
g
ate
b
ip
o
lar
t
r
an
s
is
to
r
(
I
GB
T
)
-
b
ased
p
u
ls
e
wid
th
m
o
d
u
latio
n
(
PW
M)
co
n
v
er
ter
is
th
en
u
s
ed
to
r
ec
tify
th
is
AC
o
u
tp
u
t
in
to
DC
p
o
wer
,
wh
ich
is
th
en
s
en
t
in
to
a
DC
lin
k
.
A
PW
M
co
n
v
er
ter
is
u
s
ed
to
m
o
d
if
y
th
e
AC
o
u
tp
u
t
v
o
ltag
es
o
f
th
e
I
PMSG
in
o
r
d
er
to
co
n
tr
o
l
its
p
o
wer
o
u
tp
u
t.
W
ith
th
e
aid
o
f
an
ad
d
itio
n
al
en
er
g
y
in
v
er
ter
,
th
is
en
ab
les
th
e
I
PMSG
to
d
eliv
er
AC
p
o
w
er
in
to
th
e
g
r
id
o
r
a
lo
ad
wh
ile
p
r
eser
v
in
g
its
s
tead
y
v
o
ltag
e
an
d
f
r
eq
u
e
n
cy
.
W
h
er
ea
s
d
ef
in
es
th
e
air
d
en
s
ity
;
im
p
lies
th
e
ar
ea
s
wep
t in
th
e
r
o
to
r
b
la
d
e
wh
ich
is
=
2
; L
et
th
e
r
o
to
r
r
ad
iu
s
b
e
in
m
eter
s
,
s
tates
th
e
W
T
v
elo
city
an
d
ch
ar
ac
ter
izes
th
e
win
d
p
o
wer
f
ac
to
r
wh
ic
h
is
a
f
u
n
cti
o
n
o
f
tip
s
p
ee
d
r
atio
(
)
an
d
t
h
e
b
lad
e
p
itch
an
g
le
(
)
[
2
6
]
.
W
h
ile
d
em
o
n
s
tr
ate
th
e
win
d
r
o
tatin
g
s
p
ee
d
in
r
ad
/s
ec
.
T
h
e
m
ec
h
a
n
ical
o
u
tp
u
t
p
o
wer
o
f
win
d
is
,
=
1
2
3
(
,
)
(
1
)
=
0
.
5
(
−
0
.
022
2
−
5
.
6
)
−
0
.
17
(
2
)
=
(
3
)
3
.
1
.
M
o
dellin
g
o
f
I
P
M
SG
T
h
e
d
ev
el
o
p
ed
I
PMSG
m
o
d
el
in
g
s
y
s
tem
,
wh
ich
tak
es
in
to
co
n
s
id
er
atio
n
b
o
th
co
p
p
er
lo
s
s
an
d
co
r
e
lo
s
s
in
th
e
s
ta
to
r
,
is
s
h
o
wn
in
Fig
u
r
e
1
.
Mo
r
eo
v
er
,
co
r
e
l
o
s
s
attr
ib
u
ted
to
h
y
s
ter
esis
an
d
ed
d
y
cu
r
r
en
t
lo
s
s
es
ar
e
r
ep
r
esen
ted
b
y
an
e
q
u
iv
ale
n
t
co
r
e
-
lo
s
s
r
esis
tan
ce
.
Ad
d
itio
n
ally
,
a
p
r
o
ce
s
s
is
d
ev
elo
p
ed
to
co
m
p
u
te
t
h
e
v
alu
e
o
f
,
wh
ich
v
ar
ies lin
ea
r
l
y
with
I
PMSG r
o
to
r
s
p
ee
d
ac
co
r
d
in
g
to
th
e
m
o
d
el
in
[
2
7
]
.
=
∗
(
4
)
wh
er
e,
=
0
.
2083
(
Ω
/r
p
m
)
.
B
y
u
s
in
g
p
er
m
an
e
n
t
m
ag
n
et
s
to
cr
ea
te
t
h
e
r
o
to
r
'
s
m
ag
n
etic
f
ield
,
a
PMSG
is
a
s
y
n
ch
r
o
n
o
u
s
g
en
er
ato
r
th
at
d
o
es
n
o
t
r
e
q
u
i
r
e
an
ex
ter
n
al
DC
f
ield
.
T
h
i
s
tech
n
o
lo
g
y
o
f
f
er
s
s
ev
er
al
a
d
v
an
tag
es,
m
a
k
in
g
PMSG
a
p
r
ef
er
r
ed
c
h
o
ice
f
o
r
v
ar
io
u
s
ap
p
licatio
n
s
.
T
h
e
b
r
u
s
h
less
an
d
s
lip
r
in
g
-
f
r
ee
d
es
ig
n
o
f
th
e
PMSG
en
ab
les
a
co
m
p
ac
t
s
ize,
h
ig
h
r
eliab
ilit
y
,
an
d
r
e
d
u
ce
d
m
ec
h
a
n
ical
f
r
ictio
n
lo
s
s
es.
I
ts
p
o
wer
d
en
s
ity
,
o
r
en
er
g
y
o
u
tp
u
t p
er
u
n
it v
o
lu
m
e,
is
o
p
ti
m
ized
.
Ad
d
itio
n
ally
,
th
e
PMSGs
ar
e
m
o
r
e
ef
f
icien
t b
ec
au
s
e
th
ey
d
o
n
o
t
r
e
q
u
ir
e
r
o
to
r
p
o
wer
c
o
n
v
er
s
io
n
o
r
a
lo
s
s
,
wh
ich
in
cr
ea
s
es
th
e
g
en
e
r
ato
r
'
s
to
tal
ef
f
icien
cy
[
2
7
]
.
Ad
d
itio
n
ally
,
PMSGs
h
av
e
a
g
r
ea
ter
wo
r
k
in
g
r
an
g
e
,
wh
ich
en
ab
les
t
h
em
to
f
u
n
ctio
n
b
etter
in
lo
w
win
d
s
p
ee
d
s
it
u
atio
n
s
wh
en
o
th
er
g
en
er
ato
r
ty
p
es
ca
n
h
av
e
tr
o
u
b
le.
B
ec
au
s
e
PMSGs
d
o
n
o
t
d
ep
en
d
o
n
p
o
wer
elec
tr
o
n
i
cs
o
r
s
lip
r
in
g
s
f
o
r
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
Hyb
r
id
o
p
timiz
a
tio
n
tu
n
ed
d
e
ep
n
eu
r
a
l n
etw
o
r
k
-
b
a
s
ed
w
in
d
…
(
P
r
a
s
h
a
n
t Ku
ma
r
S
.
C
h
in
a
ma
lli
)
2603
co
n
tr
o
l,
th
ey
also
h
av
e
b
ett
er
g
r
id
co
m
p
atib
ilit
y
,
wh
ich
lo
wer
s
m
ain
ten
an
ce
co
s
ts
an
d
b
o
o
s
ts
s
y
s
tem
d
ep
en
d
a
b
ilit
y
[
2
8
]
.
Fig
u
r
e
1
.
I
PMSG sy
s
tem
m
o
d
el
T
h
e
PMSG
m
o
d
el
d
ev
elo
p
ed
in
th
is
s
tu
d
y
is
b
ased
o
n
s
ev
e
r
al
s
im
p
lify
in
g
ass
u
m
p
tio
n
s
c
o
m
m
o
n
l
y
em
p
lo
y
ed
to
r
ed
u
ce
m
o
d
el
c
o
m
p
lex
ity
[
2
9
]
,
[
3
0
]
.
First
o
f
a
ll,
a
k
e
y
ass
u
m
p
tio
n
is
th
e
ab
s
en
ce
o
f
m
ag
n
etic
cir
cu
it
s
atu
r
atio
n
.
Seco
n
d
ly
,
n
eg
lectin
g
ed
d
y
cu
r
r
en
t
lo
s
s
es
an
d
h
y
s
ter
esis
lin
ea
r
izes
th
e
f
lu
x
-
cu
r
r
e
n
t
r
elatio
n
s
h
ip
.
Ad
d
itio
n
ally
,
a
s
in
u
s
o
id
al
m
ag
n
eto
m
o
tiv
e
f
o
r
ce
(
MM
F)
d
is
tr
ib
u
tio
n
is
ass
u
m
ed
.
Giv
en
t
h
e
s
m
o
o
th
p
o
le
d
esig
n
,
r
o
to
r
d
a
m
p
in
g
is
co
n
s
id
er
ed
m
in
im
al.
T
h
e
PMSG
s
tato
r
v
o
ltag
e
eq
u
atio
n
s
ar
e
e
x
p
r
ess
ed
in
th
e
d
-
q
s
y
n
ch
r
o
n
o
u
s
r
ef
er
e
n
ce
f
r
am
e
[
3
1
]
.
=
−
ℎ
+
ℎ
(
5
)
=
−
ℎ
+
ℎ
+
(
6
)
h
er
e,
d
e
f
in
es
th
e
f
lu
x
lin
k
ag
e
o
f
PM;
an
d
s
tates
th
e
d
-
q
v
o
l
tag
es
o
f
th
e
s
tato
r
;
,
im
p
lies
th
e
d
-
q
cu
r
r
en
t
o
f
th
e
s
tato
r
;
ℎ
,
ℎ
d
ep
icts
t
h
e
in
d
u
ctan
ce
o
f
d
-
q
a
x
is
.
Ma
g
n
etic
s
atu
r
atio
n
ca
n
b
e
in
co
r
p
o
r
ated
in
to
th
e
an
aly
s
is
b
y
m
o
d
elin
g
ℎ
in
ter
m
s
o
f
.
W
h
ile
w
d
ef
in
es
a
p
o
s
itiv
e
co
n
s
tan
t
in
teg
e
r
,
t
h
e
n
o
n
lin
ea
r
it
y
in
tr
o
d
u
ce
d
b
y
th
e
I
PMSG
s
y
s
tem
's
s
u
s
ce
p
tib
ili
ty
to
m
ag
n
etic
s
atu
r
atio
n
lim
its
th
e
a
p
p
licab
ilit
y
o
f
lo
s
s
m
in
im
izatio
n
m
eth
o
d
s
an
d
lin
ea
r
co
n
tr
o
l
th
eo
r
y
.
T
o
o
v
er
c
o
m
e
th
is
co
n
s
tr
ain
t,
th
e
s
tu
d
y
e
m
p
lo
y
s
a
n
o
n
lin
ea
r
co
n
tr
o
l stra
teg
y
.
T
h
e
e
x
p
r
ess
io
n
f
o
r
th
e
elec
tr
o
m
ag
n
etic
to
r
q
u
e
is
g
iv
en
b
y
(
7
)
,
(
8
)
,
ℎ
=
ℎ
−
|
|
(
7
)
=
−
3
4
[
+
(
ℎ
−
ℎ
)
]
(
8
)
3
.
2
.
M
a
x
im
izing
wind
t
urbi
ne
o
utput
T
o
m
ax
im
ize
en
er
g
y
o
u
tp
u
t
,
th
e
win
d
t
u
r
b
in
e'
s
r
o
tatio
n
al
s
p
ee
d
is
ca
r
ef
u
lly
co
n
tr
o
lled
.
Acr
o
s
s
a
r
an
g
e
o
f
win
d
s
p
ee
d
s
,
th
e
s
y
s
tem
ca
n
m
ax
im
ize
its
ef
f
icien
cy
b
y
m
ai
n
tain
in
g
th
e
i
d
ea
l
tip
s
p
ee
d
r
atio
.
T
h
is
y
ield
s
h
ig
h
m
ec
h
an
ical
o
u
tp
u
t
en
e
r
g
y
f
r
o
m
th
e
win
d
tu
r
b
in
e.
As
in
d
icate
d
b
y
(
9
)
,
th
e
id
ea
l
I
PMSG
r
o
to
r
s
p
ee
d
is
d
ir
ec
tly
p
r
o
p
o
r
tio
n
al
to
win
d
s
p
ee
d
,
with
th
e
co
n
s
tan
t
co
n
s
tr
ain
ed
b
y
win
d
t
u
r
b
in
e
lim
itatio
n
s
[
3
2
]
.
0
=
(
9
)
3
.
3
.
Co
pp
er
a
nd
co
re
lo
s
s
re
du
ct
i
o
n in IPM
SG
T
h
er
e
ar
e
f
o
u
r
ty
p
es
o
f
PMS
G
lo
s
s
es:
m
ec
h
an
ical,
s
tr
ay
-
lo
ad
,
c
o
r
e,
a
n
d
s
tato
r
c
o
p
p
e
r
.
Of
th
ese,
s
tato
r
cu
r
r
en
t
f
u
n
d
a
m
en
tals
d
i
r
ec
tly
af
f
ec
t
an
d
co
n
t
r
o
l
o
n
ly
s
tato
r
co
p
p
er
a
n
d
c
o
r
e
lo
s
s
es.
C
o
n
s
eq
u
en
tly
,
th
is
p
ap
er
d
eter
m
in
es
th
e
I
PMSG’
s
m
ax
im
u
m
ef
f
icien
cy
p
o
in
t
t
h
r
o
u
g
h
a
n
o
f
f
lin
e
n
o
n
lin
ea
r
o
p
tim
izatio
n
p
r
o
ce
s
s
aim
ed
at
m
in
im
izin
g
co
m
b
i
n
e
d
co
p
p
er
a
n
d
co
r
e
l
o
s
s
es.
wh
e
r
ea
s
,
=
0
.
246
;
=
0
.
1764
,
s
ig
n
if
ies
th
e
DC
lo
ad
I
n
v
er
ter
an
d
Gr
id
W
in
d
T
u
r
b
in
e
Gea
r
B
o
x
IP
M
S
G
P
W
M
co
n
v
er
ter
V
a
-
+
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
25
:
2
5
9
9
-
2
6
1
5
2604
o
v
er
all
s
ig
n
als,
d
ef
in
es
th
e
o
p
tim
al
v
alu
e
with
a
lo
wer
b
o
u
n
d
o
f
-
5
0
an
d
h
ig
h
er
b
o
u
n
d
o
f
1
0
.
s
y
m
b
o
lizes
th
e
b
est
v
alu
e
with
a
h
ig
h
e
r
lim
it
o
f
3
0
an
d
lo
wer
lim
it
o
f
6
.
an
d
b
e
t
h
e
s
t
ato
r
r
esis
tan
ce
an
d
co
r
e
-
l
o
s
s
r
esis
tan
ce
.
Mo
r
eo
v
er
,
s
tates
th
e
v
is
co
u
s
d
am
p
in
g
co
ef
f
icien
t
in
Kg
s
q
u
a
r
e
m
etr
e/s;
in
d
icate
s
th
e
in
p
u
t
m
ec
h
an
ica
l
to
r
q
u
e.
W
h
er
ea
s
,
ℎ
an
d
ℎ
s
tates
th
e
co
p
p
er
as
well
as
co
r
e
lo
s
s
es
[
3
0
]
.
an
d
im
p
lies
th
e
m
ax
im
al
I
P
MSG
cu
r
r
en
t a
n
d
v
o
ltag
es.
Mi
n
im
al
lo
s
s
,
ℎ
=
ℎ
+
ℎ
(
1
0
)
Su
b
jecte
d
to
ℎ
=
1
.
5
(
2
+
2
)
(
1
1
)
=
−
−
1
(
2
+
3
+
4
−
)
(
1
2
)
=
3
+
2
−
2
2
2
4
2
(
+
)
(
1
3
)
=
∗
=
(
0
.
2083
/
60
)
∗
(
1200
/
60
)
(
1
4
)
=
∗
2
=
6
∗
(
1200
/
60
)
2
(
1
5
)
=
ℎ
ℎ
=
20
.
5822
∗
1
0
−
3
6
.
24
∗
1
0
−
3
(
1
6
)
=
3
(
(
2
∗
ℎ
2
∗
(
+
)
∗
2
∗
)
+
(
1
+
)
∗
(
ℎ
2
)
−
(
(
2
∗
2
∗
2
∗
∗
ℎ
5
∗
3
∗
∗
2
)
(
+
)
)
)
(
1
7
)
=
(
(
3
∗
ℎ
2
∗
∗
2
∗
)
+
(
3
∗
2
∗
ℎ
4
∗
2
)
(
+
)
∗
(
1
+
)
)
(
1
8
)
=
(
ℎ
2
∗
3
∗
)
+
(
ℎ
5
∗
3
∗
2
)
(
+
)
(
1
9
)
=
−
(
2
∗
2
∗
2
∗
ℎ
3
∗
∗
2
∗
)
(
2
0
)
ℎ
=
1
.
5
(
2
+
2
)
(
)
(
2
1
)
=
1
.
5
2
[
(
ℎ
+
)
2
+
(
ℎ
(
)
)
2
]
/
(
)
(
2
2
)
Fro
m
(
1
)
-
(
3
)
,
(
7
)
an
d
(
8
)
=
−
ℎ
(
)
(
2
3
)
=
+
ℎ
+
(
2
4
)
=
−
(
−
)
/
(
)
(
2
5
)
=
−
(
−
)
/
(
)
(
2
6
)
−
−
=
0
(
2
7
)
=
0
/
(
2
8
)
=
/
2
(
2
9
)
2
+
2
≤
2
(
3
0
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
Hyb
r
id
o
p
timiz
a
tio
n
tu
n
ed
d
e
ep
n
eu
r
a
l n
etw
o
r
k
-
b
a
s
ed
w
in
d
…
(
P
r
a
s
h
a
n
t Ku
ma
r
S
.
C
h
in
a
ma
lli
)
2605
2
+
2
≤
2
(
3
1
)
3
.
4
.
Da
t
a
s
et
co
llect
io
n
Fo
llo
win
g
th
e
d
ata
co
llectio
n
m
eth
o
d
o
lo
g
y
o
u
tlin
ed
in
p
ap
er
,
th
is
s
tu
d
y
em
p
lo
y
s
two
d
atasets
co
r
r
esp
o
n
d
in
g
to
r
o
to
r
s
p
ee
d
s
o
f
5
0
an
d
1
0
0
r
p
m
[
3
3
]
.
T
h
e
p
r
im
ar
y
f
o
cu
s
is
to
an
aly
ze
th
e
co
r
e
lo
s
s
es
ass
o
ciate
d
with
th
ese
r
o
to
r
s
p
ee
d
s
,
alo
n
g
with
d
eter
m
i
n
in
g
o
p
tim
al
p
ar
am
eter
s
.
T
o
ac
h
ie
v
e
th
is
,
k
ey
m
etr
ics
s
u
ch
as
q
u
ad
r
atu
r
e
cu
r
r
e
n
t
′
′
an
d
tip
s
p
ee
d
r
atio
′
′
wer
e
ca
r
ef
u
lly
r
ec
o
r
d
e
d
f
o
r
a
r
an
g
e
o
f
r
o
to
r
s
p
ee
d
s
,
p
r
o
v
id
i
n
g
a
c
o
m
p
r
e
h
en
s
iv
e
u
n
d
er
s
tan
d
in
g
o
f
p
e
r
f
o
r
m
an
ce
d
y
n
am
ics.
3
.
5
.
Da
t
a
pre
-
pro
ce
s
s
ing
Data
n
o
r
m
aliza
tio
n
is
a
p
r
ep
r
o
ce
s
s
in
g
tech
n
iq
u
e
th
at
t
r
an
s
f
o
r
m
s
n
u
m
er
ical
f
ea
t
u
r
es
in
to
a
co
m
m
o
n
s
ca
le.
T
h
is
p
r
o
ce
s
s
is
e
s
s
en
ti
al
to
p
r
o
v
id
e
f
ea
t
u
r
es
with
h
ig
h
er
v
alu
es
f
r
o
m
d
o
m
i
n
atin
g
ca
lcu
latio
n
s
.
B
y
s
tan
d
ar
d
izin
g
th
e
d
ata,
we
r
e
d
u
ce
b
ias
to
war
d
s
p
ar
ticu
lar
v
ar
iab
les
b
y
s
tan
d
ar
d
izin
g
th
e
d
ata,
g
u
ar
a
n
teein
g
th
at
ea
ch
ch
ar
ac
ter
is
tic
co
n
tr
i
b
u
tes
f
air
ly
to
th
e
a
n
aly
s
is
.
Data
n
o
r
m
aliza
tio
n
is
a
cr
itical
p
r
ep
r
o
ce
s
s
in
g
s
tep
in
th
e
d
ev
el
o
p
m
en
t
o
f
r
o
b
u
s
t
D
NN.
E
x
ten
s
iv
e
r
esear
c
h
h
as
h
i
g
h
lig
h
ted
its
r
o
le
i
n
en
h
an
cin
g
m
o
d
el
ac
cu
r
ac
y
.
A
wid
ely
em
p
lo
y
ed
n
o
r
m
aliza
ti
o
n
m
eth
o
d
is
m
i
n
-
m
ax
s
ca
lin
g
,
wh
ich
lin
ea
r
ly
tr
a
n
s
f
o
r
m
s
d
ata
to
a
s
p
ec
if
ie
d
r
an
g
e,
ty
p
ically
b
etwe
en
0
an
d
1
[
3
3
]
.
W
h
er
ea
s
,
s
im
u
lates th
e
n
o
r
m
aliza
tio
n
s
ig
n
al
an
d
in
d
icate
s
th
e
co
llectio
n
o
f
s
ig
n
als ac
r
o
s
s
all
ch
ar
g
in
g
cy
cles.
=
−
−
(
3
2
)
3
.
6
.
Dee
p neura
l net
wo
rk
A
DNN
ar
ch
itectu
r
e
co
m
p
r
i
s
es
an
in
p
u
t
lay
er
,
f
o
llo
wed
b
y
o
n
e
o
r
m
o
r
e
h
id
d
en
la
y
er
s
,
an
d
cu
lm
in
atin
g
i
n
a
n
o
u
tp
u
t
lay
er
as
d
e
p
icted
in
Fig
u
r
e
2
(
a)
[
3
4
]
.
E
ac
h
lay
er
co
n
tain
s
in
t
er
co
n
n
ec
ted
n
o
d
es,
f
o
r
m
in
g
a
h
ier
ar
ch
ical
s
tr
u
c
tu
r
e.
I
n
f
o
r
m
atio
n
p
r
o
p
a
g
ates
f
o
r
war
d
th
r
o
u
g
h
th
ese
lay
er
s
,
g
en
er
atin
g
th
e
p
r
ed
icted
tar
g
et
v
alu
es
at
th
e
o
u
tp
u
t
lay
e
r
as
in
Fig
u
r
e
2
(
b
)
.
As
s
h
o
wn
in
(
3
3
)
an
d
(
3
4
)
o
u
tlin
e
t
h
e
m
ath
em
atica
l
co
m
p
u
tatio
n
o
f
th
e
tar
g
et
v
alu
e
b
ased
o
n
in
p
u
t
v
ar
iab
les.
Her
e,
1
an
d
2
r
ep
r
esen
t
in
p
u
t
f
ea
tu
r
es,
wh
ile
1
an
d
2
d
en
o
te
h
id
d
en
lay
e
r
n
o
d
e
v
alu
es.
T
h
e
ac
tiv
atio
n
f
u
n
ctio
n
tr
an
s
f
o
r
m
s
th
ese
v
alu
es.
T
h
e
f
in
al
o
u
tp
u
t
o
r
p
r
ed
icted
v
alu
e
is
r
ep
r
esen
ted
b
y
̂
,
wh
ich
is
in
f
lu
en
ce
d
b
y
weig
h
ts
(
)
an
d
b
iases
(
)
ap
p
lied
th
r
o
u
g
h
o
u
t
th
e
n
etwo
r
k
.
T
h
e
weig
h
ted
s
u
m
o
f
th
e
in
p
u
ts
is
co
n
v
er
ted
in
to
a
n
o
n
-
li
n
ea
r
o
u
tp
u
t
v
ia
th
e
ac
tiv
atio
n
f
u
n
ctio
n
.
A
co
m
m
o
n
ly
em
p
lo
y
e
d
ac
tiv
atio
n
f
u
n
ct
io
n
in
r
eg
r
ess
io
n
m
o
d
els
is
th
e
R
ec
tifie
d
L
in
ea
r
Un
it
(
R
eL
U)
[
3
5
]
.
A
th
r
esh
o
l
d
o
f
ze
r
o
is
ap
p
lied
b
y
t
h
e
n
o
n
-
lin
ea
r
ac
tiv
atio
n
f
u
n
ctio
n
R
eL
U.
I
f
th
e
in
p
u
t
(
)
is
n
eg
ativ
e,
it r
etu
r
n
s
ze
r
o
.
Oth
er
wis
e,
it r
etu
r
n
s
th
e
in
p
u
t v
alu
e
u
n
ch
a
n
g
ed
.
1
=
(
1
11
+
2
21
+
1
)
2
=
(
2
12
+
2
22
+
2
)
}
(
3
3
)
̂
=
(
1
31
+
2
32
+
3
)
(
3
4
)
=
{
<
0
(
)
=
0
≥
0
(
)
=
(
3
5
)
(
a)
(
b
)
Fig
u
r
e
2
.
Ov
e
r
v
iew
o
f
th
e
DN
N
m
o
d
el
an
d
its
o
u
tp
u
t
(
a)
s
tr
u
ctu
r
e
o
f
t
h
e
DNN
m
o
d
el
a
n
d
(
b
)
p
r
e
d
icted
o
u
tp
u
t
o
f
th
e
DNN
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
25
:
2
5
9
9
-
2
6
1
5
2606
T
o
en
h
a
n
ce
m
o
d
el
ac
cu
r
ac
y
,
th
e
b
ac
k
p
r
o
p
a
g
atio
n
al
g
o
r
it
h
m
iter
ativ
ely
ad
j
u
s
ts
lay
er
weig
h
ts
b
y
m
in
im
izin
g
th
e
d
is
cr
ep
a
n
cy
b
etwe
en
p
r
ed
icted
a
n
d
ac
tu
al
v
alu
es
[
3
6
]
.
T
h
is
o
p
tim
izatio
n
p
r
o
ce
s
s
in
v
o
lv
es
ca
lcu
latin
g
th
e
e
r
r
o
r
f
u
n
ctio
n
'
s
g
r
ad
ien
t
u
s
in
g
g
r
ad
ien
t
d
e
s
ce
n
t,
s
u
b
s
eq
u
en
tly
u
p
d
atin
g
n
eu
r
o
n
weig
h
ts
to
r
ed
u
ce
t
h
e
o
v
er
all
e
r
r
o
r
.
T
h
e
p
r
o
p
o
s
ed
m
eth
o
d
o
lo
g
y
u
tili
ze
s
co
r
e
lo
s
s
an
d
r
o
to
r
s
p
ee
d
as
in
p
u
t
p
a
r
am
eter
s
f
o
r
a
DNN
m
o
d
el,
w
h
ile
th
e
c
o
r
r
e
s
p
o
n
d
in
g
tip
s
p
ee
d
r
atio
′
′
an
d
q
u
ad
r
atu
r
e
cu
r
r
en
t
′
′
ar
e
d
esig
n
ated
as
tar
g
et
o
u
tp
u
ts
.
T
o
o
p
tim
ize
th
e
m
o
d
el's
p
er
f
o
r
m
a
n
ce
a
n
d
m
a
x
im
i
ze
win
d
p
o
wer
g
en
er
atio
n
,
th
e
h
y
b
r
id
L
y
r
eb
ir
d
-
b
ased
co
ati
o
p
tim
izatio
n
(
L
B
-
C
OA)
alg
o
r
ith
m
is
em
p
lo
y
e
d
t
o
r
ef
in
e
r
elev
an
t m
o
d
el
p
a
r
am
eter
s
.
4.
P
ARAM
E
T
E
R
T
UNING
V
I
A
H
YB
RID
LB
-
CO
A
A
L
G
O
RIT
H
M
4
.
1
.
O
bje
ct
iv
e
f
un
ct
io
n
T
h
e
s
t
u
d
y
ai
m
e
d
t
o
o
p
tim
iz
e
PMS
G
-
b
ase
d
W
E
C
S
b
y
i
n
te
g
r
ati
n
g
d
e
ep
lea
r
n
i
n
g
a
n
d
o
p
tim
i
za
t
i
o
n
alg
o
r
it
h
m
s
.
T
h
e
wei
g
h
ts
o
f
t
h
e
D
NN
cl
ass
i
f
ie
r
a
r
e
o
p
ti
m
al
l
y
t
u
n
ed
u
s
i
n
g
a
p
r
o
p
o
s
e
d
h
y
b
r
i
d
L
B
-
C
OA.
T
h
is
alg
o
r
it
h
m
s
e
r
v
es
as
t
h
e
o
b
j
ec
ti
v
e
f
u
n
c
ti
o
n
,
e
n
s
u
r
in
g
t
h
a
t
th
e
s
o
l
u
ti
o
n
b
o
u
n
d
s
r
e
m
ai
n
wi
th
in
t
h
e
r
a
n
g
e
o
f
0
to
1
.
B
y
le
v
e
r
a
g
i
n
g
t
h
is
p
r
o
p
o
s
ed
L
B
-
C
OA
a
p
p
r
o
a
ch
,
t
h
e
s
t
u
d
y
ai
m
s
to
en
h
a
n
ce
t
h
e
ac
c
u
r
a
cy
an
d
e
f
f
ici
en
c
y
o
f
t
h
e
DNN
cl
ass
i
f
ie
r
,
d
e
m
o
n
s
tr
ati
n
g
t
h
e
e
f
f
ec
ti
v
e
n
ess
o
f
th
e
L
B
-
C
OA
i
n
o
p
ti
m
i
zi
n
g
co
m
p
l
ex
n
e
u
r
al
n
et
wo
r
k
p
a
r
a
m
e
te
r
s
.
T
h
e
ef
f
ec
ti
v
en
ess
o
f
t
h
e
o
p
ti
m
i
za
t
io
n
was
ev
al
u
ate
d
t
h
r
o
u
g
h
s
i
m
u
la
ti
o
n
s
c
o
m
p
a
r
i
n
g
th
e
s
y
s
t
em
’
s
p
e
r
f
o
r
m
a
n
c
e
wit
h
th
e
e
x
is
ti
n
g
m
o
d
els
li
k
e
DNN
-
PMS
G
,
l
y
r
e
b
i
r
d
o
p
ti
m
i
za
t
io
n
al
g
o
r
ith
m
+
d
ee
p
n
e
u
r
al
n
et
wo
r
k
s
(
L
O
A+
DN
N
)
,
co
ati
o
p
ti
m
iz
ati
o
n
a
lg
o
r
it
h
m
+
d
e
e
p
n
e
u
r
al
n
etw
o
r
k
s
(
C
OA
+D
N
N
)
,
s
a
n
d
c
at
s
w
ar
m
o
p
ti
m
iz
ati
o
n
+
d
ee
p
n
e
u
r
a
l
n
et
wo
r
k
s
(
SC
SO+D
NN
)
,
a
n
d
ze
b
r
a
o
p
t
im
iza
ti
o
n
a
lg
o
r
it
h
m
+
d
e
ep
n
e
u
r
al
n
etw
o
r
k
s
(
Z
O
A+
D
NN
)
[
3
4
]
,
[
3
7
]
–
[
4
0
]
.
4
.
2
.
Dev
elo
ped hy
brid L
B
-
CO
A
t
ec
hn
iqu
e
L
OA
[
3
7
]
is
a
p
o
p
u
latio
n
-
b
ased
m
etah
eu
r
is
tic
tech
n
iq
u
e
em
p
lo
y
in
g
a
p
o
p
u
latio
n
o
f
ar
tific
ial
ly
r
eb
ir
d
s
.
T
h
e
L
y
r
eb
ir
d
o
p
ti
m
izatio
n
alg
o
r
ith
m
(
L
OA)
c
an
b
e
ap
p
lied
to
v
ar
i
o
u
s
ty
p
es
o
f
o
p
tim
izatio
n
p
r
o
b
lem
s
,
p
ar
ticu
la
r
ly
th
o
s
e
in
v
o
lv
in
g
co
m
p
le
x
,
h
ig
h
-
d
im
e
n
s
io
n
al,
an
d
n
o
n
lin
ea
r
o
b
jectiv
e
f
u
n
ctio
n
s
.
E
ac
h
ly
r
eb
ir
d
r
ep
r
esen
ts
a
p
o
ten
ti
al
s
o
lu
tio
n
,
ch
ar
ac
ter
ize
d
b
y
its
p
o
s
itio
n
with
in
th
e
p
r
o
b
lem
d
o
m
ain
.
E
ac
h
ly
r
eb
ir
d
'
s
b
eh
av
io
r
ca
n
b
e
m
o
d
elled
as
a
p
o
in
t
i
n
d
ec
is
io
n
s
p
ac
e.
T
h
e
h
y
b
r
i
d
L
B
-
C
OA
is
an
ad
v
an
ce
d
m
o
d
if
icatio
n
o
f
th
e
b
asic
L
O
A
m
eth
o
d
.
I
t
c
o
m
b
in
es
th
e
p
r
in
cip
les
o
f
t
h
e
L
OA
with
a
n
o
th
er
o
p
tim
izatio
n
tech
n
iq
u
e,
co
ati
o
p
tim
izatio
n
alg
o
r
ith
m
(
C
OA)
,
to
e
n
h
an
ce
p
er
f
o
r
m
an
ce
.
W
h
ile
th
e
b
asi
c
L
OA
is
ef
f
ec
tiv
e
f
o
r
s
ea
r
ch
in
g
g
lo
b
al
o
p
tim
a,
it
m
ay
s
tr
u
g
g
le
with
l
o
ca
l
co
n
v
er
g
en
ce
in
ce
r
tain
co
m
p
lex
o
p
tim
izatio
n
p
r
o
b
lem
s
.
B
y
in
teg
r
atin
g
C
OA,
th
e
L
B
-
C
OA
h
y
b
r
id
im
p
r
o
v
es th
e
ex
p
lo
r
atio
n
an
d
ex
p
lo
i
tatio
n
b
alan
ce
.
4
.
3
.
L
O
A
s
t
a
t
is
t
ica
l m
o
del
T
h
e
d
ev
elo
p
ed
L
OA
tech
n
iq
u
e
d
y
n
am
ically
ad
j
u
s
ts
p
o
p
u
lat
io
n
m
em
b
er
p
o
s
itio
n
s
at
ea
ch
iter
atio
n
b
y
s
im
u
latin
g
th
e
ly
r
e
b
ir
d
'
s
b
e
h
av
io
r
in
r
esp
o
n
s
e
to
p
er
ce
iv
e
d
th
r
ea
ts
.
I
n
s
p
ir
e
d
b
y
th
e
l
y
r
e
b
ir
d
'
s
b
eh
av
io
r
,
th
e
p
o
p
u
latio
n
u
p
d
ate
in
v
o
l
v
es
two
p
r
im
ar
y
ac
tio
n
s
:
escap
in
g
an
d
h
id
in
g
.
Fig
u
r
e
3
d
en
o
t
es
th
e
f
lo
wch
ar
t
o
f
LB
-
C
OA
ap
p
r
o
ac
h
.
T
h
e
L
OA
alg
o
r
ith
m
s
im
u
lates
th
e
ly
r
eb
ir
d
'
s
s
tr
ateg
ic
ch
o
ice
b
etwe
en
f
leei
n
g
a
n
d
h
id
in
g
w
h
en
f
ac
e
d
with
d
an
g
er
,
m
o
d
eled
b
y
(
3
6
)
.
C
o
n
s
eq
u
en
tly
,
ea
ch
L
OA
p
o
s
itio
n
is
ad
ju
s
ted
in
ea
ch
iter
atio
n
d
ep
en
d
in
g
o
n
o
n
e
o
f
th
e
two
p
h
ases
.
W
h
er
e,
d
ef
in
es th
e
ar
b
itra
r
y
c
o
u
n
t w
i
th
in
th
e
r
an
g
e
[
0
,
1
]
.
Up
d
ate
p
r
o
ce
s
s
f
o
r
:
{
1
,
≤
0
.
5
2
,
(
3
6
)
Stag
e
1
: stra
teg
y
f
o
r
escap
in
g
(
ex
p
lo
r
e
d
lev
el)
Du
r
in
g
th
is
L
OA
lev
el,
p
o
p
u
latio
n
m
em
b
er
s
ar
e
r
ep
o
s
itio
n
ed
with
in
th
e
s
ea
r
c
h
in
g
s
p
ac
e
b
y
m
im
ick
in
g
th
e
ly
r
e
b
ir
d
'
s
escap
e
f
r
o
m
a
p
er
ilo
u
s
l
o
ca
tio
n
to
s
af
er
g
r
o
u
n
d
s
.
Sig
n
if
ican
t
p
o
s
itio
n
al
alter
atio
n
s
ar
e
ca
u
s
ed
b
y
th
is
d
y
n
am
ic
m
o
v
em
en
t,
wh
ich
im
p
r
o
v
es
th
e
alg
o
r
ith
m
'
s
ab
ilit
y
to
s
ea
r
ch
g
l
o
b
ally
b
y
en
a
b
lin
g
it
to
in
v
esti
g
ate
v
ar
io
u
s
ar
ea
s
o
f
th
e
s
o
lu
tio
n
s
p
ac
e.
Peo
p
le
c
h
o
o
s
e
p
o
p
u
latio
n
m
em
b
e
r
s
with
h
ig
h
e
r
o
b
jectiv
e
f
u
n
ctio
n
v
alu
es
as
p
r
ef
er
r
e
d
tar
g
et
lo
ca
tio
n
s
in
th
e
L
OA
f
r
a
m
ewo
r
k
.
T
h
ese
id
e
n
tifie
d
p
o
s
i
tio
n
s
co
n
s
titu
te
th
e
s
af
e
ar
ea
s
et
f
o
r
ea
ch
m
em
b
er
,
ca
lcu
lated
u
s
in
g
(
3
7
)
.
=
{
,
<
∈
{
1
,
2
,
.
.
,
}
(
3
7
)
wh
ile,
=
1
,
2
,
.
.
.
,
.
At
th
is
tim
e,
d
ef
i
n
es
th
e
s
af
est
ar
ea
s
et
f
o
r
ℎ
ly
r
eb
i
r
d
;
im
p
lies
th
e
m
atr
ix
with
th
r
o
w
with
an
o
b
jectiv
e
f
u
n
ctio
n
o
f
wh
ich
is
b
etter
th
an
th
e
ℎ
L
OA
m
em
b
er
.
T
h
e
L
OA
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2088
-
8
7
0
8
Hyb
r
id
o
p
timiz
a
tio
n
tu
n
ed
d
e
ep
n
eu
r
a
l n
etw
o
r
k
-
b
a
s
ed
w
in
d
…
(
P
r
a
s
h
a
n
t Ku
ma
r
S
.
C
h
in
a
ma
lli
)
2607
alg
o
r
ith
m
s
im
u
lates
a
r
an
d
o
m
escap
e
o
f
th
e
ly
r
eb
ir
d
to
o
n
e
o
f
th
e
id
en
tifie
d
s
af
e
ar
ea
s
.
A
n
ew
lo
ca
tio
n
f
o
r
ea
ch
p
o
p
u
latio
n
m
em
b
e
r
is
co
m
p
u
ted
d
e
p
en
d
in
g
o
n
th
is
s
im
u
lated
d
is
p
lace
m
en
t
ac
co
r
d
i
n
g
to
(
3
8
)
.
I
f
th
e
n
ew
p
o
s
itio
n
y
ield
s
an
im
p
r
o
v
e
d
o
b
jectiv
e
f
u
n
ctio
n
v
alu
e
,
it c
h
a
n
g
es th
e
ea
r
lier
p
o
s
itio
n
ac
co
r
d
in
g
to
(
3
9
)
.
,
1
=
,
+
,
.
(
,
−
,
∗
,
)
(
3
8
)
=
{
1
,
1
≤
,
,
(
3
9
)
I
n
th
is
co
n
tex
t,
d
en
o
tes
th
e
ch
o
s
en
s
af
e
ar
ea
f
o
r
th
e
i
th
ly
r
eb
ir
d
,
with
,
r
ep
r
esen
tin
g
its
p
th
d
im
en
s
io
n
al
co
o
r
d
in
ate.
1
s
ta
tes
th
e
n
ewly
co
m
p
u
ted
lo
c
atio
n
o
f
th
e
ℎ
ly
r
eb
ir
d
ac
co
r
d
in
g
to
th
e
p
r
o
p
o
s
ed
L
OA'
s
escap
e
m
ec
h
an
is
m
.
1
ca
lcu
lates
th
e
o
b
jectiv
e
f
u
n
ctio
n
al
v
al
u
e,
,
r
ep
r
esen
t
th
e
ar
b
itra
r
y
c
o
u
n
t a
m
o
n
g
0
an
d
1
,
wh
ile
,
d
ef
in
es th
e
r
an
d
o
m
ly
ass
ig
n
ed
v
alu
es o
f
eith
er
1
o
r
2
.
Fig
u
r
e
3
.
Flo
wch
ar
t
o
f
L
B
-
C
OA
tech
n
iq
u
e
Stag
e
2
: stra
teg
y
f
o
r
h
id
in
g
(
e
x
p
lo
ited
lev
el)
T
h
is
L
OA
p
h
ase
em
u
lates
th
e
ly
r
eb
i
r
d
'
s
s
tr
ateg
y
o
f
co
n
ce
ali
n
g
its
elf
with
in
its
im
m
ed
iate
s
af
e
zo
n
e.
B
y
m
eticu
lo
u
s
ly
ex
am
in
in
g
t
h
e
s
u
r
r
o
u
n
d
in
g
e
n
v
ir
o
n
m
en
t
an
d
m
ak
in
g
in
cr
em
en
tal
m
o
v
em
en
ts
to
war
d
s
a
S
t
a
rt
S
pe
c
i
fy
p
robl
e
m
d
e
fi
n
i
t
i
on
:
i
nput
v
a
ri
a
bl
e
s
,
opt
i
m
i
z
a
t
i
on
go
a
l
,
a
nd
l
i
m
i
t
a
t
i
ons
.
S
e
t
popul
a
c
e
s
i
z
e
a
nd
i
t
e
r
a
t
i
on
c
ou
nt
Ra
ndo
m
l
y
i
ni
t
i
a
l
i
z
e
t
h
e
popu
l
a
ce
m
a
t
r
i
x
O
bt
a
i
n
t
he
n
e
w
L
O
A
l
oc
a
t
i
on
t
hrou
gh
Eq
n
.
(
40
)
E
s
t
a
b
l
i
s
h
a
s
e
t
o
f
c
a
nd
i
d
a
t
e
s
a
fe
s
po
t
s
fo
r
t
he
p
t
h
l
yre
b
i
rd
t
hr
ough
E
q
n.
(
37
)
O
bt
a
i
n
t
he
n
e
w
L
O
A
l
oc
a
t
i
o
n
t
hrough
E
q
n
.
(
38
)
M
odi
fy
by
Eq
n.
(
41
)
M
odi
fy
by
Eq
n.
(
39
)
S
a
ve
opt
i
m
a
l
ou
t
c
o
m
e
P
ri
nt
i
d
e
a
l
s
o
l
ut
i
on
E
nd
Ye
s
No
Ye
s
No
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8
7
0
8
I
n
t J E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
15
,
No
.
3
,
J
u
n
e
20
25
:
2
5
9
9
-
2
6
1
5
2608
s
u
itab
le
h
id
in
g
s
p
o
t,
t
h
e
alg
o
r
ith
m
r
ef
in
es
its
p
o
s
itio
n
s
u
b
tl
y
.
T
h
e
b
e
h
av
io
r
o
f
th
e
p
r
o
ce
s
s
o
f
escap
in
g
f
r
o
m
p
r
ed
ato
r
s
ar
e
s
am
e
f
o
r
b
o
th
L
OA
an
d
co
ati
o
p
tim
izatio
n
(
i.e
.
,
)
ex
p
l
o
itatio
n
p
h
ase.
T
h
e
r
ef
o
r
e,
c
o
ati
[
3
8
]
is
b
est
s
u
ited
f
o
r
escap
in
g
th
r
o
u
g
h
s
o
cial
b
e
h
av
io
r
,
clim
b
in
g
a
n
d
d
e
f
en
s
iv
e
a
g
g
r
ess
io
n
.
I
t
ca
n
lev
er
a
g
e
its
s
o
cial
s
tr
u
ctu
r
e
f
o
r
in
cr
ea
s
ed
v
i
g
ilan
ce
an
d
ca
n
q
u
ick
ly
m
o
v
e
to
a
s
af
e
p
lace
if
th
r
ea
ten
ed
.
So
b
ased
o
n
th
e
ab
o
v
e
s
tr
ateg
y
th
e
e
x
p
lo
itatio
n
p
h
ase
eq
u
atio
n
o
f
c
o
ati
is
u
s
ed
in
th
e
L
OA
tech
n
i
q
u
e
n
am
ed
as
ly
r
eb
ir
d
o
p
tim
izatio
n
alg
o
r
ith
m
-
b
ased
co
ati
o
p
tim
iz
atio
n
(
L
B
-
C
OA)
alg
o
r
ith
m
.
,
2
=
,
+
(
1
−
2
)
.
(
+
.
(
−
)
)
(
4
0
)
wh
ile,
=
/
=
/
=
{
2
,
2
≤
,
,
(
4
1
)
wh
ile,
2
in
d
icate
th
e
cu
r
r
e
n
t
lo
ca
tio
n
d
ep
en
d
in
g
o
n
th
e
L
OA
h
id
in
g
s
tr
ateg
y
.
T
h
e
o
b
jectiv
e
f
u
n
ctio
n
v
alu
e
2
is
ass
es
s
ed
f
o
r
th
is
n
ew
p
o
s
itio
n
.
s
tates
th
e
ar
b
itra
r
ily
ch
o
s
en
b
o
u
n
d
with
in
[
0
,
1
]
a
n
d
b
e
th
e
iter
atio
n
co
u
n
t.
Alg
o
r
ith
m
1
d
ep
icts
th
e
p
s
eu
d
o
co
d
e
o
f
th
e
p
r
esen
ted
L
B
-
C
OA
s
tr
ateg
y
.
Alg
o
r
it
h
m
1
.
Ps
e
u
d
o
c
o
d
e
o
f
L
B
-
C
OA
Provide problem definition: input variables,
optimization goal, and limitations.
Se
t
po
pu
la
c
e
si
ze
an
d
i
te
ra
t
io
n
s
(
)
Randomly initialize the populace matrix
,
←
+
.
(
−
)
Calculate the fitness
Ascertain the most suitable option
Fo
r
=
1
Fo
r
=
1
Id
e
nt
i
fy
t
h
e
ly
re
b
ir
d'
s
d
ef
e
ns
i
ve
t
a
ct
ic
a
g
ai
ns
t
pr
e
da
t
io
n
u
ti
li
zi
n
g
Eq
n
.
(3
6
)
If
≤
0
.
5
(S
ta
ge
1)
Establish a set of candidate safe spots for the
p
th
lyrebird through Eqn. (37)
Obtain the new LOA location through Eqn.
(38)
Modify the LOA member's
position by Eqn. (39)
el
s
e
(
St
ag
e
2
)
Obtain the new LOA location through Eqn.
(40)
Modify the LOA member's position by Eqn. (41)
en
d
i
f
en
d
w
h
il
e
en
d
(
F
or
=
1
)
Sa
v
e
t
he
o
p
ti
ma
l
o
ut
co
m
e
en
d
(
F
or
=
1
)
Ou
t
pu
t
t
he
op
ti
ma
l
r
es
u
lt
w
i
th
i
n
LO
A
en
d
L
O
A
5.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
NS
5
.
1
.
E
x
perim
ent
a
l set
up
A
Simu
lin
k
m
o
d
el,
as
s
ee
n
i
n
Fig
u
r
e
4
,
was
u
s
ed
to
p
r
o
d
u
ce
d
ata
f
o
r
m
an
a
g
in
g
a
P
MSG
-
b
ased
W
E
GS
co
n
s
tr
u
cted
in
th
e
M
AT
L
AB
2
0
2
1
b
en
v
ir
o
n
m
en
t
in
o
r
d
er
to
ev
alu
ate
th
e
e
f
f
ica
cy
o
f
th
e
s
u
g
g
ested
LB
-
C
OA
DNN
tech
n
iq
u
e.
A
p
er
f
o
r
m
an
ce
co
m
p
a
r
is
o
n
was
co
n
d
u
cte
d
b
etwe
en
th
e
o
b
tain
ed
r
esu
lts
an
d
th
o
s
e
o
f
alter
n
ativ
e
a
p
p
r
o
ac
h
es
s
u
c
h
as
DNN
-
PMSG,
L
OA+
DN
N,
C
OA+
DNN,
S
C
SO+
DNN
,
an
d
Z
OA+
DNN
[
3
4
]
,
[
3
7
]
–
[
4
0
]
.
T
o
h
ig
h
lig
h
t
th
e
ef
f
icien
cy
o
f
t
h
e
s
u
g
g
es
ted
co
n
tr
o
l
m
eth
o
d
,
th
e
d
y
n
a
m
ic
b
eh
av
i
o
r
a
n
d
s
tead
y
-
s
tate
o
f
th
e
W
T
G
s
y
s
t
em
is
ex
am
in
ed
in
th
is
s
ec
tio
n
f
o
r
lear
n
i
n
g
r
ates
o
f
7
0
%,
8
0
%,
an
d
9
0
%.
A
th
o
r
o
u
g
h
ev
alu
atio
n
o
f
th
e
m
o
d
el'
s
p
er
f
o
r
m
an
ce
was
d
is
p
lay
ed
u
tili
zin
g
a
r
an
g
e
o
f
er
r
o
r
m
etr
ics,
s
u
ch
as
“m
ea
n
a
b
s
o
lu
te
er
r
o
r
(
MA
E
)
,
m
ea
n
s
q
u
ar
e
d
e
r
r
o
r
(
MSE
)
,
r
o
o
t
m
ea
n
s
q
u
ar
ed
er
r
o
r
(
R
MSE
)
,
m
ea
n
ab
s
o
lu
t
e
p
er
ce
n
tag
e
er
r
o
r
(
MA
PE)
,
m
e
an
ab
s
o
lu
te
r
elativ
e
er
r
o
r
(
M
AR
E
)
,
m
ea
n
s
q
u
ar
ed
r
elativ
e
er
r
o
r
(
MSR
E
)
,
an
d
r
o
o
t
m
ea
n
s
q
u
ar
e
d
r
elativ
e
e
r
r
o
r
(
R
MSR
E
)
,
”
wh
ich
co
n
f
i
r
m
ed
its
ac
cu
r
ac
y
.
I
n
ad
d
itio
n
,
th
e
co
n
v
er
g
en
ce
g
r
ap
h
a
n
d
th
e
co
m
p
u
tatio
n
al
ti
m
e
h
av
e
also
b
ee
n
an
aly
z
ed
in
co
m
p
ar
is
o
n
with
th
e
tr
ad
itio
n
al
m
o
d
els.
Evaluation Warning : The document was created with Spire.PDF for Python.