I
AE
S
I
nte
rna
t
io
na
l J
o
urna
l o
f
Art
if
icia
l In
t
ellig
ence
(
I
J
-
AI
)
Vo
l.
4
,
No
.
4
,
Dec
em
b
er
2
0
1
5
,
p
p
.
1
2
9
~
1
3
8
I
SS
N:
2252
-
8938
129
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
jo
u
r
n
a
l.c
o
m/o
n
lin
e/in
d
ex
.
p
h
p
/I
J
AI
Na
rx
Ba
sed Sho
rt
Ter
m
Wi
nd
Pow
er F
o
recas
ti
ng
Mo
del
M
.
Na
nd
a
na
J
y
o
t
hi
1
,
V.
Dina
k
a
r
2
,
N.
S S R
a
v
i Tej
a
3
1
De
p
a
rtm
e
n
t
o
f
El
e
c
tri
c
a
l
a
n
d
El
e
c
tro
n
ics
En
g
in
e
e
rin
g
,
A
c
h
a
ry
a
Na
g
a
rju
n
a
Un
iv
e
rsit
y
,
A
n
d
h
ra
P
ra
d
e
sh
,
In
d
ia
2,
3
De
p
a
rtm
e
n
t
o
f
El
e
c
tri
c
a
l
a
n
d
E
lec
tro
n
ics
En
g
i
n
e
e
rin
g
,
K L
Un
iv
e
rsit
y
,
A
n
d
h
ra
P
ra
d
e
sh
,
I
n
d
ia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Au
g
u
s
t
4
,
2
0
1
5
R
ev
i
s
ed
Oct
8
,
2
0
1
5
A
cc
ep
ted
No
v
1
4
,
2
0
1
5
T
h
is
p
a
p
e
r
c
o
n
tri
b
u
tes
a
sh
o
rt
-
term
w
in
d
p
o
w
e
r
f
o
re
c
a
stin
g
th
ro
u
g
h
A
rti
f
icia
l
Ne
u
ra
l
Ne
tw
o
rk
w
it
h
n
o
n
li
n
e
a
r
a
u
to
re
g
re
ss
iv
e
e
x
o
g
e
n
o
u
s
in
p
u
t
s
(NA
RX
)
m
o
d
e
l.
T
h
e
m
e
teo
ro
lo
g
ica
l
p
a
ra
m
e
ters
li
k
e
w
in
d
sp
e
e
d
,
te
m
p
e
ra
tu
re
,
p
re
ss
u
re
,
a
n
d
a
ir
d
e
n
si
ty
a
r
e
c
o
n
sid
e
re
d
a
s
in
p
u
t
p
a
r
a
m
e
ters
c
o
ll
e
c
ted
f
ro
m
KL
Un
iv
e
rsit
y
a
re
a
a
n
d
th
e
c
a
lcu
late
d
g
e
n
e
ra
ted
p
o
we
r
a
s
o
u
tp
u
t
p
a
ra
m
e
ters
o
f
n
e
u
ra
l
n
e
tw
o
rk
to
p
re
d
ict
t
h
e
w
in
d
p
o
w
e
r
g
e
n
e
ra
ti
o
n
.
Ba
se
d
o
n
h
y
b
rid
f
o
re
c
a
stin
g
tec
h
n
iq
u
e
a
c
o
d
e
is
d
e
v
e
lo
p
e
d
in
M
A
TL
A
B
a
t
d
i
ff
e
r
e
n
t
h
id
d
e
n
la
y
e
rs an
d
d
e
l
a
y
ti
m
e
s.
K
ey
w
o
r
d
:
A
N
N
H
y
b
r
id
m
et
h
o
d
Nar
x
Netc
W
in
d
p
o
w
er
f
o
r
ec
asti
n
g
Co
p
y
rig
h
t
©
2
0
1
5
In
stit
u
te o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
rin
g
a
n
d
S
c
ien
c
e
.
Al
l
rig
h
ts
re
se
rv
e
d
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
M.
Nan
d
an
a
J
y
o
th
i
,
Dep
ar
t
m
en
t o
f
E
lectr
ical
an
d
E
lectr
o
n
ics
E
n
g
i
n
ee
r
in
g
,
A
c
h
ar
y
a
Na
g
ar
j
u
n
a
U
n
i
v
er
s
it
y
,
Gu
n
tu
r
,
An
d
h
r
a
P
r
ad
esh
,
5
2
2
5
0
2
-
I
n
d
ia
.
E
m
ail:
n
an
d
a
n
aj
y
o
t
h
i
@
g
m
a
il.
co
m
1.
I
NT
RO
D
UCT
I
O
N
I
n
th
e
p
r
ese
n
t
d
a
y
s
ce
n
ar
io
t
h
e
n
ee
d
f
o
r
elec
tr
ical
p
o
w
er
i
s
g
r
o
w
i
n
g
i
n
an
e
x
p
o
n
e
n
tial
m
a
n
n
er
.
T
h
e
f
u
e
l
r
eser
v
es
f
o
r
t
h
e
elec
tr
ical
p
o
w
er
g
en
er
atio
n
t
h
r
o
u
g
h
co
n
v
e
n
tio
n
al
m
et
h
o
d
s
ar
e
d
ep
let
in
g
at
a
v
er
y
f
as
ter
r
ate
an
d
ar
e
ca
u
s
in
g
s
e
v
er
e
h
ar
m
f
u
l
ef
f
ec
t
o
n
t
h
e
en
v
ir
o
n
m
e
n
t.
A
cc
o
r
d
in
g
to
th
e
Gl
o
b
al
W
in
d
E
n
er
g
y
Ou
tlo
o
k
2
0
1
4
b
y
Glo
b
al
W
in
d
E
n
er
g
y
C
o
u
n
ci
l
(
GW
E
C
)
p
o
w
er
s
ec
to
r
is
th
e
s
o
le
e
m
itter
o
f
ab
o
u
t
4
0
%
o
f
th
e
ca
r
b
o
n
d
io
x
id
e
an
d
2
5
% o
f
all
th
e
g
r
ee
n
h
o
u
s
e
g
a
s
es.
T
h
e
b
etter
s
o
lu
tio
n
th
at
ad
d
r
ess
es
m
o
s
t
o
f
t
h
e
p
r
o
b
lem
s
t
h
a
t
ar
is
es
b
ec
au
s
e
o
f
t
h
e
f
o
s
s
il
f
u
els
i
s
th
e
u
s
a
g
e
o
f
r
e
n
e
w
ab
le
e
n
er
g
y
.
Am
o
n
g
all
th
e
r
e
n
e
w
ab
le
e
n
er
g
ies
w
i
n
d
e
n
er
g
y
is
th
e
m
o
s
t
p
r
o
m
i
s
i
n
g
a
n
d
ch
ea
p
er
to
o
p
e
r
ate.
B
u
t
th
er
e
is
a
h
u
g
e
d
e
m
a
n
d
f
o
r
th
e
elec
tr
ical
en
er
g
y
w
h
ic
h
is
p
ar
tl
y
s
h
ar
ed
b
y
t
h
e
n
o
n
-
co
n
v
e
n
tio
n
al
en
er
g
y
r
eso
u
r
ce
s
.
GW
E
C
i
n
i
ts
r
ec
e
n
t
p
u
b
li
ca
tio
n
s
ta
ted
th
a
t
t
h
e
w
i
n
d
e
n
er
g
y
co
u
ld
r
ea
ch
2
0
0
0
GW
b
y
2
0
3
0
.
B
u
t
th
e
m
aj
o
r
h
in
d
r
an
ce
f
o
r
th
e
ex
p
an
s
io
n
a
n
d
in
te
g
r
atio
n
o
f
th
e
w
i
n
d
p
o
w
er
to
th
e
g
r
id
is
th
e
h
ig
h
v
o
latile
an
d
in
ter
m
itte
n
t
n
a
tu
r
e
o
f
th
e
w
i
n
d
p
o
w
er
.
B
ec
au
s
e
o
f
th
ese
n
a
tu
r
es
o
f
w
i
n
d
p
o
w
e
r
it
is
v
er
y
d
if
f
ic
u
lt
to
in
teg
r
ate
to
th
e
g
r
id
an
d
s
ch
ed
u
le
th
e
p
o
w
er
.
T
o
o
v
er
co
m
e
th
e
s
tated
p
r
o
b
lem
s
w
in
d
p
o
w
er
f
o
r
ec
asti
n
g
i
s
th
e
b
est
s
o
l
u
tio
n
.
W
in
d
p
o
wer
f
o
r
ec
asti
n
g
m
o
d
el
h
elp
s
t
h
e
p
o
w
er
s
y
s
te
m
o
p
er
ato
r
s
in
p
o
w
er
s
ch
ed
u
li
n
g
,
d
is
p
atch
,
g
r
id
s
ec
u
r
it
y
,
s
y
s
t
e
m
o
p
er
atio
n
a
n
d
m
ai
n
tai
n
i
n
g
t
h
e
r
eser
v
e
ca
p
ac
itie
s
.
T
h
er
e
ar
e
d
if
f
er
e
n
t
f
o
r
ec
asti
n
g
m
eth
o
d
s
ac
co
r
d
in
g
to
d
i
f
f
er
en
t
ti
m
e
h
o
r
izio
n
s
w
h
ic
h
ar
e
v
er
y
s
h
o
r
t
ter
m
f
o
r
ec
asti
n
g
m
e
th
o
d
i
n
th
is
m
i
n
u
te
s
to
h
o
u
r
s
ti
m
e
h
o
r
izo
n
i
s
co
n
s
id
er
ed
,
s
h
o
r
t
-
ter
m
f
o
r
ec
asti
n
g
m
e
th
o
d
i
s
h
o
u
r
s
to
d
a
y
s
ti
m
e
h
o
r
izo
n
is
co
n
s
id
er
ed
,
lo
n
g
–
ter
m
f
o
r
ec
asti
n
g
m
et
h
o
d
is
co
n
s
id
er
ed
as d
a
y
to
m
o
n
t
h
s
ar
e
co
n
s
id
er
ed
[
1
-
5
]
.
1
.
1
.
O
v
er
v
ie
w
o
f
t
he
p
re
s
ent
w
ind
f
o
re
ca
s
t
ing
m
et
ho
d
s
T
h
er
e
ar
e
v
ar
io
u
s
f
o
r
ec
asti
n
g
m
et
h
o
d
s
d
ev
elo
p
ed
to
p
r
ed
i
ct
w
in
d
s
p
ee
d
an
d
o
r
p
o
w
er
,
w
h
ic
h
ar
e
p
r
esen
ted
as th
e
f
o
llo
w
in
g
tab
u
lar
co
lu
m
n
:
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
4
,
No
.
4
,
Dec
em
b
er
2
0
1
5
:
1
2
9
–
1
3
8
130
T
ab
le
1
.
W
in
d
Fo
r
ec
asti
n
g
Me
th
o
d
s
an
d
th
e
ir
A
p
p
licatio
n
[
1
]
-
[
6
]
F
o
r
e
c
a
st
i
n
g
me
t
h
o
d
S
u
b
c
l
a
ss
T
e
c
h
n
i
q
u
e
A
d
v
a
n
t
a
g
e
s
D
i
s
-
a
d
v
a
n
t
a
g
e
P
e
r
si
st
e
n
c
e
me
t
h
o
d
-
P
(
t
+
k
)
=
p
(
t
)
B
e
n
c
h
m
a
r
k
a
p
p
r
o
a
c
h
f
o
r
V
e
r
y
sh
o
r
t
t
e
r
m
w
i
n
d
p
o
w
e
r
f
o
r
e
c
a
st
i
n
g
.
R
e
l
i
a
b
i
l
i
t
y
d
e
c
r
e
a
se
s
a
s t
i
me
l
a
p
se
i
n
c
r
e
a
se
s.
P
h
y
si
c
a
l
a
p
p
r
o
a
c
h
N
u
me
r
i
c
a
l
W
e
a
t
h
e
r
P
r
e
d
i
c
t
i
o
n
(
N
W
P
)
S
o
l
v
e
c
o
mp
l
e
x
mat
h
e
mat
i
c
a
l
mo
d
e
l
s u
si
n
g
l
a
r
g
e
me
t
e
o
r
o
l
o
g
i
c
a
l
d
a
t
a
i
s
c
o
l
l
e
c
t
e
d
t
o
d
e
v
e
l
o
p
p
a
t
t
e
r
n
s
I
t
i
s re
l
i
a
b
l
e
f
o
r
me
d
i
u
m
t
o
l
o
n
g
t
e
r
m
f
o
r
e
c
a
st
i
n
g
.
N
e
e
d
l
a
r
g
e
n
u
m
b
e
r
o
f
c
o
mp
u
t
a
t
i
o
n
s (s
u
p
e
r
c
o
mp
u
t
e
r
s n
e
e
d
e
d
)
.
S
t
a
t
i
st
i
c
a
l
a
p
p
r
o
a
c
h
A
r
t
i
f
i
c
i
a
l
N
e
u
r
a
l
N
e
t
w
o
r
k
(
A
N
N
)
T
u
n
e
t
h
e
n
e
t
w
o
r
k
w
i
t
h
t
h
e
h
e
l
p
d
i
f
f
e
r
e
n
c
e
b
e
t
w
e
e
n
a
c
t
u
a
l
d
a
t
a
a
n
d
me
a
su
r
e
d
d
a
t
a
N
o
t
c
o
mp
l
e
t
e
l
y
b
a
se
d
o
n
p
r
e
d
e
f
i
n
e
d
e
q
u
a
t
i
o
n
s
D
a
t
a
me
a
su
r
e
me
n
t
sh
o
u
l
d
b
e
a
c
c
u
r
a
t
e
f
o
r
p
r
o
p
e
r
t
r
a
i
n
i
n
g
o
f
t
h
e
n
e
t
w
o
r
k
.
T
i
me
se
r
i
e
s mo
d
e
l
s
H
y
b
r
i
d
st
r
u
c
t
u
r
e
s
N
W
P
&A
N
N
C
o
mb
i
n
a
t
i
o
n
o
f
f
o
r
e
c
a
st
i
n
g
me
t
h
o
d
s
i
s
u
se
d
t
o
i
m
p
r
o
v
e
t
h
e
e
f
f
i
c
i
e
n
c
y
o
f
f
o
r
e
c
a
st
i
n
g
.
I
t
i
s re
l
i
a
b
l
e
f
o
r
v
e
r
y
sh
o
r
t
t
e
r
m fo
r
e
c
a
st
i
n
g
.
D
a
t
a
me
a
su
r
e
me
n
t
sh
o
u
l
d
b
e
a
c
c
u
r
a
t
e
f
o
r
p
r
o
p
e
r
t
r
a
i
n
i
n
g
o
f
t
h
e
n
e
t
w
o
r
k
.
T
h
e
p
r
e
d
icatio
n
ca
r
r
ied
o
u
t
in
th
is
p
ap
er
u
s
es
t
h
e
h
y
b
r
id
s
tr
u
ctu
r
e
s
.
Statis
tical
d
ata
is
co
llected
f
r
o
m
th
e
E
n
er
g
y
Dep
ar
t
m
e
n
t
o
f
K
L
U
n
i
v
er
s
it
y
,
An
d
h
r
a
P
r
ad
esh
w
h
ich
co
n
s
is
ts
o
f
7
2
0
h
o
u
r
s
d
ata
o
f
w
h
ic
h
6
7
2
h
o
u
r
s
d
ata
is
u
s
ed
f
o
r
tr
ain
in
g
an
d
4
8
h
o
u
r
s
d
ata
i
s
u
s
ed
f
o
r
p
r
ed
ictio
n
an
d
also
th
e
p
h
y
s
i
ca
l
la
w
s
ar
e
tak
en
in
to
ac
co
u
n
t
f
o
r
th
e
ca
lc
u
lati
o
n
o
f
p
o
w
er
g
e
n
er
atio
n
.
T
h
e
d
ev
elo
p
ed
m
o
d
el
is
b
ased
o
n
th
e
n
o
n
-
l
in
ea
r
au
to
r
eg
r
ess
i
v
e
w
i
th
e
x
o
g
e
n
o
u
s
i
n
p
u
t
(
n
ar
x
)
to
o
l
w
h
ic
h
tr
ai
n
s
t
h
e
A
NN
f
o
r
t
h
e
ti
m
e
s
er
ies.
T
h
e
i
n
p
u
t
p
ar
a
m
eter
s
tak
en
i
n
to
co
n
s
id
er
atio
n
ar
e
w
i
n
d
s
p
ee
d
,
tem
p
er
at
u
r
e,
p
r
e
s
s
u
r
e,
air
d
en
s
it
y
an
d
th
e
o
u
tp
u
t
p
ar
a
m
eter
i
s
g
en
er
ated
p
o
w
er
.
Me
an
s
q
u
ar
e
er
r
o
r
an
d
r
o
o
t
m
ea
n
s
q
u
ar
e
er
r
o
r
a
r
e
ca
lcu
lated
f
r
o
m
th
e
p
r
ed
icted
an
d
k
n
o
w
n
r
esu
lt
s
.
1
.
2
.
Ca
lcula
t
i
o
n
o
f
w
ind
po
w
er
f
r
o
m
t
he
da
t
a
T
h
e
w
in
d
p
o
w
er
g
e
n
er
ated
b
y
t
h
e
tu
r
b
in
e
s
d
ep
en
d
s
u
p
o
n
th
e
f
ac
to
r
s
lik
e
w
i
n
d
s
p
ee
d
,
am
b
ien
t
te
m
p
er
atu
r
e,
w
i
n
d
p
r
es
s
u
r
e,
air
d
en
s
it
y
.
Am
o
n
g
all
th
e
f
a
cto
r
s
w
in
d
s
p
ee
d
an
d
air
d
en
s
it
y
d
o
m
i
n
ates
t
h
e
p
o
w
er
g
en
er
ated
.
W
in
d
p
o
w
er
g
en
e
r
ated
is
k
n
o
w
n
b
y
(
1
)
W
h
er
e
P
-
W
in
d
p
o
w
er
g
en
er
at
ed
Ρ
-
A
ir
d
en
s
it
y
at
th
e
g
i
v
en
te
m
p
er
atu
r
e
A
-
A
r
ea
s
w
ep
t b
y
t
h
e
t
u
r
b
in
e
b
lad
es
V
-
W
in
d
s
p
ee
d
W
in
d
p
o
w
er
g
e
n
er
ated
is
h
ig
h
l
y
a
f
f
ec
ted
b
y
t
h
e
air
d
en
s
i
t
y
a
n
d
t
h
e
w
i
n
d
s
p
ee
d
;
as
th
e
ar
ea
s
w
ep
t
b
y
th
e
t
u
r
b
in
es
b
lad
es
r
e
m
ai
n
co
n
s
ta
n
t
f
o
r
a
ta
k
en
tu
r
b
in
e
[
5
]
.
T
h
e
d
ata
is
co
llected
f
r
o
m
E
n
er
g
y
Dep
ar
t
m
en
t
o
f
K
L
Un
i
v
er
s
i
t
y
,
v
ad
d
es
w
ar
a
m
ar
ea
f
o
r
a
tim
e
s
p
a
n
o
f
o
n
e
m
o
n
t
h
w
h
ic
h
co
m
p
r
is
es
o
f
w
i
n
d
s
p
ee
d
,
am
b
ie
n
t
te
m
p
er
atu
r
e,
a
n
d
air
p
r
es
s
u
r
e.
T
h
e
ai
r
d
en
s
it
y
i
n
t
h
e
c
o
n
s
id
er
ed
ar
ea
is
n
o
t
k
n
o
wn
.
Fo
r
t
h
e
d
en
s
i
t
y
ca
lcu
latio
n
s
v
ap
o
u
r
p
r
ess
u
r
e
i
s
r
eq
u
ir
ed
ac
co
r
d
in
g
to
th
e
f
o
r
m
u
la
[
5
]
(
)
(
)
(
2
)
W
h
er
e
ρ
-
A
ir
d
en
s
it
y
at
th
e
g
i
v
en
te
m
p
er
atu
r
e
D
-
A
ir
d
en
s
it
y
at
ab
s
o
l
u
te
te
m
p
er
atu
re
T
-
Giv
en
te
m
p
er
atu
r
e
B
-
B
ar
o
m
etr
ic
(
at
m
o
s
p
h
er
ic)
p
r
ess
u
r
e
e
-
Vap
o
u
r
p
r
ess
u
r
e
o
f
t
h
e
air
a
t th
e
g
i
v
e
n
te
m
p
er
at
u
r
e
A
ll
t
h
e
d
ata
r
eq
u
ir
ed
f
o
r
th
e
d
en
s
it
y
ca
lcu
latio
n
i
s
p
r
esen
t
ex
ce
p
t
th
e
v
ap
o
u
r
p
r
ess
u
r
e.
I
n
a
clo
s
ed
s
y
s
te
m
t
h
e
p
r
ess
u
r
e
ex
er
ted
b
y
a
v
ap
o
u
r
in
t
h
er
m
o
d
y
n
a
m
ic
eq
u
ilib
r
iu
m
at
a
g
i
v
en
te
m
p
e
r
atu
r
e
is
th
e
v
ap
o
u
r
p
r
ess
u
r
e.
Vap
o
u
r
p
r
ess
u
r
e
is
c
alcu
lated
u
s
i
n
g
t
h
e
C
la
u
s
is
-
cla
p
eu
r
o
n
r
elatio
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
I
SS
N:
2252
-
8938
N
a
r
x
B
a
s
ed
S
h
o
r
t Term
W
in
d
P
o
w
er F
o
r
ec
a
s
tin
g
Mo
d
el
(
M.
N
a
n
d
a
n
a
Jy
o
th
i
)
131
(
)
(
)
(
)
(
3
)
W
h
er
e
P
1
,
P
2
–
T
h
e
v
ap
o
u
r
p
r
ess
u
r
es a
t te
m
p
er
at
u
r
es T
1
,
T
2
r
esp
ec
t
iv
el
y
H
v
ap
–
E
n
t
h
alp
y
o
f
v
ap
o
u
r
iz
atio
n
o
f
liq
u
id
R
-
R
ea
l g
a
s
co
n
s
ta
n
t (
8
.
3
1
4
J
)
T
1
-
T
em
p
er
atu
r
e
at
w
h
ic
h
th
e
v
ap
o
u
r
p
r
ess
u
r
e
is
k
n
o
w
n
T
2
-
T
em
p
er
atu
r
e
at
w
h
ic
h
th
e
v
ap
o
u
r
p
r
ess
u
r
e
to
b
e
ca
lcu
lat
ed
T
ak
e
T
1
an
d
P
1
at
STP
co
n
d
itio
n
s
an
d
P
2
is
th
e
v
ap
o
u
r
p
r
ess
u
r
e
to
b
e
ca
lcu
lated
at
w
h
ich
t
h
e
te
m
p
er
atu
r
e
is
T
2
.
B
y
t
h
is
f
o
r
m
u
la
v
ap
o
u
r
P
2
,
‘
e
’
i
n
t
h
e
d
en
s
it
y
ca
lcu
la
tio
n
is
o
b
tain
ed
.
A
ir
d
en
s
i
t
y
is
th
e
n
ca
lcu
lated
b
y
t
h
e
s
tated
f
o
r
m
u
la.
T
h
e
d
ata
r
eq
u
ir
ed
f
o
r
th
e
ca
lcu
latio
n
o
f
p
o
w
er
g
e
n
r
atio
n
is
o
b
tain
ed
an
d
t
h
e
p
o
w
er
th
at
ca
n
b
e
g
en
er
ated
b
y
u
s
i
n
g
all
th
e
s
e
p
ar
am
eter
s
is
ca
lcu
lated
b
y
i
g
n
o
r
i
n
g
t
h
e
o
p
er
atio
n
al
lo
s
ee
s
o
f
tu
r
b
in
e.
Fig
u
r
e
1
.
Diag
r
a
m
t
ical
R
ep
r
es
en
tatio
n
o
f
W
in
d
S
pe
ed
v
s
P
o
w
er
G
e
n
er
ated
2.
RE
S
E
ARCH
M
E
T
H
O
D
2
.
1
.
Art
if
ici
a
l N
eura
l N
et
wo
rk
(
ANN)
A
r
ti
f
icial
n
eu
r
al
n
e
t
w
o
r
k
s
ar
e
th
e
n
eu
r
al
n
et
w
o
r
k
s
d
er
iv
ed
f
r
o
m
th
e
i
n
s
p
ir
atio
n
o
f
b
io
lo
g
ical
n
eu
r
al
n
et
w
o
r
k
s
(
an
i
m
al
ce
n
tr
al
n
er
v
o
u
s
s
y
s
te
m
)
.
T
h
ese
ar
ti
f
icial
n
eu
r
al
n
et
w
o
r
k
s
ar
e
u
s
ed
to
ta
k
e
lo
g
ical
d
ec
i
s
io
n
s
b
ased
o
n
th
e
in
p
u
ts
.
A
NN
c
an
d
ea
l
w
it
h
n
o
n
-
l
i
n
ea
r
an
d
co
m
p
le
x
p
r
o
b
lem
s
in
ter
m
s
o
f
class
i
f
icat
io
n
o
r
f
o
r
ec
asti
n
g
b
y
ex
tr
ac
ti
n
g
t
h
e
d
ep
en
d
en
ce
b
et
w
ee
n
v
ar
iab
les
th
r
o
u
g
h
th
e
tr
ain
i
n
g
p
r
o
ce
s
s
.
So
th
e
A
NN
b
ased
m
et
h
o
d
is
a
n
ap
p
r
o
p
r
iate
m
et
h
o
d
to
ap
p
l
y
to
t
h
e
p
r
o
b
le
m
o
f
f
o
r
ec
ast
in
g
w
i
n
d
p
o
w
er
b
e
ca
u
s
e
it
i
s
d
ir
ec
tl
y
p
r
o
p
o
r
tio
n
al
to
w
in
d
s
p
ee
d
w
h
ic
h
is
h
ig
h
l
y
i
n
ter
m
i
tten
t
in
n
at
u
r
e.
Am
o
n
g
t
h
e
av
a
il
ab
le
m
e
th
o
d
s
u
s
i
n
g
ar
tif
icial
n
e
u
r
al
n
et
w
o
r
k
s
t
h
e
NARX,
a
d
y
n
a
m
ic
r
ec
u
r
r
en
t
m
et
h
o
d
,
is
u
s
ed
to
s
o
lv
e
t
h
e
ti
m
e
s
er
ie
s
p
r
o
b
lem
[
7
-
1
0
]
.
2
.
1
.
1
.
ANN
t
ra
ini
ng
On
e
o
f
th
e
k
e
y
ele
m
e
n
ts
o
f
n
eu
r
al
n
e
t
w
o
r
k
s
is
t
h
eir
ab
ilit
y
to
lear
n
.
A
n
eu
r
al
n
et
w
o
r
k
is
a
co
m
p
lex
ad
ap
tiv
e
s
y
s
te
m
,
w
h
ich
m
ea
n
s
it
ca
n
c
h
an
g
e
it
s
i
n
ter
n
al
s
tr
u
ct
u
r
e
b
ased
o
n
t
h
e
in
p
u
ts
an
d
tar
g
et
s
.
T
h
ese
A
N
Ns
n
ee
d
to
b
e
tr
ain
ed
f
o
r
d
o
in
g
a
p
ar
ticu
lar
tas
k
[
1
0
-
1
4
]
.
T
h
er
e
ar
e
th
r
ee
t
y
p
es
o
f
tr
a
in
i
n
g
p
ar
ad
ig
m
s
to
tr
ain
th
e
ar
ti
f
icial
n
e
u
r
al
n
et
wo
r
k
an
d
ar
e
as f
o
llo
w
s
:
1.
Su
p
er
v
i
s
ed
tr
ain
i
n
g
:
I
t
is
th
e
p
r
o
ce
s
s
o
f
p
r
o
v
id
in
g
t
h
e
n
et
w
o
r
k
w
it
h
a
s
er
ie
s
o
f
s
a
m
p
le
in
p
u
t
s
an
d
co
m
p
ar
i
n
g
t
h
e
o
u
tp
u
t
w
i
th
t
h
e
ex
p
ec
ted
r
esp
o
n
s
e.
T
h
e
tr
ai
n
in
g
co
n
ti
n
u
es
u
n
til
th
e
n
et
wo
r
k
is
ab
le
to
p
r
o
v
id
e
th
e
e
x
p
ec
ted
r
esp
o
n
s
e.
T
h
e
p
r
o
p
o
s
ed
w
o
r
k
is
s
u
p
er
v
i
s
ed
tr
ain
in
g
w
i
th
b
ac
k
p
r
o
p
ag
atio
n
tech
n
iq
u
e.
2.
Un
s
u
p
er
v
i
s
ed
tr
ain
i
n
g
:
I
n
t
h
is
m
et
h
o
d
o
f
tr
ain
i
n
g
,
t
h
e
i
n
p
u
t
v
ec
to
r
an
d
th
e
tar
g
et
o
u
tp
u
t
is
n
o
t
k
n
o
w
n
.
T
h
e
n
et
w
o
r
k
m
a
y
m
o
d
i
f
y
i
n
s
u
ch
a
w
a
y
th
at
t
h
e
m
o
s
t
s
i
m
il
ar
in
p
u
t
v
ec
to
r
is
as
s
i
g
n
ed
to
th
e
s
a
m
e
o
u
tp
u
t
u
n
i
t.
3.
R
ein
f
o
r
ce
m
e
n
t
tr
ain
i
n
g
:
I
t
i
s
t
h
e
p
r
o
ce
s
s
o
f
tr
ain
in
g
t
h
e
n
et
w
o
r
k
in
th
e
p
r
esen
ce
o
f
a
tea
ch
er
b
u
t
i
n
t
h
e
ab
s
en
ce
o
f
tar
g
e
t v
ec
to
r
.
T
h
e
teac
h
er
g
i
v
es o
n
l
y
t
h
e
an
s
w
er
w
h
et
h
er
it is
co
r
r
ec
t (
1
)
o
r
w
r
o
n
g
(
0
)
.
0
0
.
5
1
1
.
5
2
2
.
5
3
3
.
5
4
4
.
5
0
50
100
150
200
250
300
W
i
n
d
s
p
e
e
d
m
/
s
e
c
W
i
n
d
P
o
w
e
r
K
W
h
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
4
,
No
.
4
,
Dec
em
b
er
2
0
1
5
:
1
2
9
–
1
3
8
132
2
.
1
.
2
.
No
n linea
r
Aut
o
Reg
r
ess
iv
e
w
it
h e
x
o
g
eno
us
inp
ut
(
NARX
)
T
h
e
n
o
n
li
n
ea
r
au
to
r
eg
r
e
s
s
i
v
e
n
et
w
o
r
k
w
ith
ex
o
g
en
o
u
s
i
n
p
u
ts
(
N
AR
X)
is
a
r
ec
u
r
r
en
t
d
y
n
a
m
ic
n
et
w
o
r
k
,
w
it
h
f
ee
d
b
ac
k
co
n
n
ec
tio
n
s
en
clo
s
in
g
s
e
v
er
al
la
y
e
r
s
o
f
th
e
n
et
w
o
r
k
.
T
h
e
N
AR
X
m
o
d
el
is
b
ased
o
n
th
e
li
n
ea
r
AR
X
m
o
d
el,
w
h
ic
h
is
co
m
m
o
n
l
y
u
s
ed
i
n
ti
m
e
-
s
e
r
ies
m
o
d
eli
n
g
[
6
]
.
T
h
e
d
ef
in
i
n
g
eq
u
atio
n
f
o
r
th
e
NARX
m
o
d
el
i
s
(
)
(
(
)
)
(
)
(
)
(
)
(
)
(
)
)
(
4
)
W
h
er
e
th
e
d
ep
en
d
e
n
t
n
e
x
t
v
a
lu
e
o
u
tp
u
t
s
i
g
n
al
y
(
t)
is
r
e
g
r
ess
ed
o
n
p
r
e
v
io
u
s
v
a
lu
e
s
o
f
t
h
e
y
(
t)
s
ig
n
al
a
n
d
a
n
in
d
ep
en
d
en
t
(
ex
o
g
e
n
o
u
s
)
i
n
p
u
t
s
i
g
n
al
is
th
e
p
r
ev
io
u
s
v
alu
e
s
.
I
m
p
le
m
e
n
t
th
e
N
AR
X
m
o
d
el
t
h
r
o
u
g
h
f
ee
d
f
o
r
w
ar
d
n
e
u
r
al
n
et
w
o
r
k
is
to
ap
p
r
o
x
i
m
ate
t
h
e
f
u
n
ctio
n
f
.
A
d
iag
r
a
m
o
f
t
h
e
r
esu
lti
n
g
t
w
o
-
la
y
e
r
f
ee
d
f
o
r
w
ar
d
n
et
w
o
r
k
i
s
s
h
o
w
n
F
ig
u
r
e
2.
T
h
is
ap
p
licatio
n
also
allo
w
s
f
o
r
a
v
ec
to
r
A
R
X
m
o
d
el
f
o
r
m
u
ltid
i
m
en
s
io
n
al
i
n
p
u
t
s
an
d
o
u
tp
u
t
s
[
5
]
.
NARX
n
et
w
o
r
k
h
av
e
m
a
n
y
ap
p
licatio
n
s
to
p
r
ed
ict
th
e
n
e
x
t
v
alu
e
o
f
t
h
e
in
p
u
t
s
i
g
n
al
an
d
al
s
o
b
e
u
s
ed
f
o
r
n
o
n
li
n
ea
r
f
ilter
in
g
,
i
n
th
i
s
a
n
o
is
e
-
f
r
ee
tar
g
e
t
o
u
tp
u
t
o
f
th
e
in
p
u
t
s
ig
n
al
w
il
l
b
e
o
b
tain
ed
.
T
h
e
im
p
o
r
tan
t
u
s
e
o
f
th
e
N
AR
X
n
et
w
o
r
k
i
s
est
ab
l
is
h
ed
in
a
n
o
th
er
ap
p
licati
o
n
o
f
th
e
n
o
n
l
in
ea
r
d
y
n
a
m
ic
s
y
s
te
m
s
m
o
d
elli
n
g
.
B
ef
o
r
e
s
ig
n
if
y
i
n
g
t
h
e
tr
ain
i
n
g
o
f
th
e
N
A
R
X
n
et
w
o
r
k
s
,
v
ital
co
n
f
i
g
u
r
atio
n
s
is
u
s
e
f
u
l
in
tr
ain
i
n
g
a
n
d
co
n
s
id
er
th
e
o
u
tp
u
t
o
f
t
h
e
N
A
R
X
n
et
w
o
r
k
m
o
d
el
to
b
e
p
r
e
d
ictio
n
o
f
th
e
o
u
tp
u
t
o
f
n
o
n
l
in
ea
r
d
y
n
a
m
ic
s
y
s
te
m
.
T
h
e
s
tan
d
ar
d
N
A
R
X
ar
c
h
itect
u
r
e
is
o
u
tp
u
t
f
ed
b
ac
k
to
th
e
i
n
p
u
t
o
f
th
e
f
ee
d
f
o
r
w
ar
d
n
eu
r
al
n
et
w
o
r
k
.
Fig
u
r
e
2
.
T
w
o
-
L
a
y
er
Feed
f
o
r
w
ar
d
N
et
w
o
r
k
2
.
1
.
2
.
1
.
Series pa
ra
llel a
rc
hite
ct
ure
T
h
is
ar
ch
itectu
r
e
u
s
ed
w
h
en
t
h
e
o
u
tp
u
t
o
f
t
h
e
N
A
R
X
n
et
wo
r
k
is
co
n
s
id
er
ed
to
b
e
an
esti
m
ate
o
f
th
e
o
u
tp
u
t
o
f
s
o
m
e
n
o
n
lin
ea
r
d
y
n
a
m
ic
s
y
s
te
m
.
T
h
e
o
u
tp
u
t
is
f
ed
b
ac
k
to
th
e
in
p
u
t
o
f
t
h
e
f
ee
d
f
o
r
w
ar
d
n
e
u
r
al
n
et
w
o
r
k
as
p
ar
t
o
f
t
h
e
s
tan
d
ar
d
NARX
ar
ch
itec
tu
r
e.
B
ec
au
s
e
th
e
tr
u
e
o
u
tp
u
t
i
s
av
ai
lab
le
d
u
r
in
g
t
h
e
tr
ain
i
n
g
o
f
th
e
n
et
w
o
r
k
,
y
o
u
co
u
ld
cr
ea
te
a
s
er
ies
-
p
ar
allel
ar
ch
itect
u
r
e,
i
n
w
h
ich
th
e
tr
u
e
o
u
tp
u
t
i
s
u
s
e
d
in
s
tead
o
f
f
ee
d
i
n
g
b
ac
k
t
h
e
es
ti
m
a
ted
o
u
tp
u
t.
T
h
is
h
as
t
w
o
ad
v
a
n
ta
g
es
w
h
ich
ar
e
th
e
f
ir
s
t
i
s
t
h
at
t
h
e
in
p
u
t
t
o
th
e
f
ee
d
f
o
r
w
ar
d
n
et
w
o
r
k
is
m
o
r
e
ac
cu
r
ate.
T
h
e
s
ec
o
n
d
is
th
at
th
e
r
esu
lt
i
n
g
n
e
t
w
o
r
k
h
as
a
p
u
r
el
y
f
ee
d
f
o
r
w
a
r
d
ar
ch
itectu
r
e,
an
d
s
tatic
b
ac
k
p
r
o
p
ag
atio
n
ca
n
b
e
u
s
ed
f
o
r
tr
ain
i
n
g
.
Fig
u
r
e
3
.
Ser
ies P
ar
allel
A
r
ch
i
tectu
r
e
2
.
1
.
2
.
2
.
P
a
ra
llel a
rc
hite
ct
ur
e
L
ater
th
is
ar
ch
itec
tu
r
e
is
co
n
v
er
ted
in
to
p
ar
allel
ar
ch
itectu
r
e
f
o
r
th
e
p
r
ed
ictio
n
.
T
h
e
p
r
ed
ic
tio
n
o
f
th
e
n
ex
t
v
al
u
e
d
ep
en
d
s
o
n
th
e
in
p
u
ts
a
n
d
p
r
ev
i
o
u
s
o
u
tp
u
ts
to
th
e
n
et
w
o
r
k
.
T
h
e
d
ep
en
d
en
ce
o
n
th
e
p
r
ev
io
u
s
o
u
tp
u
t
ca
n
b
e
ad
j
u
s
ted
b
y
u
s
i
n
g
d
ela
y
s
,
in
p
u
t d
ela
y
s
a
n
d
f
ee
d
b
ac
k
d
ela
y
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
I
SS
N:
2252
-
8938
N
a
r
x
B
a
s
ed
S
h
o
r
t Term
W
in
d
P
o
w
er F
o
r
ec
a
s
tin
g
Mo
d
el
(
M.
N
a
n
d
a
n
a
Jy
o
th
i
)
133
Fig
u
r
e
4
.
B
lo
ck
Diag
r
a
m
o
f
P
ar
allel
A
r
c
h
itec
t
u
r
e
2
.
2
.
Ste
ps
f
o
r
t
r
a
in t
he
Neura
l N
et
wo
rk
B
lo
ck
d
iag
r
a
m
o
f
Ma
tlab
co
d
e
in
Fi
g
u
r
e
5
.
Fig
u
r
e
5
.
B
lo
ck
Diag
r
a
m
o
f
M
atlab
C
ode
2
.
3
.
Alg
o
rit
h
m
o
f
Neura
l N
et
w
o
rk
NARX
n
eu
r
al
n
et
w
o
r
k
i
s
u
s
ed
to
s
o
lv
e
a
ti
m
e
s
er
ie
s
p
r
o
b
lem
.
[
X,
T
]
=
s
i
m
p
leser
ies_
d
ataset;
n
et
=
n
ar
x
n
et(
1
:2
,
1
:2
,
1
0
)
[
Xs,Xi,
A
i,T
s
]
=
p
r
e
p
ar
ets(n
et,
X,
{}
,
T
)
n
et
=
tr
ain
(
n
et,
Xs,T
s
,
Xi,
A
i)
;
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
4
,
No
.
4
,
Dec
em
b
er
2
0
1
5
:
1
2
9
–
1
3
8
134
Fig
u
r
e
6
.
Neu
r
al
Net
w
o
r
k
T
r
ain
in
g
D
ia
g
r
a
m
V
ie
w
(
n
et)
Fig
u
r
e
7
.
NARX Ne
t
w
o
r
k
V
ie
w
Y
=
n
et(
Xs,Xi,
A
i)
;
p
er
f
=
p
er
f
o
r
m
(
n
et,
T
s
,
Y)
Her
e
th
e
N
A
R
X
n
et
w
o
r
k
i
s
s
i
m
u
lated
in
clo
s
ed
lo
o
p
f
o
r
m
.
n
etc
=
clo
s
elo
o
p
(
n
et)
;
v
ie
w
(
n
e
tc)
Fig
u
r
e
8
.
C
lo
s
ed
lo
o
p
o
f
NA
R
X
Net
w
o
r
k
Vie
w
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
I
SS
N:
2252
-
8938
N
a
r
x
B
a
s
ed
S
h
o
r
t Term
W
in
d
P
o
w
er F
o
r
ec
a
s
tin
g
Mo
d
el
(
M.
N
a
n
d
a
n
a
Jy
o
th
i
)
135
[
Xs,Xi,
A
i,T
s
]
=
p
r
e
p
ar
ets(n
etc,
X,
{}
,
T
)
;
y
=
n
etc(
X
s
,
Xi,
A
i)
Her
e
th
e
N
A
R
X
n
et
w
o
r
k
i
s
u
s
ed
to
p
r
e
d
ict
th
e
n
ex
t o
u
tp
u
t a
ti
m
e
s
tep
ah
ea
d
o
f
w
h
e
n
it
w
i
ll
ac
tu
all
y
ap
p
ea
r
.
n
etp
=
r
e
m
o
v
ed
ela
y
(
n
et)
;
v
ie
w
(
n
e
tp
)
Fig
u
r
e
9
.
P
r
ed
ictio
n
o
f
Nex
t O
u
tp
u
t o
f
C
lo
s
ed
L
o
o
p
o
f
NAR
X
Net
w
o
r
k
Vie
w
[
Xs,Xi,
A
i,T
s
]
=
p
r
ep
a
r
ets(n
etp
,
X,
{}
,
T
)
;
y
=
n
etp
(
Xs,Xi,
Ai)
3.
RE
SU
L
T
S
A
ND
AN
AL
Y
SI
S
3
.
1
.
P
er
f
o
rm
a
nce
o
f
t
he
AN
N
First
th
e
co
llected
d
ata
is
p
r
ep
r
o
ce
s
s
ed
i.e
.
n
o
r
m
alize
d
an
d
th
en
g
i
v
e
n
as
in
p
u
ts
,
w
h
ic
h
ar
e
te
m
p
er
atu
r
es,
p
r
es
s
u
r
e,
air
d
e
n
s
it
y
,
s
p
ee
d
an
d
p
o
w
er
g
e
n
er
ated
is
g
i
v
e
n
as
o
u
tp
u
t
to
t
h
e
NARX
Ma
t
lab
co
d
e
is
d
ev
elo
p
ed
f
o
r
tr
ain
in
g
.
Af
te
r
t
r
ain
in
g
,
th
e
n
eu
r
al
n
et
w
o
r
k
i
s
r
ea
d
y
f
o
r
th
e
p
r
ed
ictio
n
.
T
h
e
last
2
d
ay
s
d
ata
is
g
iv
e
n
to
t
h
e
n
eu
r
al
n
et
w
o
r
k
a
n
d
t
h
e
p
r
ed
icted
o
u
tp
u
t
is
o
b
tain
ed
.
T
h
e
p
r
ed
icted
o
u
tp
u
t
i
s
co
m
p
ar
ed
to
t
h
e
ca
lcu
lated
p
o
w
er
a
n
d
th
e
p
er
f
o
r
m
a
n
ce
is
m
o
n
ito
r
ed
b
y
ca
lcu
lati
n
g
t
h
e
er
r
o
r
s
b
y
v
ar
io
u
s
m
ea
n
s
o
f
er
r
o
r
ca
lcu
latio
n
s
.
Fig
u
r
e
1
0
.
C
o
llected
Data
P
lo
t
b
et
w
ee
n
air
De
n
s
i
ties
,
T
e
m
p
er
atu
r
es v
s
T
im
e
(
h
o
u
r
s
)
0
100
200
300
400
500
600
700
800
22
24
26
28
30
32
34
36
38
T
i
m
e
-
s
e
i
r
e
s
T
e
m
p
a
r
a
t
u
r
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
4
,
No
.
4
,
Dec
em
b
er
2
0
1
5
:
1
2
9
–
1
3
8
136
Fig
u
r
e
1
1
.
C
o
llected
Data
P
lo
t
b
et
w
ee
n
P
r
ess
u
r
es
v
s
T
im
e
(
h
o
u
r
s
)
Fig
u
r
e
1
2
.
C
o
llected
Data
P
lo
t
b
et
w
ee
n
W
i
n
d
Sp
ee
d
v
s
T
i
m
e
(
h
o
u
r
s
)
a.
Me
an
E
r
r
o
r
(
ME
)
:
I
t is th
e
b
a
s
ic
t
y
p
e
o
f
er
r
o
r
ca
lcu
latio
n
.
I
t is th
e
a
v
er
ag
e
o
f
t
h
e
er
r
o
r
s
.
(
5
)
b.
Me
an
Sq
u
ar
e
E
r
r
o
r
(
MSE
)
: I
t
is
o
n
e
o
f
t
h
e
b
asic t
y
p
es o
f
er
r
o
r
ca
lcu
latio
n
.
I
t is t
h
e
av
er
ag
e
o
f
th
e
s
q
u
ar
es
o
f
t
h
e
er
r
o
r
s
.
(
6
)
c.
R
o
o
t M
ea
n
Sq
u
ar
e
E
r
r
o
r
(
R
MSE
):
R
MSE
is
t
h
e
s
ta
n
d
ar
d
d
e
v
iatio
n
o
f
th
e
d
i
f
f
er
en
ce
s
b
et
wee
n
p
r
ed
icted
v
alu
e
s
an
d
ac
t
u
al
v
a
lu
e
s
.
I
t is t
h
e
s
q
u
ar
e
r
o
o
t o
f
th
e
av
er
a
g
e
o
f
s
q
u
ar
es o
f
t
h
e
er
r
o
r
s
.
(
7
)
W
h
er
e
N
-
No
.
o
f
s
a
m
p
le
s
T
-
A
ct
u
al
Ou
tp
u
t
P
-
P
r
ed
icted
Ou
tp
u
t
P
r
ed
ictio
n
is
ca
r
r
ied
o
u
t
b
y
v
ar
y
in
g
t
h
e
d
ela
y
s
o
f
th
e
i
n
p
u
t
an
d
also
th
e
n
u
m
b
er
o
f
n
e
u
r
o
n
s
in
th
e
h
id
d
en
la
y
er
.
T
h
e
er
r
o
r
s
at
d
if
f
er
en
t d
ela
y
s
a
n
d
d
if
f
er
en
t
n
u
m
b
er
o
f
n
eu
r
o
n
s
i
n
t
h
e
h
id
d
en
la
y
er
ar
e
T
ab
le
1
.
T
h
e
P
er
f
o
r
m
an
ce
o
f
t
h
e
P
r
ed
icted
A
NN
m
o
d
el
A
N
N
D
e
l
a
y
ME
M
S
E
R
M
S
E
N
R
M
S
E
S
t
a
n
d
a
r
d
D
e
v
i
a
t
i
o
n
3
2
0
.
6
5
0
.
6
4
0
.
8
0
.
3
9
0
.
1
6
4
0
.
9
8
0
.
9
9
0
.
9
3
0
.
5
7
0
.
5
3
6
0
.
2
5
0
.
7
6
0
.
2
7
0
.
7
1
0
.
0
9
8
0
.
9
1
0
.
9
9
0
.
9
0
.
5
4
0
.
2
6
5
2
0
.
6
1
0
.
5
8
0
.
7
6
0
.
4
9
0
.
1
1
4
0
.
4
3
0
.
1
6
0
.
4
6
0
.
6
4
0
.
9
6
0
.
1
6
0
.
5
1
0
.
2
2
0
.
4
3
0
.
2
8
0
.
6
7
0
.
7
7
0
.
8
8
0
.
4
0
.
2
10
2
0
.
6
8
0
.
6
9
0
.
8
3
0
.
4
8
0
.
1
4
4
0
.
7
4
0
.
8
2
0
.
9
0
.
4
5
0
.
9
6
0
.
2
4
0
.
7
9
0
.
2
8
0
.
5
9
9
0
.
3
1
8
0
.
6
2
0
.
5
8
0
.
7
6
0
.
4
9
0
.
1
1
0
100
200
300
400
500
600
700
800
294
296
298
300
302
304
306
308
310
312
T
i
m
e
i
n
h
o
u
r
s
w
i
n
d
p
r
e
s
s
u
r
e
0
100
200
300
400
500
600
700
800
0
0
.
5
1
1
.
5
2
2
.
5
3
3
.
5
4
4
.
5
W
i
n
d
s
p
e
e
d
i
n
m
/
s
e
c
T
i
m
e
-
s
e
r
i
e
s
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
I
SS
N:
2252
-
8938
N
a
r
x
B
a
s
ed
S
h
o
r
t Term
W
in
d
P
o
w
er F
o
r
ec
a
s
tin
g
Mo
d
el
(
M.
N
a
n
d
a
n
a
Jy
o
th
i
)
137
3
.
2
.
P
er
f
o
rm
a
nce
P
lo
t
s
o
f
t
he
ANN
T
h
e
tr
a
in
ed
n
eu
r
al
n
et
w
o
r
k
L
e
v
en
b
er
g
-
Ma
r
q
u
ar
d
t
(
tr
ain
l
m
)
is
e
m
p
lo
y
ed
f
o
r
t
h
e
p
r
ed
ictio
n
o
f
t
h
e
p
o
w
er
g
e
n
er
ated
f
o
r
th
e
la
s
t
4
8
d
ay
s
b
y
g
i
v
i
n
g
th
e
in
p
u
t
p
ar
a
m
eter
s
b
y
v
ar
y
in
g
t
h
e
i
n
p
u
t
t
i
m
e
d
ela
y
s
an
d
al
s
o
ch
an
g
i
n
g
th
e
n
u
m
b
er
o
f
n
eu
r
o
n
s
i
n
th
e
h
id
d
en
la
y
er
.
I
n
ca
s
e
o
f
3
n
e
u
r
o
n
s
i
n
t
h
e
h
id
d
en
la
y
er
th
e
m
i
n
i
m
u
m
er
r
o
r
is
attai
n
ed
w
h
en
t
h
e
ti
m
e
d
ela
y
g
i
v
e
n
i
s
8
a
n
d
i
n
ca
s
e
o
f
5
n
e
u
r
o
n
s
i
n
t
h
e
h
id
d
en
la
y
er
t
h
e
m
i
n
i
m
u
m
er
r
o
r
a
r
e
attain
ed
w
h
en
t
h
e
d
el
a
y
g
i
v
e
n
is
6
ar
e
s
h
o
w
n
.
Fig
u
r
e
1
3
.
T
h
e
A
NN
p
r
ed
icte
d
o
u
tp
u
t p
lo
t
w
h
en
3
h
id
d
en
la
y
er
n
e
u
r
o
n
s
a
n
d
ti
m
e
d
ela
y
o
f
8
Fig
u
r
e
1
4
.
T
h
e
A
NN
P
r
ed
icted
Ou
tp
u
t P
lo
t
w
h
e
n
5
Hid
d
en
L
a
y
er
Neu
r
o
n
s
w
it
h
T
im
e
D
el
a
y
o
f
6
Fo
r
s
h
o
r
t
ter
m
p
r
ed
ictio
n
o
f
w
i
n
d
p
o
w
er
p
er
s
is
te
n
ce
m
et
h
o
d
s
ee
m
s
to
b
e
th
e
b
en
c
h
m
a
r
k
,
b
u
t
b
y
co
n
s
id
er
in
g
w
in
d
d
ir
ec
tio
n
an
d
u
p
o
n
p
r
o
p
er
tr
ain
in
g
o
f
th
e
n
et
w
o
r
k
t
h
e
A
N
N
m
o
d
el
b
ec
o
m
e
s
m
o
r
e
r
eliab
le
ac
co
r
d
in
g
to
th
e
r
e
s
ea
r
ch
a
n
d
r
esu
lt
s
ar
e
p
r
esen
t
i
n
[
6
]
.
An
d
as
th
e
ti
m
e
p
er
io
d
in
cr
ea
s
es
t
h
e
ef
f
ic
ien
c
y
o
f
th
e
m
et
h
o
d
d
ec
r
ea
s
es
d
r
asti
ca
l
l
y
,
b
u
t
th
e
m
o
d
el
p
r
esen
ted
i
n
th
i
s
p
ap
er
ca
n
ad
o
p
t
its
elf
to
th
e
v
ar
ied
ti
m
e(
s
)
u
p
o
n
s
u
cc
e
s
s
f
u
l
tr
ain
i
n
g
.
T
h
e
m
o
d
el
is
les
s
co
m
p
le
x
,
d
o
esn
’
t
co
n
tain
a
n
y
f
ilter
s
,
w
h
ic
h
m
ak
es
it
c
h
ea
p
er
i
n
r
ea
lizatio
n
o
f
th
e
m
o
d
el
b
u
t
w
it
h
ap
p
r
o
x
i
m
atel
y
s
a
m
e
ef
f
i
cien
cies
w
h
en
co
m
p
ar
ed
to
th
e
o
th
er
m
o
d
els
s
u
c
h
as B
a
y
esia
n
ap
p
r
o
ac
h
,
My
c
iel
s
k
i,
f
u
zz
y
lo
g
ic,
g
r
e
y
m
o
d
el
an
d
etc.
[
1
1
]
-
[
1
2
]
.
4.
CO
NCLU
SI
O
N
T
h
is
p
ap
er
a
d
d
r
ess
es
th
e
p
r
o
b
le
m
a
w
in
d
p
o
w
er
f
o
r
ec
asti
n
g
m
o
d
el
w
it
h
th
e
h
elp
o
f
ar
tifi
cial
n
eu
r
al
n
et
w
o
r
k
s
(
A
NN)
.
I
t
is
d
e
v
elo
p
ed
s
o
th
at
th
e
w
i
n
d
p
o
w
er
c
an
b
e
f
o
r
ec
asted
an
h
o
u
r
b
ef
o
r
e,
w
h
ich
h
elp
s
i
n
m
ai
n
tai
n
in
g
g
r
id
in
ter
co
n
n
ec
t
io
n
an
d
also
s
c
h
ed
u
lin
g
o
f
u
n
its
.
T
h
is
p
ap
er
n
o
t
o
n
l
y
co
m
p
ar
e
s
th
e
d
i
f
f
er
en
t
la
y
er
s
an
d
d
ela
y
s
b
u
t
al
s
o
p
r
o
v
id
es
co
m
p
ar
ati
v
e
an
al
y
s
is
o
f
d
if
f
er
e
n
t
m
o
d
els.
Am
o
n
g
t
h
e
d
if
f
er
e
n
t
m
e
th
o
d
s
,
th
e
p
r
o
p
o
s
ed
is
m
o
r
e
r
eliab
le
f
o
r
an
y
t
y
p
e
o
f
p
r
ed
ictio
n
a
n
d
ch
ea
p
est
s
o
lu
t
io
n
.
O
f
all,
th
e
m
o
d
el
w
it
h
h
id
d
en
la
y
er
s
ize
3
an
d
d
elay
6
an
d
also
th
e
m
o
d
el
w
it
h
h
id
d
en
la
y
er
s
ize
5
an
d
d
elay
6
h
av
e
s
h
o
w
n
b
est
p
er
f
o
r
m
a
n
c
e
a
m
o
n
g
all
t
h
e
co
m
b
in
at
io
n
s
.
0
100
200
300
400
500
600
700
800
-
5
0
0
50
100
150
200
250
T
i
m
e
s
e
r
i
e
s
w
i
n
d
p
o
w
e
r
O
r
i
g
i
n
a
l
T
a
r
g
e
t
s
N
e
t
w
o
r
k
P
r
e
d
i
c
t
i
o
n
s
E
x
p
e
c
t
e
d
O
u
t
p
u
t
s
0
100
200
300
400
500
600
700
800
-
5
0
0
50
100
150
200
250
T
i
m
e
s
e
r
i
e
s
w
i
n
d
p
o
w
e
r
O
r
i
g
i
n
a
l
T
a
r
g
e
t
s
N
e
t
w
o
r
k
P
r
e
d
i
c
t
i
o
n
s
E
x
p
e
c
t
e
d
O
u
t
p
u
t
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
4
,
No
.
4
,
Dec
em
b
er
2
0
1
5
:
1
2
9
–
1
3
8
138
RE
F
E
R
E
NC
E
S
[1
]
Zh
a
o
,
X
W
a
n
g
,
S
X
L
i,
T
.
Re
v
i
e
w
o
f
e
v
a
lu
a
ti
o
n
c
rit
e
ria
a
n
d
m
a
in
m
e
th
o
d
s
o
f
w
in
d
p
o
w
e
r
f
o
re
c
a
stin
g
.
En
e
rg
y
Pro
c
e
d
ia
.
2
0
1
1
;
1
2
:
7
6
1
–
7
6
9
.
[2
]
Ja
e
su
n
g
Ju
n
g
,
Ro
b
e
rt
P
Br
o
a
d
w
a
ter.
Cu
rre
n
t
sta
tu
s
a
n
d
f
u
tu
re
a
d
v
a
n
c
e
s
f
o
r
w
in
d
sp
e
e
d
a
n
d
p
o
w
e
r
f
o
re
c
a
stin
g
.
Ren
e
wa
b
le
a
n
d
S
u
sta
in
a
b
le E
n
e
rg
y
Rev
iews
.
2
0
1
4
;
3
1
:
7
6
2
-
7
7
7
.
[3
]
Ya
o
Zh
a
n
g
,
Jia
n
x
u
e
W
a
n
g
,
Xif
a
n
W
a
n
g
.
Re
v
ie
w
o
n
p
ro
b
a
b
il
isti
c
f
o
re
c
a
stin
g
o
f
w
in
d
p
o
w
e
r
g
e
n
e
ra
t
io
n
.
Ren
e
wa
b
le
a
n
d
S
u
sta
i
n
a
b
le E
n
e
rg
y
Rev
iews
.
2
0
1
4
;
3
2
:
2
5
5
-
2
7
0
.
[4
]
A
Tas
c
i
k
a
ra
o
g
lu
,
M
Uz
u
n
o
g
lu
.
A
re
v
ie
w
o
f
c
o
m
b
in
e
d
a
p
p
ro
a
c
h
e
s
f
o
r
p
re
d
ictio
n
o
f
sh
o
rt
-
term
w
in
d
sp
e
e
d
a
n
d
p
o
w
e
r
.
Ren
e
wa
b
le
a
n
d
S
u
sta
in
a
b
l
e
En
e
rg
y
Rev
iew.
2
0
1
4
;
3
4
:
2
4
3
-
2
5
4
.
[5
]
Bh
a
sk
a
r
K,
S
in
g
h
S
N.
A
W
NN
-
A
s
siste
d
W
in
d
P
o
w
e
r
F
o
re
c
a
stin
g
Us
in
g
F
e
e
d
-
F
o
rw
a
rd
Ne
u
ra
l
Ne
tw
o
rk
.
IEE
E
T
ra
n
sa
c
ti
o
n
s
o
n
S
u
sta
i
n
a
b
le E
n
e
rg
y
.
2
0
1
2
;
3
:
3
0
6
-
3
1
5
.
[6
]
W
e
n
-
Ye
a
u
Ch
a
n
g
.
S
h
o
rt
-
T
e
rm
W
in
d
P
o
w
e
r
F
o
re
c
a
stin
g
Us
in
g
th
e
En
h
a
n
c
e
d
P
a
rti
c
le
S
w
a
r
m
Op
t
im
iz
a
ti
o
n
Ba
se
d
H
y
b
rid
M
e
th
o
d
.
E
n
e
rg
ies
.
2
0
1
3
;
6
:
4
8
7
9
-
4
8
9
6
.
[7
]
Ca
talã
o
,
JP
S
Os
ó
ri
o
,
G
J
P
o
u
si
n
h
o
,
HM
I.
S
h
o
rt
-
T
e
rm
W
in
d
Po
we
r
Fo
re
c
a
stin
g
Us
in
g
a
Hy
b
ri
d
Evo
lu
t
io
n
a
ry
In
telli
g
e
n
t
A
p
p
ro
a
c
h
.
In
P
ro
c
e
e
d
in
g
s
o
f
th
e
1
6
th
In
tern
a
ti
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
In
telli
g
e
n
t
S
y
ste
m
A
p
p
li
c
a
ti
o
n
t
o
P
o
w
e
r
S
y
ste
m
s (IS
A
P
),
He
rso
n
is
so
s,
G
re
e
c
e
.
2
0
1
1
;
2
5
:
1
–
5
.
[8
]
P
a
n
Zh
a
o
,
Jia
n
g
f
e
n
g
W
a
n
g
,
J
u
n
r
o
n
g
X
ia,
Yi
n
g
x
in
S
h
e
n
g
,
J
ie
Yu
e.
P
e
rf
o
rm
a
n
c
e
e
v
a
lu
ti
o
n
a
n
d
a
c
c
u
ra
c
y
e
n
h
a
n
c
e
m
e
n
t
o
f
a
d
a
y
-
a
h
e
a
d
w
in
d
p
o
w
e
r
f
o
re
c
a
stin
g
in
c
h
in
a
.
Ren
e
wa
b
le E
n
e
rg
y
.
2
0
1
2
;
4
3
:
2
3
4
-
2
4
1
.
[9
]
M
Ca
ro
li
n
M
a
b
e
l
,
E
F
e
rn
a
n
d
e
z
.
A
n
a
l
y
sis
o
f
w
in
d
p
o
w
e
r
g
e
n
e
ra
ti
o
n
a
n
d
p
re
d
icti
o
n
u
sin
g
A
NN
:
A
c
a
se
stu
d
y
.
Ren
e
wa
b
le E
n
e
rg
y
.
2
0
0
8
;
3
3
:
9
8
6
-
9
9
2
.
[1
0
]
H
L
iu
,
HQ
T
ian
,
C
Ch
e
n
,
Y
L
i.
A
h
y
b
rid
sta
ti
stica
l
m
e
th
o
d
to
p
re
d
ict
w
in
d
sp
e
e
d
a
n
d
w
in
d
p
o
w
e
r.
R
e
n
e
wa
b
l
e
En
e
rg
y
.
2
0
1
0
;
3
5
:
1
8
5
7
-
1
8
6
1
.
[1
1
]
IJ
Ra
m
irez
-
Ro
sa
d
o
,
LA
F
e
rn
a
n
d
e
z
-
Ji
m
e
n
e
z
,
C
M
o
n
teiro
,
J
S
o
u
sa
,
R
Be
ss
a
.
Co
m
p
a
riso
n
o
f
t
w
o
n
e
w
sh
o
rt
-
ter
m
w
in
d
-
p
o
w
e
r
f
o
re
c
a
stin
g
s
y
ste
m
s
.
Ren
e
wa
b
le
En
e
rg
y
.
2
0
0
9
;
3
4
:
1
8
4
8
-
1
8
5
4
.
[1
2
]
P
F
l
o
re
s,
A
T
a
p
ia,
G
T
a
p
ia.
Ap
p
li
c
a
ti
o
n
o
f
a
c
o
n
tr
o
l
a
lg
o
rit
h
m
f
o
r
w
in
d
sp
e
e
d
p
re
d
ict
io
n
a
n
d
a
c
ti
v
e
p
o
w
e
r
g
e
n
e
ra
ti
o
n
.
Ren
e
w
a
b
le E
n
e
rg
y
2
0
0
5
;
3
0
:
5
2
3
-
5
3
6
.
[1
3
]
F
O
Ho
c
a
o
g
lu
,
M
F
id
a
n
,
O.
Ge
re
k
.
M
y
c
ielsk
i
a
p
p
ro
a
c
h
f
o
r
w
in
d
sp
e
e
d
p
re
d
ictio
n
.
E
n
e
rg
y
Co
n
v
e
rs
io
n
a
n
d
M
a
n
a
g
e
me
n
t.
2
0
0
9
;
5
0
:
1
4
3
6
-
1
4
4
3
.
[1
4
]
M
M
o
n
f
a
re
d
,
H
Ra
ste
g
a
r,
H
M
Ko
jab
a
d
i.
A
n
e
w
stra
te
g
y
f
o
r
win
d
sp
e
e
d
f
o
re
c
a
stin
g
u
sin
g
a
rti
f
icia
l
in
telli
g
e
n
t
m
e
th
o
d
s.
Ren
e
wa
b
le E
n
e
rg
y
.
2
0
0
9
;
3
4
:
8
4
5
-
8
4
8
.
[1
5
]
M
Na
n
d
a
n
a
Jy
o
th
i,
V
Di
n
a
k
a
r
.
S
h
o
rt
-
term
W
in
d
S
p
e
e
d
F
o
re
c
a
stin
g
th
ro
u
g
h
A
NN
.
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
A
p
p
li
e
d
En
g
i
n
e
e
rin
g
Res
e
a
rc
h
.
2
0
1
5
;
1
0
;
2
1
4
7
5
-
2
1
4
8
6
.
Evaluation Warning : The document was created with Spire.PDF for Python.