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
.
2857
~
2
8
7
3
I
SS
N:
2088
-
8
7
0
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijece.
v
15
i
3
.
pp
2
8
5
7
-
2
8
7
3
2857
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ec
e.
ia
esco
r
e.
co
m
A hybrid a
da
ptiv
e neuro
-
fuzzy in
fe
rence
sy
stem
and
reptile
sea
rch alg
o
rithm
mo
del f
o
r wi
nd p
o
wer f
o
recas
ting
M
o
ha
m
a
d I.
Al
-
Widy
a
n
1
,
L
a
it
h Abua
lig
a
h
2
,
G
ha
it
h M
.
J
a
ra
da
t
3
,
M
uta
s
em
K
ha
lil
Als
m
a
di
4
1
D
e
p
a
r
t
me
n
t
o
f
M
e
c
h
a
n
i
c
a
l
E
n
g
i
n
e
e
r
i
n
g
,
J
o
r
d
a
n
U
n
i
v
e
r
si
t
y
o
f
S
c
i
e
n
c
e
a
n
d
Te
c
h
n
o
l
o
g
y
,
I
r
b
i
d
,
J
o
r
d
a
n
2
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
D
e
p
a
r
t
me
n
t
,
A
l
a
l
-
B
a
y
t
U
n
i
v
e
r
s
i
t
y
,
A
l
a
l
-
B
a
y
t
U
n
i
v
e
r
s
i
t
y
,
M
a
f
r
a
q
,
J
o
r
d
a
n
3
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
I
n
f
o
r
ma
t
i
o
n
S
y
st
e
ms,
C
o
l
l
e
g
e
o
f
C
o
mp
u
t
e
r
S
c
i
e
n
c
e
s
a
n
d
I
n
f
o
r
mat
i
c
s
,
A
m
man
A
r
a
b
U
n
i
v
e
r
si
t
y
,
A
mm
a
n
,
J
o
r
d
a
n
4
D
e
p
a
r
t
me
n
t
o
f
M
a
n
a
g
e
me
n
t
I
n
f
o
r
ma
t
i
o
n
S
y
st
e
ms
,
C
o
l
l
e
g
e
o
f
A
p
p
l
i
e
d
S
t
u
d
i
e
s
a
n
d
C
o
mm
u
n
i
t
y
S
e
r
v
i
c
e
,
I
mam
A
b
d
u
l
r
a
h
m
a
n
B
i
n
F
a
i
sa
l
U
n
i
v
e
r
si
t
y
,
D
a
mm
a
m
,
S
a
u
d
i
A
r
a
b
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Feb
1
3
,
2
0
2
4
R
ev
is
ed
Dec
1
1
,
2
0
2
4
Acc
ep
ted
J
an
1
6
,
2
0
2
5
Esti
m
a
ti
n
g
t
h
e
n
u
m
b
e
r
o
f
wi
n
d
r
a
n
c
h
e
s
g
e
n
e
ra
ted
in
th
e
u
p
c
o
m
in
g
m
in
u
tes
,
h
o
u
rs,
o
r
d
a
y
s
is
t
h
e
fo
c
u
s
o
f
win
d
p
o
we
r
fo
re
c
a
stin
g
.
De
e
p
lea
rn
in
g
h
a
s
g
a
rn
e
re
d
a
l
o
t
o
f
in
tere
st
i
n
wi
n
d
c
o
n
tro
l
e
stim
a
ti
o
n
b
e
c
a
u
se
o
f
h
o
w
we
ll
th
e
y
p
e
rf
o
rm
c
las
sifica
ti
o
n
,
g
r
o
u
p
in
g
,
a
n
d
re
c
u
rre
n
c
e
.
Th
e
a
d
a
p
ti
v
e
n
e
u
ro
-
fu
z
z
y
i
n
fe
re
n
c
e
sy
ste
m
wa
s
su
c
c
e
ss
fu
ll
y
a
p
p
li
e
d
i
n
wi
n
d
p
o
we
r
f
o
re
c
a
stin
g
.
Ho
we
v
e
r,
it
s
p
e
rf
o
rm
a
n
c
e
re
li
e
s
o
n
o
p
t
ima
l
se
lec
ti
o
n
o
f
h
y
p
e
rp
a
ra
m
e
ters
.
Th
is
stu
d
y
in
tr
o
d
u
c
e
s
a
n
o
v
e
l
p
r
e
d
ictiv
e
m
o
d
e
l
b
y
in
c
o
r
p
o
ra
ti
n
g
th
e
re
p
ti
le
se
a
rc
h
a
lg
o
rit
h
m
wit
h
a
d
a
p
ti
v
e
n
e
u
ro
-
f
u
z
z
y
in
fe
re
n
c
e
s
y
ste
m
(AN
F
IS
)
f
o
r
sh
o
rt
-
term
wi
n
d
p
o
we
r
fo
re
c
a
stin
g
.
It
e
m
p
lo
y
s
re
p
t
il
e
se
a
rc
h
a
lg
o
rit
h
m
(RS
A),
k
n
o
wn
f
o
r
a
d
ju
sta
b
le
p
a
ra
m
e
ters
,
d
ise
n
tan
g
led
se
a
rc
h
,
a
n
d
c
o
n
siste
n
t
o
u
tco
m
e
s,
to
o
p
ti
m
ize
AN
F
I
S
’s
h
y
p
e
rp
a
ra
m
e
ters
.
Ad
d
it
i
o
n
a
ll
y
,
v
ia
e
x
p
lo
it
a
ti
o
n
d
u
r
in
g
trai
n
i
n
g
,
RS
A
p
e
rfo
rm
s
a
se
lec
ti
o
n
o
f
b
e
st
fe
a
t
u
re
s
in
t
h
e
d
a
tas
e
t
th
a
t
c
o
n
tr
ib
u
tes
to
t
h
e
c
las
sifica
ti
o
n
a
c
c
u
ra
c
y
o
f
AN
F
I
S
.
T
h
is
a
ims
to
e
n
h
a
n
c
e
p
re
c
isio
n
o
f
th
e
a
n
ti
c
ip
a
t
e
d
y
ield
.
Emp
lo
y
i
n
g
a
u
t
h
e
n
ti
c
wi
n
d
p
o
we
r
d
a
ta
fro
m
Jo
rd
a
n
is
u
n
d
e
rtak
e
n
t
o
e
v
a
lu
a
te
e
fficie
n
c
y
.
T
h
e
p
e
rfo
r
m
a
n
c
e
is
c
o
m
p
a
re
d
wit
h
a
lt
e
rn
a
ti
v
e
tec
h
n
i
q
u
e
s,
in
c
lu
d
i
n
g
a
rti
ficia
l
n
e
u
ra
l
n
e
two
rk
s,
ra
n
d
o
m
fo
re
sts,
a
n
d
s
u
p
p
o
rt
v
e
c
t
o
r
m
a
c
h
in
e
s.
F
in
d
in
g
s
sh
o
we
d
t
h
a
t
AN
F
IS
-
RS
A
p
e
rfo
rm
s
c
o
m
p
e
ti
ti
v
e
ly
f
o
r
t
h
e
we
ll
-
k
n
o
wn
C
h
in
e
se
b
e
n
c
h
m
a
rk
d
a
tas
e
t
(9
9
.
9
%
a
c
c
u
ra
c
y
;
0
.
9
9
R
2
;
1
0
.
5
4
M
AE;
1
1
.
6
2
RM
S
E)
a
n
d
is
m
o
r
e
ro
b
u
stly
a
c
c
u
ra
te
th
a
n
o
th
e
rs
o
v
e
r
th
e
Jo
r
d
a
n
ian
d
a
tas
e
t
(0
.
8
4
.
6
%
a
c
c
u
ra
c
y
;
0
.
9
6
R
2
;
0
.
0
9
8
M
AE;
0
.
2
0
3
RM
S
E)
.
K
ey
w
o
r
d
s
:
Ad
ap
tiv
e
n
eu
r
o
-
f
u
zz
y
in
f
er
en
c
e
s
y
s
tem
Dee
p
lear
n
in
g
E
x
p
lo
itatio
n
Pre
d
ictiv
e
m
o
d
el
R
ep
tile sear
ch
alg
o
r
ith
m
W
in
d
p
o
wer
f
o
r
ec
asti
n
g
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
:
Gh
aith
M.
J
ar
ad
at
Dep
ar
tm
e
n
t
o
f
C
o
m
p
u
t
er
Sc
ie
n
c
e
an
d
I
n
f
o
r
m
at
io
n
S
y
s
te
m
s
,
C
o
ll
eg
e
o
f
C
o
m
p
u
t
er
Sc
ie
n
c
es
a
n
d
I
n
f
o
r
m
ati
cs,
Am
m
a
n
A
r
a
b
U
n
i
v
e
r
s
it
y
1
1
9
5
3
,
Am
m
an
,
J
o
r
d
a
n
E
m
ail: g
.
jar
ad
at@
aa
u
.
ed
u
.
j
o
1.
I
NT
RO
D
UCT
I
O
N
I
n
r
ec
en
t
tim
es,
r
esear
ch
er
s
h
av
e
in
cr
ea
s
in
g
ly
f
o
c
u
s
ed
o
n
win
d
en
er
g
y
s
o
u
r
ce
s
,
attr
ac
ted
b
y
th
eir
ad
v
an
tag
es in
c
o
m
p
ar
is
o
n
t
o
o
th
er
f
o
r
m
s
o
f
en
e
r
g
y
,
s
u
c
h
as f
o
s
s
il f
u
els.
Ho
wev
er
,
a
d
r
awb
ac
k
o
f
win
d
en
er
g
y
is
its
r
elian
ce
o
n
wea
th
er
c
o
n
d
itio
n
s
f
o
r
th
e
ef
f
ec
tiv
e
o
p
er
at
io
n
o
f
win
d
f
ar
m
s
.
T
h
er
e
f
o
r
e,
ac
cu
r
ate
p
r
ed
ictio
n
o
f
win
d
p
o
wer
b
ec
o
m
es
h
i
g
h
ly
s
ig
n
if
ican
t
f
o
r
o
p
tim
izin
g
th
e
u
tili
za
tio
n
o
f
p
o
ten
tial
r
en
ewa
b
le
en
er
g
y
s
o
u
r
ce
s
.
Ma
n
y
wo
r
k
s
h
av
e
in
v
ested
a
g
r
ea
t
d
ea
l
o
f
tim
e
a
n
d
en
er
g
y
in
to
im
p
r
o
v
in
g
win
d
p
o
wer
f
o
r
ec
asti
n
g
m
eth
o
d
s
b
y
co
m
b
i
n
in
g
s
tatis
tical
an
d
p
h
y
s
ical
m
eth
o
d
s
.
Ph
y
s
ical
m
eth
o
d
s
,
lik
e
th
e
n
u
m
er
ical
wea
th
er
p
r
ed
ictio
n
m
o
d
el,
e
m
p
lo
y
r
u
le
s
r
eg
u
latin
g
atm
o
s
p
h
er
ic
b
e
h
a
v
io
r
to
a
n
ticip
ate
lo
ca
l
win
d
s
p
ee
d
an
d
d
ir
ec
tio
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
8
5
7
-
2
8
7
3
2858
b
ased
o
n
wea
th
e
r
d
ata
[
1
]
.
On
th
e
o
th
er
h
an
d
,
s
tatis
tical
m
eth
o
d
s
m
in
im
ize
th
e
d
if
f
e
r
en
ce
b
etwe
en
p
r
e
d
icted
an
d
o
b
s
er
v
ed
v
alu
es
b
y
u
til
izin
g
lar
g
e
am
o
u
n
ts
o
f
h
is
to
r
ical
d
ata
an
d
o
p
tim
izin
g
m
o
d
el
p
ar
a
m
eter
s
.
Statis
t
ical
-
b
ased
m
o
d
els
ar
e
t
h
e
b
est
c
h
o
ice
f
o
r
s
h
o
r
t
-
ter
m
win
d
p
o
wer
f
o
r
ec
asti
n
g
b
ec
au
s
e
th
ey
ar
e
less
ex
p
en
s
iv
e
to
ac
q
u
ir
e
an
d
p
r
o
ce
s
s
th
an
m
ath
em
atica
l
we
ath
er
p
r
e
d
ictio
n
d
ata
[
2
]
,
[
3
]
.
Sev
er
al
s
tatis
tical
m
eth
o
d
s
h
a
v
e
b
ee
n
s
u
g
g
ested
f
o
r
p
r
o
jectin
g
win
d
p
o
wer
in
th
e
n
ea
r
f
u
t
u
r
e.
T
im
e
s
er
ies
tech
n
iq
u
es
wer
e
th
e
f
o
cu
s
o
f
ea
r
l
y
m
eth
o
d
o
l
o
g
ies,
as
[
4
]
,
[
5
]
h
av
e
s
h
o
wn
.
Ho
wev
er
,
th
ese
a
p
p
r
o
ac
h
es
f
ac
e
s
ev
er
al
s
ig
n
if
ican
t
d
r
awb
ac
k
s
,
o
n
e
o
f
wh
ich
is
th
eir
lack
o
f
ac
c
u
r
ac
y
[
6
]
.
L
ately
,
co
n
v
en
tio
n
al
d
ee
p
lear
n
in
g
-
b
ased
m
eth
o
d
s
lik
e
ar
tifi
cial
n
eu
r
al
n
etwo
r
k
(
ANN)
[
7
]
,
d
ec
is
io
n
tr
ee
(
DT
)
[
8
]
,
ar
b
itra
r
y
tim
b
er
[
9
]
,
an
d
u
tili
za
tio
n
o
f
a
d
ap
tiv
e
n
eu
r
o
-
f
u
zz
y
in
f
e
r
en
ce
s
y
s
tem
(
ANFI
S)
in
win
d
p
o
wer
f
o
r
ec
asti
n
g
h
av
e
b
ee
n
ex
ten
s
iv
e
[
1
0
]
.
Alth
o
u
g
h
co
n
v
en
tio
n
al
d
ee
p
lear
n
i
n
g
m
et
h
o
d
s
ar
e
ty
p
ically
co
m
p
u
tatio
n
ally
s
tr
aig
h
tf
o
r
w
ar
d
,
th
eir
ca
p
ac
ity
to
ex
tr
ac
t
co
m
p
lex
f
ea
tu
r
es
an
d
atta
in
h
ig
h
p
r
ed
ictio
n
ac
cu
r
ac
y
is
ty
p
ically
co
n
s
tr
ain
ed
[
1
1
]
.
Ma
n
y
im
p
r
o
v
em
e
n
ts
h
av
e
b
ee
n
m
ad
e
in
th
e
liter
atu
r
e
[
1
2
]
b
u
t
it
was
lim
ited
to
s
o
m
e
p
er
f
o
r
m
a
n
ce
m
ea
s
u
r
es/ev
alu
atio
n
m
etr
ics
(
e.
g
.
,
p
r
ed
ictio
n
ac
cu
r
ac
y
a
n
d
/
o
r
r
o
o
t
-
m
ea
n
-
s
q
u
ar
e
er
r
o
r
(
R
MSE
)
)
.
T
h
er
e
f
o
r
e,
th
i
s
s
tu
d
y
aim
s
to
u
tili
ze
th
e
s
im
p
licity
o
f
im
p
lem
en
tin
g
th
e
A
NFI
S
m
o
d
el
wh
ile
im
p
r
o
v
in
g
th
e
f
o
r
ec
asti
n
g
ac
c
u
r
ac
y
u
s
in
g
a
p
o
wer
f
u
l
o
p
tim
i
ze
r
,
n
am
el
y
,
th
e
r
ep
tile
s
ea
r
ch
alg
o
r
ith
m
(
R
SA)
.
T
h
e
R
SA c
an
ef
f
icien
tly
ex
p
lo
it th
e
h
y
p
er
-
p
a
r
am
eter
s
o
f
th
e
ANFI
S f
o
r
an
ef
f
ec
tiv
e
f
ea
tu
r
e
s
elec
tio
n
p
r
o
ce
s
s
,
wh
ich
ev
en
tu
ally
m
ay
lead
to
a
r
o
b
u
s
t a
n
d
ac
cu
r
ate
p
er
f
o
r
m
an
ce
o
f
th
e
f
o
r
ec
asti
n
g
m
o
d
el.
T
h
e
m
o
tiv
atio
n
b
eh
i
n
d
in
tr
o
d
u
cin
g
ANFI
S
-
R
SA
in
th
is
s
tu
d
y
lies
in
th
e
d
esire
to
e
n
h
an
ce
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
ANFI
S
m
o
d
el
an
d
r
ec
tify
th
e
is
s
u
e
o
f
lo
w
ac
cu
r
ac
y
ar
is
in
g
f
r
o
m
th
e
u
n
s
tab
le
s
elec
tio
n
o
f
its
h
y
p
er
ac
tiv
e
p
ar
am
eter
s
.
T
h
is
n
ee
d
was
ad
d
r
ess
ed
b
y
n
u
m
er
o
u
s
o
p
tim
izatio
n
m
eth
o
d
s
p
r
o
p
o
s
ed
in
th
e
liter
atu
r
e.
Acc
o
r
d
in
g
to
Am
r
o
u
n
e
[
3
]
,
tr
ad
itio
n
al
ap
p
r
o
ac
h
e
s
s
u
ch
as
g
r
i
d
s
ea
r
ch
an
d
g
r
a
d
ien
t
d
escen
t
we
r
e
em
p
lo
y
ed
to
o
p
tim
ize
th
e
h
y
p
er
p
ar
am
eter
s
o
f
a
f
o
r
ec
asti
n
g
m
o
d
el,
s
u
ch
as
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM)
.
Sti
ll,
th
ey
p
r
o
v
ed
u
n
f
it
to
p
er
f
o
r
m
lar
g
e
-
s
ca
le
co
m
p
u
tatio
n
s
with
h
ig
h
p
r
ec
is
io
n
.
Ma
n
y
s
tu
d
ies
in
clu
d
in
g
[
3
]
r
ec
o
m
m
en
d
ed
d
ea
lin
g
with
t
h
is
p
r
o
b
le
m
b
y
c
o
n
s
id
er
in
g
m
etah
eu
r
is
tics
(
e.
g
.
,
n
atu
r
e
-
i
n
s
p
ir
ed
alg
o
r
ith
m
s
in
clu
d
in
g
p
a
r
ticle
s
war
m
o
p
tim
izatio
n
)
wh
ich
ar
e
m
o
r
e
s
tab
le
an
d
r
o
b
u
s
t th
e
win
d
p
o
wer
f
o
r
ec
asti
n
g
.
T
h
e
u
s
e
o
f
a
m
etah
e
u
r
is
tic
lead
s
to
th
e
o
p
tim
izatio
n
o
f
a
m
o
d
el’
s
h
y
p
er
-
p
ar
am
eter
s
to
i
n
cr
ea
s
e
th
e
f
o
r
ec
ast ac
cu
r
ac
y
.
Ho
wev
er
,
it
is
wo
r
th
n
o
tin
g
t
h
at
th
e
p
er
f
o
r
m
an
ce
o
f
s
u
ch
m
etah
eu
r
is
tics
with
m
u
ltip
le
p
ar
am
eter
s
ca
n
in
f
lu
en
ce
t
h
e
ef
f
icien
c
y
o
f
th
e
f
o
r
ec
asti
n
g
m
o
d
el.
Hen
ce
,
it is
im
p
o
r
tan
t to
s
elec
t a
m
etah
eu
r
is
tic
ca
r
ef
u
lly
co
n
s
id
er
in
g
th
e
n
u
m
b
er
o
f
p
ar
am
eter
s
(
e.
g
.
,
in
v
o
lv
ed
in
t
h
e
s
p
ac
e
ex
p
lo
r
atio
n
m
ec
h
an
is
m
)
as
wel
l
as
th
e
s
tr
u
ctu
r
e
o
f
in
ter
n
al
m
ec
h
a
n
is
m
s
(
e.
g
.
,
lo
ca
l
s
ea
r
ch
m
eth
o
d
)
.
T
h
is
m
a
y
af
f
ec
t
th
e
o
p
tim
iz
atio
n
o
f
th
e
h
y
p
e
r
-
p
ar
am
eter
s
o
f
t
h
e
m
o
d
el
(
e.
g
.
,
ANN)
.
Mo
r
eo
v
e
r
,
L
iu
et
a
l.
[
1
2
]
h
i
g
h
lig
h
ted
th
r
ee
m
ai
n
d
r
awb
ac
k
s
ass
o
ciate
d
with
s
war
m
o
p
tim
izatio
n
alg
o
r
ith
m
s
,
wh
ich
in
clu
d
e
co
n
c
er
n
s
r
elate
d
to
p
a
r
am
eter
tu
n
in
g
,
i
n
itializatio
n
p
r
o
ce
s
s
,
an
d
b
o
u
n
d
ar
y
is
s
u
es.
A
b
u
a
l
i
g
a
h
e
t
a
l
.
[
1
3
]
s
u
g
g
e
s
t
ed
a
m
o
d
e
l
t
h
a
t
e
m
p
l
o
y
s
R
S
A
f
o
r
f
i
n
e
-
t
u
n
i
n
g
t
h
e
h
y
p
e
r
p
a
r
a
m
e
t
e
r
s
o
f
t
h
e
A
N
F
I
S
m
o
d
e
l
,
a
i
m
i
n
g
t
o
e
n
h
an
c
e
i
t
s
p
r
e
d
i
ct
i
o
n
a
c
c
u
r
a
c
y
.
T
h
e
p
e
r
f
o
r
m
a
n
c
e
a
s
s
ess
m
e
n
t
o
f
th
e
d
e
v
e
l
o
p
e
d
w
i
n
d
p
o
w
e
r
f
o
r
e
c
a
s
te
r
i
n
v
o
l
v
es
t
h
e
u
t
i
l
i
z
at
i
o
n
o
f
r
e
al
-
l
i
f
e
w
i
n
d
f
a
r
m
s
i
n
J
o
r
d
a
n
a
n
d
s
t
a
n
d
a
r
d
i
z
e
d
C
h
i
n
e
s
e
b
e
n
c
h
m
a
r
k
d
a
t
a
s
et
s
,
s
u
c
h
as
H
e
x
i
C
o
r
r
i
d
o
r
.
T
h
i
s
i
s
t
h
e
n
c
o
m
p
a
r
e
d
t
o
r
e
s
u
l
t
s
o
b
t
ai
n
e
d
w
i
t
h
o
t
h
e
r
e
x
i
s
ti
n
g
m
o
d
e
l
s
o
f
f
o
r
e
c
a
s
t
i
n
g
:
d
ec
i
s
i
o
n
t
r
e
e
(
D
T
)
,
r
a
n
d
o
m
f
o
r
e
s
t
(
R
F
)
o
r
l
i
n
e
a
r
r
e
g
r
e
s
s
i
o
n
(
L
R
)
a
n
d
/
o
r
h
y
b
r
i
d
ap
p
r
o
a
c
h
e
s
.
T
h
e
p
r
im
a
r
y
g
o
al
an
d
co
n
tr
ib
u
tio
n
s
o
f
t
h
is
s
tu
d
y
ar
e
s
u
m
m
ed
u
p
as
f
o
llo
ws:
i)
Pre
s
en
t
a
n
en
h
a
n
ce
d
ANFI
S
m
o
d
el
in
c
o
r
p
o
r
atin
g
th
e
u
s
e
o
f
R
SA
f
o
r
o
p
tim
izi
n
g
ANFI
S
h
y
p
er
-
p
ar
am
ete
r
s
,
th
er
eb
y
en
h
an
ci
n
g
p
r
ed
ictio
n
ef
f
icien
c
y
;
ii)
Ap
p
ly
th
e
d
ev
elo
p
ed
ANFI
S
-
R
S
A
m
o
d
el
f
o
r
s
h
o
r
t
-
ter
m
win
d
p
o
wer
f
o
r
ec
asti
n
g
u
s
in
g
h
is
to
r
ical
d
atasets
o
b
ta
in
ed
f
r
o
m
win
d
f
ar
m
s
in
J
o
r
d
an
a
n
d
C
h
in
a
;
iii)
E
v
alu
at
e
an
d
an
aly
ze
th
e
p
r
ed
ictiv
e
p
er
f
o
r
m
an
ce
o
f
t
h
e
s
u
g
g
ested
ANFI
S
-
R
SA
m
o
d
el
u
s
in
g
m
etr
ics
s
u
ch
as
co
r
r
elatio
n
co
e
f
f
icien
t
(R
2
)
,
m
ea
n
a
b
s
o
lu
te
er
r
o
r
(
MA
E
)
,
m
ea
n
ab
s
o
lu
te
p
er
ce
n
tag
e
er
r
o
r
(
MA
PE)
,
an
d
r
o
o
t
-
m
ea
n
-
s
q
u
a
r
e
er
r
o
r
(
R
MSE
)
;
an
d
iv
)
C
o
m
p
ar
e
th
e
ac
cu
r
ac
y
o
f
th
e
p
r
o
p
o
s
ed
m
o
d
el
with
alter
n
ativ
e
tech
n
iq
u
es.
T
h
is
p
a
p
er
is
s
tr
u
ctu
r
ed
as
f
o
llo
ws.
Sectio
n
2
p
r
o
v
id
es
th
e
b
ac
k
g
r
o
u
n
d
f
o
r
th
e
ap
p
r
o
ac
h
es.
T
h
e
g
en
e
r
al
f
r
am
ewo
r
k
o
f
th
e
p
r
o
p
o
s
ed
f
o
r
ec
asti
n
g
m
o
d
el
is
p
r
esen
ted
in
s
ec
tio
n
3
.
Sectio
n
4
in
clu
d
es
co
m
p
a
r
is
o
n
s
an
d
s
im
u
latio
n
r
esu
lts
.
Fin
ally
,
s
ec
tio
n
5
co
n
cl
u
d
es w
ith
k
ey
f
in
d
in
g
s
.
2.
L
I
E
RA
T
UR
E
RE
V
I
E
W
T
h
is
s
ec
tio
n
will
h
ig
h
lig
h
t
th
e
r
ec
en
t
d
e
v
elo
p
m
e
n
ts
o
f
s
ev
er
al
f
o
r
ec
asti
n
g
m
o
d
els
with
th
eir
wea
k
n
ess
es
an
d
lim
itatio
n
s
,
wh
ich
ar
e
u
s
ed
b
y
v
ar
io
u
s
r
esear
ch
er
s
.
T
h
en
,
th
e
d
ev
elo
p
m
en
t
o
f
a
n
o
v
el
f
o
r
ec
asti
n
g
m
o
d
el
will
co
n
s
id
er
th
o
s
e
wea
k
n
ess
es
an
d
lim
itatio
n
s
in
te
r
m
s
o
f
win
d
ch
a
r
a
cter
is
tics
,
ac
cu
r
ac
y
,
an
d
s
tab
ilit
y
.
T
h
e
m
ain
p
o
in
t
o
f
s
tar
tin
g
th
is
p
r
o
ject
is
to
r
e
f
er
to
th
e
liter
atu
r
e
t
o
f
in
d
th
e
m
o
s
t
r
ec
en
t
wo
r
k
s
an
d
h
o
w
we
ca
n
s
u
p
p
o
r
t
o
u
r
ar
g
u
m
en
ts
r
e
g
ar
d
in
g
th
e
f
o
r
ec
asti
n
g
m
o
d
el
in
th
is
r
esear
c
h
.
As
we
f
o
u
n
d
,
t
h
er
e
ar
e
m
an
y
m
o
d
els
in
th
e
liter
atu
r
e.
T
h
e
m
o
d
els
h
av
e
n
o
t
r
e
ac
h
ed
th
e
b
est
ac
cu
r
ac
y
v
al
u
e
in
th
e
f
o
r
ec
asti
n
g
p
r
o
ce
s
s
,
p
ar
ticu
lar
ly
f
o
r
t
h
e
win
d
p
o
we
r
p
r
o
b
lem
.
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
A
h
yb
r
id
a
d
a
p
tive
n
eu
r
o
-
fu
z
z
y
in
feren
ce
s
ystem
a
n
d
r
ep
tile
…
(
Mo
h
a
ma
d
I
.
A
l
-
Wid
ya
n
)
2859
2
.
1
.
ANF
I
S pre
dict
io
n m
o
dels
f
o
r
wind
po
wer
f
o
re
ca
s
t
ing
Sev
er
al
r
ec
en
t
an
d
in
ter
esti
n
g
p
u
b
licatio
n
s
h
av
e
im
p
lem
en
ted
th
e
p
r
ed
ictio
n
o
f
win
d
p
o
we
r
f
o
r
ec
asti
n
g
an
d
o
th
er
r
elate
d
co
n
ce
p
ts
.
T
h
is
s
u
b
s
ec
tio
n
tack
les
o
n
ly
th
e
im
p
lem
en
tatio
n
o
f
th
e
tr
ad
itio
n
al
ANFI
S
p
r
ed
ictio
n
m
o
d
el
as
well
as
th
e
en
h
an
ce
d
v
e
r
s
io
n
s
o
f
th
e
m
o
d
el
b
y
in
c
o
r
p
o
r
ati
n
g
th
e
o
p
tim
izatio
n
p
r
o
ce
s
s
f
o
r
d
ata
d
im
e
n
s
io
n
ality
r
ed
u
ctio
n
(
a
ka
f
ea
tu
r
e
s
elec
tio
n
)
.
W
h
ich
h
as
a
g
r
ea
t
im
p
ac
t
o
n
h
ig
h
-
ac
c
u
r
ac
y
p
r
ed
ictio
n
s
;
th
at
is
o
u
r
r
esear
ch
h
as
b
ee
n
i
n
s
p
ir
ed
b
y
th
e
s
u
f
f
icien
t
co
n
ce
p
t
o
f
in
co
r
p
o
r
a
tin
g
ANFI
S
an
d
an
o
p
tim
izatio
n
alg
o
r
ith
m
.
So
m
e
ex
am
p
les f
r
o
m
th
e
liter
atu
r
e
a
r
e
th
e
f
o
llo
win
g
.
I
n
th
e
s
tu
d
ies
o
f
[
1
4
]
,
[
1
5
]
,
t
h
e
au
th
o
r
s
co
n
ce
n
tr
ated
o
n
f
o
r
ec
asti
n
g
win
d
tu
r
b
in
e
o
u
tp
u
t
p
o
wer
b
y
in
co
r
p
o
r
atin
g
win
d
s
p
ee
d
an
d
o
p
er
atio
n
al
f
ea
tu
r
es.
T
h
e
y
em
p
lo
y
ed
an
ANFI
S
m
o
d
el
with
m
u
lti
-
s
o
u
r
ce
d
ata
f
u
s
io
n
o
n
a
d
y
n
a
m
ic
win
d
o
w,
co
n
d
u
ctin
g
co
m
p
u
tatio
n
al
a
n
aly
s
es
o
n
tim
e
s
er
ies
win
d
o
w
d
atasets
d
er
iv
ed
f
r
o
m
a
win
d
f
ar
m
in
Ma
u
r
i
tan
ia
with
a
3
0
-
m
o
v
in
g
win
d
o
w.
T
h
eir
m
o
d
el
g
ain
ed
s
u
p
er
io
r
p
r
ed
ictio
n
s
co
m
p
ar
ed
to
n
o
n
p
ar
a
m
etr
ic
m
o
d
els
an
d
o
th
er
s
im
ilar
m
ac
h
in
e
an
d
d
ee
p
lear
n
in
g
te
ch
n
iq
u
es
f
r
o
m
th
e
liter
atu
r
e.
I
n
s
u
m
m
ar
y
,
th
eir
f
in
d
in
g
s
s
u
g
g
ested
th
at
th
e
p
r
ed
ictio
n
m
o
d
el
s
h
o
ws
p
r
o
m
is
e
in
ac
cu
r
ately
esti
m
atin
g
o
u
tp
u
t,
co
n
tr
i
b
u
ti
n
g
to
en
h
a
n
ce
d
s
u
s
tain
ab
ilit
y
th
r
o
u
g
h
im
p
r
o
v
ed
c
o
n
tr
o
l
o
f
win
d
tu
r
b
in
e
o
p
er
atio
n
s
.
W
ith
en
co
u
r
ag
in
g
o
u
tco
m
es,
Kay
s
al
et
a
l.
[
1
6
]
u
s
ed
th
e
c
o
n
v
en
tio
n
al,
s
tan
d
-
alo
n
e
v
er
s
io
n
o
f
ANFI
S
to
f
o
r
ec
ast
T
u
r
k
ey
'
s
elec
tr
ical
lo
ad
.
L
ao
u
af
i
et
a
l.
[
1
7
]
d
is
cu
s
s
ed
n
eu
r
o
-
f
u
zz
y
s
y
s
tem
s
,
wav
elet
p
ac
k
et
d
ec
o
m
p
o
s
itio
n
p
r
in
cip
les,
th
e
ad
v
an
tag
es
o
f
d
ata
p
r
ep
r
o
ce
s
s
in
g
,
an
d
f
o
r
ec
ast
co
m
b
in
ati
o
n
f
r
am
ew
o
r
k
s
f
o
r
win
d
p
o
wer
f
o
r
ec
asti
n
g
.
T
h
e
s
u
g
g
ested
ap
p
r
o
ac
h
is
u
tili
ze
d
to
g
u
ar
a
n
tee
f
o
r
ec
asts
f
o
r
a
ti
m
e
h
o
r
izo
n
o
f
o
n
e
h
o
u
r
ah
ea
d
o
f
tim
e,
u
s
in
g
q
u
a
r
ter
-
h
o
u
r
ly
m
ea
s
u
r
em
en
ts
o
f
win
d
p
o
we
r
g
e
n
er
atio
n
in
Fra
n
ce
as
th
e
s
o
u
r
ce
o
f
d
ata.
T
h
e
r
esu
lts
s
h
o
w
th
at
th
e
s
u
g
g
ested
m
o
d
el
is
m
o
r
e
ac
cu
r
ate
th
an
o
t
h
er
s
with
an
av
er
ag
e
MA
PE
o
f
3
.
4
0
8
%.
2
.
2
.
O
t
her
predict
io
n m
o
dels
f
o
r
wind
po
wer
f
o
re
ca
s
t
ing
Sev
er
al
r
ec
en
t
an
d
in
ter
esti
n
g
p
u
b
licatio
n
s
h
av
e
im
p
lem
en
ted
th
e
p
r
ed
ictio
n
o
f
win
d
p
o
we
r
f
o
r
ec
asti
n
g
an
d
o
t
h
er
r
elate
d
co
n
ce
p
ts
.
T
h
is
s
u
b
s
ec
tio
n
d
is
cu
s
s
e
s
o
n
ly
th
e
im
p
lem
en
t
atio
n
o
f
th
e
o
th
er
p
r
ed
ictio
n
m
o
d
els
th
an
ANFI
S
as
well
as
th
e
en
h
an
ce
d
v
er
s
io
n
s
u
s
in
g
well
-
k
n
o
wn
o
p
tim
izer
s
.
So
m
e
ex
am
p
les f
r
o
m
th
e
liter
atu
r
e
a
r
e
th
e
f
o
llo
win
g
.
Xio
n
g
et
a
l.
[
1
8
]
p
r
o
p
o
s
ed
a
h
y
b
r
id
p
r
e
d
ictio
n
m
o
d
el
co
n
s
is
tin
g
o
f
a
s
h
u
f
f
le
d
f
r
o
g
leap
in
g
alg
o
r
ith
m
with
a
b
ac
k
p
r
o
p
ag
atio
n
n
eu
r
a
l
n
etwo
r
k
f
o
r
s
h
o
r
t
-
ter
m
p
r
ed
i
ctio
n
o
f
win
d
p
o
wer
f
o
r
ec
asti
n
g
.
T
h
eir
m
o
d
el
is
b
ased
o
n
a
g
r
ad
ien
t
d
escen
t
alg
o
r
ith
m
f
o
r
s
ea
r
ch
s
p
ac
e
e
x
p
lo
r
atio
n
to
im
p
r
o
v
e
th
e
ac
cu
r
ac
y
s
co
r
e
o
f
th
e
n
eu
r
al
n
etwo
r
k
.
T
h
ey
test
ed
th
eir
p
r
e
d
ictio
n
m
o
d
el
o
n
a
C
h
in
ese
win
d
p
o
wer
f
o
r
ec
asti
n
g
d
a
taset.
T
h
eir
m
o
d
el
s
h
o
wed
a
b
etter
p
er
f
o
r
m
a
n
ce
th
an
o
t
h
er
m
o
d
els
i
n
clu
d
in
g
b
ac
k
p
r
o
p
ag
atio
n
,
g
ate
d
r
ec
u
r
r
e
n
t
u
n
it
(
GR
U)
,
a
n
d
lo
n
g
s
h
o
r
t
-
te
r
m
m
em
o
r
y
(
L
S
T
M)
in
ter
m
s
o
f
ac
cu
r
ac
y
,
s
t
ab
ilit
y
,
an
d
ef
f
icien
cy
.
Similar
wo
r
k
with
m
o
r
e
o
p
tim
izatio
n
ef
f
o
r
ts
b
ased
o
n
aq
u
ila
o
p
tim
izer
with
L
STM
was p
r
o
p
o
s
ed
b
y
Z
h
e
n
g
et
a
l.
[
1
0
]
.
Am
r
o
u
n
e
[
3
]
p
r
o
p
o
s
ed
two
h
y
b
r
id
m
o
d
els,
a
h
y
b
r
id
s
u
p
p
o
r
t
v
ec
to
r
r
eg
r
ess
io
n
g
r
ay
wo
lf
o
p
tim
izatio
n
alg
o
r
ith
m
,
an
d
a
h
y
b
r
id
s
u
p
p
o
r
t
v
ec
to
r
r
eg
r
ess
io
n
(
SVR
)
an
d
m
a
n
ta
r
ay
f
o
r
a
g
in
g
o
p
tim
izer
.
T
h
e
two
h
y
b
r
id
m
o
d
els
o
u
t
p
er
f
o
r
m
ed
s
ev
er
al
tr
ad
itio
n
al
m
o
d
el
s
an
d
o
th
er
h
y
b
r
i
d
m
o
d
els
in
clu
d
in
g
th
e
p
ar
ticle
s
war
m
o
p
tim
izatio
n
.
I
n
s
tu
d
y
[
1
9
]
,
a
h
y
b
r
id
im
p
r
o
v
ed
c
u
ck
o
o
s
ea
r
ch
alg
o
r
ith
m
(
HI
C
S)
is
p
r
o
p
o
s
ed
to
o
p
tim
ize
th
e
h
y
p
er
p
ar
am
eter
s
o
f
th
e
s
u
p
p
o
r
t
v
ec
to
r
r
eg
r
ess
io
n
m
ac
h
in
e,
em
p
lo
y
ed
f
o
r
s
h
o
r
t
-
ter
m
win
d
p
o
wer
o
u
tp
u
t
p
r
e
d
ictio
n
(
HI
C
S
-
SVR
)
.
T
h
e
s
u
g
g
ested
HI
C
S
-
SV
R
s
h
o
wed
g
r
ea
ter
ca
p
ac
ity
to
f
o
r
e
ca
s
t
s
h
o
r
t
-
ter
m
win
d
p
o
wer
o
u
t
p
u
t
in
co
m
p
ar
is
o
n
to
p
r
ev
io
u
s
SVR
v
ar
iatio
n
s
.
T
h
e
HI
C
S
-
SVR
d
o
es,
h
o
wev
er
,
h
a
v
e
s
ev
er
al
d
r
awb
ac
k
s
,
s
u
ch
as
its
in
tr
icate
s
tr
u
ctu
r
e.
Oth
er
wo
r
th
y
ap
p
r
o
ac
h
es
with
th
e
s
am
e
ad
v
a
n
tag
es
(
e.
g
.
,
s
p
ee
d
)
an
d
lim
itatio
n
s
(
e.
g
.
,
m
o
d
el
’
s
o
v
er
f
itti
n
g
)
ar
e
p
r
o
p
o
s
ed
b
y
S
h
ah
id
et
a
l.
[
2
0
]
an
d
Xia
a
n
d
W
an
g
[
2
1
]
.
All
o
f
th
ese
h
av
e
b
ee
n
ap
p
lied
t
o
th
e
Hex
i Co
r
r
id
o
r
C
h
in
ese
d
ataset
.
F
o
r
m
o
r
e
i
n
f
o
r
m
a
t
i
o
n
a
n
d
r
e
c
e
n
t
a
d
v
a
n
c
e
s
i
n
p
r
e
d
i
c
t
i
o
n
f
o
r
w
i
n
d
p
o
w
e
r
,
p
l
e
a
s
e
r
e
f
e
r
t
o
[
2
2
]
–
[
2
7
]
.
T
h
e
y
p
r
o
v
i
d
e
d
c
o
m
p
r
e
h
e
n
s
i
v
e
s
y
s
te
m
a
t
i
c
r
e
v
i
ew
s
o
f
s
t
a
t
is
ti
c
al
a
n
d
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
n
d
d
e
ep
l
e
a
r
n
i
n
g
m
e
t
h
o
d
s
f
o
r
w
i
n
d
p
o
w
e
r
f
o
r
e
c
a
s
t
i
n
g
.
O
th
e
r
u
s
e
f
u
l
r
e
a
d
i
n
g
s
o
f
wi
n
d
f
a
r
m
a
n
d
r
e
s
o
u
r
c
e
d
a
t
a
s
e
ts
a
r
e
f
o
u
n
d
i
n
[
2
8
]
,
[
2
9
]
.
3.
M
E
T
H
O
D
3
.
1
.
Da
t
a
c
o
llect
io
n
I
n
th
is
r
esear
ch
,
th
e
p
r
o
p
o
s
ed
f
o
r
ec
asti
n
g
m
o
d
el
will
b
e
test
ed
o
n
a
b
en
ch
m
ar
k
d
ataset
f
o
r
th
e
win
d
p
o
wer
f
o
r
ec
asti
n
g
p
r
o
b
lem
to
p
r
o
v
e
its
ab
ilit
y
.
Mo
r
eo
v
er
,
th
e
p
r
o
p
o
s
ed
m
o
d
el
will
b
e
f
u
r
t
h
er
ap
p
lied
to
a
r
ea
l
d
ataset
f
r
o
m
d
if
f
er
en
t
s
o
u
r
c
es
to
illu
s
tr
ate
its
ef
f
ec
tiv
en
ess
in
r
ea
l
p
o
wer
g
en
er
atio
n
.
T
o
en
s
u
r
e
b
etter
g
en
er
aliza
tio
n
,
a
cr
o
s
s
-
v
alid
at
io
n
ap
p
r
o
ac
h
is
ap
p
lied
.
T
h
e
i
n
p
u
t
d
ata
is
r
an
d
o
m
l
y
d
iv
id
ed
in
to
two
p
o
r
tio
n
s
f
o
r
th
e
cr
o
s
s
-
v
alid
atio
n
p
r
o
ce
s
s
:
tr
ain
in
g
an
d
v
alid
atio
n
(
test
in
g
)
.
T
h
e
d
ata'
s
tr
ain
in
g
p
o
r
tio
n
is
u
s
ed
to
tr
ain
th
e
n
etwo
r
k
.
T
h
e
v
alid
atio
n
d
a
ta
co
llectio
n
au
th
en
ticates th
e
n
etwo
r
k
an
d
co
n
tr
o
ls
n
etwo
r
k
d
esig
n
s
ettin
g
s
.
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
8
5
7
-
2
8
7
3
2860
3
.
2
.
Ada
ptiv
e
neuro
-
f
uzzy
i
nfe
re
nce
s
y
s
t
em
T
h
e
ANFI
S
m
o
d
el
is
a
c
o
m
b
in
atio
n
o
f
n
e
u
r
al
n
etwo
r
k
s
an
d
f
u
zz
y
s
y
s
tem
s
,
wh
ich
in
clu
d
es
tim
e
s
er
ies
p
r
ed
ictio
n
an
d
f
o
r
ec
asti
n
g
.
T
h
is
m
o
d
el
is
c
o
n
s
id
er
ed
th
e
b
est
a
n
d
m
o
s
t
ef
f
icien
t
f
o
r
ec
asti
n
g
m
o
d
el
in
th
e
liter
atu
r
e
[
1
0
]
.
T
h
e
"I
F
-
T
HE
N"
r
u
les
ar
e
u
s
ed
in
th
e
f
u
n
d
am
e
n
tal
f
r
am
ewo
r
k
o
f
ANFI
S
to
g
en
er
ate
m
ap
p
in
g
in
p
u
ts
an
d
o
u
tp
u
ts
.
Fu
zz
y
v
ar
iab
les
at
th
e
m
o
d
el's
in
p
u
t
an
d
o
u
tp
u
t
ar
e
n
ec
ess
ar
y
f
o
r
cr
ea
tin
g
an
ANFI
S
m
o
d
el
f
o
r
a
d
y
n
am
ic
p
r
o
ce
s
s
,
as
Fig
u
r
e
1
illu
s
tr
ates.
T
h
r
o
u
g
h
th
e
f
u
zz
if
icatio
n
p
r
o
ce
s
s
,
th
e
f
u
zz
y
v
ar
iab
les
tak
e
a
d
v
an
ta
g
e
o
f
p
ast
k
n
o
wled
g
e
ab
o
u
t
th
e
p
h
y
s
ical
b
eh
av
io
r
s
o
f
th
e
s
y
s
tem
.
W
h
er
e
L1
i
is
t
h
e
n
o
d
e,
I
o
u
t
p
u
t a
n
d
x
an
d
y
ar
e
th
e
lay
er
1
i
n
p
u
ts
.
Fig
u
r
e
1
.
T
h
e
o
r
i
g
in
al
ANFI
S st
r
u
ctu
r
e
[
1
0
]
T
h
e
m
ain
m
o
d
el
o
f
ANFI
S is
d
escr
ib
ed
as
(
1
)
-
(
4
)
:
1
=
µ
(
)
(
1
)
=
1
,
2
,
…
.
,
1
=
µ
−
2
(
2
)
=
3
,
4
(
3
)
µ
(
)
=
(
(
−
i
)
/
i
)
2
(
4
)
Her
e,
A
i
a
n
d
B
i
ar
e
t
h
e
m
em
b
e
r
s
h
ip
v
alu
es
o
f
µ,
an
d
r
e
f
er
s
t
o
th
e
g
en
er
alize
d
Gau
s
s
ian
m
em
b
er
s
h
ip
f
u
n
ctio
n
.
Fu
r
th
er
m
o
r
e
,
th
e
p
r
em
is
e
p
ar
a
m
eter
s
et
is
d
en
o
ted
b
y
a
i
a
n
d
p
i
.
T
h
e
r
esu
lt
o
f
l
ay
er
2
is
ca
lc
u
lated
as
(
5
)
.
2
=
µ
(
)
∗
µ
−
2
(
)
(
5
)
T
h
e
r
esu
lt o
f
l
a
y
er
3
is
ca
lcu
la
ted
as
(
6
)
.
3
=
=
(
/
(
)
)
(
6
)
Af
ter
ch
o
o
s
in
g
th
e
in
p
u
t
a
n
d
o
u
tp
u
t
v
a
r
iab
les,
as
well
as
th
e
co
llectio
n
o
f
r
elate
d
d
ata,
th
e
in
itial
s
tep
in
d
ev
elo
p
in
g
th
e
m
o
d
el
is
to
cr
ea
te
a
s
et
o
f
p
r
im
ar
y
m
e
m
b
er
s
b
y
ANFI
S.
Fo
r
m
o
r
e
in
f
o
r
m
atio
n
o
n
ANFI
S
s
tr
u
ctu
r
e
an
d
f
o
r
m
u
latio
n
s
p
le
ase
r
ef
er
to
[
3
0
]
.
As
was
p
r
ev
io
u
s
ly
m
en
tio
n
e
d
,
ANFI
S
is
co
m
p
o
s
ed
o
f
m
u
ltip
le
lay
er
s
,
with
n
o
d
es
in
ea
ch
lay
er
ca
r
r
y
in
g
f
o
r
wa
r
d
f
u
zz
y
p
ar
am
eter
s
.
Fu
zz
y
in
f
e
r
en
ce
s
y
s
tem
s
(
FIS)
with
d
is
tr
ib
u
ted
p
ar
am
ete
r
s
ar
e
co
m
p
ar
ab
le
to
th
is
.
I
t
em
p
lo
y
s
a
b
ac
k
p
r
o
p
ag
atio
n
tech
n
iq
u
e
to
ad
ju
s
t
f
u
zz
y
p
a
r
am
eter
s
an
d
s
p
lits
p
r
ev
io
u
s
in
f
o
r
m
atio
n
in
to
s
u
b
ca
teg
o
r
ie
s
to
m
in
im
ize
s
ea
r
ch
s
p
ac
e.
An
ad
ap
tiv
e
n
e
u
r
al
n
etwo
r
k
with
a
lin
ea
r
in
p
u
t
-
o
u
tp
u
t
r
elatio
n
s
h
ip
is
th
e
f
in
al
p
r
ed
ictiv
e
m
o
d
el.
T
h
e
AN
FIS
s
tr
u
ctu
r
e
ty
p
ically
co
n
s
is
ts
o
f
f
iv
e
d
is
tin
c
t
lay
er
s
:
a.
L
ay
er
1
in
v
o
lv
es
f
u
zz
if
icatio
n
,
wh
er
ein
in
p
u
ts
ar
e
class
if
ied
u
s
in
g
ad
j
u
s
tab
le
f
u
zz
y
m
e
m
b
e
r
s
h
ip
f
u
n
ctio
n
s
(
MFs)
s
u
ch
as
Gau
s
s
ian
-
s
h
ap
ed
o
n
es.
T
o
b
u
ild
th
e
ANFI
S
m
o
d
el,
th
e
in
p
u
t
an
d
o
u
tp
u
t
d
ata
ar
e
p
ar
titi
o
n
ed
in
to
r
u
le
p
atc
h
es
u
s
in
g
f
u
zz
y
c
-
m
ea
n
(
FC
M)
.
FC
M,
a
wid
ely
u
s
ed
f
u
zz
y
l
o
g
ic
clu
s
ter
in
g
m
eth
o
d
,
p
ar
titi
o
n
s
in
p
u
t
d
ata
in
to
clu
s
ter
s
ess
en
tial
f
o
r
f
u
zz
y
lo
g
ic.
I
t
d
eter
m
in
es
th
e
ce
n
ter
o
f
ea
c
h
clu
s
ter
,
f
ac
ilit
atin
g
s
u
b
s
eq
u
en
t
o
p
er
atio
n
s
o
n
th
e
d
ata.
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
A
h
yb
r
id
a
d
a
p
tive
n
eu
r
o
-
fu
z
z
y
in
feren
ce
s
ystem
a
n
d
r
ep
tile
…
(
Mo
h
a
ma
d
I
.
A
l
-
Wid
ya
n
)
2861
b.
L
ay
er
2
in
v
o
l
v
es
r
u
les
,
wh
er
e
th
e
r
u
le
o
p
er
ato
r
(
AND/OR)
is
em
p
lo
y
ed
to
o
b
tain
a
s
in
g
le
o
u
tp
u
t
r
ep
r
esen
tin
g
t
h
e
an
tece
d
e
n
t
r
e
s
u
lts
f
o
r
a
f
u
zz
y
r
u
le
b
y
m
u
lti
p
ly
in
g
th
e
in
co
m
i
n
g
s
ig
n
als.
I
n
ess
en
ce
,
th
is
lay
er
d
eter
m
in
es th
e
weig
h
ts
o
f
th
e
MFs.
c.
L
ay
er
3
k
n
o
wn
as
n
o
r
m
aliza
ti
o
n
,
s
er
v
es
as
th
e
in
ter
p
o
latio
n
lay
er
,
i
n
d
icatin
g
t
h
e
f
ir
in
g
s
tr
en
g
th
o
f
a
r
u
le
r
elativ
e
to
th
e
to
tal
f
ir
i
n
g
s
tr
en
g
th
o
f
all
r
u
les.
d.
L
ay
er
4
d
ef
u
zz
if
icatio
n
,
ca
r
r
ie
s
o
u
t
th
e
ex
ec
u
tio
n
o
f
th
e
o
u
t
p
u
t
g
en
e
r
ated
f
r
o
m
r
u
le
in
f
er
en
ce
,
m
u
ltip
ly
i
n
g
th
em
with
th
e
Su
g
e
n
o
f
u
zz
y
r
u
le'
s
f
u
n
ctio
n
.
e.
L
ay
er
5
c
u
m
u
lativ
e,
in
v
o
lv
es
th
e
ap
p
licatio
n
o
f
th
e
weig
h
te
d
av
er
a
g
e
s
u
m
m
atio
n
m
et
h
o
d
f
o
r
d
eter
m
in
in
g
th
e
n
etwo
r
k
o
u
tp
u
t.
3
.
3
.
Rept
ile
s
ea
rc
h a
lg
o
ri
t
hm
Ab
u
alig
ah
et
a
l.
[
1
3
]
in
tr
o
d
u
c
ed
th
e
R
SA,
a
r
ev
o
lu
tio
n
ar
y
n
atu
r
e
-
in
s
p
ir
e
d
m
eta
-
h
eu
r
is
tic
o
p
tim
ize
r
in
s
p
ir
ed
b
y
cr
o
c
o
d
ile
h
u
n
tin
g
b
eh
av
io
r
as
s
h
o
wn
in
Fig
u
r
e
2
.
Usi
n
g
well
-
k
n
o
wn
b
en
c
h
m
ar
k
d
atasets
th
at
in
clu
d
ed
class
ical
f
u
n
ctio
n
s
a
n
d
ac
tu
al
en
g
i
n
ee
r
in
g
ch
allen
g
es,
th
e
R
SA
was
ass
ess
ed
.
T
h
e
ass
ess
m
en
t
s
h
o
ws
a
b
etter
p
er
f
o
r
m
a
n
ce
th
a
n
v
a
r
io
u
s
o
p
tim
izatio
n
alg
o
r
ith
m
s
in
th
e
liter
atu
r
e
.
As
all
m
eta
-
h
eu
r
is
tics
ar
e
k
n
o
w
n
to
b
e
v
u
ln
e
r
ab
le
t
o
ea
r
l
y
c
o
n
v
er
g
en
ce
t
o
war
d
s
lo
ca
l
o
p
tim
a,
th
e
R
SA
h
as
u
n
d
er
g
o
n
e
a
v
ar
iety
o
f
im
p
r
o
v
em
e
n
ts
ac
r
o
s
s
d
if
f
er
en
t
o
p
tim
izatio
n
p
r
o
b
lem
s
b
y
h
y
b
r
id
izatio
n
,
m
o
d
if
icatio
n
,
o
r
p
ar
am
eter
-
tu
r
n
i
n
g
ap
p
r
o
ac
h
es.
So
m
e
ex
am
p
les o
f
r
ec
en
t im
p
r
o
v
em
e
n
ts
ar
e:
a.
Hy
b
r
id
R
SA a
n
d
s
lap
s
war
m
a
lg
o
r
ith
m
[
3
1
]
,
[
3
2
]
f
o
r
m
ed
ica
l im
ag
e
s
eg
m
en
tatio
n
.
b.
E
n
h
a
n
c
e
d
R
S
A
u
ti
l
iz
i
n
g
a
m
u
lt
i
-
h
u
n
t
i
n
g
c
o
o
r
d
i
n
a
t
i
o
n
s
t
r
at
e
g
y
f
o
r
a
d
d
r
e
s
s
i
n
g
g
l
o
b
a
l
o
p
ti
m
iz
a
t
i
o
n
p
r
o
b
l
e
m
s
[
3
3
]
.
c.
I
m
p
r
o
v
ed
R
SA f
o
r
r
e
g
r
ess
io
n
an
d
class
if
icatio
n
p
r
o
b
lem
s
[
3
4
]
,
[
3
5
]
.
d.
B
o
o
s
ted
R
SA f
o
r
en
g
in
ee
r
in
g
p
r
o
b
lem
s
[
3
6
]
.
e.
C
h
ao
tic
b
in
ar
y
R
SA a
n
d
a
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
f
o
r
class
if
icatio
n
p
r
o
b
lem
s
[
3
7
]
.
f.
I
m
p
r
o
v
ed
R
SA f
o
r
in
d
u
s
tr
ial
e
n
g
in
ee
r
in
g
p
r
o
b
lem
s
[
3
8
]
.
g.
E
n
h
an
ce
d
R
SA w
ith
s
im
u
late
d
an
n
ea
lin
g
f
o
r
f
ea
tu
r
e
s
elec
tio
n
in
th
e
m
ed
ical
f
ield
[
3
9
]
.
h.
R
SA
an
d
r
em
o
r
a
o
p
tim
izatio
n
alg
o
r
ith
m
with
q
u
an
tu
m
m
u
tatio
n
f
o
r
s
o
lv
in
g
d
ata
clu
s
t
er
in
g
p
r
o
b
lem
s
[
4
0
]
.
i.
Mo
d
if
ied
R
SA f
o
r
n
u
m
er
ical
o
p
tim
izatio
n
p
r
o
b
lem
s
[
4
1
]
.
j.
L
ev
y
f
lig
h
t
-
b
ased
R
SA w
ith
lo
ca
l sear
ch
f
o
r
p
o
wer
s
y
s
tem
s
en
g
in
ee
r
in
g
d
esig
n
[
4
2
]
.
C
h
au
h
an
et
a
l.
[
4
3
]
,
[
4
4
]
h
av
e
co
n
d
u
cted
o
t
h
er
a
p
p
lica
tio
n
s
o
f
th
e
t
r
ad
itio
n
al
R
SA
s
u
ch
as
o
p
tim
izatio
n
an
d
p
r
ed
ictio
n
.
T
h
e
R
SA
is
u
tili
ze
d
as
an
en
h
a
n
ce
m
en
t
s
tr
ateg
y
f
o
r
o
th
er
m
e
tah
eu
r
is
tics
s
u
ch
as
th
e
h
y
b
r
id
izatio
n
o
f
a
n
t
co
lo
n
y
o
p
tim
izatio
n
with
R
SA
f
o
r
C
h
u
r
n
p
r
ed
ictio
n
co
n
d
u
cte
d
b
y
Al
-
Sh
o
u
r
b
aji
et
a
l.
[
4
5
]
.
Mo
s
t
im
p
o
r
tan
tly
,
th
e
R
SA
h
as
b
ee
n
u
s
ed
to
e
n
h
an
ce
th
e
ad
a
p
tiv
e
n
eu
r
o
-
f
u
zz
y
i
n
f
er
en
ce
s
y
s
tem
f
o
r
r
elatin
g
s
wellin
g
p
o
ten
tiality
co
n
d
u
cte
d
b
y
E
l
Sh
in
awi
et
a
l.
[
3
0
]
wh
ic
h
also
tr
ig
g
er
ed
o
u
r
m
o
tiv
atio
n
to
ad
o
p
t th
eir
a
p
p
r
o
ac
h
f
o
r
th
e
win
d
p
o
wer
f
o
r
ec
asti
n
g
p
r
o
b
le
m
.
As
m
en
tio
n
ed
ea
r
lier
,
th
e
R
SA
is
a
m
etah
eu
r
is
tic
in
s
p
ir
ed
b
y
th
e
m
o
v
e
m
en
t
p
atter
n
s
o
f
r
ep
tiles
in
s
ea
r
ch
o
f
f
o
o
d
.
Gen
er
ally
,
th
e
R
SA b
eg
in
s
b
y
r
an
d
o
m
ly
in
iti
alizin
g
a
p
o
p
u
latio
n
o
f
s
o
lu
tio
n
s
,
r
ep
r
esen
tin
g
t
h
e
in
d
iv
id
u
als
o
r
p
ar
am
eter
s
o
f
a
p
r
o
b
lem
.
E
ac
h
s
o
lu
tio
n
is
t
h
en
ev
alu
ated
b
ased
o
n
t
h
e
p
r
o
b
lem
’
s
o
b
jectiv
e
f
u
n
ctio
n
.
T
h
e
R
SA
th
en
ap
p
lies
a
s
er
ie
s
o
f
iter
ativ
e
u
p
d
ates
to
th
e
s
o
lu
tio
n
s
,
wh
ich
ar
e
b
ased
o
n
th
e
m
o
v
em
en
t
p
atter
n
s
o
f
r
ep
tiles
.
T
h
e
R
SA
h
as
s
ev
er
al
ad
v
an
tag
es,
m
ak
in
g
it
a
p
r
o
m
is
in
g
o
p
tim
izatio
n
alg
o
r
ith
m
f
o
r
s
o
lv
in
g
v
ar
i
o
u
s
o
p
tim
izatio
n
p
r
o
b
lem
s
an
d
m
a
ch
in
e
lear
n
in
g
ap
p
licatio
n
s
.
So
m
e
o
f
t
h
ese
ar
e:
a.
E
f
f
icien
cy
:
it
is
s
im
p
le
a
n
d
c
o
m
p
u
tatio
n
ally
ef
f
icien
t,
wh
ich
r
eq
u
ir
es
o
n
l
y
b
asic
m
ath
em
at
ical
o
p
er
atio
n
s
.
T
h
is
m
ak
es
it
s
u
itab
le
f
o
r
s
o
l
v
in
g
lar
g
e
-
s
ca
le
o
p
tim
izatio
n
p
r
o
b
lem
s
with
m
a
n
y
d
ec
is
io
n
v
ar
iab
les
(
e.
g
.
,
f
ea
tu
r
e
s
elec
tio
n
in
class
if
icatio
n
task
s
)
.
b.
Div
er
s
if
icatio
n
:
it
in
co
r
p
o
r
ate
s
a
m
ec
h
an
is
m
f
o
r
d
i
v
er
s
if
y
i
n
g
th
e
s
ea
r
ch
p
r
o
ce
s
s
b
y
in
tr
o
d
u
cin
g
r
an
d
o
m
p
er
tu
r
b
atio
n
s
to
a
s
u
b
s
et
o
f
s
o
lu
tio
n
s
.
T
h
is
h
elp
s
av
o
id
s
tag
n
atio
n
in
l
o
ca
l
o
p
tim
a
an
d
e
n
h
an
ce
s
its
g
lo
b
al
s
ea
r
ch
ca
p
ab
ilit
ies.
c.
C
o
n
v
er
g
en
ce
:
it
h
as
s
h
o
wn
to
co
n
v
er
g
e
to
h
ig
h
-
q
u
ality
s
o
l
u
tio
n
s
co
n
v
er
g
e
q
u
ic
k
ly
an
d
ef
f
ec
tiv
ely
,
ev
en
f
o
r
co
m
p
lex
an
d
m
u
ltimo
d
al
o
p
tim
izatio
n
p
r
o
b
lem
s
.
T
h
is
i
s
d
u
e
to
its
ab
ilit
y
to
b
alan
ce
ex
p
lo
itin
g
th
e
b
est
-
p
er
f
o
r
m
in
g
s
o
lu
tio
n
s
an
d
ex
p
lo
r
in
g
th
e
s
ea
r
ch
s
p
ac
e
.
d.
R
o
b
u
s
tn
ess
:
it
i
s
r
o
b
u
s
t
to
n
o
i
s
y
an
d
u
n
ce
r
tai
n
o
b
jectiv
e
f
u
n
ctio
n
s
an
d
n
o
n
-
d
i
f
f
er
en
tiab
le
f
u
n
ctio
n
s
.
T
h
is
m
ak
es it su
itab
le
f
o
r
s
o
lv
in
g
r
ea
l
-
wo
r
ld
ap
p
licatio
n
s
th
at
o
f
t
en
in
v
o
l
v
e
u
n
ce
r
tain
ties
an
d
n
o
is
e.
e.
Flex
ib
ilit
y
:
it
ca
n
b
e
ea
s
ily
ad
ap
ted
to
d
if
f
er
en
t
o
p
tim
izatio
n
p
r
o
b
lem
s
b
y
ad
j
u
s
tin
g
its
p
ar
am
eter
s
ettin
g
s
.
T
h
is
allo
ws
it
to
b
e
a
p
p
lied
t
o
v
ar
io
u
s
o
p
tim
izatio
n
p
r
o
b
lem
s
in
v
ar
io
u
s
d
o
m
ai
n
s
(
e.
g
.
,
m
ac
h
in
e
lear
n
i
n
g
p
r
ed
ictio
n
m
o
d
els).
Ho
wev
er
,
s
o
m
e
d
r
awb
ac
k
s
o
f
th
e
R
SA
ar
e
also
co
n
ce
r
n
ed
with
im
p
r
o
v
em
en
ts
th
at
m
a
y
af
f
ec
t
its
p
er
f
o
r
m
an
ce
.
T
h
e
m
ain
c
o
n
ce
r
n
is
p
r
em
atu
r
e
co
n
v
e
r
g
en
ce
.
T
h
e
R
SA
r
elian
ce
o
n
th
e
lead
er
’
s
m
o
v
em
en
t
m
a
y
ca
u
s
e
th
e
alg
o
r
ith
m
to
co
n
v
er
g
e
p
r
em
atu
r
ely
to
a
lo
ca
l
o
p
tim
u
m
,
esp
ec
ially
if
th
e
lead
er
’
s
m
o
v
em
e
n
t
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
8
5
7
-
2
8
7
3
2862
b
ec
o
m
es
s
tag
n
an
t.
T
h
is
m
ay
a
f
f
ec
t
th
e
alg
o
r
ith
m
’
s
ab
ilit
y
to
ex
p
lo
r
e
th
e
s
ea
r
ch
s
p
ac
e
ef
f
e
ctiv
ely
an
d
f
in
d
th
e
g
lo
b
al
o
p
tim
u
m
.
I
n
tech
n
ical
te
r
m
s
,
th
e
R
SA
is
elu
cid
ated
c
o
n
ce
r
n
in
g
its
ex
p
lo
r
atio
n
o
f
th
e
s
ea
r
ch
s
p
ac
e
(
r
ef
er
r
ed
t
o
as
g
lo
b
al
s
ea
r
ch
)
an
d
ex
p
lo
itat
io
n
o
f
th
e
s
o
lu
tio
n
s
p
ac
e
(
r
ef
e
r
r
ed
to
as
lo
ca
l
s
ea
r
ch
)
p
h
ases
.
T
h
ese
p
h
ases
tak
e
in
s
p
ir
atio
n
f
r
o
m
th
e
s
o
cial
b
eh
av
io
r
,
s
u
r
r
o
u
n
d
in
g
m
ec
h
a
n
ics,
an
d
h
u
n
tin
g
tech
n
iq
u
es
o
b
s
er
v
ed
in
r
ea
l
cr
o
co
d
iles
[
1
3
]
.
T
h
e
two
p
r
i
m
ar
y
s
tag
es
o
f
c
r
o
co
d
ile
ac
t
iv
ity
in
v
o
lv
e
e
n
cir
clin
g
(
ex
p
lo
r
atio
n
)
,
ac
h
iev
e
d
th
r
o
u
g
h
h
ig
h
walk
i
n
g
o
r
b
elly
walk
in
g
,
an
d
h
u
n
tin
g
(
ex
p
lo
itatio
n
)
,
ac
co
m
p
lis
h
ed
th
r
o
u
g
h
h
u
n
tin
g
co
o
p
er
atio
n
o
r
co
o
r
d
in
atio
n
.
Acc
o
r
d
in
g
to
A
b
u
alig
ah
et
a
l.
[
1
3
]
,
d
u
r
in
g
th
e
e
n
cir
clin
g
p
h
ase,
wh
ich
is
p
ar
t
o
f
th
e
R
SA
'
s
ex
p
lo
r
ato
r
y
b
e
h
av
io
r
,
cr
o
co
d
i
les
walk
in
two
d
if
f
e
r
en
t
wa
y
s
wh
ile
en
cir
clin
g
a
n
d
u
p
d
a
tin
g
th
eir
p
o
s
itio
n
:
h
ig
h
walk
in
g
an
d
b
elly
walk
in
g
.
T
h
e
R
SA
alter
n
ates
b
et
wee
n
ex
p
lo
r
atio
n
an
d
ex
p
l
o
itatio
n
s
ea
r
ch
s
tag
es
b
ased
o
n
d
if
f
e
r
en
t
s
ce
n
ar
io
s
,
s
u
ch
as
d
iv
id
in
g
th
e
to
tal
n
u
m
b
er
o
f
iter
atio
n
s
in
to
f
o
u
r
p
ar
ts
o
r
s
p
litt
in
g
th
e
n
u
m
b
er
o
f
iter
atio
n
s
in
to
f
o
u
r
s
eg
m
e
n
ts
.
T
h
e
e
x
p
lo
r
ati
o
n
m
ec
h
an
is
m
s
,
g
u
id
e
d
b
y
two
m
ain
s
ea
r
c
h
ap
p
r
o
ac
h
es,
f
o
cu
s
o
n
th
e
s
ea
r
ch
s
p
ac
e
an
d
m
eth
o
d
o
lo
g
y
to
id
e
n
tify
im
p
r
o
v
e
d
ca
n
d
id
ate
s
o
lu
tio
n
s
.
T
h
r
o
u
g
h
o
u
t
th
is
s
ea
r
ch
p
h
ase,
a
s
p
ec
if
ic
r
eq
u
ir
em
en
t
m
u
s
t
b
e
m
et.
On
th
e
o
th
er
h
a
n
d
,
th
e
h
u
n
tin
g
p
h
ase
in
v
o
lv
es
s
ea
r
ch
i
n
g
a
r
o
u
n
d
th
e
s
o
lu
tio
n
’
s
s
p
ac
e
wh
er
e
cr
o
co
d
iles
u
s
e
two
h
u
n
tin
g
tech
n
iq
u
es
wh
ile
u
p
d
atin
g
th
e
p
o
s
itio
n
o
f
cr
o
co
d
iles
:
h
u
n
tin
g
co
o
r
d
in
atio
n
a
n
d
h
u
n
tin
g
c
o
llab
o
r
atio
n
,
ac
c
o
r
d
i
n
g
to
th
eir
h
u
n
tin
g
b
eh
av
io
r
.
Fo
r
a
co
m
p
r
eh
e
n
s
iv
e
d
is
cu
s
s
io
n
a
n
d
d
etailed
in
f
o
r
m
atio
n
o
n
R
SA
s
tr
u
ctu
r
e
an
d
f
o
r
m
u
latio
n
s
in
clu
d
in
g
e
q
u
atio
n
s
o
f
th
e
e
x
p
lo
r
atio
n
an
d
ex
p
l
o
itatio
n
p
h
ases
,
p
lease
r
ef
er
to
[
1
3
]
.
3
.
4
.
T
he
ANF
I
S
-
RSA
pred
ic
t
io
n m
o
del
T
h
i
s
s
tu
d
y
p
r
e
s
en
t
s
t
h
e
A
N
FI
S
-
R
SA
p
r
ed
i
ct
io
n
m
o
d
e
l,
wh
e
r
e
th
e
AN
FI
S
i
s
a
s
s
i
s
ted
b
y
th
e
R
SA
to
o
b
ta
in
a
h
i
g
h
a
cc
u
r
ac
y
o
f
p
r
e
d
ic
tio
n
s
f
o
r
wi
n
d
p
o
wer
f
o
r
e
ca
s
t
in
g
.
T
h
e
R
S
A’
s
ex
p
l
o
r
a
ti
o
n
an
d
ex
p
l
o
i
ta
tio
n
ab
i
li
ti
e
s
ar
e
u
ti
li
ze
d
in
th
e
p
r
e
d
ic
tio
n
m
o
d
e
l
to
d
iv
er
s
if
y
an
d
/o
r
in
t
en
s
if
y
th
e
l
ea
r
n
in
g
p
r
o
ce
s
s
o
f
th
e
A
NFI
S.
Fig
u
r
e
2
i
llu
s
tr
a
te
s
t
h
e
co
m
p
le
te
p
r
o
ce
d
u
r
e
o
f
th
e
A
NF
I
S
-
R
S
A
p
r
ed
ic
ti
o
n
m
o
d
e
l
f
o
r
w
in
d
p
o
w
er
f
o
r
ec
as
t
in
g
.
I
n
ess
en
ce
,
th
e
co
n
s
tr
u
cted
p
r
ed
ictio
n
m
o
d
el
aim
s
to
e
n
h
an
ce
ANFI
S
p
er
f
o
r
m
an
ce
b
y
in
c
o
r
p
o
r
ati
n
g
R
SA
o
p
er
ato
r
s
to
d
eter
m
i
n
e
t
h
e
s
u
itab
le
ANFI
S
p
ar
am
eter
s
.
T
h
e
c
o
m
p
lete
p
r
o
ce
d
u
r
e
o
f
t
h
e
p
r
e
d
ictio
n
m
o
d
el
is
o
u
tlin
ed
as f
o
llo
ws:
a.
I
n
itializatio
n
p
h
ase:
−
Def
in
e
p
ar
am
ete
r
s
o
f
R
SA,
m
ain
ly
th
e
β
a
n
d
α
.
I
n
w
h
ich
,
β
is
th
e
ex
p
lo
r
atio
n
co
n
t
r
o
ller
s
et
to
0
.
0
0
5
o
n
ly
in
th
e
i
n
itializatio
n
p
h
as
e,
wh
ile
α
is
th
e
ex
p
lo
r
atio
n
c
o
n
tr
o
ller
(
co
n
tr
o
ls
s
ea
r
ch
ac
c
u
r
ac
y
)
s
et
to
0
.
1
o
n
ly
in
th
e
h
u
n
tin
g
p
h
ase
to
d
eter
m
in
e
th
e
d
if
f
e
r
en
ce
b
e
twee
n
th
e
b
est
s
o
lu
tio
n
(
e.
g
.
,
n
ea
r
-
o
p
tim
al
at
ea
ch
iter
atio
n
)
f
o
u
n
d
s
o
f
a
r
an
d
th
e
c
u
r
r
en
t so
lu
tio
n
.
−
T
h
e
d
ataset
u
n
d
er
g
o
es
p
r
e
p
r
o
ce
s
s
in
g
,
in
v
o
lv
in
g
its
d
iv
is
io
n
in
to
two
s
ets:
a
tr
ain
in
g
s
et
(
7
0
%)
an
d
a
test
in
g
s
et
(
3
0
%),
d
eter
m
in
e
d
th
r
o
u
g
h
ex
p
er
im
en
tal
ev
al
u
atio
n
.
−
R
SA
b
eg
in
s
with
r
an
d
o
m
ly
g
en
er
atin
g
a
n
in
itial
p
o
p
u
latio
n
o
f
ca
n
d
id
ate
s
o
lu
tio
n
s
b
a
s
ed
o
n
th
e
d
ataset.
b.
T
r
ain
in
g
p
h
ase:
−
I
n
itial
co
n
f
ig
u
r
atio
n
s
ar
e
g
en
e
r
ated
to
r
e
p
r
esen
t
th
e
R
SA'
s
p
o
p
u
latio
n
,
with
ea
ch
c
o
n
f
ig
u
r
a
tio
n
d
er
iv
e
d
f
r
o
m
th
e
ANFI
S
s
ettin
g
s
.
T
h
e
tr
ain
in
g
s
et
u
n
d
er
g
o
es
tr
ain
in
g
u
s
in
g
t
h
e
ANFI
S
m
o
d
el
tailo
r
ed
s
p
ec
if
ically
f
o
r
th
at
s
et.
−
T
h
en
,
th
e
q
u
ality
(
th
e
p
r
ed
ic
ted
o
u
tp
u
t
o
f
ANFI
S
m
o
d
el
with
th
e
s
m
alles
t
f
itn
ess
v
alu
e)
o
f
ea
ch
co
n
f
ig
u
r
atio
n
(
f
o
r
ea
ch
s
o
lu
tio
n
in
t
h
e
p
o
p
u
latio
n
)
is
e
v
alu
at
ed
b
y
co
m
p
u
tin
g
th
e
f
itn
ess
v
alu
e
d
ef
in
e
d
as th
e
R
MSE
.
−
R
SA
ex
am
in
es
th
e
s
to
p
p
in
g
c
o
n
d
itio
n
(
e.
g
.
,
r
ea
ch
in
g
m
a
x
i
m
u
m
iter
atio
n
s
)
.
I
f
t
h
e
s
to
p
p
in
g
co
n
d
itio
n
is
m
et,
th
e
t
r
ain
in
g
p
h
ase
is
ter
m
in
ated
an
d
th
e
test
in
g
p
h
ase
is
ex
ec
u
ted
.
Oth
er
wis
e,
t
h
e
R
SA
s
elec
ts
th
e
b
est s
o
lu
tio
n
f
o
u
n
d
s
o
f
ar
,
an
d
th
en
th
e
R
SA u
p
d
ates th
e
p
o
p
u
latio
n
an
d
r
ep
ea
ts
th
e
tr
ain
i
n
g
p
h
ase.
c.
T
esti
n
g
p
h
ase:
o
n
ce
t
h
e
s
to
p
p
in
g
co
n
d
itio
n
is
m
et,
th
e
test
in
g
p
h
ase
is
ex
ec
u
ted
b
y
co
n
s
id
er
in
g
th
e
b
est
co
n
f
ig
u
r
atio
n
f
o
r
th
e
b
u
ilt AN
FIS
m
o
d
el
to
ap
p
ly
it to
th
e
te
s
tin
g
s
et
f
o
r
th
e
p
r
ed
ictio
n
p
r
o
ce
s
s
th
at
will b
e
ev
alu
ated
b
ased
o
n
th
e
R
MSE
.
T
h
e
p
r
ed
ictiv
e
m
o
d
el
th
en
ter
m
in
ates
th
e
wh
o
le
p
r
o
ce
s
s
with
th
e
f
in
al
p
r
ed
ictio
n
o
u
tp
u
t.
T
h
e
d
ev
elo
p
ed
ANFI
S
-
R
SA
p
r
ed
ictiv
e
m
o
d
el
s
ee
k
s
to
en
h
an
ce
ANFI
S
p
er
f
o
r
m
an
ce
b
y
h
ar
n
ess
in
g
th
e
s
ea
r
ch
an
d
iter
ativ
e
ca
p
a
b
ilit
ies
o
f
th
e
R
S
A.
T
h
is
in
v
o
lv
es
tu
n
in
g
th
e
MFs
to
ac
h
iev
e
lo
wer
er
r
o
r
r
ates
an
d
im
p
r
o
v
e
p
r
ed
ictio
n
q
u
ality
in
th
e
ANFI
S
c
o
n
f
ig
u
r
atio
n
.
T
h
e
R
SA
is
em
p
lo
y
ed
in
th
e
lear
n
in
g
s
tag
e
(
tr
ain
in
g
p
h
ase)
to
id
en
tif
y
s
u
itab
le
co
n
f
ig
u
r
atio
n
s
(
h
y
p
er
-
p
ar
am
eter
s
)
f
o
r
th
e
weig
h
ts
b
e
twee
n
lay
er
4
an
d
lay
er
5
o
f
ANFI
S.
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
A
h
yb
r
id
a
d
a
p
tive
n
eu
r
o
-
fu
z
z
y
in
feren
ce
s
ystem
a
n
d
r
ep
tile
…
(
Mo
h
a
ma
d
I
.
A
l
-
Wid
ya
n
)
2863
Fig
u
r
e
2
.
T
h
e
p
r
o
p
o
s
ed
ANFI
S
-
R
SA p
r
ed
ictio
n
m
o
d
el
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
I
t
is
ex
p
ec
ted
to
d
e
v
elo
p
an
ef
f
icien
t
f
o
r
ec
asti
n
g
m
o
d
el
u
s
in
g
th
e
p
r
o
p
o
s
ed
ar
tific
ial
i
n
tellig
en
ce
o
p
tim
izatio
n
tech
n
iq
u
e
t
o
ef
f
i
cien
tly
d
ea
l
with
win
d
p
o
wer
f
o
r
ec
asti
n
g
.
T
h
e
tar
g
eted
(
u
n
k
n
o
wn
)
r
esu
lts
ca
n
b
en
ef
it
wh
en
u
s
in
g
r
ea
l
d
ata
f
r
o
m
J
o
r
d
a
n
an
d
/o
r
o
th
er
co
u
n
tr
ies.
Als
o
,
im
p
r
o
v
e
th
e
p
r
ed
icti
o
n
p
e
r
f
o
r
m
an
ce
o
f
th
e
b
en
ch
m
ar
k
d
ata
an
d
m
o
d
el
s
.
4
.
1
.
Descript
io
n o
f
da
t
a
s
et
s
T
h
e
ef
f
ec
tiv
en
ess
o
f
th
e
h
y
b
r
id
ANFI
S
-
R
SA
m
o
d
el
was
ass
ess
ed
u
s
in
g
ac
tu
al
p
o
wer
g
en
e
r
atio
n
d
at
a
f
r
o
m
win
d
f
a
r
m
s
in
J
o
r
d
an
an
d
an
ad
d
itio
n
al
b
en
c
h
m
ar
k
d
ataset,
th
e
C
h
in
ese
(
Hex
i
C
o
r
r
id
o
r
)
d
ataset.
T
h
e
J
o
r
d
an
ian
d
ataset
h
as
li
m
itatio
n
s
th
at
m
ak
e
it
p
ar
tially
u
n
r
eliab
le
f
o
r
test
in
g
th
e
p
r
o
p
o
s
ed
f
o
r
ec
asti
n
g
m
eth
o
d
.
T
h
ese
lim
itatio
n
s
ar
e
d
u
e
to
th
e
lack
o
f
in
f
o
r
m
atio
n
an
d
d
etails o
f
th
e
J
o
r
d
an
ian
win
d
f
ar
m
s
an
d
th
eir
o
b
s
er
v
atio
n
s
,
in
cl
u
d
in
g
h
u
b
te
m
p
er
atu
r
e,
ac
tiv
e
p
o
wer
r
ate,
b
ea
r
in
g
s
h
a
f
t
an
d
g
ea
r
b
o
x
o
il
tem
p
er
atu
r
e,
b
lad
e
p
itch
an
g
le,
an
d
o
th
er
attr
ib
u
te
s
th
at
m
ig
h
t p
lay
a
s
ig
n
if
ican
t r
o
le
in
im
p
r
o
v
in
g
th
e
ac
cu
r
ac
y
o
f
th
e
f
o
r
ec
asti
n
g
task
.
I
t
o
n
ly
co
n
s
is
ts
o
f
th
e
i
n
f
o
r
m
atio
n
an
d
r
ea
d
in
g
s
o
f
1
3
tu
r
b
i
n
es
s
u
ch
as
lo
n
g
itu
d
es
an
d
latitu
d
es,
m
ast
h
eig
h
t,
in
s
tallatio
n
an
d
ter
m
i
n
atio
n
d
ates,
an
d
win
d
s
p
ee
d
m
o
n
th
ly
as
well
as
y
ea
r
ly
b
asis
f
o
r
4
y
ea
r
s
in
a
r
o
w.
Hen
ce
,
th
e
well
-
k
n
o
wn
C
h
in
ese
d
ataset
is
co
n
s
id
er
ed
f
o
r
f
u
r
th
er
e
x
p
er
im
e
n
tatio
n
t
o
war
d
r
o
b
u
s
t
a
n
d
r
eliab
le
test
in
g
an
d
ev
alu
atio
n
o
f
th
e
p
r
o
p
o
s
ed
f
o
r
ec
asti
n
g
m
o
d
el.
T
h
e
J
o
r
d
an
ian
d
ataset
h
as
f
ew
attr
ib
u
tes
th
at
p
r
o
v
id
e
litt
le
in
f
o
r
m
atio
n
th
at
m
a
y
b
e
ex
p
lo
ited
b
y
th
e
f
o
r
ec
asti
n
g
p
r
e
d
ictio
n
m
o
d
el
to
o
b
tain
a
h
ig
h
s
co
r
e
o
f
ac
cu
r
ac
y
.
On
th
e
o
th
er
h
an
d
,
alth
o
u
g
h
it
h
as
2
8
.
5
%
m
is
s
in
g
v
alu
es,
th
e
Hex
i
co
r
r
id
o
r
d
ataset
h
as
1
9
o
u
t
o
f
2
1
n
u
m
er
ical
attr
ib
u
tes
o
f
h
ig
h
i
m
p
ac
t
in
f
o
r
m
atio
n
th
at
m
ay
l
ea
d
to
war
d
s
h
ig
h
ac
cu
r
ac
y
s
c
o
r
es.
T
h
er
e
f
o
r
e,
th
e
id
ea
b
eh
in
d
a
h
ig
h
ly
ac
c
u
r
at
e
an
d
s
tab
le
m
o
d
el
is
th
e
a
b
ilit
y
to
ef
f
ec
tiv
ely
em
p
l
o
y
a
f
ea
tu
r
e
s
elec
tio
n
m
ec
h
an
is
m
th
at
elim
in
ates
f
ewe
r
in
ter
esti
n
g
attr
ib
u
tes
wh
ile
ex
p
lo
itin
g
f
ew
attr
ib
u
tes
y
et
p
o
ten
tially
h
av
e
an
im
p
ac
t
o
n
im
p
r
o
v
in
g
th
e
p
r
ed
ictio
n
ac
cu
r
ac
y
.
I
t
ty
p
ically
h
as
a
v
ar
iety
o
f
r
o
to
r
,
tu
r
b
in
e,
an
d
wea
th
er
ch
ar
ac
ter
is
tics
.
Data
was
co
llected
b
etwe
en
J
an
u
ar
y
2
0
1
8
an
d
Ma
r
c
h
2
0
2
0
.
T
e
n
-
m
in
u
te
in
ter
v
als
h
a
v
e
b
ee
n
u
s
ed
to
r
ec
o
r
d
r
ea
d
in
g
s
.
4
.
2
.
E
v
a
lua
t
i
o
n pro
ce
du
re
a
nd
co
m
pa
ra
t
iv
e
m
et
ho
ds
T
h
e
co
m
m
er
cial
MA
T
L
AB
R
2
0
2
0
b
s
o
f
twar
e
is
u
s
ed
f
o
r
th
e
ANFI
S
-
R
S
A
m
o
d
el
an
aly
s
is
o
n
MS
W
in
d
o
ws
1
0
with
an
I
n
tel
C
o
r
e
-
i7
C
PU
an
d
1
6
GB
o
f
R
A
M.
T
h
e
s
ea
r
ch
m
eth
o
d
s
em
p
lo
y
ed
in
th
e
p
r
o
p
o
s
ed
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
8
5
7
-
2
8
7
3
2864
R
SA
ar
e
d
is
t
in
ctiv
e
an
d
will
b
e
co
n
tr
asted
with
th
o
s
e
o
f
o
t
h
er
ex
is
tin
g
alg
o
r
ith
m
s
.
T
h
er
e
ar
e
s
ev
er
al
m
o
d
els
p
u
b
lis
h
ed
in
th
e
liter
atu
r
e,
a
n
d
o
n
ly
h
ig
h
ly
s
u
cc
ess
f
u
l
m
o
d
els
will
b
e
s
elec
ted
f
o
r
th
e
p
u
r
p
o
s
es
o
f
th
e
co
m
p
ar
is
o
n
s
.
I
n
th
is
in
v
esti
g
atio
n
,
co
m
m
o
n
l
y
ass
o
ciate
d
s
tan
d
ar
d
s
tatis
tical
er
r
o
r
m
etr
ics
f
o
u
n
d
i
n
th
e
liter
atu
r
e,
s
u
ch
as M
AE
,
MA
P
E
,
R
MSE
,
an
d
R
2
,
ar
e
u
tili
ze
d
to
ass
es
s
th
e
p
r
ed
ictiv
e
ac
cu
r
ac
y
o
f
th
e
p
r
o
p
o
s
ed
m
o
d
el.
T
h
is
s
tu
d
y
h
as
co
n
d
u
cted
s
ev
er
al
tr
ad
itio
n
al
m
ac
h
in
e
l
ea
r
n
in
g
tech
n
i
q
u
es
to
b
u
ild
a
b
aselin
e
ac
cu
r
ac
y
s
co
r
e
f
o
r
th
e
J
o
r
d
an
ian
b
en
c
h
m
ar
k
th
at
en
a
b
les
an
ef
f
icien
t
e
v
alu
atio
n
o
f
t
h
e
ANFI
S
-
R
SA
m
o
d
el.
T
ab
le
1
s
h
o
ws
a
co
m
p
ar
is
o
n
o
f
th
e
p
r
o
p
o
s
ed
m
o
d
el
ag
ain
s
t
m
ac
h
in
e
lear
n
in
g
m
o
d
els.
A
cc
u
r
ac
y
,
p
r
ec
is
io
n
,
an
d
r
ec
all
a
r
e
ass
o
ciate
d
with
ev
alu
atin
g
th
e
p
er
f
o
r
m
an
ce
o
f
class
if
icatio
n
task
s
.
T
h
ey
a
r
e
u
s
ed
to
m
ea
s
u
r
e
th
e
q
u
ality
o
f
p
r
e
d
ictio
n
s
o
b
t
ain
ed
b
y
th
e
m
o
d
els,
esp
ec
ially
wh
en
d
ea
lin
g
with
im
b
ala
n
ce
d
d
atasets
(
e.
g
.
,
J
o
r
d
an
ian
d
ataset)
wh
er
e
o
n
e
class
m
ay
b
e
m
o
r
e
p
r
ev
al
en
t
th
an
th
e
o
th
er
.
Ho
wev
er
,
in
th
e
co
n
tex
t
o
f
f
o
r
ec
asti
n
g
,
p
r
ec
is
io
n
an
d
r
ec
all
ar
e
k
n
o
wn
t
o
b
e
in
d
ir
ec
tly
ap
p
licab
le
to
f
o
r
ec
asti
n
g
task
s
in
m
an
y
s
ce
n
ar
i
o
s
.
Fo
r
ec
asti
n
g
ty
p
ically
in
v
o
lv
es
p
r
ed
ictin
g
co
n
tin
u
o
u
s
v
alu
es
(
e.
g
.
,
tem
p
er
atu
r
e
an
d
s
ales)
r
ath
er
th
an
d
is
cr
ete
class
lab
els.
Hen
ce
,
p
r
ec
is
io
n
an
d
r
ec
all
ar
e
u
s
ed
in
th
i
s
s
tu
d
y
as
s
id
e
v
iews
o
f
t
h
e
r
esu
lts
with
litt
le
in
d
icatio
n
.
T
h
er
ef
o
r
e,
in
f
o
r
ec
asti
n
g
,
d
i
f
f
er
en
t
m
et
r
ics
th
at
ar
e
ap
p
r
o
p
r
iate
f
o
r
e
v
alu
atin
g
ac
cu
r
ac
y
an
d
p
er
f
o
r
m
an
ce
a
r
e
u
s
ed
d
ep
en
d
i
n
g
o
n
s
p
ec
if
ic
g
o
als
an
d
c
h
ar
ac
ter
is
tics
o
f
th
e
d
ataset.
So
m
e
co
m
m
o
n
m
etr
ics
f
o
r
f
o
r
ec
asti
n
g
task
s
in
clu
d
e:
−
MA
E
: c
alcu
lates th
e
m
ea
n
ab
s
o
lu
te
d
if
f
er
e
n
ce
b
etwe
en
p
r
ed
i
cted
an
d
ac
tu
al
v
alu
es.
−
MA
PE:
q
u
an
tifie
s
th
e
p
er
ce
n
t
ag
e
d
if
f
er
e
n
ce
b
etwe
en
p
r
ed
ic
ted
an
d
ac
tu
al
v
alu
es.
−
R
MSE
: i
s
th
e
s
q
u
ar
e
r
o
o
t o
f
th
e
MSE
an
d
g
iv
es a
s
en
s
e
o
f
th
e
ty
p
ical
s
ize
o
f
er
r
o
r
s
.
−
Acc
u
r
ac
y
: m
ea
s
u
r
es th
e
p
r
o
p
o
r
tio
n
o
f
co
r
r
ec
t
p
r
ed
ictio
n
s
.
−
R
2
:
aid
s
in
ev
alu
atin
g
th
e
d
eg
r
ee
to
wh
ich
th
e
m
o
d
el'
s
p
r
ed
ictio
n
s
alig
n
with
th
e
v
ar
iab
ili
ty
in
th
e
ac
tu
al
d
ata.
T
h
e
f
o
r
m
u
las o
f
th
e
er
r
o
r
m
ea
s
u
r
em
en
ts
ar
e
ex
p
r
ess
ed
as
(
7
)
,
(
8
)
:
=
1
∙
∑
|
−
|
=
1
(
7
)
=
1
∙
∑
(
−
)
2
=
1
(
8
)
wh
er
e
N
is
th
e
n
u
m
b
er
o
f
p
r
e
d
ictio
n
s
,
Σ
i
s
th
e
s
u
m
o
f
ab
s
o
lu
te
er
r
o
r
s
,
an
d
x
i
r
ep
r
esen
ts
th
e
i
-
th
ac
tu
al
v
alu
e
o
f
a
tim
e
s
er
ies
an
d
mi
is
th
e
v
alu
e
th
at
was
f
o
r
ec
asted
,
f
o
r
th
e
s
am
e
p
o
s
itio
n
in
th
e
s
er
ie
s
,
b
y
th
e
m
o
d
el.
No
te
th
at
th
e
l
o
wer
th
e
v
alu
e
(
t
o
war
d
s
0
)
,
th
e
b
etter
th
e
f
o
r
ec
as
tin
g
m
o
d
el
is
.
I
n
ad
d
itio
n
,
p
r
ec
is
io
n
,
r
ec
all,
a
n
d
F1
-
s
co
r
e
f
o
r
m
ea
s
u
r
in
g
th
e
p
e
r
f
o
r
m
a
n
ce
ar
e
ca
lcu
lated
as:
=
+
+
=
+
(
+
)
+
(
+
)
(
9
)
=
+
+
=
+
(
+
)
+
(
−
)
(
1
0
)
1
=
2
1
+
1
(
1
1
)
wh
er
e
T+
is
a
tr
u
e
p
o
s
itiv
e
d
i
r
ec
tio
n
,
an
ac
tu
al
p
o
s
itiv
e,
p
r
ed
icted
p
o
s
itiv
e,
o
n
th
e
o
th
er
h
an
d
,
F+
is
a
f
alse
p
o
s
itiv
e
d
ir
ec
tio
n
,
a
n
ac
tu
al
n
eg
ativ
e,
p
r
ed
icted
p
o
s
itiv
e.
W
h
ils
t
F
-
is
a
f
alse
n
eg
ativ
e
d
ir
ec
tio
n
,
a
n
ac
tu
al
p
o
s
itiv
e,
p
r
e
d
icted
n
eg
ativ
e.
x+
is
th
e
ac
tu
al
v
alu
e
o
f
t
h
e
ti
m
e
s
er
ies,
an
d
m+
is
th
e
p
r
e
d
icted
(
f
o
r
ec
asted
)
v
alu
e
o
f
t
h
e
s
am
e
tim
e
s
er
ies.
No
te
th
at
th
e
h
ig
h
er
th
e
v
alu
e,
th
e
b
etter
th
e
m
o
d
el’
s
p
e
r
f
o
r
m
an
ce
is
.
4
.
3
.
E
x
perim
ent
a
l
r
esu
lt
s
Acc
o
r
d
in
g
to
T
ab
le
1
,
it
is
s
h
o
wn
th
at
th
e
p
r
o
p
o
s
ed
ANFI
S
-
R
SA
m
o
d
el
h
as
less
ac
cu
r
ac
y
s
co
r
e
th
an
e.
g
.
,
L
STM
,
n
aïv
e
B
ay
es
(
NB
)
,
LR
,
an
d
RF
.
Ho
wev
er
,
it h
as c
o
m
p
etitiv
e
s
co
r
es a
n
d
s
tab
il
ity
in
d
icatio
n
s
s
u
ch
as
r
ec
all,
F1
-
s
co
r
e,
MA
PE,
an
d
s
p
ec
if
ically
R
MSE
,
an
d
R
2
.
All
p
r
esen
ted
m
o
d
els
in
th
e
tab
le
h
a
v
e
n
o
t
r
ea
ch
ed
a
g
r
ea
ter
ac
c
u
r
ac
y
s
co
r
e
th
a
n
9
0
%,
b
ec
au
s
e
th
e
J
o
r
d
an
ian
d
ataset
is
lim
ited
to
a
s
m
all
s
ize
an
d
lo
w
d
is
tr
ib
u
tio
n
ar
o
u
n
d
tar
g
et
f
o
r
e
ca
s
tin
g
lab
els,
wh
ich
ca
n
b
e
s
ee
n
in
R
2
.
Fo
r
a
m
o
r
e
th
o
r
o
u
g
h
co
m
p
r
e
h
en
s
io
n
o
f
th
e
o
u
tco
m
es
o
f
th
e
p
r
o
p
o
s
ed
m
o
d
el,
th
e
g
r
a
p
h
ical
r
e
p
r
esen
tatio
n
s
in
Fig
u
r
es
3
to
7
elu
cid
ate
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
ANFI
S
-
R
SA
m
o
d
el
o
n
th
e
J
o
r
d
an
ian
d
ataset.
Fig
u
r
e
3
d
is
p
lay
s
er
r
o
r
s
d
u
r
in
g
th
e
tr
ain
in
g
an
d
e
v
alu
atio
n
p
h
ases
,
en
a
b
l
in
g
tr
ac
k
i
n
g
o
f
th
e
m
o
d
el'
s
ef
f
icien
cy
o
v
er
tim
e.
Sp
ec
i
f
ically
,
it
id
en
tifie
s
p
atter
n
s
th
at
h
elp
s
u
n
d
er
s
tan
d
i
n
g
tem
p
o
r
al
d
ep
e
n
d
en
cies in
t
h
e
d
ataset.
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
A
h
yb
r
id
a
d
a
p
tive
n
eu
r
o
-
fu
z
z
y
in
feren
ce
s
ystem
a
n
d
r
ep
tile
…
(
Mo
h
a
ma
d
I
.
A
l
-
Wid
ya
n
)
2865
Fig
u
r
e
4
illu
s
tr
ates th
e
co
m
p
ar
is
o
n
b
etwe
en
ac
tu
al
win
d
s
p
ee
d
v
alu
es a
n
d
th
e
m
o
d
el’
s
esti
m
ates o
v
er
tim
e.
I
t
aid
s
in
u
n
d
e
r
s
tan
d
in
g
th
e
m
o
d
el'
s
ca
p
ac
ity
to
ca
p
tu
r
e
tem
p
o
r
al
tr
en
d
s
.
Ho
wev
e
r
,
t
h
e
p
r
e
d
ictio
n
s
lin
e
u
p
d
if
f
er
e
n
tly
th
an
in
p
r
ev
io
u
s
m
o
d
els
(
ac
tu
al
v
s
p
r
ed
icted
)
.
Hen
ce
,
th
e
p
r
ed
ictio
n
s
ar
e
m
an
y
tim
es
o
n
e
-
tim
e
s
tep
b
eh
in
d
th
e
ac
tu
al
v
alu
es,
wh
ich
is
u
n
co
m
m
o
n
.
Fig
u
r
e
5
s
h
o
wca
s
es
er
r
o
r
s
o
v
er
th
e
tim
e
s
p
an
.
Mo
n
ito
r
in
g
er
r
o
r
s
o
v
e
r
tim
e
h
elp
s
id
e
n
tify
p
er
io
d
s
o
f
ac
cu
r
ate
a
n
d
in
ac
c
u
r
ate
p
r
e
d
ictio
n
s
.
T
ab
le
1
.
Per
f
o
r
m
an
ce
ev
alu
ati
o
n
o
f
th
e
p
r
o
p
o
s
ed
ANFI
S
-
R
S
A
vs
.
o
th
er
class
if
ier
s
f
o
r
th
e
J
o
r
d
an
ian
d
ataset
M
o
d
e
l
M
A
E
M
A
P
E
R
M
S
E
R
2
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F1
-
S
c
o
r
e
A
c
c
u
r
a
c
y
NB
1
.
5
2
9
7
.
1
8
7
1
.
1
1
8
0
.
8
5
0
.
9
1
3
0
.
8
4
0
.
8
8
0
.
8
9
8
3
LR
2
.
9
7
1
8
.
3
6
3
2
.
1
1
1
0
.
8
7
0
.
7
7
2
7
0
.
8
9
4
7
0
.
8
3
0
.
8
8
1
4
K
-
n
e
a
r
e
s
t
n
e
i
g
h
b
o
r
(K
-
NN)
1
.
1
3
6
3
.
9
6
5
1
.
1
1
2
0
.
7
3
3
3
0
.
7
3
3
3
0
.
7
3
3
3
0
.
7
3
0
.
8
6
6
7
RF
1
.
6
5
5
4
.
3
3
3
1
.
3
4
2
0
.
8
3
0
.
9
1
3
0
.
8
0
7
7
0
.
8
6
0
.
8
8
1
4
S
V
M
3
.
8
1
1
9
.
3
6
3
3
.
2
1
8
0
.
8
2
0
.
7
6
9
2
0
.
8
3
3
3
0
.
8
0
0
.
8
3
3
3
LSTM
1
.
1
2
2
4
.
0
1
2
1
.
0
8
0
0
.
8
9
0
.
8
3
3
3
0
.
9
0
9
1
0
.
8
7
0
.
9
0
M
u
l
t
i
-
l
a
y
e
r
p
e
r
c
e
p
t
r
o
n
3
.
8
7
1
8
.
4
7
1
2
.
9
8
0
0
.
6
8
0
.
7
5
0
.
6
6
6
7
0
.
7
1
0
.
8
3
3
3
A
N
F
I
S
1
.
0
0
1
0
.
8
5
1
4
5
.
5
1
4
1
0
.
7
6
0
.
7
8
2
6
0
.
7
5
0
.
7
7
0
.
8
1
3
6
Th
e
p
r
o
p
o
s
e
d
A
N
F
I
S
-
R
S
A
0
.
0
9
8
9
7
.
0
8
1
5
0
.
2
0
3
8
0
.
9
6
7
0
.
7
7
7
7
1
.
0
0
.
8
7
5
0
.
8
4
6
1
Fig
u
r
e
3
.
Actu
al
v
s
.
p
r
ed
icted
win
d
s
p
ee
d
o
f
th
e
m
o
d
el
f
o
r
th
e
J
o
r
d
an
ian
d
ataset
Fig
u
r
e
4
.
Actu
al
v
s
.
p
r
ed
icted
win
d
s
p
ee
d
o
f
th
e
m
o
d
el
f
o
r
th
e
J
o
r
d
an
ian
d
ataset
(
Mo
n
th
ly
wis
e)
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
8
5
7
-
2
8
7
3
2866
Fig
u
r
e
5
.
Mo
d
el’
s
MA
E
f
o
r
tu
r
b
in
es
T1
–
T
1
2
as a
f
u
n
ctio
n
o
f
win
d
d
ir
ec
tio
n
b
y
tim
e
f
o
r
th
e
J
o
r
d
an
ian
d
ataset
Fig
u
r
e
6
.
R
MSE
b
y
e
p
o
ch
s
o
f
th
e
m
o
d
el
f
o
r
th
e
J
o
r
d
an
ia
n
d
a
taset
Fig
u
r
e
7
.
R
OC
cu
r
v
e
f
o
r
test
in
g
th
e
m
o
d
el
Evaluation Warning : The document was created with Spire.PDF for Python.