I
nte
rna
t
io
na
l J
o
urna
l o
f
P
o
wer
E
lect
ro
nics
a
nd
Driv
e
S
y
s
t
em
s
(
I
J
P
E
DS)
Vo
l.
1
5
,
No
.
4
,
Dec
em
b
er
20
2
4
,
p
p
.
2
5
5
3
~
2
5
6
9
I
SS
N:
2
0
8
8
-
8
6
9
4
,
DOI
: 1
0
.
1
1
5
9
1
/ijp
ed
s
.
v
1
5
.i
4
.
p
p
2
5
5
3
-
2
5
6
9
2553
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
p
e
d
s
.
ia
esco
r
e.
co
m
Foreca
sting
hourl
y
sho
rt
-
ter
m so
la
r photo
v
o
ltaic po
wer using
ma
chine learning
mo
dels
Sra
v
a
nk
um
a
r
J
o
g
un
uri
1,
2
, F.
T.
J
o
s
h
1
,
J.
J
ency
J
o
s
eph
3
,
R.
M
ee
na
l
4
,
R.
M
o
ha
n Da
s
5
,
S.
K
a
nn
a
dh
a
s
a
n
6
1
D
i
v
i
si
o
n
o
f
E
l
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
,
K
a
r
u
n
y
a
I
n
st
i
t
u
t
e
o
f
T
e
c
h
n
o
l
o
g
y
a
n
d
S
c
i
e
n
c
e
s
,
C
o
i
m
b
a
t
o
r
e
,
I
n
d
i
a
2
D
e
p
a
r
t
me
n
t
o
f
R
e
n
e
w
a
b
l
e
E
n
e
r
g
y
En
g
i
n
e
e
r
i
n
g
,
C
o
l
l
e
g
e
o
f
A
g
r
i
c
u
l
t
u
r
a
l
E
n
g
i
n
e
e
r
i
n
g
a
n
d
Te
c
h
n
o
l
o
g
y
,
A
n
a
n
d
A
g
r
i
c
u
l
t
u
r
a
l
U
n
i
v
e
r
si
t
y
,
G
u
j
a
r
a
t
,
I
n
d
i
a
3
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 E
n
g
i
n
e
e
r
i
n
g
,
S
r
i
K
r
i
s
h
n
a
C
o
l
l
e
g
e
o
f
T
e
c
h
n
o
l
o
g
y
,
C
o
i
m
b
a
t
o
r
e
,
I
n
d
i
a
4
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 E
n
g
i
n
e
e
r
i
n
g
,
V
.
S
.
B
E
n
g
i
n
e
e
r
i
n
g
C
o
l
l
e
g
e
,
K
a
r
u
r
,
I
n
d
i
a
5
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 E
n
g
i
n
e
e
r
i
n
g
,
N
e
w
H
o
r
i
z
o
n
C
o
l
l
e
g
e
o
f
E
n
g
i
n
e
e
r
i
n
g
,
B
e
n
g
a
l
u
r
u
,
I
n
d
i
a
6
D
e
p
a
r
t
me
n
t
o
f
El
e
c
t
r
o
n
i
c
s a
n
d
C
o
m
mu
n
i
c
a
t
i
o
n
En
g
i
n
e
e
r
i
n
g
,
S
t
u
d
y
W
o
r
l
d
C
o
l
l
e
g
e
o
f
E
n
g
i
n
e
e
r
i
n
g
,
C
o
i
mb
a
t
o
r
e
,
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
Dec
1
8
,
2
0
2
3
R
ev
is
ed
Ma
y
2
2
,
2
0
2
4
Acc
ep
ted
J
u
n
2
,
2
0
2
4
F
o
re
c
a
stin
g
so
lar
p
h
o
t
o
v
o
lt
a
ic
p
o
we
r
e
n
su
re
s
a
sta
b
le
a
n
d
d
e
p
e
n
d
a
b
le
p
o
we
r
g
rid
.
G
iv
e
n
it
s d
e
p
e
n
d
e
n
c
e
o
n
st
o
c
h
a
stic
we
a
th
e
r
c
o
n
d
it
i
o
n
s,
p
re
d
i
c
ti
n
g
so
lar
p
h
o
to
v
o
lt
a
ic
p
o
we
r
a
c
c
u
ra
tely
d
e
m
a
n
d
s
a
p
p
ly
in
g
in
telli
g
e
n
t
a
n
d
so
p
h
isti
c
a
ted
tec
h
n
iq
u
e
s
c
a
p
a
b
le
o
f
h
a
n
d
li
n
g
it
s
in
h
e
re
n
t
n
o
n
li
n
e
a
rit
y
a
n
d
v
o
latil
it
y
.
Co
n
tr
o
ll
i
n
g
e
lec
tri
c
a
l
e
n
e
rg
y
s
o
u
r
c
e
s
is
a
n
imp
o
rta
n
t
stra
te
g
y
fo
r
re
a
c
h
in
g
t
h
is
e
n
e
rg
y
b
a
lan
c
e
b
e
c
a
u
se
g
rid
o
p
e
ra
to
rs
o
fte
n
h
a
v
e
n
o
c
o
n
tr
o
l
o
v
e
r
u
s
e
p
a
tt
e
rn
s.
Ac
c
u
ra
tely
fo
re
c
a
stin
g
p
h
o
to
v
o
l
taic
(P
V)
p
o
we
r
g
e
n
e
ra
ti
o
n
fro
m
h
ig
h
ly
in
teg
ra
ted
s
o
lar
p
lan
ts
to
t
h
e
g
ri
d
i
s
e
ss
e
n
ti
a
l
fo
r
g
ri
d
sta
b
il
it
y
.
Th
is
s
tu
d
y
a
ims
to
imp
r
o
v
e
fo
re
c
a
stin
g
a
c
c
u
ra
c
y
a
n
d
m
a
k
e
a
c
c
u
ra
te
p
re
d
ictio
n
s
o
f
s
o
lar
p
o
we
r
o
u
t
p
u
t
fro
m
th
e
se
lec
ted
g
ri
d
-
c
o
n
n
e
c
ted
P
V
s
y
ste
m
.
In
t
h
is
st
u
d
y
,
t
h
e
we
a
th
e
r
d
a
ta
wa
s
c
o
ll
e
c
ted
o
n
-
site
a
n
d
r
e
c
o
rd
e
d
P
V
p
o
we
r
fro
m
a
2
0
k
W
o
n
-
g
r
id
sy
ste
m
fo
r
o
n
e
y
e
a
r,
a
n
d
d
iffere
n
t
m
a
c
h
i
n
e
lea
rn
i
n
g
tec
h
n
i
q
u
e
s
li
k
e
d
e
e
p
n
e
u
ra
l
n
e
two
rk
s,
ra
n
d
o
m
f
o
re
sts,
a
n
d
a
rti
ficia
l
n
e
u
ra
l
n
e
two
rk
s
we
re
e
v
a
lu
a
ted
a
n
d
b
e
n
c
h
m
a
rk
e
d
a
g
a
in
st
re
fe
r
e
n
c
e
su
p
p
o
rt
v
e
c
to
r
re
g
re
ss
io
n
m
o
d
e
l.
Wi
t
h
imp
ro
v
e
m
e
n
ts
in
f
o
re
c
a
stin
g
a
c
c
u
ra
c
y
o
f
2
t
o
3
7
%
o
v
e
r
t
h
e
re
fe
re
n
c
e
m
o
d
e
l
a
t
stu
d
y
lo
c
a
ti
o
n
(
2
2
.
7
8
°
N,
7
3
.
6
5
°
E)
,
Co
l
leg
e
o
f
A
g
ricu
lt
u
ra
l
E
n
g
i
n
e
e
rin
g
a
n
d
Tec
h
n
o
lo
g
y
,
An
a
n
d
Ag
ricu
l
t
u
ra
l
Un
iv
e
rsit
y
,
G
o
d
h
ra
,
I
n
d
ia,
s
imu
latio
n
re
su
lt
s
sh
o
we
d
th
a
t
th
e
ra
n
d
o
m
fo
re
st
tec
h
n
iq
u
e
is
e
ffe
c
ti
v
e
f
o
r
t
h
e
f
o
re
c
a
stin
g
h
o
rizo
n
s o
f
1
to
4
h
o
u
rs.
K
ey
w
o
r
d
s
:
Ma
ch
in
e
lear
n
in
g
R
an
d
o
m
f
o
r
est
Sh
o
r
t
-
ter
m
f
o
r
ec
asti
n
g
Sk
ill s
co
r
e
So
lar
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
:
Sra
v
an
k
u
m
a
r
J
o
g
u
n
u
r
i
Div
is
io
n
o
f
E
lectr
ical
an
d
E
le
ctr
o
n
ics E
n
g
in
ee
r
in
g
,
Kar
u
n
y
a
I
n
s
titu
te
o
f
T
ec
h
n
o
lo
g
y
an
d
S
cien
ce
s
C
o
im
b
ato
r
e,
I
n
d
ia
E
m
ail:
s
r
av
an
jo
g
u
n
u
r
i@
g
m
ail
.
co
m
1.
I
NT
RO
D
UCT
I
O
N
Har
n
ess
in
g
p
h
o
to
v
o
ltaic
(
PV)
p
o
wer
h
o
ld
s
p
r
o
m
is
e
in
ad
d
r
ess
in
g
th
e
in
c
r
ea
s
in
g
g
lo
b
al
d
em
an
d
f
o
r
clea
n
en
er
g
y
d
u
e
to
its
r
en
ewa
b
le
n
atu
r
e,
ec
o
-
f
r
ien
d
lin
ess
,
an
d
v
er
s
atility
as
a
d
is
tr
ib
u
ted
en
er
g
y
s
o
u
r
ce
[
1
]
.
Po
wer
g
r
id
s
co
m
m
o
n
l
y
co
n
s
is
t o
f
p
o
wer
p
lan
ts
th
at
g
en
er
ate
co
n
s
is
ten
t e
n
er
g
y
f
lo
ws,
in
clu
d
in
g
co
al,
g
as,
an
d
n
u
clea
r
f
ac
ilit
ies,
alo
n
g
s
id
e
t
h
o
s
e
p
r
o
d
u
cin
g
f
l
u
ctu
atin
g
en
er
g
y
s
o
u
r
ce
s
,
lik
e
win
d
an
d
p
h
o
t
o
v
o
ltaic
p
la
n
ts
.
T
h
e
latter
's
o
u
tp
u
t
h
in
g
es
o
n
l
o
ca
l
we
ath
er
co
n
d
itio
n
s
at
s
p
ec
if
ic
tim
es
an
d
l
o
ca
tio
n
s
.
Pre
s
er
v
i
n
g
g
r
id
s
tab
ilit
y
n
ec
ess
itates
eq
u
ilib
r
iu
m
b
etwe
en
en
er
g
y
p
r
o
d
u
ce
d
b
y
s
o
u
r
ce
s
an
d
th
at
co
n
s
u
m
ed
b
y
u
s
er
s
.
S
u
s
tain
ed
ex
p
an
s
io
n
with
in
th
e
en
er
g
y
s
ec
to
r
o
f
an
y
n
atio
n
is
a
cr
u
cial
co
m
p
o
n
e
n
t
f
o
r
b
o
th
ec
o
n
o
m
ic
an
d
tech
n
o
lo
g
ical
p
r
o
g
r
ess
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2088
-
8
6
9
4
I
n
t J Po
w
E
lec
&
Dr
i Sy
s
t
,
Vo
l.
1
5
,
No
.
4
,
Dec
em
b
er
20
2
4
:
2553
-
2
5
6
9
2554
[
2
]
.
As
tech
n
o
l
o
g
ical
ad
v
a
n
ce
m
en
ts
ac
ce
ler
ate,
en
er
g
y
co
n
s
u
m
p
tio
n
is
ex
p
er
ien
cin
g
a
m
u
c
h
m
o
r
e
r
ap
id
g
r
o
wth
co
m
p
ar
ed
to
th
e
in
c
r
ea
s
e
in
p
o
p
u
latio
n
s
ize
[
3
]
.
C
u
r
r
e
n
tly
,
en
s
u
r
in
g
e
n
er
g
y
b
alan
ce
lar
g
e
ly
r
ev
o
lv
es
ar
o
u
n
d
r
eg
u
latin
g
elec
tr
ical
en
er
g
y
s
o
u
r
ce
s
d
u
e
to
c
o
n
s
u
m
er
co
n
s
u
m
p
tio
n
ly
i
n
g
m
o
s
tly
o
u
ts
id
e
g
r
id
o
p
e
r
ato
r
s
'
co
n
tr
o
l
[
4
]
,
[
5
]
.
So
lar
p
o
wer
p
r
ed
ictio
n
h
o
l
d
s
p
iv
o
tal
s
ig
n
if
ican
ce
in
s
h
ap
in
g
th
e
f
u
tu
r
e
o
f
r
en
ewa
b
le
en
er
g
y
f
ac
ilit
ies
an
d
th
eir
ex
ten
s
iv
e
in
te
g
r
atio
n
in
to
g
r
id
s
.
T
h
e
p
r
ec
is
io
n
o
f
p
h
o
to
v
o
ltaic
p
o
wer
g
e
n
er
atio
n
f
o
r
ec
asts
is
h
ig
h
ly
r
ely
in
g
o
n
th
e
d
y
n
am
ic
a
n
d
e
v
er
-
ch
an
g
in
g
wea
th
er
co
n
d
itio
n
s
[
6
]
,
[
7
]
.
Pre
cise
s
o
lar
p
o
wer
p
r
ed
ictio
n
is
ess
en
tial
to
m
in
im
ize
e
n
er
g
y
c
o
s
ts
an
d
u
p
h
o
l
d
p
r
em
i
u
m
p
o
wer
s
u
p
p
ly
with
in
elec
tr
ical
g
r
id
s
d
ep
en
d
e
n
t
o
n
d
is
p
er
s
ed
p
h
o
to
v
o
ltaic
g
en
er
atio
n
.
Fo
r
h
o
u
s
eh
o
ld
s
an
d
s
m
all
b
u
s
in
ess
es
u
tili
zin
g
o
n
-
s
ite
p
h
o
to
v
o
ltaic
g
e
n
er
atio
n
,
d
ir
ec
tly
ac
ce
s
s
in
g
h
is
to
r
ical
ir
r
ad
ian
ce
d
ata
is
ch
allen
g
in
g
d
u
e
to
co
s
tly
s
o
lar
ir
r
ad
ia
n
ce
m
eter
s
.
Ho
wev
er
,
wh
ile
lo
ca
l
m
eteo
r
o
lo
g
ical
in
s
titu
tio
n
s
h
a
v
e
en
h
an
ce
d
wea
th
er
f
o
r
ec
asti
n
g
s
er
v
ices,
p
r
o
v
id
in
g
o
n
lin
e
in
f
o
r
m
atio
n
s
u
ch
as
t
em
p
er
atu
r
e,
d
ew
p
o
i
n
t,
h
u
m
i
d
i
ty
,
v
is
ib
ilit
y
,
win
d
s
p
ee
d
,
an
d
c
o
m
p
r
eh
en
s
iv
e
we
ath
er
s
u
m
m
ar
ies,
d
ata
f
o
r
s
o
la
r
p
o
wer
f
o
r
ec
asti
n
g
is
f
r
e
q
u
en
t
ly
u
n
av
ailab
le
[
8
]
.
O
n
e
o
f
t
h
e
m
o
s
t
i
m
p
o
r
t
a
n
t
s
o
u
r
c
e
s
o
f
e
l
e
c
t
r
i
c
i
t
y
o
n
t
h
e
g
r
i
d
i
s
s
o
l
a
r
p
o
w
e
r
,
a
n
d
f
o
r
i
t
t
o
b
e
u
s
e
d
t
o
i
t
s
f
u
l
l
p
o
t
e
n
t
i
a
l
,
p
r
e
c
i
s
e
i
n
f
o
r
m
a
t
i
o
n
o
n
t
h
e
q
u
a
n
t
i
t
y
o
f
s
o
l
a
r
p
o
w
e
r
t
h
a
t
w
i
l
l
b
e
p
r
o
d
u
c
e
d
f
r
o
m
v
a
r
i
o
u
s
s
o
u
r
c
e
s
a
n
d
a
t
v
a
r
i
o
u
s
t
i
m
e
s
–
m
i
n
u
t
e
s
,
h
o
u
r
s
,
a
n
d
d
a
y
s
–
i
s
r
e
q
u
i
r
e
d
.
T
h
e
f
o
r
e
c
a
s
t
h
o
r
i
z
o
n
p
e
r
t
a
i
n
s
t
o
t
h
e
t
i
m
e
i
n
t
e
r
v
a
l
t
h
a
t
l
i
e
s
b
e
t
w
e
e
n
t
h
e
p
r
e
s
e
n
t
m
o
m
e
n
t
o
f
p
r
e
d
i
c
t
i
o
n
a
n
d
t
h
e
f
o
r
t
h
c
o
m
i
n
g
p
e
r
i
o
d
e
a
r
m
a
r
k
e
d
f
o
r
o
u
t
p
u
t
p
r
e
d
i
c
t
i
o
n
.
S
o
m
e
r
e
s
e
a
r
c
h
e
r
s
c
l
a
s
s
i
f
y
t
h
e
f
o
r
e
c
a
s
t
h
o
r
i
z
o
n
i
n
t
o
t
h
r
e
e
d
i
s
t
i
n
c
t
c
a
t
e
g
o
r
i
e
s
[
9
]
:
s
h
o
r
t
-
t
e
r
m
,
m
e
d
i
u
m
-
t
e
r
m
,
a
n
d
l
o
n
g
-
t
e
r
m
.
I
n
th
e
r
ea
lm
o
f
p
o
wer
s
y
s
tem
s
an
d
s
m
ar
t
g
r
id
p
lan
n
in
g
,
v
e
r
y
s
h
o
r
t
-
ter
m
an
d
s
h
o
r
t
-
ter
m
f
o
r
e
ca
s
tin
g
ar
e
d
ep
lo
y
e
d
,
en
c
o
m
p
ass
in
g
p
r
ed
i
ctio
n
p
er
i
o
d
s
r
a
n
g
in
g
f
r
o
m
m
e
r
e
s
ec
o
n
d
s
u
p
t
o
less
th
an
3
0
m
in
u
tes.
Sh
o
r
t
-
ter
m
f
o
r
ec
asti
n
g
f
i
n
d
s
u
tili
ty
with
in
th
e
elec
tr
icit
y
m
a
r
k
et,
in
f
lu
en
cin
g
d
ec
is
io
n
s
r
e
g
ar
d
in
g
ec
o
n
o
m
ic
lo
ad
d
is
p
atch
an
d
p
o
wer
s
y
s
tem
o
p
er
atio
n
.
Fu
r
th
er
m
o
r
e
,
it
p
lay
s
a
p
iv
o
tal
r
o
le
in
co
n
tr
o
llin
g
p
o
wer
m
an
ag
em
en
t
s
y
s
tem
s
in
teg
r
ated
with
r
en
ewa
b
le
en
e
r
g
y
s
o
u
r
ce
s
.
T
h
e
tem
p
o
r
al
s
p
an
f
o
r
s
h
o
r
t
-
ter
m
f
o
r
ec
asti
n
g
ty
p
i
ca
lly
f
alls
b
etwe
en
3
0
a
n
d
3
6
0
m
in
u
tes.
Me
d
iu
m
-
ter
m
f
o
r
ec
asti
n
g
,
co
v
e
r
in
g
a
s
p
an
o
f
6
to
2
4
h
o
u
r
s
,
h
o
ld
s
s
ig
n
if
ican
ce
f
o
r
s
ch
ed
u
lin
g
m
ain
te
n
an
ce
ac
tiv
i
ties
with
in
co
n
v
en
tio
n
al
o
r
s
o
lar
en
er
g
y
-
in
teg
r
ate
d
p
o
wer
s
y
s
tem
s
.
T
h
is
in
clu
d
es
s
y
s
tem
s
eq
u
ip
p
ed
with
s
o
p
h
is
ticated
tr
an
s
f
o
r
m
er
s
a
n
d
d
i
v
er
s
e
ele
ctr
o
-
m
ec
h
an
ical
m
a
ch
in
er
y
.
L
o
n
g
-
ter
m
f
o
r
ec
asts
ex
ten
d
th
eir
p
r
e
d
ictio
n
s
b
ey
o
n
d
th
e
2
4
-
h
o
u
r
m
a
r
k
,
f
o
r
ec
asti
n
g
s
ce
n
ar
io
s
th
at
ex
t
en
d
in
to
th
e
d
is
tan
t
f
u
tu
r
e.
T
h
is
p
r
e
d
ictio
n
h
o
r
izo
n
f
in
d
s
a
p
p
licab
ilit
y
in
s
tr
ateg
i
c
asp
ec
ts
s
u
ch
as
lo
n
g
-
ter
m
p
o
wer
g
en
er
atio
n
,
tr
an
s
m
is
s
io
n
,
d
is
tr
ib
u
tio
n
,
an
d
ev
en
s
o
lar
en
er
g
y
r
atio
n
in
g
.
I
t
is
wo
r
th
n
o
tin
g
,
h
o
wev
er
,
th
at
th
e
p
r
ed
ictiv
e
ac
cu
r
ac
y
o
f
th
ese
m
o
d
els
ten
d
s
to
d
im
in
is
h
d
u
e
to
th
e
in
h
er
en
t
ch
allen
g
e
o
f
p
r
ed
ictin
g
wea
th
er
f
lu
ct
u
atio
n
s
s
p
an
n
in
g
o
v
er
a
f
ew
d
ay
s
wh
e
n
u
s
in
g
s
u
ch
e
x
ten
d
e
d
h
o
r
izo
n
s
.
B
ased
o
n
t
h
e
ti
m
e
h
o
r
iz
o
n
s
,
tw
o
m
ai
n
te
ch
n
i
q
u
es
a
r
e
u
s
e
d
t
o
f
o
r
ec
ast
s
o
la
r
e
n
e
r
g
y
:
s
tat
is
ti
ca
l
tim
e
s
er
ies
f
o
r
ec
as
ti
n
g
f
o
r
s
h
o
r
t
-
to
m
i
d
-
te
r
m
i
n
t
e
r
v
als
,
a
n
d
n
u
m
e
r
ic
al
we
ath
e
r
p
r
e
d
i
cti
o
n
f
o
r
l
o
n
g
-
t
o
m
e
d
i
u
m
-
t
er
m
i
n
t
er
v
a
ls
[
8
]
,
[
1
0
]
.
W
i
th
t
h
e
t
r
e
m
e
n
d
o
u
s
im
p
r
o
v
em
e
n
t
in
t
h
e
ir
a
b
i
lit
y
a
n
d
a
cc
u
r
ac
y
t
o
p
r
o
v
id
e
tr
u
s
two
r
t
h
y
f
o
r
ec
asts
,
m
ac
h
i
n
e
l
ea
r
n
i
n
g
-
b
as
ed
(
M
L
)
alg
o
r
it
h
m
s
h
a
v
e
r
e
ce
n
tl
y
e
m
e
r
g
e
d
as
a
tr
u
s
t
wo
r
t
h
y
alt
er
n
ati
v
e
t
o
o
r
s
u
p
p
le
m
en
t
to
n
u
m
e
r
i
ca
l
w
ea
t
h
e
r
p
r
o
j
ec
t
i
o
n
s
(
NW
P
)
i
n
s
o
la
r
e
n
e
r
g
y
p
r
e
d
i
cti
o
n
ch
all
e
n
g
es
[
1
1
]
.
Ma
ch
i
n
e
lea
r
n
i
n
g
is
a
s
u
b
f
i
el
d
o
f
a
r
t
if
ici
al
in
tel
li
g
e
n
c
e
t
h
a
t,
wit
h
o
u
t
t
h
e
u
s
e
o
f
e
x
p
lic
it
p
r
o
g
r
a
m
m
i
n
g
,
u
s
es
d
atas
ets
t
o
cr
ea
t
e
a
n
o
n
li
n
ea
r
m
a
p
p
i
n
g
b
etw
ee
n
i
n
p
u
t
an
d
o
u
t
p
u
t
d
at
a.
W
h
i
le
th
e
l
ite
r
a
t
u
r
e
h
as
u
s
e
d
s
t
atis
t
ica
ll
y
b
as
e
d
m
a
ch
in
e
l
ea
r
n
i
n
g
f
o
r
ec
as
ti
n
g
t
ec
h
n
i
q
u
es,
t
h
e
m
o
s
t
c
o
m
m
o
n
tec
h
n
i
q
u
es
a
r
e
SV
M,
ANN
,
o
r
DNN
,
th
er
e
ar
e
o
n
l
y
a
s
m
a
ll
n
u
m
b
er
o
f
s
tu
d
i
es
t
h
a
t
u
s
e
r
a
n
d
o
m
f
o
r
e
s
t
(
R
F
)
al
g
o
r
it
h
m
s
f
o
r
s
o
la
r
p
h
o
t
o
v
o
l
tai
c
p
o
we
r
f
o
r
e
ca
s
t
in
g
,
a
n
d
t
h
ese
al
g
o
r
it
h
m
s
s
till
r
eq
u
i
r
e
ex
te
n
s
i
v
e
e
x
p
l
o
r
ati
o
n
o
f
t
h
eir
f
o
r
ec
as
ti
n
g
a
cc
u
r
a
c
y
f
o
r
v
a
r
i
o
u
s
s
i
te
-
s
p
ec
if
ic
a
n
d
s
ea
s
o
n
al
d
ata
.
Sar
m
as
et
a
l.
[
1
2
]
,
a
m
eta
-
lear
n
in
g
tech
n
i
q
u
e
to
e
n
h
an
ce
o
n
e
-
hour
-
ah
ea
d
f
o
r
ec
asts
o
f
PV
s
y
s
tem
s
b
y
b
len
d
in
g
p
r
e
d
ictio
n
s
f
r
o
m
d
iv
er
s
e
d
ee
p
lea
r
n
in
g
m
o
d
els wa
s
p
r
esen
ted
.
W
ith
o
u
t r
el
y
in
g
o
n
n
u
m
e
r
ical
wea
th
er
p
r
ed
ictio
n
s
,
th
e
ap
p
r
o
ac
h
d
y
n
am
ically
s
elec
ts
th
e
m
o
s
t
ef
f
e
ctiv
e
m
o
d
el
f
o
r
s
p
ec
if
ic
co
n
d
i
tio
n
s
.
E
v
alu
atio
n
in
L
is
b
o
n
,
Po
r
tu
g
al,
d
em
o
n
s
tr
ated
u
p
to
5
%
im
p
r
o
v
e
m
en
t
o
v
e
r
th
e
b
est
-
p
er
f
o
r
m
in
g
b
ase
m
o
d
el
p
er
s
ite
an
d
u
p
to
4
.
5
%
o
v
er
a
n
e
q
u
al
-
weig
h
ted
co
m
b
i
n
atio
n
o
f
f
o
r
ec
asts
.
T
h
e
s
tu
d
y
e
m
p
lo
y
e
d
in
[
1
3
]
p
r
o
p
o
s
ed
an
o
p
tim
ized
ex
tr
em
e
lear
n
in
g
m
ac
h
in
e
(
E
L
M)
to
f
o
r
ec
ast
r
ea
l
-
tim
e
s
o
la
r
p
o
we
r
g
e
n
er
atio
n
(
SP
G)
in
C
h
h
attis
g
ar
h
,
I
n
d
ia,
in
teg
r
atin
g
wea
th
er
c
o
n
d
itio
n
s
.
B
y
r
ef
in
i
n
g
p
ar
am
eter
s
lik
e
weig
h
ts
an
d
b
iases
,
E
L
M
ex
h
ib
ited
e
n
h
an
c
ed
p
er
f
o
r
m
an
ce
.
C
o
m
p
u
tatio
n
al
tech
n
iq
u
es
ad
ep
t
at
h
an
d
lin
g
h
ig
h
-
d
im
e
n
s
io
n
al
p
r
o
b
lem
s
ar
e
em
p
lo
y
ed
.
T
h
e
co
llab
o
r
atio
n
o
f
m
o
d
if
ied
tea
ch
in
g
-
lear
n
in
g
-
b
ased
o
p
tim
iz
atio
n
(
MT
L
B
O)
with
o
p
tim
ized
E
L
M
im
p
r
o
v
ed
s
o
lar
p
o
wer
g
e
n
er
atio
n
f
o
r
ec
a
s
tin
g
f
o
r
v
a
r
io
u
s
tim
ef
r
am
es,
i
n
clu
d
in
g
o
n
e
h
o
u
r
,
o
n
e
d
ay
,
o
n
e
m
o
n
th
,
an
d
th
r
ee
m
o
n
th
s
a
h
ea
d
,
s
h
o
wca
s
in
g
s
u
p
er
io
r
p
er
f
o
r
m
a
n
ce
in
s
im
u
lat
io
n
s
.
E
x
tr
e
m
e
lear
n
in
g
m
ac
h
i
n
e
tech
n
i
q
u
es
a
n
d
o
th
er
r
ec
en
tly
r
ep
o
r
ted
wo
r
k
s
in
v
o
l
v
e
co
m
p
lex
co
m
p
u
tatio
n
al
tech
n
iq
u
es
an
d
a
b
i
g
v
o
lu
m
e
o
f
d
ata.
I
n
r
ea
l
co
n
d
itio
n
s
,
it
is
n
o
t
p
o
s
s
ib
le
to
g
et
h
u
m
o
n
g
o
u
s
d
ata
f
r
o
m
ea
ch
p
lace
.
T
h
e
r
e
is
a
n
ee
d
to
w
o
r
k
o
n
s
u
ch
tech
n
iq
u
es
wh
ich
will o
f
f
er
b
etter
ac
cu
r
ac
y
with
less
er
d
ata.
I
n
th
e
cu
r
r
en
t
s
tu
d
y
,
u
s
in
g
th
e
en
s
em
b
le
r
an
d
o
m
f
o
r
est
(
R
F)
tech
n
iq
u
e,
a
f
o
r
ec
asti
n
g
m
o
d
el
with
m
u
lti
-
s
tep
s
h
o
r
t
-
ter
m
(
o
n
e
h
o
u
r
to
f
o
u
r
h
o
u
r
s
in
a
d
v
an
ce
)
s
o
la
r
PV
p
o
we
r
f
o
r
ec
asti
n
g
f
o
r
t
h
e
ch
o
s
en
s
ite
was
d
ev
elo
p
e
d
.
Ma
ch
in
e
lear
n
in
g
alg
o
r
ith
m
s
lik
e
s
u
p
p
o
r
t
v
ec
t
o
r
m
ac
h
in
e/r
eg
r
ess
io
n
(
SVM/SVR
)
an
d
ar
tific
ia
l
n
eu
r
al
n
etwo
r
k
o
r
d
ee
p
lear
n
i
n
g
n
e
u
r
al
n
etwo
r
k
s
(
ANN/DN
N)
,
th
e
two
co
m
m
o
n
l
y
u
s
ed
s
t
atis
tical
m
ac
h
in
e
lear
n
in
g
-
b
ased
tech
n
iq
u
es,
wer
e
co
m
p
ar
ed
to
th
e
r
esu
lts
attain
ed
f
r
o
m
th
e
r
a
n
d
o
m
f
o
r
est
te
ch
n
iq
u
e.
A
c
ar
ef
u
l
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Po
w
E
lec
&
Dr
i Sy
s
t
I
SS
N:
2088
-
8
6
9
4
F
o
r
ec
a
s
tin
g
h
o
u
r
ly
s
h
o
r
t
-
term so
la
r
p
h
o
to
vo
lta
ic
p
o
w
er u
s
in
g
ma
ch
in
e
…
(
S
r
a
va
n
ku
ma
r
J
o
g
u
n
u
r
i
)
2555
co
m
p
ar
is
o
n
an
d
ev
alu
atio
n
o
f
th
e
f
o
r
ec
asti
n
g
m
o
d
els
u
s
in
g
v
ar
io
u
s
p
e
r
f
o
r
m
an
ce
m
etr
ics
f
o
r
v
a
r
io
u
s
s
ea
s
o
n
al
in
ter
v
als
o
f
th
e
d
ata
co
n
d
u
cte
d
f
o
r
s
elec
tin
g
an
ap
p
r
o
p
r
iate
f
o
r
ec
asti
n
g
f
o
r
th
e
f
o
r
ec
asti
n
g
h
o
r
izo
n
o
f
1
h
o
u
r
ah
ea
d
to
4
h
o
u
r
s
ah
ea
d
.
T
h
e
r
est
o
f
th
is
p
ap
er
is
d
iv
id
ed
in
to
s
ev
er
al
s
ec
tio
n
s
.
I
n
s
e
ctio
n
2
,
a
b
r
ief
in
tr
o
d
u
ctio
n
to
R
F,
SVR
,
an
d
ANN/DN
N
m
ac
h
in
e
-
lear
n
in
g
s
tatis
tical
m
o
d
els
f
o
r
p
r
ed
ictin
g
s
h
o
r
t
-
ter
m
s
o
lar
p
h
o
to
v
o
ltaic
p
o
wer
is
p
r
esen
ted
.
Sectio
n
3
d
elv
es
i
n
to
th
e
s
p
ec
if
ics
o
f
th
e
d
ataset
em
p
lo
y
e
d
in
th
is
r
esear
c
h
,
th
e
m
et
h
o
d
o
lo
g
ies
ap
p
lied
f
o
r
d
ata
an
al
y
s
is
,
an
d
t
h
e
f
o
r
ec
asti
n
g
o
f
s
o
lar
p
h
o
t
o
v
o
ltai
c
p
o
wer
o
u
tp
u
ts
ac
r
o
s
s
v
a
r
io
u
s
tim
e
in
ter
v
als
(
r
an
g
in
g
f
r
o
m
1
5
to
6
0
m
in
u
t
es)
u
tili
zin
g
r
ec
o
r
d
ed
wea
th
er
p
ar
am
ete
r
d
ata.
Sectio
n
4
p
r
esen
ts
p
er
f
o
r
m
an
ce
m
etr
ics
b
ased
o
n
wh
ich
th
e
a
n
aly
s
is
was
d
o
n
e,
an
d
th
e
r
esu
l
ts
o
b
tain
ed
f
r
o
m
th
e
s
tu
d
y
p
r
e
s
en
ted
in
s
ec
tio
n
5
an
d
s
ec
tio
n
6
c
o
n
clu
d
e
th
e
p
a
p
er
.
2.
M
ACH
I
N
E
L
E
AR
NING
M
O
DE
L
S E
M
P
L
O
Y
E
D
I
N
T
H
I
S WO
RK
Pro
p
o
s
als
f
o
r
v
ar
io
u
s
f
o
r
ec
asti
n
g
tech
n
iq
u
es
h
av
e
b
ee
n
d
escr
i
b
ed
in
v
ar
i
o
u
s
p
ap
er
s
in
o
r
d
er
t
o
f
o
r
ec
ast
PV
p
o
wer
at
v
ar
io
u
s
tim
e
h
o
r
izo
n
s
an
d
m
o
s
t
im
p
o
r
tan
tly
,
s
h
o
r
t
-
ter
m
PV
p
o
wer
f
o
r
ec
a
s
tin
g
is
v
er
y
m
u
ch
ess
en
tial
in
co
n
tr
o
llin
g
,
d
is
p
at
ch
in
g
,
a
n
d
s
ch
ed
u
lin
g
p
o
wer
[
1
4
]
.
Ma
ch
in
e
lear
n
in
g
(
ML
)
in
v
o
lv
es
tr
ain
in
g
a
co
m
p
u
ter
s
y
s
tem
to
g
ain
e
x
p
er
tis
e
b
y
p
r
o
ce
s
s
in
g
an
d
an
aly
zi
n
g
d
ata
co
llected
o
v
er
tim
e,
ai
m
in
g
to
im
p
r
o
v
e
its
p
er
f
o
r
m
an
ce
o
v
er
tim
e
[
1
5
]
.
I
n
th
e
p
r
esen
t
s
tu
d
y
m
ac
h
in
e
l
ea
r
n
in
g
m
o
d
els
R
an
d
o
m
Fo
r
e
s
t
(
R
F)
tech
n
iq
u
e,
ar
tific
ial
n
eu
r
al
n
etwo
r
k
s
(
A
NN)
,
an
d
d
ee
p
n
eu
r
al
n
etwo
r
k
s
(
DNN)
wer
e
b
e
n
ch
m
ar
k
ed
with
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
es (
SVM)
.
C
o
d
e
f
o
r
an
aly
zin
g
th
e
d
ata
is
s
cr
ip
ted
in
P
y
th
o
n
co
d
e
u
s
in
g
J
u
p
y
ter
N
o
teb
o
o
k
.
2
.
1
.
Ra
nd
o
m
f
o
re
s
t
R
en
o
wn
ed
f
o
r
its
s
im
p
licity
,
r
an
d
o
m
f
o
r
est
(
R
F)
s
tan
d
s
o
u
t
as
o
n
e
o
f
th
e
m
o
s
t
p
o
p
u
la
r
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
.
W
ith
ap
p
licatio
n
s
in
b
o
th
r
eg
r
ess
io
n
an
d
class
if
icatio
n
task
s
,
R
F
b
elo
n
g
s
to
th
e
ca
teg
o
r
y
o
f
s
u
p
er
v
is
ed
lear
n
in
g
alg
o
r
it
h
m
s
,
alo
n
g
s
id
e
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
es
(
SVMs),
n
aiv
e
B
a
y
es,
an
d
o
th
er
tr
ee
-
b
ased
m
eth
o
d
s
lik
e
Ad
ab
o
o
s
t
[
1
6
]
.
I
n
itially
in
tr
o
d
u
ce
d
a
n
d
p
r
o
p
o
s
ed
b
y
L
eo
B
r
eim
an
at
th
e
Un
i
v
er
s
ity
o
f
C
alif
o
r
n
ia
in
2
0
0
1
,
r
an
d
o
m
f
o
r
est
r
eg
r
ess
io
n
r
ep
r
esen
ts
an
en
s
em
b
le
lear
n
in
g
ap
p
r
o
ac
h
.
B
y
am
alg
am
atin
g
p
r
ed
ictio
n
s
f
r
o
m
d
iv
er
s
e
m
ac
h
in
e
lear
n
in
g
alg
o
r
ith
m
s
,
it a
c
h
iev
es g
r
ea
ter
ac
cu
r
ac
y
in
p
r
ed
i
ctio
n
s
co
m
p
ar
e
d
to
in
d
iv
id
u
al
m
o
d
els
[
1
7
]
,
[
1
8
]
.
T
h
e
tech
n
i
q
u
e
i
n
v
o
l
v
es
co
n
s
tr
u
ctin
g
tr
ee
s
i
n
d
iv
id
u
ally
b
y
e
m
p
lo
y
in
g
b
o
o
ts
tr
ap
d
ata
s
am
p
les,
cr
ea
tin
g
a
f
o
r
e
s
t
co
m
p
r
is
in
g
n
u
m
er
o
u
s
d
ec
i
s
io
n
tr
ee
s
.
As
m
o
r
e
tr
ee
s
ar
e
in
co
r
p
o
r
ated
,
th
e
f
o
r
ec
ast'
s
ac
cu
r
ac
y
is
en
h
an
ce
s
,
r
esu
ltin
g
in
en
h
an
ce
d
p
r
ec
is
io
n
[
1
9
]
.
Fig
u
r
e
1
d
em
o
n
s
tr
ate
s
th
e
co
n
f
ig
u
r
atio
n
o
f
th
e
r
a
n
d
o
m
f
o
r
est m
o
d
el.
Fig
u
r
e
1
.
R
an
d
o
m
f
o
r
est m
o
d
e
l
T
o
e
x
e
c
u
t
e
r
a
n
d
o
m
f
o
r
e
s
t
r
e
g
r
e
s
s
i
o
n
o
n
t
h
e
t
r
a
i
n
i
n
g
d
a
t
a
s
e
t
,
t
h
e
s
u
b
s
e
q
u
e
n
t
a
c
t
i
o
n
s
m
u
s
t
b
e
t
ak
e
n
:
i
)
T
o
b
e
g
i
n
,
a
s
e
t
o
f
'
k
'
d
a
t
a
p
o
i
n
ts
s
el
e
c
t
e
d
f
r
o
m
t
h
e
i
n
p
u
t
(
t
r
ai
n
i
n
g
)
d
a
t
a
s
et
,
l
a
b
e
le
d
a
s
'
x
'
;
ii
)
A
d
e
ci
s
i
o
n
t
r
e
e
g
e
n
e
r
a
t
e
d
t
o
r
e
p
r
e
s
e
n
t
t
h
e
s
e
c
h
o
s
e
n
'
k
'
d
a
ta
p
o
i
n
t
s
;
i
ii
)
T
h
e
f
i
r
s
t
a
n
d
s
e
c
o
n
d
s
t
e
p
s
r
e
i
t
e
r
a
t
e
d
u
n
t
il
'
N'
d
e
c
is
i
o
n
t
r
e
es
g
e
n
e
r
a
te
d
d
u
r
i
n
g
t
h
e
t
r
a
i
n
i
n
g
p
h
a
s
e
;
i
v
)
W
h
e
n
p
r
e
s
e
n
t
e
d
wi
t
h
a
n
e
w
d
a
t
a
p
o
i
n
t
,
e
a
c
h
o
f
t
h
e
g
e
n
e
r
a
t
ed
t
r
e
e
s
p
r
o
d
u
c
e
s
a
p
r
e
d
i
c
t
i
o
n
v
a
l
u
e
'
y
'
;
a
n
d
v
)
T
h
e
d
a
t
a
p
o
i
n
t
is
t
h
e
n
a
t
t
r
i
b
u
t
e
d
t
o
t
h
e
a
v
e
r
a
g
e
o
f
a
l
l
p
r
e
d
i
ct
e
d
'
y
'
v
al
u
e
s
.
R
a
n
d
o
m
f
o
r
e
s
t
r
e
g
r
e
s
s
i
o
n
p
e
r
f
o
r
m
s
we
l
l
o
n
d
i
v
e
r
s
i
f
i
e
d
p
r
o
b
l
e
m
s
w
i
t
h
th
e
p
o
t
e
n
t
i
a
l
o
f
h
a
n
d
l
i
n
g
n
o
n
-
l
in
e
a
r
r
e
l
a
t
i
o
n
s
h
i
p
s
.
I
n
r
e
c
e
n
t
t
i
m
e
s
,
r
a
n
d
o
m
f
o
r
e
s
t
(
RF
)
h
a
s
g
a
r
n
e
r
e
d
i
n
c
r
e
a
s
e
d
i
n
t
e
r
e
s
t
a
m
o
n
g
r
e
s
e
a
r
c
h
e
r
s
i
n
t
h
e
r
e
a
l
m
o
f
p
h
o
t
o
v
o
l
t
a
i
c
(
PV
)
p
o
w
e
r
f
o
r
e
ca
s
t
i
n
g
,
o
w
i
n
g
t
o
i
ts
b
e
n
e
f
i
t
s
i
n
e
n
s
e
m
b
l
e
l
e
a
r
n
i
n
g
a
n
d
s
u
p
e
r
i
o
r
p
e
r
f
o
r
m
a
n
c
e
w
h
e
n
c
o
m
p
a
r
e
d
t
o
a
l
t
e
r
n
a
t
i
v
e
s
t
at
is
t
ic
a
l
-
b
a
s
e
d
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
m
s
[
2
0
]
.
I
n
t
h
i
s
s
t
u
d
y
,
R
F
te
c
h
n
i
q
u
e
s
w
e
r
e
u
s
e
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2088
-
8
6
9
4
I
n
t J Po
w
E
lec
&
Dr
i Sy
s
t
,
Vo
l.
1
5
,
No
.
4
,
Dec
em
b
er
20
2
4
:
2553
-
2
5
6
9
2556
t
o
f
o
r
e
c
a
s
t
PV
p
o
w
e
r
o
u
t
p
u
t
f
r
o
m
a
g
r
i
d
-
c
o
n
n
e
c
t
e
d
P
V
p
l
a
n
t
.
T
o
a
s
s
es
s
t
h
e
e
f
f
e
c
t
i
v
e
n
es
s
o
f
R
F
,
o
t
h
e
r
c
o
m
m
o
n
l
y
u
s
e
d
m
o
d
e
l
s
s
u
c
h
a
s
SVR
a
n
d
A
N
N
/
DN
N
w
e
r
e
a
ls
o
e
v
a
l
u
a
t
e
d
a
n
d
c
o
m
p
a
r
e
d
w
i
t
h
t
h
e
R
F
t
ec
h
n
i
q
u
e
.
S
VR
t
e
c
h
n
i
q
u
e
is
u
s
e
d
as
a
b
e
n
c
h
m
ar
k
m
o
d
e
l
f
o
r
e
v
a
l
u
at
i
n
g
t
h
e
s
k
i
ll
s
c
o
r
e
o
f
t
h
e
m
o
d
e
l
p
r
o
p
o
s
e
d
f
o
r
t
h
e
s
e
l
e
ct
e
d
s
i
t
e
.
2
.
2
.
Su
pp
o
rt
v
ec
t
o
r
m
a
chines
Vap
n
ik
in
tr
o
d
u
ce
d
th
e
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM)
,
an
ad
v
an
ce
d
m
ac
h
in
e
-
lear
n
in
g
tech
n
iq
u
e
ac
claim
ed
f
o
r
its
ex
ce
p
tio
n
al
p
er
f
o
r
m
an
ce
[
2
1
]
.
A
s
u
p
er
v
is
ed
lear
n
in
g
alg
o
r
ith
m
k
n
o
wn
as
a
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM)
is
u
tili
ze
d
f
o
r
b
o
th
class
if
icatio
n
an
d
r
eg
r
ess
io
n
an
aly
s
is
task
s
[
2
2
]
.
Su
p
p
o
r
t
v
ec
to
r
m
ac
h
in
es
(
SVMs)
d
eter
m
in
e
th
e
o
p
tim
al
h
y
p
er
p
lan
e
to
s
ep
ar
ate
d
ata
in
to
d
if
f
er
en
t
class
es,
u
s
in
g
a
k
er
n
el
f
u
n
ctio
n
to
m
ap
d
ata
in
to
h
ig
h
er
d
im
en
s
io
n
s
.
Su
p
p
o
r
t
v
ec
to
r
s
d
ef
in
e
th
is
h
y
p
er
p
lan
e
,
u
tili
ze
d
f
o
r
class
if
icatio
n
(
SVC
)
an
d
r
eg
r
ess
io
n
(
SVR
)
.
SVR
aim
s
to
m
in
im
iz
e
e
r
r
o
r
s
b
etwe
en
p
r
e
d
icted
a
n
d
ac
tu
al
v
alu
es,
co
n
tr
o
llin
g
m
o
d
e
l
co
m
p
lex
ity
th
r
o
u
g
h
m
ar
g
i
n
s
,
with
h
ea
v
ier
p
en
alties
o
n
d
is
ta
n
t
p
o
in
ts
t
o
r
e
d
u
ce
o
u
tlier
s
en
s
itiv
ity
.
SVC
f
o
cu
s
es
o
n
f
i
n
d
in
g
a
h
y
p
er
p
lan
e
th
at
b
est
class
if
ies
d
ata,
p
en
alizin
g
m
is
class
if
icatio
n
s
to
en
h
a
n
c
e
m
o
d
el
s
en
s
itiv
ity
.
SVMs
em
p
lo
y
s
tr
u
ctu
r
al
r
is
k
m
in
im
izatio
n
f
o
r
g
e
n
er
aliza
tio
n
,
en
s
u
r
in
g
p
o
ten
tial
f
o
r
g
l
o
b
al
o
p
tim
u
m
s
o
lu
tio
n
s
,
an
d
ap
p
l
y
to
b
o
th
class
if
icatio
n
an
d
r
eg
r
ess
io
n
task
s
[
2
3
]
.
a.
Featu
r
e
s
p
ac
e
an
d
k
er
n
el
f
u
n
ct
io
n
s
T
h
e
f
u
n
d
am
en
tal
p
r
in
ci
p
le
o
f
SVMs
in
v
o
lv
es
m
ap
p
in
g
d
ata
in
to
a
f
ea
tu
r
e
s
p
ac
e
u
s
in
g
n
o
n
-
lin
ea
r
m
ap
p
in
g
,
f
o
llo
wed
b
y
t
h
e
ap
p
licatio
n
o
f
a
lin
ea
r
alg
o
r
ith
m
.
Ho
wev
er
,
th
is
f
ea
tu
r
e
s
p
ac
e
o
f
ten
r
eq
u
ir
es
h
ig
h
-
d
im
en
s
io
n
al
d
o
t
p
r
o
d
u
ct
ev
al
u
atio
n
,
d
em
an
d
in
g
s
u
b
s
tan
tial
co
m
p
u
tatio
n
al
r
eso
u
r
ce
s
an
d
tim
e.
Occ
asio
n
ally
,
s
im
p
ler
k
er
n
els
ar
e
ex
p
lo
r
ed
f
o
r
ef
f
icac
y
.
I
n
r
ea
l
-
wo
r
l
d
s
ce
n
ar
io
s
,
co
m
p
lex
p
r
o
b
lem
s
n
ec
ess
itate
m
o
r
e
s
o
p
h
is
ticated
h
y
p
o
th
esis
s
p
ac
es
th
an
th
o
s
e
o
f
f
er
ed
b
y
lin
ea
r
l
ea
r
n
in
g
m
ac
h
i
n
es,
co
n
s
tr
ain
ed
b
y
co
m
p
u
ta
tio
n
al
lim
itatio
n
s
.
L
in
ea
r
lear
n
in
g
m
ac
h
in
es
p
o
s
s
ess
th
e
ad
v
an
tag
e
o
u
s
ch
ar
ac
ter
is
tic
o
f
b
ein
g
ex
p
r
ess
ib
le
in
a
d
u
al
f
o
r
m
,
allo
win
g
th
e
h
y
p
o
th
esis
to
b
e
r
ep
r
esen
ted
as
a
lin
ea
r
co
m
b
in
atio
n
o
f
t
r
ain
in
g
p
o
i
n
ts
.
T
h
is
f
ac
ilit
ates
d
ec
is
io
n
r
u
le
ev
alu
atio
n
b
ased
s
o
lely
o
n
in
n
er
p
r
o
d
u
cts b
etwe
en
test
an
d
tr
ain
in
g
p
o
in
ts
.
I
n
ca
s
es
wh
er
e
d
ir
ec
t
ca
lcu
latio
n
o
f
in
n
er
p
r
o
d
u
cts
in
f
ea
tu
r
e
s
p
ac
e
u
s
in
g
o
r
ig
in
al
in
p
u
t
p
o
in
ts
is
f
ea
s
ib
le,
a
n
o
n
-
lin
ea
r
lear
n
in
g
m
ac
h
in
e
ter
m
ed
d
ir
ec
t c
o
m
p
u
t
atio
n
m
eth
o
d
o
f
t
h
e
k
er
n
el
f
u
n
ctio
n
(
d
en
o
ted
b
y
K)
m
a
y
b
e
c
o
n
s
tr
u
cted
[
2
3
]
.
SVM
m
o
d
els
u
tili
ze
in
p
u
t
v
ar
iab
les
co
r
r
elate
d
with
t
h
e
tar
g
et
v
ar
iab
le,
wh
ic
h
is
th
e
v
ar
ia
b
le
to
b
e
p
r
e
d
icted
.
T
h
is
en
tails
r
ep
r
esen
tin
g
th
e
d
ata
th
r
o
u
g
h
a
n
o
n
-
lin
ea
r
f
u
n
ctio
n
,
d
e
n
o
ted
as
f
(
x
)
in
(
1
)
,
an
d
v
is
u
alizin
g
it.
(
)
=
.
(
)
+
(
1
)
W
h
er
e
is
th
e
n
o
r
m
al
v
ec
to
r
;
is
a
co
n
s
tan
t
o
r
b
iased
ter
m
;
an
d
(
)
is
a
lar
g
e
d
im
en
s
io
n
al
s
p
ec
ial
ch
ar
ac
ter
is
tic
m
ap
p
ed
b
y
a
s
p
ac
e
v
ec
to
r
x
.
T
o
d
eter
m
in
e
t
h
e
co
ef
f
icien
ts
an
d
,
an
o
p
tim
iza
tio
n
p
r
o
b
lem
is
s
o
lv
ed
u
s
in
g
(
2
)
-
(
6
)
th
r
o
u
g
h
m
in
im
izatio
n
.
(
)
(
)
=
1
∑
=
1
=
(
(
)
,
)
+
1
2
‖
‖
2
=
1
(
2
)
(
(
)
,
)
=
|
(
(
)
,
)
|
−
|
(
(
)
,
)
|
≥
(
3
)
(
(
)
,
)
=
0
ℎ
(
4
)
W
h
er
e
is
th
e
p
ar
am
eter
o
f
th
e
m
o
d
el
;
(
(
)
,
)
is
d
escr
ib
es
th
e
ℎ
m
is
s
in
g
f
u
n
ctio
n
,
th
is
r
ef
er
s
to
t
h
e
f
ac
t
th
at
an
y
er
r
o
r
s
th
at
f
all
u
n
d
er
th
e
v
alu
e
o
f
ep
s
ilo
n
will
n
o
t
b
e
s
u
b
ject
to
p
en
alty
,
r
ep
r
e
s
en
ts
th
e
s
o
lar
PV
p
o
wer
in
th
e
p
er
io
d
i
;
a
n
d
1
∑
(
(
)
,
)
=
1
d
ef
in
es
th
e
em
p
ir
ical
er
r
o
r
o
f
t
h
e
SVM
m
o
d
el.
1
2
‖
‖
2
is
th
e
r
eg
u
lar
izatio
n
te
r
m
,
C
is
th
e
p
en
alty
f
u
n
ctio
n
ass
ess
ed
to
b
alan
ce
th
e
c
o
m
p
en
s
atio
n
b
et
wee
n
th
e
er
r
o
r
an
d
em
p
ir
ical
r
is
k
b
y
u
tili
zin
g
s
lack
v
ar
iab
les
an
d
∗
.
T
h
ese
v
ar
iab
les in
d
icate
th
e
p
r
esen
ce
o
f
ex
ce
s
s
iv
e
to
p
an
d
b
o
tto
m
s
k
ews,
r
esp
ec
tiv
ely
.
In
(
2
)
ca
n
b
e
f
o
r
m
u
lated
as d
em
o
n
s
tr
ated
b
elo
w
b
y
u
tili
zin
g
t
h
e
ch
ar
ac
ter
is
tics
o
f
th
e
f
u
n
ctio
n
th
at
n
ee
d
to
b
e
o
p
tim
ized
(
illu
s
tr
atio
n
s
h
o
wn
in
F
ig
u
r
e
2
).
Min
im
ize:
1
2
‖
‖
2
+
1
∑
(
+
∗
)
=
1
(
5
)
o
n
ly
wh
en
:
{
|
−
(
⟨
|
⟩
+
)
≥
+
∗
|
(
⟨
|
⟩
+
.
≤
+
∗
)
,
∗
≥
0
(
6
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Po
w
E
lec
&
Dr
i Sy
s
t
I
SS
N:
2088
-
8
6
9
4
F
o
r
ec
a
s
tin
g
h
o
u
r
ly
s
h
o
r
t
-
term so
la
r
p
h
o
to
vo
lta
ic
p
o
w
er u
s
in
g
ma
ch
in
e
…
(
S
r
a
va
n
ku
ma
r
J
o
g
u
n
u
r
i
)
2557
B
y
u
tili
zin
g
L
ag
r
an
g
e
an
d
o
p
tim
al
co
n
s
tr
ain
ts
,
it
i
s
f
ea
s
ib
le
to
d
er
iv
e
a
n
o
n
-
lin
ea
r
r
eg
r
ess
io
n
f
u
n
ctio
n
s
h
o
wn
in
(
7
)
a
n
d
(
8
)
to
s
o
l
v
e
in
(
1
)
:
(
)
=
∑
(
∝
−
∝
∗
)
=
1
(
−
)
+
(
7
)
wh
er
e
∝
,
∝
∗
ar
e
L
ag
r
an
g
e
m
u
ltip
lie
r
s
.
T
h
e
ter
m
(
−
)
is
d
ef
in
ed
as a
k
e
r
n
el
f
u
n
ctio
n
.
(
−
)
=
∑
(
)
+
(
)
=
1
(
8
)
T
h
er
e
ar
e
f
o
u
r
m
ain
f
u
n
ctio
n
s
av
ailab
le
f
o
r
SVM:
lin
ea
r
(
9
)
,
p
o
ly
n
o
m
ial
(
1
0
)
,
r
a
d
ial
b
asis
f
u
n
ctio
n
(
1
1
)
,
an
d
s
ig
m
o
id
(
1
2
)
.
[
1
5
]
.
b.
L
in
ea
r
k
er
n
el
f
u
n
ctio
n
(
,
)
=
.
(
9
)
wh
er
e
,
ar
e
th
e
in
p
u
ts
to
th
e
i
th
an
d
j
th
d
im
en
s
io
n
s
r
esp
ec
tiv
ely
.
c.
Po
ly
n
o
m
ial
k
er
n
el
f
u
n
ctio
n
(
,
)
=
(
.
)
(
1
0
)
wh
er
e
is
a
d
eg
r
ee
o
f
t
h
e
p
o
l
y
n
o
m
ial.
d.
R
ad
ial
B
as
is
k
er
n
el
f
u
n
ctio
n
(
,
)
=
(
‖
−
‖
2
2
)
(
.
)
(
1
1
)
wh
er
e
is
k
er
n
el
weig
h
t.
e.
Sig
m
o
id
k
er
n
el
f
u
n
ctio
n
(
,
)
=
ta
n
h
(
(
.
)
+
)
(
1
2
)
wh
er
e
ae
ad
ju
s
tab
le
k
er
n
el
f
u
n
ctio
n
s
r
ely
in
g
o
n
t
h
e
d
ata.
Fig
u
r
e
2
.
I
ll
u
s
tr
atio
n
o
f
a
n
SVR
2
.
3
.
Art
if
icia
l
neura
l net
wo
rk
s
(
ANN)
a
nd
deep
ne
ura
l net
wo
rk
s
(
DNN)
ANNs
m
im
ic
th
e
s
tr
u
ctu
r
e
a
n
d
f
u
n
ctio
n
o
f
n
atu
r
al
n
eu
r
al
n
etwo
r
k
s
f
o
u
n
d
in
th
e
h
u
m
an
b
o
d
y
.
T
h
ey
p
o
s
s
ess
th
e
ab
ilit
y
to
a
u
to
n
o
m
o
u
s
ly
r
ec
o
g
n
ize
p
atter
n
s
with
i
n
p
r
e
v
io
u
s
ly
in
clu
d
e
d
d
ata,
m
ak
in
g
t
h
em
h
ig
h
ly
ef
f
ec
tiv
e
f
o
r
m
o
d
elin
g
c
o
m
p
le
x
an
d
n
o
n
lin
ea
r
r
elatio
n
s
h
ip
s
b
etwe
en
in
p
u
t a
n
d
o
u
tp
u
t v
ar
i
ab
les in
co
m
p
ar
is
o
n
to
o
th
er
f
o
r
ec
asti
n
g
m
eth
o
d
s
.
Fig
u
r
e
3
d
ep
icts
th
e
b
asic
ar
ch
itectu
r
e
o
f
an
ANN,
wh
er
e
n
e
u
r
o
n
s
p
r
o
ce
s
s
in
p
u
t
d
ata
an
d
g
en
e
r
ate
o
u
t
p
u
t
u
s
in
g
in
d
iv
i
d
u
al
ac
tiv
atio
n
f
u
n
ctio
n
s
.
C
r
u
cial
p
ar
am
eter
s
lik
e
le
ar
n
in
g
r
ate,
n
u
m
b
e
r
o
f
h
id
d
e
n
lay
er
s
,
an
d
m
ax
i
m
u
m
iter
atio
n
co
u
n
t
r
eg
u
late
th
e
lear
n
in
g
p
r
o
ce
s
s
in
ANNs.
Ad
ju
s
tm
en
ts
to
ac
tiv
atio
n
f
u
n
ctio
n
weig
h
ts
an
d
p
ar
am
eter
s
o
cc
u
r
th
r
o
u
g
h
a
lear
n
in
g
p
r
o
ce
s
s
.
ANNs
m
ay
v
ar
y
in
th
e
n
u
m
b
er
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2088
-
8
6
9
4
I
n
t J Po
w
E
lec
&
Dr
i Sy
s
t
,
Vo
l.
1
5
,
No
.
4
,
Dec
em
b
er
20
2
4
:
2553
-
2
5
6
9
2558
o
f
n
eu
r
o
n
s
ac
r
o
s
s
in
p
u
t,
h
i
d
d
e
n
,
an
d
o
u
tp
u
t
lay
er
s
,
em
p
lo
y
i
n
g
v
ar
i
o
u
s
ac
tiv
atio
n
f
u
n
ctio
n
s
s
u
ch
as
Sig
m
o
id
,
R
ec
tifie
d
L
in
ea
r
Un
it,
an
d
S
o
f
tMa
x
f
o
r
co
m
p
u
tatio
n
.
Desp
ite
ad
v
an
tag
es
lik
e
f
au
lt
to
ler
an
ce
an
d
p
ar
allel
p
r
o
ce
s
s
in
g
ca
p
ab
ilit
y
,
ANNs
h
av
e
lim
itatio
n
s
in
clu
d
in
g
h
a
r
d
war
e
d
ep
e
n
d
en
c
y
an
d
lack
o
f
in
ter
p
r
etab
ilit
y
,
wh
ich
n
ec
ess
itate
p
r
o
ce
s
s
o
r
s
with
p
ar
allel
p
r
o
ce
s
s
in
g
p
o
we
r
.
Ad
d
itio
n
ally
,
th
e
n
etwo
r
k
'
s
in
ter
p
r
et
ab
ilit
y
an
d
th
e
p
r
ed
ictab
ilit
y
o
f
its
d
u
r
atio
n
ar
e
s
ig
n
if
ican
t c
o
n
ce
r
n
s
[
2
4
]
.
Dee
p
n
eu
r
al
n
etwo
r
k
s
(
DNNs)
,
a
v
ar
ia
n
t
o
f
ar
tific
ial
n
e
u
r
al
n
etwo
r
k
s
(
ANNs)
,
d
if
f
er
b
y
in
c
o
r
p
o
r
ati
n
g
m
u
ltip
le
h
id
d
en
lay
e
r
s
in
s
tead
o
f
ju
s
t
o
n
e.
T
h
ese
lay
er
s
ar
e
a
d
ep
t
at
ca
p
tu
r
in
g
an
d
lev
er
a
g
i
n
g
th
e
in
h
er
e
n
t
o
n
e
-
o
r
two
-
d
im
en
s
io
n
al
s
tr
u
ctu
r
e
with
in
th
e
n
etwo
r
k
.
Giv
en
th
e
p
r
o
life
r
atio
n
o
f
in
ter
n
et
o
f
th
in
g
s
an
d
th
e
escalatin
g
v
o
lu
m
e
o
f
b
ig
d
ata,
DNN
m
o
d
els
h
av
e
g
ar
n
e
r
ed
c
o
n
s
id
er
a
b
le
atten
tio
n
ac
r
o
s
s
v
ar
i
o
u
s
r
esear
ch
d
o
m
ain
s
.
A
n
o
tab
le
ad
v
an
tag
e
lies
in
th
ei
r
p
r
o
f
icien
cy
at
d
is
ce
r
n
in
g
n
o
n
lin
ea
r
r
elatio
n
s
h
i
p
s
b
etwe
en
in
p
u
t
f
ea
tu
r
es
an
d
o
u
tp
u
t
ta
r
g
ets
.
T
h
is
in
v
o
lv
es
p
r
o
g
r
ess
iv
ely
lea
r
n
in
g
m
u
ltip
le
lay
er
s
o
f
r
ep
r
esen
tatio
n
s
f
r
o
m
d
ata,
g
r
a
d
u
all
y
r
ef
in
in
g
th
em
i
n
to
m
o
r
e
m
ea
n
i
n
g
f
u
l
d
ep
ictio
n
s
.
W
ith
d
ee
p
er
ex
p
lo
r
atio
n
,
DNNs g
ain
th
e
ca
p
ab
ilit
y
to
i
d
en
tify
in
cr
ea
s
in
g
ly
s
o
p
h
is
ticated
r
ep
r
esen
tatio
n
s
,
th
er
eb
y
estab
lis
h
in
g
p
r
ec
is
e
co
r
r
elatio
n
s
b
etwe
en
in
p
u
t
ch
ar
ac
ter
is
tics
an
d
th
eir
in
ten
d
ed
tar
g
ets [
2
5
]
.
(
a)
(
b
)
Fig
u
r
e
3
.
B
asic a
r
ch
itectu
r
e
o
f
(
a)
ar
tific
ial
n
eu
r
al
n
etwo
r
k
s
an
d
(
b
)
d
ee
p
n
eu
r
al
n
etwo
r
k
s
3.
M
AT
E
R
I
AL
S AN
D
M
E
T
H
O
DS
3
.
1
.
Study
a
re
a
des
cr
iptio
n,
da
t
a
co
llect
io
n a
nd
prepa
ra
t
io
n
Data
co
llectio
n
was
d
o
n
e
o
v
e
r
a
1
2
-
m
o
n
th
p
e
r
io
d
s
p
an
n
in
g
f
r
o
m
Octo
b
er
2
0
2
1
to
Sep
te
m
b
er
2
0
2
2
.
T
h
is
d
ata
co
llected
f
r
o
m
a
2
0
k
W
s
o
lar
p
o
wer
p
lan
t
c
o
n
n
ec
te
d
to
th
e
g
r
id
,
al
o
n
g
with
a
n
ea
r
b
y
wea
th
er
s
tatio
n
s
itu
ated
at
co
o
r
d
in
ates
2
2
.
7
8
°
N,
7
3
.
6
5
°
E
,
s
p
ec
if
ica
lly
at
th
e
C
o
lleg
e
o
f
Ag
r
ic
u
ltu
r
al
E
n
g
in
ee
r
i
n
g
a
n
d
T
ec
h
n
o
lo
g
y
with
in
An
an
d
Ag
r
icu
ltu
r
al
Un
iv
er
s
ity
in
Go
d
h
r
a,
I
n
d
ia.
T
h
e
s
tu
d
y
s
ite
ex
p
er
i
en
ce
s
f
o
u
r
d
is
tin
ct
s
ea
s
o
n
s
.
Au
tu
m
n
f
r
o
m
Octo
b
er
to
No
v
em
b
e
r
,
f
o
llo
we
d
b
y
win
ter
f
r
o
m
Dec
em
b
e
r
to
Feb
r
u
ar
y
.
T
h
e
s
u
m
m
er
s
ea
s
o
n
en
co
m
p
ass
es th
e
m
o
n
t
h
s
o
f
Ma
r
ch
t
o
Ma
y
,
w
h
ile
th
e
r
ain
y
s
ea
s
o
n
ex
te
n
d
s
f
r
o
m
J
u
n
e
to
Sep
tem
b
er
.
Data
co
llected
at
a
1
5
-
m
i
n
u
te
tim
e
r
eso
lu
tio
n
,
b
u
t
o
n
l
y
d
at
a
b
etwe
en
7
:
0
0
AM
a
n
d
5
:0
0
PM
was
co
n
s
id
er
ed
d
u
e
to
s
o
lar
r
a
d
iatio
n
av
ailab
ilit
y
an
d
av
er
a
g
ed
to
h
o
u
r
ly
v
al
u
es
as
th
e
f
o
r
ec
asti
n
g
in
ter
est
is
in
s
tep
s
o
f
h
o
u
r
s
.
Ho
wev
er
,
m
is
s
in
g
v
a
lu
es
o
cc
u
r
r
ed
i
n
th
e
d
ataset
d
u
e
to
p
o
wer
f
ailu
r
es
af
f
ec
tin
g
t
h
e
d
ata
lo
g
g
e
r
s
.
T
h
is
r
esu
lted
in
a
d
ataset
o
f
2
4
6
0
s
am
p
les,
as
s
h
o
wn
in
T
ab
le
1
.
On
e
d
ata
lo
g
g
er
o
f
wea
th
er
m
o
n
ito
r
in
g
s
tatio
n
p
r
o
cu
r
ed
f
r
o
m
E
n
g
i
n
ee
r
in
g
a
n
d
E
n
v
ir
o
n
m
en
tal
So
lu
tio
n
s
Pv
t
lim
ited
m
o
d
el
(
W
MS1
0
1
5
8
)
was
u
s
ed
to
r
ec
o
r
d
v
ar
io
u
s
wea
th
er
p
ar
a
m
eter
s
,
i
n
clu
d
in
g
tim
e
o
f
d
a
y
,
am
b
ien
t
tem
p
er
atu
r
e,
r
elativ
e
h
u
m
id
i
ty
,
s
o
lar
r
a
d
iatio
n
,
win
d
s
p
ee
d
,
an
d
win
d
d
ir
ec
tio
n
.
Simu
ltan
e
o
u
s
ly
,
a
s
ec
o
n
d
d
ata
lo
g
g
e
r
in
te
g
r
ated
with
th
e
i
n
v
er
ter
ca
p
tu
r
e
d
t
h
e
tar
g
et
v
ar
iab
le,
w
h
ic
h
is
s
o
lar
p
h
o
to
v
o
ltaic
p
o
wer
.
Sp
ec
ial
atten
tio
n
was
g
iv
en
to
e
n
s
u
r
in
g
th
e
s
y
n
ch
r
o
n
izatio
n
o
f
d
ata
co
llected
f
r
o
m
b
o
th
lo
g
g
er
s
in
ter
m
s
o
f
d
ata
co
llectio
n
tim
in
g
.
T
h
e
d
is
tr
ib
u
tio
n
o
f
t
h
e
wea
th
er
v
ar
iab
les
is
illu
s
tr
ated
in
Fig
u
r
e
4
.
Acc
o
u
n
tin
g
f
o
r
d
ata
av
ailab
il
ity
an
d
co
r
r
elatio
n
b
etwe
en
t
h
em
p
lay
s
a
v
ital
r
o
le
wh
en
s
elec
tin
g
v
ar
iab
les
f
o
r
p
r
ed
ictiv
e
m
o
d
elin
g
.
C
o
n
s
eq
u
en
tly
,
a
co
m
p
r
eh
en
s
iv
e
s
tatis
tical
an
aly
s
is
was
co
n
d
u
cted
to
ev
alu
ate
th
e
r
elatio
n
s
h
ip
b
et
wee
n
ea
ch
ac
ce
s
s
ib
le
wea
th
er
v
ar
iab
le
a
n
d
s
o
lar
p
h
o
to
v
o
ltaic
p
o
wer
.
T
h
is
ex
am
in
atio
n
is
illu
s
tr
ated
in
Fig
u
r
e
5
,
wh
ich
s
h
o
wca
s
es
th
e
co
r
r
elatio
n
co
e
f
f
icien
ts
am
o
n
g
all
f
iv
e
wea
th
er
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Po
w
E
lec
&
Dr
i Sy
s
t
I
SS
N:
2088
-
8
6
9
4
F
o
r
ec
a
s
tin
g
h
o
u
r
ly
s
h
o
r
t
-
term so
la
r
p
h
o
to
vo
lta
ic
p
o
w
er u
s
in
g
ma
ch
in
e
…
(
S
r
a
va
n
ku
ma
r
J
o
g
u
n
u
r
i
)
2559
p
ar
am
eter
s
an
d
th
e
tim
e
o
f
d
a
y
(
d
e
n
o
ted
as
i
n
p
u
t v
ar
iab
les X1
to
X6
)
co
n
ce
r
n
in
g
PV
output
(
th
e
o
u
t
p
u
t
o
r
tar
g
et
v
ar
iab
le
Yt)
.
T
h
e
an
aly
s
is
en
c
o
m
p
ass
ed
th
e
en
tire
d
ataset.
Du
r
in
g
t
h
e
an
aly
s
is
,
a
n
eg
ativ
e
co
r
r
elatio
n
em
er
g
ed
b
etwe
e
n
tem
p
er
at
u
r
e
a
n
d
h
u
m
id
ity
.
No
tab
ly
,
th
e
d
ata
h
ig
h
lig
h
ted
a
r
o
b
u
s
t
co
r
r
elatio
n
b
etwe
en
s
o
lar
r
ad
iatio
n
an
d
PV
output
.
W
h
ile
a
d
ir
ec
t
an
d
s
u
b
s
tan
tial
co
n
n
ec
tio
n
b
etwe
en
tim
e,
tem
p
er
atu
r
e,
an
d
PV
output
was n
o
t
im
m
ed
iately
ev
i
d
en
t,
it b
ec
am
e
ap
p
ar
en
t th
at
tim
e
d
o
es
ex
er
t
a
co
n
s
id
er
a
b
le
in
f
l
u
en
ce
o
n
tem
p
e
r
atu
r
e.
T
h
is
,
i
n
tu
r
n
,
co
n
tr
ib
u
tes
to
t
h
e
co
r
r
elatio
n
with
s
o
lar
r
ad
iatio
n
.
As
a
r
esu
lt,
th
e
s
tu
d
y
in
c
o
r
p
o
r
ated
t
h
e
tim
e
o
f
d
ay
as
an
ad
d
itio
n
al
in
p
u
t
v
a
r
i
ab
le,
alo
n
g
s
id
e
th
e
wea
th
er
p
ar
am
eter
s
.
I
n
th
e
cu
r
r
en
t
s
tu
d
y
,
en
tire
o
n
e
-
y
ea
r
d
ata
s
et
was
d
iv
id
e
d
in
to
s
ea
s
o
n
al
d
ata,
f
o
r
au
tu
m
n
s
ea
s
o
n
d
ata
co
n
s
id
er
ed
f
r
o
m
1
st
Octo
b
er
2
0
2
1
to
3
0
th
No
v
em
b
er
2
0
2
1
,
f
o
r
win
ter
s
ea
s
o
n
f
r
o
m
1
st
Dec
em
b
er
2
0
2
1
to
2
8
th
Feb
r
u
ar
y
2
0
2
2
,
f
o
r
s
u
m
m
er
s
ea
s
o
n
f
r
o
m
1
st
Ma
r
ch
2
0
2
2
to
3
1
st
Ma
y
2
0
2
2
a
n
d
,
f
o
r
r
ain
y
s
e
aso
n
f
r
o
m
1
st
J
u
n
e
2
0
2
2
to
3
0
th
Sep
tem
b
er
2
0
2
2
.
T
h
e
d
ata
was
s
am
p
led
ev
er
y
1
5
m
in
u
tes
an
d
av
er
a
g
ed
o
v
er
an
h
o
u
r
f
o
r
th
e
p
r
ed
ictio
n
o
f
m
u
ltip
le
s
tep
s
ah
ea
d
with
ea
ch
s
tep
eq
u
alin
g
to
an
h
o
u
r
.
Af
ter
d
ata
cu
r
atio
n
,
a
n
d
cl
u
s
ter
in
g
as
p
er
th
e
s
ea
s
o
n
,
b
y
em
p
lo
y
i
n
g
d
if
f
e
r
en
t
s
tatis
tical
m
ac
h
in
e
lear
n
in
g
tech
n
iq
u
es,
f
o
r
ec
asti
n
g
o
f
th
e
s
o
lar
p
h
o
to
v
o
ltaic
p
o
wer
g
e
n
er
atio
n
f
o
r
o
n
1
h
o
u
r
ah
ea
d
to
4
h
o
u
r
s
ah
ea
d
w
as
ca
r
r
ied
o
u
t.
T
h
e
p
er
f
o
r
m
a
n
ce
o
f
th
e
s
tatis
tical
tech
n
iq
u
es
co
m
p
ar
ed
a
n
d
b
es
t
m
o
d
el
o
f
f
er
i
n
g
e
n
h
an
ce
d
a
cc
u
r
ac
y
was
d
ec
i
d
ed
b
ased
o
n
th
e
p
er
f
o
r
m
an
ce
m
etr
ics co
n
s
id
er
ed
in
t
h
e
f
o
llo
win
g
s
ec
tio
n
.
T
ab
le
1
.
Descr
ip
tio
n
o
f
th
e
d
at
a
co
llected
ev
er
y
1
5
m
in
u
tes a
n
d
av
er
a
g
ed
h
o
u
r
ly
f
o
r
th
e
s
tu
d
y
p
er
i
o
d
.
Ti
me
Te
mp
H
u
m
Wd
Ws
R
a
d
PV
out
put
C
o
u
n
t
2
4
6
0
2
4
6
0
2
4
6
0
2
4
6
0
2
4
6
0
2
4
6
0
2
4
6
0
M
e
a
n
1
2
.
5
0
3
0
.
5
3
5
3
.
4
8
1
2
1
.
0
6
2
.
0
1
3
4
5
.
4
1
7
.
2
7
S
td
2
.
8
7
5
.
4
2
2
6
.
9
5
6
6
.
0
5
1
.
9
6
1
8
3
.
0
9
3
.
9
1
M
in
8
.
0
0
8
.
8
5
0
.
7
5
1
.
2
5
0
.
0
0
0
.
0
0
0
.
0
7
2
5
%
1
0
.
0
0
2
7
.
7
8
3
1
.
7
5
7
2
.
7
5
0
.
0
0
1
9
5
.
9
5
3
.
8
7
5
0
%
1
2
.
5
0
3
1
.
3
8
5
0
.
2
5
9
0
.
9
4
1
.
8
0
3
5
1
.
5
0
7
.
3
1
7
5
%
1
5
.
0
0
3
3
.
6
5
7
4
.
7
5
1
8
2
.
2
5
2
.
7
0
5
0
2
.
5
6
1
0
.
6
3
M
ax
1
7
.
0
0
4
4
.
8
8
1
0
0
.
0
0
3
3
3
.
5
0
1
2
.
6
0
7
9
8
.
5
0
1
6
.
9
5
*
S
t
u
d
y
p
e
r
i
o
d
(
O
c
t
o
b
e
r
2
0
2
1
t
o
S
e
p
t
e
mb
e
r
2
0
2
2
)
a
t
1
5
m
i
n
u
t
e
s’
t
i
m
e
r
e
s
o
l
u
t
i
o
n
a
v
e
r
a
g
e
d
h
o
u
r
l
y
(
a)
(
b
)
(
a)
(
b
)
Fig
u
r
e
4
.
Dis
tr
ib
u
tio
n
o
f
d
if
f
er
en
t w
ea
th
er
p
ar
a
m
eter
s
p
r
esen
ted
in
v
io
lin
p
lo
ts
:
(
a)
am
b
ie
n
t
tem
p
er
atu
r
e
(
T
em
p
)
,
(
b
)
r
ad
iatio
n
(
R
ad
)
,
(c
)
h
u
m
id
ity
(
Hu
m
)
,
an
d
(
d
)
win
d
s
p
ee
d
(
W
s
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2088
-
8
6
9
4
I
n
t J Po
w
E
lec
&
Dr
i Sy
s
t
,
Vo
l.
1
5
,
No
.
4
,
Dec
em
b
er
20
2
4
:
2553
-
2
5
6
9
2560
Fig
u
r
e
5
.
Hea
t m
a
p
o
f
co
r
r
elat
io
n
co
ef
f
icien
ts
o
f
in
p
u
t v
ar
ia
b
les with
s
o
lar
p
h
o
to
v
o
ltaic
p
o
wer
o
u
tp
u
t
(
Pv
output
)
u
s
in
g
all
o
n
e
-
y
ea
r
d
ata
4.
P
E
RF
O
RM
A
NCE
M
E
T
R
I
C
S
T
h
e
r
2
(R
-
Sq
u
ar
ed
)
co
e
f
f
icien
t
o
f
r
eg
r
ess
io
n
,
r
o
o
t
m
ea
n
s
q
u
a
r
e
er
r
o
r
(
R
MSE
)
as
in
(
1
2
)
an
d
th
e
m
ea
n
ab
s
o
lu
te
p
er
ce
n
tag
e
er
r
o
r
(
MA
PE)
as
in
(
1
3
)
wer
e
ca
lcu
lated
an
d
ar
e
u
s
ed
as
ev
alu
atio
n
cr
iter
ia
to
v
alid
ate
th
e
er
r
o
r
an
d
ass
ess
h
o
w
well
th
e
p
r
o
p
o
s
ed
m
o
d
el
is
p
e
r
f
o
r
m
in
g
.
Als
o
,
s
k
ill
s
co
r
e
to
o
ca
lcu
late
d
b
y
co
n
s
id
er
ed
o
n
e
o
f
th
e
s
tatis
tical
m
o
d
els as
a
r
ef
er
en
ce
.
Her
e
i
n
th
is
s
tu
d
y
,
S
VR
is
th
e
r
ef
er
en
ce
m
o
d
el.
R
o
o
t m
ea
n
s
q
u
ar
e
e
r
r
o
r
(
R
MSE
)
:
=
√
1
∑
|
−
|
=
1
(
1
2
)
W
h
en
co
n
v
ey
in
g
m
o
d
el
p
er
f
o
r
m
an
ce
to
in
d
iv
i
d
u
als
with
o
u
t
a
b
ac
k
g
r
o
u
n
d
in
d
ata
an
al
y
s
is
,
MA
PE
p
r
o
v
es
s
u
p
er
io
r
to
R
MSE
d
u
e
to
its
s
im
p
licity
.
E
x
p
r
ess
ed
as
a
p
er
ce
n
tag
e,
MA
PE
is
m
o
r
e
s
tr
aig
h
tf
o
r
war
d
an
d
u
n
d
er
s
tan
d
a
b
le
f
o
r
en
d
u
s
er
s
,
r
en
d
er
in
g
it a
cc
ess
ib
le
ev
en
to
th
o
s
e
u
n
f
a
m
iliar
with
d
ata
co
n
ce
p
ts
.
Me
an
ab
s
o
lu
te
p
er
ce
n
tag
e
er
r
o
r
(
MA
PE)
:
=
1
∑
|
−
|
=
1
.
100%
(
1
3
)
T
h
er
e
is
a
co
m
m
o
n
ar
g
u
m
e
n
t th
at
m
ea
s
u
r
es o
f
f
o
r
ec
ast ac
cu
r
ac
y
o
u
g
h
t to
b
e
p
r
esen
ted
as a
s
k
ill s
co
r
e
(
1
4
)
.
Sk
ill s
co
r
e
(
SS
)
:
=
−
−
(
1
4
)
w
h
e
r
e
a
n
d
d
e
n
o
t
e
t
h
e
a
cc
u
r
a
c
y
o
f
t
h
e
f
o
r
e
c
a
s
ti
n
g
s
y
s
t
e
m
b
e
i
n
g
e
v
a
l
u
a
t
e
d
a
n
d
a
r
e
f
e
r
e
n
c
e
f
o
r
e
c
a
s
ti
n
g
s
y
s
te
m
,
r
e
s
p
ec
t
i
v
e
l
y
,
b
a
s
e
d
o
n
a
s
p
e
c
i
f
i
c
m
e
as
u
r
e
.
T
h
e
q
u
a
n
ti
ty
r
e
p
r
e
s
e
n
ts
t
h
e
p
e
r
s
is
t
e
n
t
m
o
d
e
l
a
c
c
u
r
a
c
y
v
al
u
e
o
f
t
h
e
m
e
a
s
u
r
e
;
s
i
g
n
i
f
y
i
n
g
t
h
e
m
e
t
r
i
c'
s
v
a
l
u
e
i
f
t
h
e
o
u
t
c
o
m
e
w
e
r
e
p
e
r
f
e
c
t
l
y
k
n
o
w
n
.
W
h
e
n
th
e
p
e
r
s
i
s
te
n
t
m
o
d
e
l
a
c
c
u
r
a
c
y
i
s
e
q
u
a
l
t
o
ze
r
o
,
a
d
i
f
f
e
r
e
n
t
s
t
a
t
is
ti
c
a
l
m
e
a
s
u
r
e
ca
n
b
e
u
s
e
d
t
o
c
o
m
p
a
r
e
t
h
e
p
e
r
f
o
r
m
a
n
c
e
o
f
t
wo
f
o
r
e
c
a
s
t
i
n
g
s
y
s
te
m
s
.
T
h
is
m
e
asu
r
e
i
s
d
e
f
i
n
e
d
as
r
e
l
at
i
v
e
s
k
i
l
l
s
c
o
r
e
(
1
5
)
,
a
n
a
l
t
e
r
n
a
t
i
v
e
t
o
a
s
k
i
l
l
s
c
o
r
e
[
2
6
]
-
[
2
8
]
.
R
elativ
e
s
k
ill s
co
r
e
(
SS
)
:
=
−
−
(
1
5
)
Fo
r
ca
lcu
latin
g
s
k
ill
s
co
r
e/r
el
ativ
e
s
k
ill
s
co
r
e
in
th
is
wo
r
k
,
th
e
ac
cu
r
ac
y
p
ar
am
eter
u
s
ed
is
MA
PE
o
v
er
its
ad
v
an
tag
es a
s
m
en
tio
n
ed
ab
o
v
e.
5.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
Fig
u
r
e
6
s
h
o
ws
th
e
r
e
g
r
ess
io
n
p
lo
t
f
o
r
th
e
ac
tu
al
a
n
d
p
r
ed
icted
v
alu
es
f
o
r
ec
asted
th
r
o
u
g
h
d
if
f
er
e
n
t
m
ac
h
in
e
lear
n
in
g
tech
n
iq
u
es
(
SVM,
ANN,
DNN
an
d
R
F)
f
o
r
m
u
ltip
le
s
tep
s
ah
ea
d
f
o
r
ec
asti
n
g
(
i.e
.
,
o
n
e
h
o
u
r
ah
ea
d
,
two
h
o
u
r
s
,
t
h
r
ee
h
o
u
r
s
an
d
f
o
u
r
h
o
u
r
s
a
h
ea
d
f
o
r
ec
asti
n
g
)
o
f
s
o
lar
p
h
o
t
o
v
o
ltaic
p
o
wer
f
o
r
d
if
f
e
r
en
t
s
ea
s
o
n
s
o
f
a
ty
p
ical
y
ea
r
with
a
m
en
tio
n
o
f
c
o
ef
f
icien
t
o
f
r
eg
r
ess
io
n
r
2
v
alu
es.
T
h
e
m
o
d
el
h
av
in
g
r
2
v
alu
es
clo
s
er
to
o
n
e
is
th
e
ac
cu
r
ate
m
o
d
el.
B
ased
o
n
th
e
v
al
u
es
o
f
r
2
,
all
m
o
d
els
ar
e
g
iv
in
g
b
est
r
es
u
lts
in
win
ter
s
ea
s
o
n
,
f
o
llo
wed
b
y
s
u
m
m
er
a
n
d
/o
r
au
tu
m
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Po
w
E
lec
&
Dr
i Sy
s
t
I
SS
N:
2088
-
8
6
9
4
F
o
r
ec
a
s
tin
g
h
o
u
r
ly
s
h
o
r
t
-
term so
la
r
p
h
o
to
vo
lta
ic
p
o
w
er u
s
in
g
ma
ch
in
e
…
(
S
r
a
va
n
ku
ma
r
J
o
g
u
n
u
r
i
)
2561
T
ab
le
2
c
o
n
tain
s
er
r
o
r
m
etr
ics
f
o
r
d
i
f
f
er
en
t
f
o
r
ec
asti
n
g
h
o
r
izo
n
s
an
d
v
ar
io
u
s
m
ac
h
in
e
lear
n
i
n
g
m
o
d
els
(
r
an
d
o
m
f
o
r
est
-
RF
,
d
ee
p
n
e
u
r
a
l
n
etwo
r
k
-
DNN,
ar
tific
ial
n
e
u
r
al
n
etwo
r
k
-
ANN
,
Su
p
p
o
r
t
v
ec
t
o
r
m
ac
h
i
n
e
-
SVM
).
T
h
e
er
r
o
r
m
etr
ics u
s
ed
ar
e:
-
r
2
(R
-
s
q
u
ar
ed
)
: T
h
e
s
tatis
tical
m
ea
s
u
r
e
th
at
r
ep
r
esen
ts
th
e
p
r
o
p
o
r
tio
n
o
f
v
a
r
ian
ce
in
th
e
d
e
p
en
d
en
t v
a
r
iab
le
th
at
is
p
r
ed
ictab
le
f
r
o
m
th
e
in
d
ep
en
d
en
t
v
a
r
iab
le.
I
t
in
d
icate
s
h
o
w
well
th
e
m
o
d
el
f
its
th
e
d
at
a.
Hig
h
er
v
alu
es
in
d
icate
a
b
etter
f
it.
-
Me
an
s
q
u
ar
e
d
e
r
r
o
r
(
MSE
)
:
I
t
m
ea
s
u
r
es
th
e
av
e
r
ag
e
s
q
u
ar
e
d
d
if
f
er
en
ce
b
etwe
en
th
e
ac
tu
al
an
d
p
r
e
d
icted
v
alu
es.
L
o
wer
v
al
u
es in
d
icate
b
etter
m
o
d
el
p
er
f
o
r
m
an
ce
.
-
Me
an
ab
s
o
lu
te
p
er
ce
n
tag
e
e
r
r
o
r
(
MA
PE
)
:
I
t
m
ea
s
u
r
es
th
e
p
er
ce
n
tag
e
d
if
f
e
r
en
ce
b
etwe
e
n
th
e
ac
tu
al
a
n
d
p
r
ed
icted
v
al
u
es.
L
o
wer
v
alu
e
s
in
d
icate
b
etter
m
o
d
el
p
er
f
o
r
m
an
ce
.
A
s
tep
ah
ea
d
(
1
h
o
u
r
ah
ea
d
)
f
o
r
ec
asti
n
g
’
s
p
lo
ts
s
h
o
wn
i
n
F
ig
u
r
e
6
,
f
o
r
d
if
f
er
en
t
s
ea
s
o
n
s
.
Du
r
in
g
au
tu
m
n
s
ea
s
o
n
,
th
e
r
2
v
alu
es
ar
e
0
.
8
8
5
,
0
.
8
8
5
,
0
.
8
7
7
an
d
0
.
8
4
7
f
o
r
m
o
d
els
R
F,
DNN
,
SVM
an
d
ANN,
r
esp
ec
tiv
ely
.
T
h
e
R
F
m
o
d
el
h
a
s
th
e
h
ig
h
est
v
alu
e
an
d
p
er
f
o
r
m
s
b
etter
th
an
th
e
o
th
e
r
m
o
d
el
s
o
f
th
e
s
tu
d
y
.
Du
r
i
n
g
win
ter
s
ea
s
o
n
,
th
e
r
2
v
alu
es
ar
e
0
.
9
5
5
,
0
.
9
4
9
,
0
.
9
4
5
a
n
d
0
.
9
3
6
f
o
r
m
o
d
els
R
F,
DNN,
SVM
an
d
ANN,
r
esp
ec
tiv
ely
.
T
h
e
R
F
m
o
d
el
h
a
s
th
e
h
ig
h
est
v
alu
e
an
d
p
er
f
o
r
m
s
b
etter
th
an
th
e
o
th
e
r
m
o
d
el
s
o
f
th
e
s
tu
d
y
.
Du
r
i
n
g
s
u
m
m
er
s
ea
s
o
n
,
th
e
r
2
v
alu
es
ar
e
0
.
8
9
7
,
0
.
8
8
2
,
0
.
8
7
6
a
n
d
0
.
8
6
5
f
o
r
m
o
d
els
DNN,
ANN,
R
F
an
d
SVM,
r
esp
ec
tiv
ely
.
T
h
e
DNN
m
o
d
el
h
as
th
e
h
ig
h
est
v
alu
e
an
d
p
er
f
o
r
m
s
b
etter
th
an
th
e
o
th
e
r
m
o
d
els
o
f
th
e
s
tu
d
y
.
Du
r
in
g
m
o
n
s
o
o
n
s
ea
s
o
n
,
th
e
r
2
v
alu
es
ar
e
e
x
tr
em
ely
lo
w,
b
u
t
R
F
m
o
d
el
h
as
th
e
h
ig
h
est
v
alu
e
an
d
p
er
f
o
r
m
s
b
etter
th
an
th
e
o
th
er
m
o
d
els o
f
th
e
s
tu
d
y
.
T
wo
s
tep
s
ah
ea
d
(
2
h
o
u
r
s
ah
e
ad
)
f
o
r
ec
asti
n
g
’
s
p
lo
ts
s
h
o
wn
in
Fig
u
r
e
7
,
f
o
r
d
if
f
er
e
n
t
s
ea
s
o
n
s
.
Du
r
in
g
au
tu
m
n
s
ea
s
o
n
,
th
e
r
2
v
alu
es
ar
e
0
.
8
6
6
,
0
.
8
3
6
,
0
.
8
3
4
an
d
0
.
8
3
4
f
o
r
m
o
d
els
R
F,
ANN
,
SVM
an
d
ANN,
r
esp
ec
tiv
ely
.
T
h
e
R
F
m
o
d
el
h
a
s
th
e
h
ig
h
est
v
alu
e
an
d
p
er
f
o
r
m
s
b
etter
th
an
th
e
o
th
e
r
m
o
d
el
s
o
f
th
e
s
tu
d
y
.
Du
r
i
n
g
win
ter
s
ea
s
o
n
,
th
e
r
2
v
alu
es
ar
e
0
.
9
2
1
,
0
.
9
2
1
,
0
.
9
1
6
a
n
d
0
.
9
0
6
f
o
r
m
o
d
els
R
F,
SVM,
DNN
an
d
ANN,
r
esp
ec
tiv
ely
.
T
h
e
R
F
m
o
d
el
h
a
s
th
e
h
ig
h
est
v
alu
e
an
d
p
er
f
o
r
m
s
b
etter
th
an
th
e
o
th
e
r
m
o
d
el
s
o
f
th
e
s
tu
d
y
.
Du
r
i
n
g
s
u
m
m
er
s
ea
s
o
n
,
th
e
r
2
v
alu
es
ar
e
0
.
8
4
5
,
0
.
8
3
8
,
0
.
8
3
7
an
d
0
.
8
2
8
f
o
r
m
o
d
els
R
F,
SV
M,
ANN
an
d
DNN,
r
esp
ec
tiv
ely
.
T
h
e
R
F
m
o
d
el
h
a
s
th
e
h
ig
h
est
v
alu
e
an
d
p
er
f
o
r
m
s
b
etter
th
an
th
e
o
th
e
r
m
o
d
el
s
o
f
th
e
s
tu
d
y
.
Du
r
i
n
g
m
o
n
s
o
o
n
s
ea
s
o
n
,
t
h
e
r
2
v
alu
e
s
ar
e
0
.
3
1
6
,
0
.
3
1
1
,
0
.
2
9
2
an
d
0
.
2
6
0
f
o
r
m
o
d
els
ANN,
R
F,
S
VM
an
d
DNN.
T
h
e
ANN
m
o
d
el
h
as th
e
h
ig
h
est v
alu
e
an
d
p
er
f
o
r
m
s
b
etter
th
an
th
e
o
th
er
m
o
d
els o
f
th
e
s
tu
d
y
.
Fig
u
r
e
8
illu
s
tr
ates
p
lo
ts
f
o
r
T
h
r
ee
-
h
o
u
r
-
a
h
ea
d
f
o
r
ec
asti
n
g
ac
r
o
s
s
d
if
f
er
en
t
s
ea
s
o
n
s
.
I
n
au
tu
m
n
,
r
2
v
alu
es
ar
e
0
.
8
7
7
,
0
.
8
6
5
,
0
.
8
4
4
,
an
d
0
.
8
1
1
f
o
r
DNN,
R
F,
ANN,
an
d
SVM
m
o
d
els
r
esp
ec
tiv
ely
,
with
DNN
o
u
tp
er
f
o
r
m
in
g
o
th
e
r
s
.
W
in
ter
s
h
o
ws
r
2
v
alu
es
o
f
0
.
9
3
1
,
0
.
8
9
1
,
0
.
8
8
8
,
an
d
0
.
8
8
7
f
o
r
DNN,
R
F,
SVM,
an
d
ANN
m
o
d
els
r
esp
ec
tiv
ely
,
with
DN
N
ag
ain
lead
in
g
.
Su
m
m
er
s
ea
s
o
n
d
ep
icts
r
2
v
alu
es
o
f
0
.
8
5
4
,
0
.
8
3
1
,
0
.
8
2
8
,
an
d
0
.
8
1
6
f
o
r
R
F,
DNN,
ANN,
an
d
SVM
m
o
d
els
r
esp
ec
tiv
ely
,
with
R
F
p
er
f
o
r
m
in
g
b
est.
Mo
n
s
o
o
n
s
ea
s
o
n
p
r
esen
t
s
r
2
v
alu
es
o
f
0
.
3
8
4
,
0
.
3
6
9
,
0
.
3
2
3
,
an
d
0
.
2
6
0
f
o
r
ANN,
R
F,
DNN,
an
d
SVM
m
o
d
els
r
esp
ec
tiv
ely
,
with
ANN
ex
h
ib
itin
g
th
e
h
ig
h
est p
e
r
f
o
r
m
an
ce
.
I
n
Fig
u
r
e
9
,
p
l
o
ts
f
o
r
f
o
u
r
-
h
o
u
r
-
ah
ea
d
f
o
r
ec
asti
n
g
ac
r
o
s
s
d
if
f
er
en
t
s
ea
s
o
n
s
ar
e
d
is
p
lay
ed
.
I
n
au
tu
m
n
,
r
2
v
alu
es
ar
e
0
.
8
7
5
,
0
.
8
7
0
,
0
.
8
6
8
,
a
n
d
0
.
8
6
0
f
o
r
DNN,
ANN,
R
F,
an
d
SVM
m
o
d
els
r
esp
ec
tiv
ely
,
with
DNN
d
em
o
n
s
tr
atin
g
s
u
p
er
io
r
p
er
f
o
r
m
an
ce
.
W
in
ter
ex
h
ib
its
r
2
v
alu
es
o
f
0
.
8
1
6
,
0
.
8
1
0
,
0
.
7
9
1
,
a
n
d
0
.
7
7
9
f
o
r
DNN,
ANN,
SVM,
an
d
R
F
m
o
d
els
r
e
s
p
ec
tiv
ely
,
with
DNN
lead
in
g
ag
ain
.
Du
r
in
g
s
u
m
m
er
,
r
2
v
alu
es
ar
e
0
.
8
9
7
,
0
.
8
8
2
,
0
.
8
7
6
,
a
n
d
0
.
8
6
5
f
o
r
DNN,
ANN,
SVM,
an
d
R
F
m
o
d
els
r
esp
ec
tiv
ely
,
with
DNN
o
u
t
p
er
f
o
r
m
in
g
o
th
er
s
.
Mo
n
s
o
o
n
s
ea
s
o
n
s
h
o
ws
r
2
v
alu
es
o
f
0
.
4
0
3
,
0
.
3
6
8
,
0
.
3
4
5
,
a
n
d
0
.
3
1
7
f
o
r
ANN,
R
F,
DNN
,
an
d
SVM
m
o
d
els
r
esp
ec
tiv
ely
,
with
ANN
d
is
p
lay
in
g
th
e
h
ig
h
est p
er
f
o
r
m
a
n
ce
.
T
h
e
m
o
s
t sig
n
if
ican
t f
in
d
in
g
s
f
r
o
m
th
e
r
esu
lts
p
r
esen
ted
ab
o
v
e
ar
e:
-
Acr
o
s
s
all
f
o
r
ec
asti
n
g
h
o
r
izo
n
s
(
1
to
4
h
o
u
r
s
ah
ea
d
)
a
n
d
s
ea
s
o
n
s
,
th
e
r
an
d
o
m
f
o
r
est
(
R
F)
m
o
d
el
co
n
s
is
ten
tly
o
u
tp
er
f
o
r
m
s
o
th
e
r
m
o
d
els,
ex
h
ib
itin
g
h
ig
h
e
r
R
-
s
q
u
ar
ed
(
r
2
)
v
alu
es.
T
h
is
in
d
icate
s
th
at
R
F
p
r
o
v
id
es
a
b
etter
f
it
f
o
r
th
e
d
ata
an
d
y
ield
s
m
o
r
e
ac
cu
r
ate
p
r
e
d
ictio
n
s
co
m
p
ar
ed
to
s
u
p
p
o
r
t v
ec
to
r
m
ac
h
i
n
e
(
SVM)
,
ar
tific
ial
n
eu
r
al
n
etwo
r
k
(
ANN)
,
a
n
d
d
e
ep
n
eu
r
al
n
etwo
r
k
(
DNN)
m
o
d
els.
-
Seaso
n
al
v
ar
iatio
n
s
ig
n
if
ica
n
tly
im
p
ac
ts
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
m
o
d
els,
with
a
u
tu
m
n
an
d
win
ter
s
ea
s
o
n
s
g
en
er
ally
r
esu
ltin
g
in
h
i
g
h
er
r
2
v
alu
es
an
d
l
o
wer
er
r
o
r
m
etr
ics
,
in
d
icatin
g
b
etter
p
r
e
d
ictab
ilit
y
an
d
ac
c
u
r
ac
y
.
I
n
co
n
tr
ast,
th
e
s
u
m
m
er
s
ea
s
o
n
s
h
o
ws
lo
wer
r
2
v
alu
es
an
d
h
ig
h
er
er
r
o
r
m
etr
ics,
wh
ile
th
e
m
o
n
s
o
o
n
s
ea
s
o
n
ex
h
ib
its
th
e
lo
west
r
2
v
alu
es
an
d
h
ig
h
est
er
r
o
r
m
etr
ics,
s
u
g
g
esti
n
g
d
ec
r
ea
s
ed
p
r
ed
ictab
ili
ty
an
d
ac
cu
r
ac
y
d
u
r
in
g
th
ese
p
er
io
d
s
.
-
Desp
ite
th
e
ch
allen
g
in
g
co
n
d
itio
n
s
p
r
esen
ted
b
y
th
e
m
o
n
s
o
o
n
s
ea
s
o
n
,
th
e
R
an
d
o
m
Fo
r
est
m
o
d
el
co
n
s
is
ten
tly
d
em
o
n
s
tr
ates
s
u
p
er
io
r
p
er
f
o
r
m
a
n
ce
,
h
ig
h
lig
h
tin
g
its
r
o
b
u
s
tn
ess
an
d
ef
f
ec
tiv
en
ess
in
h
an
d
lin
g
s
ea
s
o
n
al
v
ar
iatio
n
s
an
d
p
r
o
v
id
i
n
g
ac
cu
r
ate
f
o
r
ec
asts
f
o
r
s
o
lar
p
h
o
to
v
o
ltaic
p
o
wer
.
Ad
d
itio
n
a
lly
,
th
e
f
in
d
in
g
s
u
n
d
er
s
co
r
e
t
h
e
im
p
o
r
tan
ce
o
f
co
n
s
id
er
in
g
s
ea
s
o
n
al
d
y
n
am
ic
s
wh
en
d
ev
elo
p
in
g
an
d
e
v
alu
a
tin
g
f
o
r
ec
asti
n
g
m
o
d
els f
o
r
r
en
ewa
b
le
e
n
er
g
y
ap
p
licatio
n
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2088
-
8
6
9
4
I
n
t J Po
w
E
lec
&
Dr
i Sy
s
t
,
Vo
l.
1
5
,
No
.
4
,
Dec
em
b
er
20
2
4
:
2553
-
2
5
6
9
2562
Fo
r
ec
asti
n
g
o
n
e
h
o
u
r
ah
ea
d
Au
tu
m
n
W
in
ter
Su
m
m
er
Mo
n
s
o
o
n
(
a)
r
2
=0
.
8
7
7
r
2
=0
.
9
4
5
r
2
=0
.
8
6
5
r
2
=0
.
2
0
9
(
b
)
r
2
=
0
.
8
4
7
r
2
=
0
.
9
3
6
r
2
=
0
.
8
8
2
r
2
=
0
.
2
1
4
(
c)
r
2
=
0
.
8
8
5
r
2
=
0
.
9
4
9
r
2
=
0
.
8
9
7
r
2
=
0
.
1
9
1
(
d
)
r
2
=
0
.
8
8
5
r
2
=
0
.
9
5
5
r
2
=
0
.
8
7
6
r
2
=
0
.
2
5
3
Fig
u
r
e
6
.
R
eg
r
ess
io
n
p
l
o
ts
f
o
r
(
a)
SVM,
(
b
)
ANN,
(
c
)
DNN,
an
d
(
d
)
R
F f
o
r
d
if
f
er
en
t seaso
n
s
f
o
r
o
n
e
s
tep
ah
ea
d
(
1
h
o
u
r
ah
ea
d
f
o
r
ec
asti
n
g
)
Evaluation Warning : The document was created with Spire.PDF for Python.