I
nte
rna
t
io
na
l J
o
urna
l o
f
Appl
ied P
o
wer
E
ng
i
neer
ing
(
I
J
AP
E
)
Vo
l.
14
,
No
.
2
,
J
u
n
e
20
25
,
p
p
.
264
~
2
7
4
I
SS
N:
2252
-
8
7
9
2
,
DOI
:
1
0
.
1
1
5
9
1
/ijap
e
.
v
1
4
.
i
2
.
pp
264
-
274
264
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
p
e
.
ia
esco
r
e.
co
m
Powering
t
he f
utu
re of e
l
ectr
i
ca
l loa
d f
o
recas
ting usi
n
g
a
regres
sio
n learne
r in
ma
chin
e lear
ning
Su
s
ha
m
a
D.
Wa
nk
ha
de
1,
2
,
B
a
ba
s
a
heb R.
P
a
t
il
3
1
D
e
p
a
r
t
me
n
t
o
f
El
e
c
t
r
i
c
a
l
En
g
i
n
e
e
r
i
n
g
,
S
a
r
d
a
r
P
a
t
e
l
C
o
l
l
e
g
e
o
f
E
n
g
i
n
e
e
r
i
n
g
,
A
n
d
h
e
r
i
,
I
n
d
i
a
2
D
e
p
a
r
t
me
n
t
o
f
I
n
st
r
u
m
e
n
t
a
t
i
o
n
E
n
g
i
n
e
e
r
i
n
g
,
B
h
a
r
a
t
i
V
i
d
y
a
p
e
e
t
h
C
o
l
l
e
g
e
o
f
En
g
i
n
e
e
r
i
n
g
,
N
a
v
i
M
u
m
b
a
i
,
I
n
d
i
a
3
V
i
sh
w
a
n
i
k
e
t
a
n
’
s I
n
st
i
t
u
t
e
o
f
M
a
n
a
g
e
men
t
E
n
t
r
e
p
r
e
n
e
u
r
s
h
i
p
a
n
d
En
g
i
n
e
e
r
i
n
g
Te
c
h
n
o
l
o
g
y
,
K
u
mb
h
i
v
a
l
i
,
I
n
d
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Feb
2
8
,
2
0
2
4
R
ev
is
ed
Oct
1
0
,
2
0
2
4
Acc
ep
ted
Oct
2
3
,
2
0
2
4
Th
e
p
rima
ry
in
te
n
t
o
f
th
e
p
re
se
n
t
re
se
a
rc
h
wa
s
to
d
e
sig
n
a
n
d
e
x
e
c
u
te
a
n
e
lec
tri
c
a
l
lo
a
d
fo
re
c
a
stin
g
s
y
ste
m
u
sin
g
m
a
c
h
in
e
lea
rn
i
n
g
(M
L)
te
c
h
n
iq
u
e
s.
Th
e
imp
lem
e
n
tatio
n
o
f
a
n
a
d
v
a
n
c
e
d
p
re
d
icti
v
e
m
e
th
o
d
,
sp
e
c
ifi
c
a
l
ly
a
n
ML
a
lg
o
rit
h
m
,
h
e
lp
e
d
i
n
a
c
c
u
ra
te
lo
a
d
f
o
re
c
a
stin
g
,
wh
ich
is
c
ru
c
ial
fo
r
e
fficie
n
t
p
o
we
r
g
rid
m
a
n
a
g
e
m
e
n
t,
a
n
d
o
p
ti
m
izin
g
re
so
u
rc
e
a
ll
o
c
a
ti
o
n
.
El
e
c
tr
icity
l
o
a
d
flu
c
tu
a
tes
d
u
e
to
v
a
ri
o
u
s
c
o
m
p
l
e
x
fa
c
to
rs
,
m
a
k
in
g
trad
it
i
o
n
a
l
f
o
re
c
a
stin
g
m
e
th
o
d
s
str
u
g
g
le.
T
h
is
is
wh
e
re
ML
sh
i
n
e
s.
M
L
a
l
g
o
ri
th
m
s
c
a
n
l
e
a
rn
fro
m
h
isto
rica
l
d
a
ta,
id
e
n
ti
fy
i
n
g
in
tr
ica
te
p
a
tt
e
rn
s
a
n
d
re
latio
n
sh
i
p
s
th
a
t
in
flu
e
n
c
e
e
lec
tri
c
it
y
d
e
m
a
n
d
.
T
h
is
a
ll
o
ws
t
h
e
m
to
m
a
k
e
m
o
re
a
c
c
u
ra
te
p
re
d
ict
io
n
s
th
a
n
sta
ti
c
m
o
d
e
ls.
In
t
h
is
wo
rk
,
re
g
re
ss
io
n
lea
rn
i
n
g
m
o
d
e
ls
in
ML
a
re
u
se
d
with
th
e
M
ATLAB
p
latf
o
rm
.
Th
re
e
y
e
a
rs
o
f
re
a
l
-
ti
m
e
d
a
ta
fro
m
th
e
Wav
i
su
b
sta
ti
o
n
in
In
d
ia
a
re
u
se
d
.
Co
n
sid
e
rin
g
d
a
y
,
d
a
te,
h
o
u
r
o
f
d
a
y
,
m
a
x
a
n
d
m
i
n
tem
p
e
ra
tu
re
o
f
th
e
d
a
y
,
a
n
d
v
o
l
tag
e
a
n
d
c
u
rre
n
t
a
re
tak
e
n
a
s in
p
u
t
p
a
ra
m
e
ters
to
tes
t
f
o
u
rtee
n
d
iffere
n
t
m
o
d
e
l
s
o
f
a
ss
o
rted
re
g
re
ss
io
n
a
lg
o
rit
h
m
s.
Th
e
p
e
rfo
rm
a
n
c
e
o
f
th
e
se
m
o
d
e
ls
is ev
a
lu
a
ted
u
sin
g
c
o
m
m
o
n
ly
u
se
d
m
e
tri
c
s,
ro
o
t
m
e
a
n
sq
u
a
re
e
rro
r
(
RM
S
E
)
,
m
e
a
n
sq
u
a
r
e
d
e
rro
r
(
M
S
E
)
,
a
n
d
m
e
a
n
a
b
s
o
lu
te
e
rro
r
(
M
AE
)
,
a
lo
n
g
wit
h
a
fe
w
o
th
e
r
p
a
ra
m
e
ters
.
Th
e
o
p
ti
m
ize
d
trai
n
e
d
m
o
d
e
l
is
th
e
n
tes
ted
with
re
a
l
d
a
ta
to
o
b
tai
n
th
e
f
o
re
c
a
ste
d
lo
a
d
.
Th
e
c
o
rre
latio
n
b
e
twe
e
n
th
e
Ac
tu
a
l
l
o
a
d
a
n
d
f
o
re
c
a
ste
d
lo
a
d
is f
o
u
n
d
to
b
e
0
.
9
9
9
9
6
2
.
K
ey
w
o
r
d
s
:
Ar
tific
ial
n
eu
r
al
n
etwo
r
k
E
lectr
ical
lo
ad
f
o
r
ec
asti
n
g
Ma
ch
in
e
lear
n
in
g
R
eg
r
ess
io
n
lear
n
er
R
o
o
t m
ea
n
s
q
u
ar
e
e
r
r
o
r
Su
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
W
id
e
n
eu
r
al
n
etwo
r
k
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
:
Su
s
h
am
a
D.
W
an
k
h
ad
e
Dep
ar
tm
en
t o
f
I
n
s
tr
u
m
e
n
tatio
n
E
n
g
in
ee
r
in
g
,
B
h
ar
ati
Vid
y
ap
ee
th
C
o
lleg
e
o
f
E
n
g
in
ee
r
in
g
C
B
D
B
elap
u
r
,
Nav
i M
u
m
b
ai,
Ma
h
ar
ash
tr
a
4
0
0
6
1
4
,
I
n
d
ia
E
m
ail:
s
u
s
h
am
awa
n
k
h
ad
e1
3
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
L
o
ad
f
o
r
ec
asti
n
g
is
a
cr
u
cial
co
m
p
o
n
e
n
t
o
f
b
o
th
th
e
p
lan
n
in
g
an
d
o
p
er
atio
n
o
f
p
o
wer
s
y
s
tem
s
.
Acc
u
r
ate
lo
ad
p
r
ed
ictio
n
is
cr
u
cial
f
o
r
d
eter
m
in
i
n
g
th
e
s
c
h
ed
u
lin
g
o
f
p
o
wer
u
n
its
,
p
la
n
n
in
g
f
o
r
ca
p
ac
ity
,
im
p
r
o
v
in
g
th
e
n
etwo
r
k
,
an
d
m
an
ag
in
g
d
em
a
n
d
f
r
o
m
c
o
n
s
u
m
er
s
[
1
]
.
Giv
e
n
th
e
ch
allen
g
es
ass
o
ciate
d
with
s
to
r
in
g
lar
g
e
am
o
u
n
ts
o
f
elec
tr
ical
en
er
g
y
an
d
th
e
f
lu
ct
u
atio
n
s
in
p
o
wer
d
em
an
d
,
it
is
n
ec
ess
ar
y
to
en
s
u
r
e
th
at
th
e
p
o
wer
g
e
n
er
atio
n
s
y
s
tem
is
ab
le
to
a
d
ap
t
t
o
ch
a
n
g
es
in
lo
ad
in
r
ea
l
-
tim
e.
L
o
ad
f
o
r
ec
asti
n
g
is
cr
u
cial
in
p
o
wer
in
f
r
astru
ctu
r
e
p
lan
n
in
g
an
d
g
r
id
o
p
er
atio
n
.
Pre
cis
e
lo
ad
f
o
r
ec
asti
n
g
h
elp
s
r
ed
u
ce
th
e
d
is
cr
ep
an
cy
b
etwe
en
elec
tr
icity
s
u
p
p
ly
a
n
d
d
em
an
d
,
th
er
eb
y
en
h
an
cin
g
th
e
s
tab
ilit
y
o
f
p
o
wer
s
y
s
tem
s
.
T
h
e
p
r
ec
is
io
n
o
f
th
e
f
o
r
ec
asti
n
g
m
o
d
els
is
cr
u
cial
in
m
an
ag
in
g
th
e
em
er
g
in
g
e
n
er
g
y
g
en
e
r
atio
n
an
d
co
n
s
u
m
p
tio
n
.
Ar
tific
ial
in
telli
g
en
ce
(
AI
)
-
b
ased
m
eth
o
d
s
ar
e
b
ein
g
r
esear
ch
ed
an
d
u
s
ed
in
a
wid
e
r
an
g
e
o
f
ap
p
licatio
n
s
ar
o
u
n
d
t
h
e
wo
r
ld
b
ec
au
s
e
th
ey
a
r
e
b
etter
at
h
an
d
lin
g
co
m
p
le
x
in
p
u
t
-
o
u
t
p
u
t
r
ela
tio
n
s
h
ip
s
.
Ma
ch
in
e
lear
n
in
g
(
ML
)
is
r
ev
o
l
u
tio
n
izi
n
g
elec
tr
ical
lo
ad
f
o
r
ec
asti
n
g
,
p
av
in
g
t
h
e
way
f
o
r
a
m
o
r
e
ef
f
icien
t,
r
eliab
le,
an
d
s
u
s
tain
ab
le
f
u
tu
r
e
f
o
r
t
h
e
p
o
w
er
g
r
id
[
2
]
,
[
3
]
.
T
r
a
d
itio
n
ally
,
p
r
ed
ictin
g
elec
tr
icity
d
em
a
n
d
r
elied
o
n
s
tatis
tical
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ap
p
l Po
wer
E
n
g
I
SS
N:
2252
-
8
7
9
2
P
o
w
erin
g
th
e
fu
tu
r
e
o
f e
lectrica
l lo
a
d
f
o
r
ec
a
s
tin
g
u
s
in
g
a
r
eg
r
ess
io
n
lea
r
n
er
…
(
S
u
s
h
a
ma
D.
Wa
n
kh
a
d
e
)
265
m
o
d
els
an
d
h
is
to
r
ical
tr
e
n
d
s
.
Ho
wev
er
,
th
e
r
is
e
o
f
ML
h
as
r
ev
o
lu
tio
n
ized
th
e
f
ield
,
o
f
f
er
i
n
g
m
o
r
e
ac
c
u
r
ate
an
d
f
lex
ib
le
f
o
r
ec
asti
n
g
.
Her
e'
s
h
o
w
ML
is
ch
an
g
in
g
th
e
g
am
e:
-
C
ap
tu
r
in
g
c
o
m
p
lex
ity
:
E
lectr
icity
d
em
an
d
is
in
f
lu
en
ce
d
b
y
a
m
y
r
iad
o
f
f
ac
to
r
s
,
lik
e
wea
th
er
,
tim
e
o
f
d
ay
,
h
o
lid
ay
s
,
a
n
d
ev
en
s
o
cial
ev
e
n
ts
.
ML
alg
o
r
ith
m
s
ca
n
u
n
r
av
el
th
ese
co
m
p
lex
r
elatio
n
s
h
i
p
s
an
d
cr
ea
te
m
o
d
els
th
at
ad
ap
t to
d
y
n
am
ic
c
o
n
d
itio
n
s
[
3
]
.
-
I
n
cr
ea
s
ed
ac
cu
r
ac
y
:
ML
m
o
d
els
h
av
e
th
e
ab
ilit
y
to
ac
q
u
ir
e
k
n
o
wled
g
e
f
r
o
m
ex
ten
s
iv
e
d
atasets
,
wh
ich
en
co
m
p
ass
h
is
to
r
ical
lo
ad
p
att
er
n
s
,
wea
th
er
f
o
r
ec
asts
,
an
d
r
ea
l
-
tim
e
g
r
id
i
n
f
o
r
m
atio
n
.
T
h
i
s
lead
s
to
m
o
r
e
p
r
ec
is
e
p
r
ed
ictio
n
s
,
r
e
d
u
cin
g
t
h
e
r
is
k
o
f
e
n
er
g
y
s
h
o
r
tag
es
o
r
o
v
er
p
r
o
d
u
ct
i
o
n
[
4
]
.
-
E
n
h
an
ce
d
s
ca
lab
ilit
y
:
ML
m
o
d
els
ca
n
h
a
n
d
le
lar
g
e
d
atasets
ef
f
icien
tly
,
m
a
k
in
g
th
em
i
d
ea
l
f
o
r
f
o
r
ec
asti
n
g
at
d
if
f
er
e
n
t
lev
els,
f
r
o
m
in
d
i
v
id
u
al
b
u
ild
in
g
s
to
e
n
tire
p
o
wer
g
r
id
s
.
T
h
is
f
lex
ib
ilit
y
e
m
p
o
wer
s
tailo
r
e
d
s
o
lu
tio
n
s
f
o
r
d
i
v
er
s
e
s
ce
n
ar
io
s
[
4
]
.
-
Pro
ac
tiv
e
p
lan
n
in
g
:
Acc
u
r
ate
f
o
r
ec
asts
en
ab
le
ef
f
icien
t
r
eso
u
r
ce
allo
ca
tio
n
,
o
p
tim
izin
g
p
o
wer
g
en
er
atio
n
an
d
d
is
tr
ib
u
tio
n
.
T
h
is
tr
an
s
late
s
to
co
s
t sav
in
g
s
,
r
ed
u
ce
d
em
i
s
s
io
n
s
,
an
d
im
p
r
o
v
ed
g
r
id
r
eli
ab
ilit
y
.
-
I
m
p
r
o
v
ed
g
r
id
m
an
a
g
em
en
t:
Pre
d
ictin
g
p
ea
k
d
em
a
n
d
a
llo
ws
u
tili
ties
to
o
p
tim
ize
g
en
er
atio
n
an
d
d
is
tr
ib
u
tio
n
,
r
e
d
u
cin
g
co
s
ts
an
d
en
h
a
n
cin
g
r
eliab
ilit
y
.
-
R
en
ewa
b
le
en
er
g
y
i
n
teg
r
atio
n
:
ML
ca
n
h
el
p
in
te
g
r
ate
t
h
e
v
ar
iab
le
o
u
tp
u
t
o
f
r
en
ewa
b
le
s
o
u
r
ce
s
lik
e
s
o
lar
an
d
win
d
in
t
o
th
e
g
r
id
,
m
a
x
im
izin
g
th
eir
co
n
tr
ib
u
tio
n
.
-
Dem
an
d
-
s
id
e
m
an
a
g
em
en
t:
Pr
ed
ictin
g
p
ea
k
d
em
an
d
allo
ws
u
tili
ties
to
o
p
tim
ize
g
en
er
atio
n
a
n
d
d
is
tr
ib
u
tio
n
,
r
ed
u
cin
g
co
s
ts
an
d
en
h
an
cin
g
r
eliab
ilit
y
.
B
y
u
n
d
er
s
tan
d
i
n
g
f
u
tu
r
e
lo
ad
,
u
tili
ties
ca
n
in
ce
n
tiv
ize
co
n
s
u
m
er
s
to
s
h
if
t c
o
n
s
u
m
p
tio
n
p
atter
n
s
,
s
m
o
o
th
in
g
d
e
m
an
d
p
ea
k
s
an
d
r
ed
u
cin
g
s
tr
ess
o
n
th
e
g
r
id
.
A
d
iv
er
s
e
s
et
o
f
ML
to
o
ls
is
at
p
lay
,
lik
e
r
eg
r
ess
io
n
m
o
d
els
,
wh
ich
in
clu
d
e
m
o
d
els
lik
e
r
an
d
o
m
f
o
r
ests
an
d
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
es
(
SVM)
.
T
h
ese
m
o
d
els
ca
p
tu
r
e
n
o
n
-
lin
ea
r
r
elatio
n
s
h
ip
s
b
et
wee
n
m
u
ltip
le
in
p
u
t
f
ea
tu
r
es
(
wea
th
er
,
tim
e
o
f
d
a
y
)
an
d
th
e
elec
tr
icity
l
o
ad
.
Dee
p
lear
n
in
g
tech
n
iq
u
es
lik
e
lo
n
g
s
h
o
r
t
-
ter
m
m
em
o
r
y
(
L
STM
)
n
etwo
r
k
s
ex
ce
l
at
h
a
n
d
lin
g
te
m
p
o
r
al
d
ata,
ef
f
ec
tiv
ely
ca
p
tu
r
i
n
g
d
ay
-
to
-
d
ay
an
d
s
ea
s
o
n
al
p
atter
n
s
in
elec
tr
icity
co
n
s
u
m
p
tio
n
.
Hy
b
r
id
ap
p
r
o
ac
h
es,
c
o
m
b
in
i
n
g
d
if
f
er
en
t
ML
alg
o
r
ith
m
s
ca
n
lev
er
ag
e
th
eir
s
tr
en
g
th
s
,
b
o
o
s
tin
g
o
v
er
all
f
o
r
ec
asti
n
g
a
cc
u
r
ac
y
.
R
eg
r
ess
io
n
r
ef
er
s
to
a
s
et
o
f
s
t
atis
tical
m
eth
o
d
s
u
s
ed
to
an
aly
z
e
th
e
r
elatio
n
s
h
ip
b
etwe
en
a
d
ep
en
d
e
n
t
v
ar
iab
le
an
d
o
n
e
o
r
m
o
r
e
in
d
ep
en
d
en
t
v
a
r
iab
les
.
A
r
eg
r
ess
i
o
n
m
o
d
el
ca
n
d
eter
m
in
e
if
th
er
e
is
a
r
elatio
n
s
h
ip
b
etwe
en
ch
an
g
es
in
th
e
d
ep
en
d
en
t
v
ar
iab
le
a
n
d
c
h
an
g
es
in
o
n
e
o
r
m
o
r
e
o
f
th
e
ex
p
lan
at
o
r
y
v
a
r
iab
les.
R
eg
r
ess
io
n
m
eth
o
d
s
ar
e
co
m
m
o
n
ly
em
p
lo
y
ed
in
elec
tr
ical
l
o
ad
f
o
r
ec
asti
n
g
to
p
r
ed
ict
f
u
tu
r
e
elec
tr
icity
u
s
ag
e
ac
cu
r
ately
[
5
]
.
Var
i
o
u
s
s
tu
d
ies
h
av
e
h
ig
h
lig
h
ted
th
e
ef
f
ec
tiv
en
ess
o
f
r
eg
r
ess
io
n
m
o
d
els
in
th
is
d
o
m
ain
.
Fo
r
in
s
tan
ce
,
a
s
tu
d
y
u
tili
ze
d
lin
ea
r
r
eg
r
ess
io
n
e
q
u
atio
n
s
t
o
f
o
r
ec
ast
elec
tr
icity
lo
ad
s
,
ac
h
iev
in
g
an
av
er
a
g
e
f
o
r
ec
asti
n
g
er
r
o
r
o
f
3
.
8
6
%
f
o
r
ac
tiv
e
p
o
wer
a
n
d
3
.
7
7
%
f
o
r
a
p
p
ar
en
t p
o
wer
[
6
]
.
Ad
d
itio
n
all
y
,
an
o
t
h
er
r
esear
c
h
p
ap
er
ev
al
u
ated
2
4
r
eg
r
ess
io
n
m
o
d
el
-
b
ased
alg
o
r
ith
m
s
f
o
r
h
a
lf
-
h
o
u
r
ly
lo
ad
f
o
r
ec
asti
n
g
,
wit
h
G
au
s
s
ian
p
r
o
ce
s
s
r
eg
r
ess
io
n
m
o
d
els d
em
o
n
s
tr
atin
g
th
e
b
est p
er
f
o
r
m
an
ce
[
7
]
.
F
u
r
th
er
m
o
r
e,
a
m
eta
-
r
e
g
r
ess
io
n
an
aly
s
is
id
en
tifie
d
th
e
L
STM
ap
p
r
o
ac
h
an
d
n
eu
r
al
n
etwo
r
k
s
co
m
b
in
ed
with
o
t
h
er
m
eth
o
d
s
as
ef
f
ec
tiv
e
f
o
r
e
ca
s
tin
g
tech
n
iq
u
es
,
em
p
h
asizin
g
th
e
im
p
o
r
tan
ce
o
f
m
o
d
el
s
elec
tio
n
in
lo
a
d
f
o
r
ec
asti
n
g
[
8
]
.
T
h
ese
f
in
d
i
n
g
s
u
n
d
er
s
co
r
e
t
h
e
s
ig
n
if
ican
ce
o
f
r
eg
r
ess
io
n
m
eth
o
d
s
in
ac
cu
r
ately
p
r
ed
icti
n
g
elec
tr
ical
lo
ad
s
,
aid
in
g
i
n
ef
f
icien
t
en
er
g
y
m
an
ag
em
en
t
an
d
r
eso
u
r
ce
allo
ca
tio
n
.
Utilized
ar
tific
ial
in
telli
g
en
ce
,
n
eu
r
al
n
etwo
r
k
,
AR
I
MA
m
o
d
els,
B
ay
esian
m
o
d
els,
an
d
r
eg
r
ess
io
n
m
o
d
e
ls
f
o
r
f
o
r
ec
asti
n
g
an
d
p
r
o
p
o
s
ed
a
s
o
lu
tio
n
t
o
th
e
p
r
o
b
lem
o
f
s
elec
tin
g
th
r
ee
p
ar
am
eter
s
f
o
r
th
e
s
u
p
p
o
r
t
v
ec
to
r
r
eg
r
ess
io
n
(
SVR
)
m
o
d
el
u
s
in
g
a
ch
a
o
tic
alg
o
r
ith
m
to
en
h
a
n
ce
g
lo
b
al
o
p
tim
izatio
n
an
d
p
r
e
v
en
t
f
all
in
g
in
to
lo
ca
l
o
p
tim
izatio
n
[
9
]
,
[
1
0
]
.
Gau
s
s
ian
p
r
o
ce
s
s
r
eg
r
ess
io
n
m
eth
o
d
is
r
ec
o
m
m
en
d
ed
f
o
r
lo
a
d
p
r
e
d
ictio
n
[
1
1
]
.
T
h
is
p
ap
er
f
o
c
u
s
es
o
n
r
eg
r
ess
io
n
lear
n
er
s
f
o
r
elec
tr
ical
lo
ad
p
r
ed
ictio
n
u
s
in
g
ML
,
u
tili
zin
g
r
ea
l
tim
e
h
o
u
r
ly
d
ata
f
r
o
m
J
an
u
ar
y
2
0
1
9
an
d
J
u
ly
,
2
0
2
2
f
r
o
m
th
e
3
3
/1
1
k
V
s
u
b
s
tatio
n
at
W
av
i,
I
n
d
ia
f
o
r
an
aly
s
is
to
co
m
p
ar
e
1
4
r
e
g
r
ess
io
n
m
o
d
el
s
lik
e
lin
ea
r
r
eg
r
ess
io
n
,
SVM
,
an
d
n
eu
r
al
n
etwo
r
k
s
.
T
h
e
m
a
in
co
n
tr
ib
u
tio
n
s
o
f
th
e
p
ap
er
in
clu
d
e:
i
)
p
r
o
p
o
s
e
lo
ad
f
o
r
ec
asti
n
g
ap
p
r
o
ac
h
f
o
r
W
av
i
s
u
b
s
tatio
n
an
d
ii
)
d
em
o
n
s
tr
ate
p
r
o
b
ab
ilis
tic
f
o
r
ec
asti
n
g
m
o
d
els
.
R
eg
r
ess
io
n
m
o
d
els'
p
er
f
o
r
m
an
ce
is
ev
alu
ated
u
s
in
g
r
o
o
t
m
ea
n
s
q
u
ar
e
er
r
o
r
(
R
MSE
)
,
m
ea
n
s
q
u
ar
ed
e
r
r
o
r
(
MSE
)
,
a
n
d
m
ea
n
ab
s
o
lu
te
e
r
r
o
r
(
MA
E
)
m
etr
i
cs
.
A
co
n
clu
s
io
n
is
d
r
awn
b
y
i
d
en
tify
in
g
in
tr
icate
p
atter
n
s
an
d
r
elatio
n
s
h
ip
s
af
f
e
ctin
g
elec
tr
icity
d
em
an
d
f
o
r
ac
cu
r
ate
p
r
ed
ictio
n
s
.
T
h
e
o
p
tim
ized
m
o
d
el
s
h
o
ws a
h
ig
h
co
r
r
elatio
n
b
etwe
en
ac
tu
al
an
d
f
o
r
ec
asted
lo
ad
.
2.
M
E
T
H
O
DO
L
O
G
Y
T
h
e
ex
p
er
im
e
n
tatio
n
is
d
o
n
e
with
r
eg
r
ess
io
n
lear
n
er
s
u
s
in
g
ML
o
n
MA
T
L
AB
p
latf
o
r
m
.
R
eg
r
ess
io
n
lear
n
er
is
a
MA
T
L
A
B
to
o
l th
a
t c
an
b
e
u
s
ed
to
tr
ain
d
if
f
er
en
t
r
eg
r
ess
io
n
m
o
d
els with
s
u
p
er
v
is
ed
ML
.
I
n
itially
,
th
e
r
ea
l
-
tim
e
d
ata
is
ac
q
u
ir
ed
f
r
o
m
th
e
s
u
b
s
tatio
n
,
a
n
d
,
ar
r
a
n
g
ed
o
n
a
d
aily
b
asis
f
o
r
twe
n
ty
-
f
o
u
r
h
o
u
r
s
.
T
h
e
f
ea
tu
r
es
s
elec
ted
f
o
r
d
ata
a
r
r
an
g
em
en
t
ar
e
d
ate
,
d
ay
o
f
th
e
we
ek
,
h
o
u
r
o
f
th
e
wee
k
,
a
n
d
m
a
x
im
u
m
an
d
m
in
im
u
m
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
9
2
I
n
t J Ap
p
l Po
wer
E
n
g
,
Vo
l.
14
,
No
.
2
,
J
u
n
e
20
25
:
26
4
-
27
4
266
tem
p
er
atu
r
e.
Data
is
th
en
test
ed
f
o
r
d
if
f
er
e
n
t
m
o
d
els.
I
n
th
i
s
ex
p
er
im
en
t
to
tal
o
f
f
o
u
r
teen
m
o
d
els
ar
e
test
ed
.
Af
ter
v
alid
atin
g
th
e
s
ch
e
m
e
a
n
d
p
ar
a
m
eter
o
p
tim
izatio
n
o
f
h
y
p
er
p
ar
am
eter
s
m
o
d
el
p
er
f
o
r
m
an
ce
is
ass
ess
ed
ag
ain
.
I
n
all
f
o
u
r
teen
m
o
d
els
f
r
o
m
f
i
v
e
r
e
g
r
ess
io
n
f
am
ilies
T
ab
le
1
a
r
e
test
ed
h
er
e
a
n
d
t
h
e
r
esu
lts
o
b
tain
ed
ar
e
tab
u
lated
as
s
h
o
wn
in
T
ab
le
2
.
T
h
e
o
p
tim
ized
m
o
d
el
is
th
en
test
ed
f
o
r
f
o
r
ec
asti
n
g
o
f
th
e
l
o
ad
.
T
h
e
d
if
f
er
en
t
m
o
d
els
u
s
ed
f
o
r
r
eg
r
ess
io
n
ar
e
d
is
cu
s
s
ed
b
elo
w.
T
h
e
c
o
m
p
lete
p
r
o
ce
s
s
f
lo
w
o
f
th
e
wo
r
k
is
as
s
h
o
wn
in
Fig
u
r
e
1.
Fig
u
r
e
1
.
Pro
ce
s
s
f
lo
wch
a
r
t f
o
r
f
o
r
ec
asti
n
g
T
ab
le
1
.
R
eg
r
ess
io
n
m
o
d
els u
s
ed
in
th
e
wo
r
k
F
a
mi
l
y
o
f
r
e
g
r
e
ss
i
o
n
mo
d
e
l
s
S
e
l
e
c
t
e
d
r
e
g
r
e
ssi
o
n
m
o
d
e
l
Li
n
e
a
r
r
e
g
r
e
s
si
o
n
Li
n
e
a
r
r
e
g
r
e
s
si
o
n
mo
d
e
l
R
e
g
r
e
ssi
o
n
t
r
e
e
s
M
e
d
i
u
m
t
r
e
e
C
o
a
r
se
t
r
e
e
F
i
n
e
t
r
e
e
S
u
p
p
o
r
t
v
e
c
t
o
r
ma
c
h
i
n
e
s
Li
n
e
a
r
S
V
M
Q
u
a
d
r
a
t
i
c
S
V
M
C
u
b
i
c
S
V
M
M
e
d
i
u
m
G
a
u
ss
i
a
n
S
V
M
C
o
a
r
se
G
a
u
ss
i
a
n
S
V
M
En
se
mb
l
e
o
f
t
r
e
e
s
B
o
o
st
e
d
t
r
e
e
s
B
a
g
g
e
d
t
r
e
e
s
N
e
u
r
a
l
n
e
t
w
o
r
k
s
N
a
r
r
o
w
n
e
u
r
a
l
n
e
t
w
o
r
k
M
e
d
i
u
m
n
e
u
r
a
l
n
e
t
w
o
r
k
W
i
d
e
n
e
u
r
a
l
n
e
t
w
o
r
k
2
.
1
.
L
inea
r
re
g
re
s
s
io
n m
o
de
l
A
lin
ea
r
r
eg
r
ess
io
n
m
o
d
el
is
a
s
tatis
t
ical
m
o
d
el
th
at
elu
cid
at
es
th
e
co
n
n
ec
tio
n
b
etwe
en
a
d
ep
en
d
en
t
v
ar
iab
le
a
n
d
o
n
e
o
r
m
o
r
e
in
d
ep
en
d
en
t
v
ar
ia
b
les.
T
h
e
d
ep
e
n
d
en
t
v
ar
iab
le
is
alter
n
ativ
ely
r
ef
er
r
ed
to
as
t
h
e
r
esp
o
n
s
e
v
ar
ia
b
le.
A
li
n
ea
r
m
o
d
el
ex
am
p
le
is
a
v
er
b
al
s
ce
n
ar
i
o
th
at
ca
n
b
e
m
o
d
elled
u
s
in
g
a
lin
ea
r
e
q
u
atio
n
o
r
v
ice
v
er
s
a
[
1
2
]
,
[
1
3
]
.
L
in
ea
r
r
eg
r
ess
io
n
is
em
p
lo
y
ed
to
p
r
e
cisely
ascer
tain
th
e
n
atu
r
e
an
d
m
ag
n
itu
d
e
o
f
th
e
r
elatio
n
s
h
ip
b
etwe
en
a
d
ep
e
n
d
en
t
v
a
r
iab
le
a
n
d
a
s
et
o
f
in
d
ep
en
d
en
t
v
ar
ia
b
les.
I
t
f
ac
ilit
ates
th
e
g
e
n
er
atio
n
o
f
m
o
d
els f
o
r
t
h
e
p
u
r
p
o
s
e
o
f
m
ak
in
g
p
r
e
d
ic
tio
n
s
[
1
4
]
.
2
.
2
.
SVM
re
g
re
s
s
io
n
SVM
r
eg
r
ess
io
n
,
also
k
n
o
wn
as
SVR
,
is
an
ML
alg
o
r
ith
m
u
tili
ze
d
f
o
r
r
eg
r
ess
io
n
an
aly
s
is
.
Un
lik
e
tr
ad
itio
n
al
lin
ea
r
r
e
g
r
ess
io
n
m
eth
o
d
s
,
th
is
ap
p
r
o
ac
h
s
ee
k
s
to
id
en
tify
a
h
y
p
er
p
la
n
e
th
at
o
p
tim
ally
alig
n
s
with
th
e
d
ata
p
o
i
n
ts
in
a
co
n
tin
u
o
u
s
s
p
ac
e,
r
ath
er
t
h
an
f
itti
n
g
a
lin
e
to
th
e
d
ata
p
o
in
ts
.
SVMs
ca
n
u
s
e
d
if
f
er
en
t
k
er
n
el
f
u
n
ctio
n
s
to
tr
an
s
f
o
r
m
th
e
d
ata
in
to
a
h
ig
h
e
r
-
d
im
e
n
s
io
n
al
s
p
ac
e,
allo
win
g
f
o
r
n
o
n
-
lin
ea
r
d
ec
is
io
n
b
o
u
n
d
ar
ies
[
1
5
]
.
T
h
e
ac
c
u
r
ac
y
o
f
th
e
test
r
esu
lts
with
th
e
SVM
m
eth
o
d
is
b
etter
th
an
th
e
lin
ea
r
r
eg
r
ess
io
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ap
p
l Po
wer
E
n
g
I
SS
N:
2252
-
8
7
9
2
P
o
w
erin
g
th
e
fu
tu
r
e
o
f e
lectrica
l lo
a
d
f
o
r
ec
a
s
tin
g
u
s
in
g
a
r
eg
r
ess
io
n
lea
r
n
er
…
(
S
u
s
h
a
ma
D.
Wa
n
kh
a
d
e
)
267
m
eth
o
d
.
T
h
e
K
er
n
el
tr
ick
is
th
e
p
r
im
ar
y
co
m
p
o
n
e
n
t
o
f
SVM
th
at
is
r
en
o
w
n
ed
f
o
r
its
s
ig
n
if
ican
ce
.
A
K
er
n
el
is
a
m
eth
o
d
f
o
r
ca
lcu
latin
g
th
e
d
o
t
p
r
o
d
u
ct
o
f
two
v
ec
to
r
s
,
x
,
an
d
y
,
in
a
f
ea
tu
r
e
s
p
ac
e
th
at
is
o
f
ten
o
f
v
er
y
h
i
g
h
d
im
en
s
io
n
ality
.
T
h
is
is
wh
y
K
er
n
el
f
u
n
ctio
n
s
ar
e
o
cc
asio
n
al
ly
r
ef
er
r
e
d
to
as
"g
en
er
alize
d
d
o
t
p
r
o
d
u
cts".
T
h
e
SVM
m
eth
o
d
ca
n
p
er
f
o
r
m
a
K
er
n
el
tr
ick
th
at
ca
n
o
v
e
r
co
m
e
t
h
e
n
o
n
-
lin
ea
r
d
is
tr
ib
u
tio
n
o
f
d
ata
[
1
5
]
,
[
1
6
]
.
T
ab
le
2
.
Per
f
o
r
m
an
ce
ev
alu
ati
o
n
o
f
d
if
f
er
en
t m
o
d
els
S
r
.
n
o
.
M
o
d
e
l
n
a
m
e
R
M
S
E
R
-
s
q
u
a
r
e
d
M
S
E
M
A
E
1
Li
n
e
a
r
0
.
0
7
5
1
4
1
0
.
9
9
0
.
0
0
5
6
4
6
2
0
.
0
5
0
3
8
1
2
F
i
n
e
t
r
e
e
0
.
0
7
3
4
4
4
0
.
9
9
0
.
0
0
5
3
9
4
1
0
.
0
3
3
6
7
1
3
M
e
d
i
u
m
t
r
e
e
0
.
0
7
6
9
7
2
0
.
9
9
0
.
0
0
5
9
2
4
7
0
.
0
3
5
5
3
1
4
C
o
a
r
se
t
r
e
e
0
.
0
8
0
9
2
5
0
.
9
9
0
.
0
0
6
5
4
8
9
0
.
0
3
9
3
3
9
5
Li
n
e
a
r
S
V
M
0
.
0
7
5
6
8
9
0
.
9
9
0
.
0
0
5
7
2
8
9
0
.
0
5
3
7
8
4
6
Q
u
a
d
r
a
t
i
c
S
V
M
0
.
0
3
6
5
5
6
1
.
0
0
0
.
0
0
1
3
3
6
3
0
.
0
3
0
9
3
7
7
C
u
b
i
c
S
V
M
0
.
0
4
3
3
6
1
1
.
0
0
0
.
0
0
1
8
8
0
2
0
.
0
3
5
8
3
3
8
M
e
d
i
u
m
G
a
u
ss
i
a
n
S
V
M
0
.
0
4
9
7
4
7
0
.
9
9
0
.
0
0
2
4
7
4
7
0
.
0
3
5
1
4
8
9
C
o
a
r
se
G
a
u
ss
i
a
n
S
V
M
0
.
0
3
2
6
0
7
1
.
0
0
0
.
0
0
1
0
6
3
2
0
.
0
2
3
2
4
5
10
B
o
o
st
e
d
t
r
e
e
s
0
.
0
9
1
9
9
0
.
9
8
0
.
0
0
8
4
6
2
1
0
.
0
7
0
6
8
7
11
B
a
g
g
e
d
t
r
e
e
s
0
.
0
9
5
2
5
2
0
.
9
8
0
.
0
0
9
0
7
3
0
.
0
5
0
1
7
7
12
N
a
r
r
o
w
n
e
u
r
a
l
n
e
t
w
o
r
k
0
.
0
2
1
9
8
3
1
.
0
0
0
.
0
0
0
4
8
3
2
6
0
.
0
0
5
4
2
5
9
13
M
e
d
i
u
m
n
e
u
r
a
l
n
e
t
w
o
r
k
0
.
0
1
7
1
9
8
1
.
0
0
0
.
0
0
0
2
9
5
7
8
0
.
0
0
3
3
3
9
14
W
i
d
e
n
e
u
r
a
l
n
e
t
w
o
r
k
0
.
0
1
2
5
7
6
1
.
0
0
0
.
0
0
0
1
5
8
1
6
0
.
0
0
2
7
1
4
3
2
.
3
.
E
ns
em
ble o
f
t
re
es
-
B
ag
g
ed
tr
ee
m
o
d
el
W
e
ca
n
cr
ea
te
a
r
an
d
o
m
f
o
r
est
b
y
co
m
b
in
in
g
m
u
ltip
le
d
ec
is
io
n
tr
ee
s
v
ia
a
tech
n
iq
u
e
ca
lle
d
b
ag
g
in
g
.
I
n
th
is
m
eth
o
d
m
u
ltip
le
d
ec
is
io
n
tr
ee
s
ar
e
t
r
ain
ed
o
n
d
if
f
er
e
n
t su
b
s
ets o
f
tr
ain
in
g
d
ata,
r
an
d
o
m
ly
s
am
p
le
d
with
r
ep
lace
m
en
t.
E
ac
h
tr
ee
u
n
d
er
g
o
es
in
d
ep
en
d
en
t
tr
ain
in
g
,
an
d
th
e
f
in
al
p
r
ed
ictio
n
is
d
er
iv
ed
b
y
av
er
a
g
in
g
th
e
p
r
ed
ictio
n
s
o
f
all
th
e
tr
ee
s
[
1
7
]
.
A
p
r
im
ar
y
co
n
s
tr
ain
t
o
f
b
ag
g
in
g
tr
ee
s
is
th
at
it
em
p
lo
y
s
th
e
co
m
p
lete
f
ea
tu
r
e
s
p
ac
e
d
u
r
in
g
th
e
p
r
o
ce
s
s
o
f
cr
e
atin
g
s
p
lits
in
th
e
tr
ee
s
.
I
f
ce
r
ta
in
v
ar
iab
les
w
ith
in
t
h
e
f
ea
tu
r
e
s
p
ac
e
ar
e
in
d
icatin
g
s
p
ec
if
ic
p
r
ed
ictio
n
s
,
th
er
e
is
a
p
o
s
s
ib
ilit
y
o
f
h
av
in
g
a
clu
s
ter
o
f
co
r
r
elate
d
tr
ee
s
,
wh
ich
u
ltima
tely
lead
s
to
an
in
cr
ea
s
e
in
b
ias an
d
a
d
ec
r
ea
s
e
in
v
ar
ian
ce
.
-
B
o
o
s
ted
tr
ee
m
o
d
el
T
h
e
p
r
im
a
r
y
b
en
ef
it
o
f
b
a
g
g
ed
tr
ee
s
lies
in
th
eir
r
elian
ce
o
n
m
u
ltip
le
d
ec
is
io
n
tr
ee
s
in
s
tead
o
f
a
s
in
g
le
o
n
e,
en
ab
lin
g
th
e
u
tili
za
tio
n
o
f
th
e
co
llectiv
e
k
n
o
wled
g
e
f
r
o
m
n
u
m
e
r
o
u
s
m
o
d
els.
Dec
r
ea
s
es
v
ar
iab
ilit
y
b
y
tak
in
g
th
e
av
er
a
g
e
o
f
p
r
ed
icti
o
n
s
m
ad
e
b
y
m
o
d
els
tr
ain
ed
o
n
d
is
tin
ct
s
u
b
s
ets
o
f
d
ata.
E
f
f
icien
t
f
o
r
m
o
d
els
ex
h
ib
itin
g
s
ig
n
if
ican
t
v
ar
iab
ili
ty
.
B
o
o
s
tin
g
,
m
itig
ates
b
ias
b
y
iter
ativ
ely
tr
ain
in
g
m
o
d
els
th
a
t
s
p
ec
if
ically
tar
g
et
th
e
er
r
o
r
s
m
a
d
e
b
y
p
r
e
v
io
u
s
m
o
d
els.
Su
itab
le
f
o
r
m
o
d
els ex
h
ib
itin
g
s
ig
n
if
ican
t b
ias
[
1
8
]
,
[
1
9
]
.
2
.
4
.
Neura
l
net
wo
r
k
s
T
h
e
n
etwo
r
k
lear
n
s
f
r
o
m
in
p
u
t
-
o
u
tp
u
t
d
ata
p
air
s
,
a
d
ju
s
tin
g
i
ts
weig
h
ts
an
d
b
iases
to
a
p
p
r
o
x
im
ate
th
e
u
n
d
er
ly
i
n
g
r
elatio
n
s
h
ip
b
etwe
en
th
e
in
p
u
t
v
a
r
iab
les
an
d
t
h
e
tar
g
et
v
ar
iab
le
[
2
0
]
.
T
h
is
e
n
ab
les
n
eu
r
al
n
etwo
r
k
s
to
p
er
f
o
r
m
r
e
g
r
ess
io
n
task
s
,
m
ak
in
g
th
e
m
v
alu
a
b
le
in
v
ar
i
o
u
s
p
r
ed
ictiv
e
a
n
d
f
o
r
ec
asti
n
g
ap
p
licatio
n
s
w
id
e
n
eu
r
al
n
etwo
r
k
s
ar
e
c
h
ar
ac
ter
ized
b
y
h
av
in
g
a
s
m
aller
n
u
m
b
er
o
f
h
id
d
en
lay
e
r
s
(
ty
p
icall
y
1
-
2
)
,
b
u
t
a
lar
g
er
n
u
m
b
er
o
f
n
e
u
r
o
n
s
p
er
lay
er
[
2
1
]
,
[
2
2
]
.
Ne
u
r
al
n
etwo
r
k
s
ar
e
an
ex
citin
g
a
n
d
p
r
o
m
is
in
g
ty
p
e
o
f
ML
alg
o
r
ith
m
th
at
can
h
elp
us
b
etter
u
n
d
er
s
ta
n
d
an
d
p
r
ed
ict
co
m
p
lex
p
atter
n
s
an
d
r
elatio
n
s
h
ip
s
.
As
th
e
n
et
wo
r
k
r
ec
eiv
es
m
o
r
e
d
ata,
it
ad
ju
s
ts
it
s
weig
h
ts
an
d
b
iases
to
ap
p
r
o
x
im
ate
th
e
u
n
d
e
r
ly
in
g
ass
o
ciatio
n
b
etwe
en
th
e
tar
g
et
v
ar
iab
le
an
d
th
e
in
p
u
t
v
ar
iab
les
[
2
3
]
.
Neu
r
al
n
etwo
r
k
s
ca
n
h
an
d
le
lar
g
e
d
atasets
ef
f
icien
tly
,
m
ak
in
g
th
em
s
u
itab
le
f
o
r
ap
p
licatio
n
s
with
ex
ten
s
iv
e
h
i
s
to
r
ical
lo
ad
d
ata
[
2
4
]
.
T
h
is
s
ca
lab
ilit
y
en
s
u
r
es
th
at
m
o
d
els
can
be
tr
ain
ed
on
co
m
p
r
eh
e
n
s
iv
e
d
atasets
,
p
o
ten
tially
lead
in
g
to
m
o
r
e
ac
cu
r
ate
f
o
r
ec
asts
.
T
h
is
m
ak
es
n
eu
r
al
n
etwo
r
k
s
u
s
ef
u
l
in
v
ar
io
u
s
p
r
ed
ictiv
e
an
d
f
o
r
ec
as
tin
g
ap
p
licatio
n
s
,
as
th
e
y
ca
n
p
er
f
o
r
m
r
eg
r
ess
io
n
task
s
.
W
id
e
n
eu
r
al
n
etwo
r
k
s
h
av
e
a
s
m
aller
n
u
m
b
er
of
h
id
d
en
lay
er
s
(
ty
p
ically
1
-
2
)
,
b
u
t
a
lar
g
er
n
u
m
b
e
r
o
f
n
e
u
r
o
n
s
p
er
lay
er
[
2
5
]
,
[
2
6
]
.
A
wid
e
n
eu
r
al
n
etwo
r
k
m
o
d
e
l
f
o
r
r
eg
r
ess
io
n
ty
p
ically
i
n
v
o
lv
es
a
n
e
u
r
al
n
etwo
r
k
ar
ch
it
ec
tu
r
e
with
a
lar
g
e
n
u
m
b
er
o
f
n
eu
r
o
n
s
in
its
lay
er
s
.
T
h
e
m
o
d
el
ca
n
b
e
d
escr
ib
ed
m
ath
em
atica
lly
as f
o
llo
ws:
Fo
r
in
p
u
t la
y
e
r
L
et
th
e
in
p
u
t
f
ea
tu
r
es b
e
x
=[
x
1
, x
2
,
…,
x
n
]
Su
p
p
o
s
e
th
er
e
a
r
e
L
h
i
d
d
en
la
y
er
s
,
ea
ch
with
a
lar
g
e
n
u
m
b
er
o
f
n
e
u
r
o
n
s
.
T
h
e
o
u
tp
u
t la
y
er
p
r
o
d
u
ce
s
th
e
p
r
ed
ictio
n
̂
T
h
en
in
f
o
r
war
d
p
r
o
p
ag
atio
n
,
f
o
r
f
ir
s
t h
id
d
en
lay
e
r
l
=1
z
(1)
=
W
(1)
x
+
b
(1)
a
(
1
)
=
σ
(
z
(
1
)
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
9
2
I
n
t J Ap
p
l Po
wer
E
n
g
,
Vo
l.
14
,
No
.
2
,
J
u
n
e
20
25
:
26
4
-
27
4
268
Fo
r
s
u
b
s
eq
u
en
t
h
id
d
en
lay
er
s
(
l=
2
,
3
,
…
,
L
)
z
(
l
)
=
W
(
l
)
a
(
l
−
1
)
+
b
(
l
)
Fo
r
th
e
o
u
t
p
u
t la
y
er
̂
=
W
(
L
+
1
)
a
(
L
)
+
b
(
L
+
1
)
Her
e,
W
(
l
)
an
d
b
(
l
)
ar
e
th
e
wei
g
h
t
m
atr
ix
an
d
b
ias
v
ec
to
r
f
o
r
th
e
l
-
th
lay
er
,
r
esp
ec
tiv
ely
,
a
n
d
σ
is
th
e
ac
tiv
atio
n
f
u
n
ctio
n
.
3.
M
O
DE
L
E
V
AL
U
AT
I
O
N
I
n
o
r
d
e
r
to
ass
ess
th
e
s
u
itab
ilit
y
o
f
a
m
o
d
el,
it
is
ess
en
tial
to
h
av
e
a
p
er
f
o
r
m
a
n
ce
m
etr
ic
th
a
t
m
ea
s
u
r
es
h
o
w
well
it
f
its
th
e
d
ata.
I
t
is
c
r
u
cial
to
ascer
tain
th
e
a
d
eq
u
ac
y
o
f
a
r
eg
r
ess
io
n
m
o
d
el,
w
h
ich
in
v
o
lv
es
ass
ess
in
g
wh
eth
er
th
e
m
o
d
el
ac
cu
r
ately
p
r
ed
ic
ts
th
e
tar
g
et
v
ar
ia
b
les
with
in
an
ac
ce
p
tab
le
lev
el
o
f
ac
c
u
r
ac
y
.
T
h
ese
m
etr
ics
ca
n
b
e
u
s
ed
f
o
r
e
v
alu
atio
n
to
m
ea
s
u
r
e
th
e
ac
cu
r
ac
y
o
f
a
r
eg
r
ess
io
n
m
o
d
el.
T
h
e
f
o
llo
win
g
m
etr
ics
ar
e
g
en
er
ally
em
p
lo
y
ed
f
o
r
m
o
d
el
p
er
f
o
r
m
a
n
ce
ev
alu
atio
n
.
-
R
MSE
T
h
is
is
a
f
r
eq
u
en
tly
em
p
lo
y
e
d
m
etr
ic
f
o
r
ev
alu
atin
g
th
e
ac
cu
r
ac
y
o
f
p
r
ed
ictio
n
s
b
y
m
e
asu
r
in
g
th
e
E
u
clid
ea
n
d
is
tan
ce
b
etwe
en
p
r
ed
icted
v
alu
es
an
d
tr
u
e
v
al
u
es.
I
t
is
f
r
eq
u
e
n
tly
e
m
p
lo
y
e
d
i
n
s
u
p
er
v
is
ed
lear
n
in
g
ap
p
licatio
n
s
d
u
e
t
o
its
r
elian
ce
o
n
ac
cu
r
ate
m
ea
s
u
r
em
e
n
ts
f
o
r
ea
ch
p
r
ed
icted
d
ata
p
o
in
t.
R
MSE
ca
n
b
e
r
ep
r
esen
ted
as
(
1
)
.
=
√
∑
‖
(
)
−
(
)
̂
‖
2
=
1
(
1
)
W
h
er
e
N
r
ep
r
esen
ts
th
e
s
ize
o
f
th
e
d
ataset,
(
)
is
th
e
i
-
th
m
ea
s
u
r
em
en
t,
an
d
(
)
̂
is
its
co
r
r
elativ
e
p
r
ed
ictio
n
.
Hav
in
g
a
s
in
g
le
n
u
m
er
ical
m
et
r
ic
to
ass
ess
a
m
o
d
el'
s
p
er
f
o
r
m
an
ce
is
h
ig
h
ly
a
d
v
an
tag
e
o
u
s
in
ML
,
wh
eth
e
r
it is
f
o
r
tr
ain
i
n
g
,
c
r
o
s
s
-
v
alid
atio
n
,
o
r
p
o
s
t
-
d
ep
lo
y
m
en
t
m
o
n
ito
r
in
g
.
R
MSE
is
a
h
ig
h
l
y
p
r
ev
al
en
t
m
etr
ic
f
o
r
th
is
p
u
r
p
o
s
e.
T
h
is
s
co
r
in
g
r
u
le
is
b
o
th
co
m
p
r
eh
en
s
ib
le
a
n
d
co
n
s
is
ten
t w
ith
p
r
ev
alen
t statis
tical
ass
u
m
p
tio
n
s
.
-
R
2
T
h
e
co
ef
f
icien
t
o
f
d
ete
r
m
in
ati
o
n
,
also
k
n
o
wn
as
R
2
,
is
a
m
etr
ic
u
s
ed
to
ex
am
in
e
th
e
ac
cu
r
ac
y
o
f
a
r
eg
r
ess
io
n
m
o
d
el.
I
t
m
ea
s
u
r
es
th
e
d
is
p
er
s
io
n
o
f
th
e
d
ata
p
o
i
n
ts
ar
o
u
n
d
th
e
r
e
g
r
ess
io
n
lin
e
t
h
at
h
as
b
ee
n
f
itted
.
Hig
h
er
R
-
s
q
u
ar
ed
v
alu
es
in
d
ic
ate
a
s
m
aller
d
is
cr
ep
an
cy
b
etw
ee
n
th
e
o
b
s
er
v
ed
d
ata
a
n
d
th
e
f
itted
v
alu
es
f
o
r
th
e
s
am
e
d
ata
s
et.
I
t
also
d
ep
icts
t
h
e
p
r
o
p
o
r
tio
n
o
f
th
e
v
a
r
iab
ilit
y
in
th
e
d
ep
en
d
en
t
v
a
r
iab
le
th
at
ca
n
b
e
ac
co
u
n
te
d
f
o
r
b
y
a
lin
ea
r
m
o
d
el.
T
h
e
(
2
)
d
ef
in
es R
2
.
2
=
ℎ
(
2
)
T
h
e
o
u
tp
u
t
o
f
th
is
m
eth
o
d
v
ar
ies
b
etwe
en
0
an
d
1
,
with
a
v
alu
e
o
f
1
in
d
icatin
g
a
p
er
f
ec
t
f
it
o
f
th
e
r
eg
r
ess
io
n
lin
e
to
th
e
d
ata.
A
v
alu
e
o
f
0
.
7
in
d
icate
s
th
at
7
0
%
o
f
th
e
d
ata
p
o
in
ts
ar
e
with
in
th
e
r
a
n
g
e
o
f
th
e
r
eg
r
ess
io
n
lin
e.
-
MA
E
I
n
th
e
d
o
m
ain
o
f
ML
,
ab
s
o
lu
t
e
er
r
o
r
d
en
o
tes
th
e
m
ag
n
itu
d
e
o
f
th
e
d
is
p
ar
ity
b
etwe
en
th
e
f
o
r
ec
asted
v
alu
e
o
f
an
o
b
s
er
v
atio
n
an
d
it
s
ac
tu
al
v
alu
e.
T
h
e
m
ea
n
ab
s
o
lu
te
er
r
o
r
q
u
an
tifie
s
th
e
av
er
a
g
e
s
ize
o
f
e
r
r
o
r
s
in
a
co
llectio
n
o
f
f
o
r
ec
asts
,
ir
r
esp
e
ctiv
e
o
f
th
eir
d
ir
ec
tio
n
.
I
t
q
u
an
t
if
ies
p
r
ec
is
io
n
f
o
r
v
ar
iab
les
th
a
t
h
av
e
a
c
o
n
tin
u
o
u
s
r
an
g
e
o
f
v
alu
es.
T
y
p
ically
,
a
l
o
wer
MA
E
s
p
ec
if
ies
b
etter
p
r
ed
ictiv
e
p
er
f
o
r
m
a
n
ce
o
f
th
e
m
o
d
el
.
No
n
eth
eless
,
th
e
co
r
r
elatio
n
b
etwe
en
MA
E
v
alu
es
an
d
th
e
ef
f
icac
y
o
f
a
m
o
d
el
is
co
n
tin
g
en
t
u
p
o
n
th
e
ch
ar
ac
ter
is
tics
o
f
th
e
d
at
a.
I
t is ca
lcu
late
d
u
s
in
g
(
3
)
.
=
∑
|
−
̂
|
=
1
(
3
)
W
h
er
e,
is
th
e
ac
tu
al
v
alu
e
a
n
d
̂
is
th
e
p
r
e
d
icted
v
alu
e
a
n
d
n
is
th
e
n
u
m
b
er
o
f
m
ea
s
u
r
em
en
t
p
o
in
t
.
-
MSE
MSE
is
th
e
m
ea
n
o
f
t
h
e
s
q
u
ar
ed
d
if
f
er
en
c
es
b
etwe
en
th
e
ac
tu
al
v
alu
es
an
d
th
e
p
r
ed
ic
ted
v
alu
es
o
b
tain
ed
f
r
o
m
th
e
r
eg
r
ess
io
n
m
o
d
el.
W
e
ar
e
en
d
ea
v
o
r
in
g
to
ascer
tain
th
e
d
is
p
ar
ity
b
etwe
e
n
th
e
r
ea
l
v
alu
e
a
n
d
th
e
p
r
o
jecte
d
v
alu
e,
a
n
d
th
e
n
c
o
m
p
u
te
its
s
q
u
ar
e.
T
h
e
f
o
r
m
u
l
a
f
o
r
th
e
s
am
e
is
p
r
o
v
id
e
d
as (
4
)
.
=
1
2
∗
∑
(
−
)
2
=
1
(
4
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ap
p
l Po
wer
E
n
g
I
SS
N:
2252
-
8
7
9
2
P
o
w
erin
g
th
e
fu
tu
r
e
o
f e
lectrica
l lo
a
d
f
o
r
ec
a
s
tin
g
u
s
in
g
a
r
eg
r
ess
io
n
lea
r
n
er
…
(
S
u
s
h
a
ma
D.
Wa
n
kh
a
d
e
)
269
Her
e,
r
ep
r
esen
ts
th
e
m
o
d
el'
s
er
r
o
r
.
T
h
e
v
a
r
iab
le
r
ep
r
esen
ts
th
e
m
o
d
el'
s
o
u
tp
u
t
f
o
r
th
e
g
iv
e
n
d
ata,
wh
ile
Y
r
ep
r
esen
ts
th
e
ex
p
ec
ted
o
u
tp
u
t.
d
en
o
tes
th
e
q
u
an
tity
o
f
d
ata
r
o
ws
th
at
we
f
ee
d
in
to
th
e
m
o
d
el.
T
h
e
p
u
r
p
o
s
e
o
f
s
q
u
ar
in
g
th
e
er
r
o
r
is
to
elim
in
ate
an
y
n
eg
ativ
e
v
alu
es.
MSE
ap
p
lies
a
h
ig
h
er
p
en
alty
to
th
e
er
r
o
r
c
o
m
p
ar
e
d
to
th
e
MA
E
b
y
s
q
u
ar
in
g
it.
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
ML
m
o
d
els
a
r
e
a
d
ap
tiv
e
a
n
d
co
n
tin
u
o
u
s
ly
ev
o
lv
e
b
y
ass
im
ilatin
g
n
ew
s
am
p
le
d
ata
an
d
ex
p
er
ien
ce
s
.
T
h
er
ef
o
r
e,
th
e
m
o
d
els
ar
e
ca
p
ab
le
o
f
d
is
ce
r
n
in
g
th
e
p
atter
n
s
with
in
th
e
d
ata.
I
n
itially
,
f
o
u
r
teen
m
o
d
els
f
r
o
m
f
iv
e
d
if
f
e
r
en
t f
am
ilies
wer
e
tr
a
in
ed
.
4
.
1
.
Respo
ns
e
plo
t
s
R
esp
o
n
s
e
p
lo
ts
f
o
r
th
e
s
elec
ted
f
o
u
r
teen
m
o
d
els
ar
e
p
lo
tte
d
.
Fig
u
r
e
2
s
h
o
ws
th
e
r
esp
o
n
s
e
p
lo
ts
o
f
th
ir
teen
d
if
f
er
en
t
m
o
d
els
d
u
r
i
n
g
th
e
tr
ain
in
g
o
f
ea
ch
m
o
d
el.
I
t’
s
a
d
o
t
g
r
ap
h
b
etwe
en
t
h
e
r
esp
o
n
s
e
an
d
r
ec
o
r
d
n
u
m
b
er
.
W
h
er
e
th
e
r
esp
o
n
s
e
is
p
lo
tted
ag
ain
s
t
th
e
r
ec
o
r
d
n
u
m
b
er
s
i.e
.
2
1
8
9
1
r
o
w
en
tr
ies
.
T
h
is
in
clu
d
es
th
e
d
ate,
d
a
y
h
o
u
r
o
f
d
a
y
v
o
ltag
e
an
d
c
u
r
r
en
t
at
th
at
h
o
u
r
a
n
d
m
ax
im
u
m
an
d
m
in
im
u
m
tem
p
er
atu
r
e
at
th
at
h
o
u
r
.
B
lu
e
d
o
ts
r
ep
r
esen
t
ac
tu
al
r
esp
o
n
s
e
wh
er
ea
s
y
ello
w
d
o
ts
r
ep
r
esen
t
p
r
ed
icted
r
esp
o
n
s
e.
R
ef
er
r
in
g
Fig
u
r
e
s
2
(
a
)
-
2(
n
)
we
ca
n
s
ee
th
e
r
esp
o
n
s
es o
v
er
lap
p
i
n
g
an
d
s
o
m
e
o
f
th
e
r
esp
o
n
s
es a
r
e
s
in
g
u
lar
.
T
h
is
is
v
is
ib
le
in
alm
o
s
t
all
th
e
m
o
d
els
in
m
o
r
e
o
r
less
q
u
an
tity
.
I
n
b
ag
g
ed
t
r
ee
an
d
b
o
o
s
ted
tr
ee
m
o
d
el
th
e
o
v
er
lap
p
in
g
is
less
,
s
o
d
if
f
er
e
n
t
co
l
o
r
ed
d
o
ts
a
r
e
r
e
m
ar
k
ab
ly
d
is
ti
n
g
u
is
h
ab
le.
I
n
o
th
er
m
o
d
els
lik
e
lin
ea
r
,
f
in
e
t
r
e
e,
m
ed
iu
m
tr
ee
an
d
co
ar
s
e
tr
ee
m
o
d
els
also
s
o
m
e
p
o
in
ts
ar
e
n
o
t
o
v
er
lap
p
i
n
g
,
b
u
t
th
e
p
e
r
ce
n
tag
e
is
less
er
th
at
th
e
b
ag
g
ed
a
n
d
b
o
o
s
ted
tr
ee
m
o
d
el.
T
h
er
e
is
g
r
ea
ter
s
u
p
er
im
p
o
s
itio
n
o
f
p
o
in
ts
in
n
eu
r
al
n
etwo
r
k
m
o
d
els.
W
id
e
n
eu
r
al
n
etwo
r
k
s
h
o
ws h
ig
h
est s
u
p
er
im
p
o
s
itio
n
as c
o
m
p
ar
e
d
to
n
ar
r
o
w
a
n
d
m
ed
iu
m
n
e
u
r
al
n
etwo
r
k
s
.
(
a)
(b
)
(c
)
(d
)
(
e)
(f)
(
g
)
(
h
)
(
i)
(j)
(
k
)
(l)
(m)
(n
)
Fig
u
r
e
2
.
R
esp
o
n
s
e
p
lo
ts
o
f
v
a
r
io
u
s
m
o
d
el
s
:
(a
)
lin
ea
r
,
(
b
)
f
i
n
e
tr
ee
,
(
c)
m
ed
iu
m
tr
ee
,
(
d
)
c
o
ar
s
e
tr
ee
,
(
e)
lin
ea
r
SVM,
(
f
)
q
u
a
d
r
atic
SVM,
(
g
)
cu
b
ic
SVM,
(
h
)
m
ed
iu
m
G
a
u
s
s
ian
SVM,
(
i)
co
ar
s
e
G
au
s
s
ian
SVM,
(
j)
b
o
o
s
ted
tr
ee
s
,
(
k
)
b
a
g
g
ed
tr
ee
s
,
(
l)
n
a
r
r
o
w
n
eu
r
al
n
etwo
r
k
,
(
m
)
m
ed
iu
m
n
eu
r
al
n
etwo
r
k
,
an
d
(
n
)
wid
e
n
eu
r
al
n
etwo
r
k
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
9
2
I
n
t J Ap
p
l Po
wer
E
n
g
,
Vo
l.
14
,
No
.
2
,
J
u
n
e
20
25
:
26
4
-
27
4
270
4
.
2
.
P
re
dict
io
n
v
er
s
us
a
ct
ua
l
plo
t
Fig
u
r
es
3
(
a)
-
3
(
n
)
d
is
p
lay
th
e
p
r
ed
icted
v
al
u
es
ag
ain
s
t
th
e
ac
tu
al
v
alu
es
to
ass
ess
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
m
o
d
el.
T
h
e
o
b
jectiv
e
o
f
th
i
s
p
lo
t
is
to
a
s
s
e
s
s
th
e
p
r
ed
ictiv
e
ac
cu
r
ac
y
o
f
th
e
r
e
g
r
ess
io
n
m
o
d
el
ac
r
o
s
s
v
ar
io
u
s
r
esp
o
n
s
e
v
alu
es.
T
y
p
ically
,
a
well
-
co
n
s
tr
u
cted
m
o
d
el
ex
h
ib
its
d
ata
p
o
in
ts
th
at
ar
e
d
is
tr
ib
u
ted
in
a
r
o
u
g
h
ly
s
y
m
m
etr
ical
m
an
n
er
ar
o
u
n
d
th
e
d
iag
o
n
al
lin
e.
I
n
an
id
ea
l
r
eg
r
ess
io
n
m
o
d
el,
th
e
p
r
ed
icted
r
e
s
p
o
n
s
e
m
atch
es
th
e
tr
u
e
r
esp
o
n
s
e,
r
esu
ltin
g
in
all
d
ata
p
o
in
ts
alig
n
in
g
p
er
f
ec
tly
alo
n
g
a
d
iag
o
n
al
lin
e.
T
h
e
er
r
o
r
o
f
th
e
p
r
ed
ictio
n
f
o
r
an
y
p
o
in
t
is
d
eter
m
i
n
ed
b
y
th
e
v
e
r
tical
d
is
tan
ce
f
r
o
m
t
h
e
lin
e
to
t
h
at
p
o
i
n
t.
An
e
f
f
ec
tiv
e
m
o
d
el
e
x
h
ib
i
ts
m
in
im
al
er
r
o
r
s
,
r
esu
ltin
g
in
p
r
ed
ictio
n
s
th
at
ar
e
wid
ely
d
is
p
e
r
s
ed
.
R
ef
er
r
in
g
th
e
Fig
u
r
es
3
(
a)
-
3
(
n
)
,
we
ca
n
s
ee
th
at
q
u
ad
r
atic
SVM
s
h
o
ws
th
e
b
est
d
esire
d
s
p
r
ea
d
in
g
o
f
th
e
p
o
in
ts
.
W
h
er
ea
s
b
ag
g
ed
t
r
ee
m
o
d
el
s
h
o
ws
wid
er
s
p
r
ea
d
o
f
t
h
e
p
o
i
n
ts
alo
n
g
th
e
d
iag
o
n
al.
(
a)
(
b
)
(
c)
(
d
)
(
e)
(f)
(
g
)
(
h
)
(
i)
(
j)
(
k
)
(
l)
(m)
(
n
)
Fig
u
r
e
3
.
Pre
d
ictio
n
s
b
y
v
ar
io
u
s
m
o
d
els
n
ar
r
o
w
n
eu
r
al
n
etw
o
r
k
:
(
a)
lin
ea
r
,
(
b
)
f
i
n
e
tr
ee
,
(
c
)
m
ed
iu
m
t
r
ee
,
(
d
)
co
a
r
s
e
tr
ee
,
(
e)
lin
ea
r
SVM,
(
f
)
q
u
ad
r
atic
SVM,
(
g
)
cu
b
ic
SVM,
(
h
)
m
ed
iu
m
G
au
s
s
ian
SVM,
(
i)
co
ar
s
e
G
au
s
s
ian
SVM,
(
j)
b
o
o
s
ted
tr
ee
s
,
(
k
)
b
ag
g
ed
tr
ee
s
,
(
l)
n
ar
r
o
w
n
eu
r
al
n
etwo
r
k
,
(m)
m
ed
i
u
m
n
e
u
r
al
n
etwo
r
k
,
a
n
d
(
n
)
wid
e
n
e
u
r
al
n
etwo
r
k
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ap
p
l Po
wer
E
n
g
I
SS
N:
2252
-
8
7
9
2
P
o
w
erin
g
th
e
fu
tu
r
e
o
f e
lectrica
l lo
a
d
f
o
r
ec
a
s
tin
g
u
s
in
g
a
r
eg
r
ess
io
n
lea
r
n
er
…
(
S
u
s
h
a
ma
D.
Wa
n
kh
a
d
e
)
271
4
.
3
.
Resid
ua
l
plo
t
s
Fig
u
r
e
s
4
(
a
)
-
4(
n
)
d
is
p
lay
r
esid
u
al
p
lo
ts
f
o
r
v
a
r
io
u
s
m
o
d
els.
A
r
esid
u
al
is
a
q
u
an
titativ
e
in
d
icato
r
o
f
th
e
v
er
tical
d
is
tan
ce
b
etwe
en
a
d
ata
p
o
in
t
an
d
th
e
r
eg
r
ess
i
o
n
lin
e.
E
s
s
en
tially
,
it
r
ef
er
s
to
th
e
d
is
cr
ep
an
c
y
b
etwe
en
a
p
r
o
jecte
d
v
alu
e
an
d
th
e
v
e
r
if
ied
r
ea
l
v
alu
e.
T
y
p
i
ca
lly
,
a
r
eliab
le
m
o
d
el
e
x
h
ib
it
s
r
esid
u
als
th
at
ar
e
d
is
tr
ib
u
ted
in
a
r
o
u
g
h
l
y
s
y
m
m
etr
ical
m
an
n
er
ar
o
u
n
d
ze
r
o
.
T
h
ese
p
lo
ts
in
d
icate
if
th
e
m
o
d
e
l
h
as
f
u
lly
ca
p
tu
r
e
d
th
e
p
r
ed
ictiv
e
in
f
o
r
m
atio
n
o
f
t
h
e
d
ata,
r
esu
ltin
g
in
th
e
r
esid
u
als
b
e
in
g
af
f
ec
ted
.
A
p
er
f
ec
t
r
esid
u
al
p
lo
t
s
h
o
u
l
d
ex
h
ib
it
a
co
n
ce
n
tr
ate
d
clu
s
ter
o
f
d
ata
p
o
i
n
ts
in
clo
s
e
p
r
o
x
im
ity
to
th
e
o
r
ig
in
wh
ile
d
is
p
lay
in
g
a
s
p
ar
s
e
d
is
tr
ib
u
tio
n
o
f
p
o
in
ts
f
u
r
th
er
a
way
f
r
o
m
th
e
o
r
i
g
in
.
Ad
d
itio
n
ally
,
th
e
p
lo
t
s
h
o
u
ld
d
em
o
n
s
tr
ate
s
y
m
m
etr
y
wi
th
r
esp
ec
t
to
th
e
o
r
ig
in
.
E
v
er
y
r
eg
r
ess
io
n
m
o
d
el
in
h
er
e
n
tly
p
o
s
s
ess
es
a
ce
r
tain
d
eg
r
ee
o
f
er
r
o
r
d
u
e
to
th
e
im
p
o
s
s
ib
ilit
y
o
f
ac
h
iev
in
g
1
0
0
%
ac
cu
r
ate
p
r
ed
ictio
n
s
.
T
h
er
ef
o
r
e
,
a
r
eg
r
ess
io
n
m
o
d
el
ca
n
b
e
d
ef
in
ed
as:
R
esp
o
n
s
e
=
Dete
r
m
in
is
tic
+
St
o
ch
asti
c
.
A
m
o
d
el
o
r
p
r
o
ce
s
s
is
co
n
s
id
er
ed
s
to
ch
asti
c
wh
en
it
in
co
r
p
o
r
ates
r
an
d
o
m
n
ess
,
wh
ich
m
ea
n
s
th
at
it
ca
n
g
en
er
ate
v
a
r
y
in
g
o
u
t
p
u
ts
wh
en
p
r
o
v
id
ed
with
id
en
tical
i
n
p
u
ts
.
I
n
d
eter
m
in
is
tic
m
o
d
e
ls
,
th
e
r
esu
lts
ar
e
co
m
p
letely
d
eter
m
in
ed
b
y
t
h
e
in
p
u
ts
to
th
e
m
o
d
el,
m
ea
n
in
g
t
h
at
if
th
e
s
am
e
in
p
u
ts
ar
e
u
s
ed
,
th
e
o
u
tp
u
ts
will
b
e
th
e
s
am
e.
Her
e,
th
e
r
eg
r
ess
io
n
m
o
d
el
is
em
p
lo
y
ed
to
ca
p
tu
r
e
th
e
d
eter
m
in
is
tic
co
m
p
o
n
e
n
t
o
f
th
e
m
o
d
el.
T
h
e
eq
u
atio
n
m
o
d
el
s
h
o
u
ld
id
ea
lly
p
r
ec
is
ely
ca
p
tu
r
e
th
e
p
r
e
d
ictiv
e
in
f
o
r
m
atio
n
.
T
h
e
r
em
ain
in
g
r
e
s
id
u
als
s
h
o
u
ld
b
e
en
tire
ly
s
to
ch
asti
c,
m
ea
n
in
g
th
ey
ar
e
co
m
p
letely
r
an
d
o
m
a
n
d
u
n
p
r
ed
ictab
le.
I
n
o
u
r
r
esu
l
ts
,
th
e
cu
b
ic
SV
M
m
o
d
el
s
h
o
ws
a
g
o
o
d
r
esid
u
al
p
lo
t.
T
h
e
p
e
r
f
o
r
m
an
ce
o
f
all
th
e
m
o
d
els
tr
ain
ed
ar
e
ev
alu
ated
c
o
n
s
id
er
in
g
d
if
f
er
e
n
t
p
ar
am
eter
s
lik
e
R
MSE
,
R
2
,
MSE
,
MA
E
,
p
r
ed
ictio
n
s
p
ee
d
(
o
b
s
er
v
atio
n
/s
ec
)
,
an
d
tr
ain
i
n
g
tim
e
r
eq
u
ir
ed
i
n
s
ec
o
n
d
s
,
wh
ich
ar
e
ta
b
u
lated
i
n
T
ab
le
2
.
(
a)
(
b
)
(
c)
(
d
)
(
e)
(f)
(
g
)
(
h
)
(
i)
(
j)
(
k
)
(
l)
(m)
(
n
)
Fig
u
r
e
4
.
R
esid
u
al
p
lo
ts
f
o
r
v
a
r
io
u
s
m
o
d
els:
(
a)
lin
ea
r
,
(
b
)
f
i
n
e
tr
ee
,
(
c)
m
ed
iu
m
tr
ee
,
(
d
)
c
o
ar
s
e
tr
ee
,
(
e)
lin
ea
r
SVM,
(
f
)
q
u
a
d
r
atic
SVM,
(
g
)
cu
b
ic
SVM,
(
h
)
m
e
d
iu
m
G
au
s
s
ian
SVM,
(
i)
co
ar
s
e
G
au
s
s
ian
SVM,
(
j)
b
o
o
s
ted
t
r
ee
s
,
(
k
)
b
ag
g
e
d
tr
ee
s
,
(
l)
n
ar
r
o
w
n
eu
r
al
n
etwo
r
k
,
(
m
)
m
e
d
iu
m
n
eu
r
al
n
etwo
r
k
,
an
d
(
n
)
wid
e
n
e
u
r
al
n
etwo
r
k
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
9
2
I
n
t J Ap
p
l Po
wer
E
n
g
,
Vo
l.
14
,
No
.
2
,
J
u
n
e
20
25
:
26
4
-
27
4
272
T
ab
le
2
s
h
o
ws
th
at
th
e
wid
e
n
eu
r
al
n
etwo
r
k
m
o
d
el
h
as
t
h
e
l
o
west
R
MSE
v
alu
e
o
f
0
.
0
1
2
5
7
6
d
u
r
in
g
v
alid
atio
n
,
with
th
e
m
ed
i
u
m
n
eu
r
al
n
etwo
r
k
m
o
d
el
f
o
llo
win
g
clo
s
ely
b
eh
in
d
.
I
n
b
o
th
s
ce
n
ar
io
s
,
th
e
R
-
s
q
u
ar
ed
v
alu
e
is
1
.
T
h
e
n
ar
r
o
w
n
eu
r
al
n
etwo
r
k
m
o
d
el
h
as
th
e
lo
west
MSE
v
alu
e
o
f
0
.
0
0
0
1
5
8
1
6
an
d
a
MA
E
v
alu
e
o
f
0
.
0
0
2
7
1
4
3
.
Alth
o
u
g
h
th
is
m
o
d
el
ca
n
h
an
d
le
3
5
0
0
0
d
ata
p
o
in
ts
,
it
tak
es
th
e
lo
n
g
est
tr
ain
in
g
tim
e
o
f
1
9
4
1
.
4
s
ec
o
n
d
s
.
T
h
e
tr
ain
i
n
g
tim
e
f
o
r
th
e
co
ar
s
e
tr
ee
an
d
m
e
d
iu
m
tr
e
e
m
o
d
els
is
s
ig
n
if
ican
tly
s
h
o
r
t,
o
n
ly
2
.
1
5
8
s
ec
o
n
d
s
with
a
m
in
im
u
m
leaf
s
ize
o
f
3
6
an
d
2
.
2
2
5
1
s
ec
o
n
d
s
with
a
m
in
im
u
m
leaf
s
ize
o
f
1
2
.
T
h
e
q
u
ad
r
atic
SVM
m
o
d
el
d
em
o
n
s
tr
ates
th
e
to
p
p
er
f
o
r
m
an
ce
am
o
n
g
SVM
m
o
d
els,
with
R
MSE
o
f
0
.
0
3
6
5
5
6
,
MSE
o
f
0
.
0
0
1
3
3
6
3
,
an
d
MA
E
o
f
0
.
0
3
0
9
3
7
.
T
h
e
d
u
r
atio
n
o
f
t
h
e
tr
ain
i
n
g
p
r
o
ce
s
s
is
4
1
.
1
1
8
s
ec
o
n
d
s
.
Af
ter
an
al
y
zin
g
all
th
e
m
o
d
els,
we
d
is
co
v
er
ed
th
at
wid
e
n
eu
r
al
n
e
two
r
k
s
p
r
o
d
u
ce
d
th
e
m
o
s
t
s
u
p
er
io
r
r
esu
lts
with
m
in
im
al
ef
f
o
r
t.
Af
ter
ca
lcu
latin
g
R
MSE
an
d
MSE
v
alu
es f
o
r
all
m
o
d
els,
we
c
o
n
clu
d
ed
t
h
at
th
e
wid
e
n
e
u
r
al
n
etwo
r
k
p
r
o
d
u
c
ed
th
e
m
o
s
t o
p
tim
al
r
esu
lt
with
th
e
lo
west
v
alu
e.
Valu
es
f
o
r
R
MSE
an
d
MSE
.
On
ce
th
e
m
o
d
el
is
s
elec
ted
,
i
t
ca
n
b
e
tu
n
ed
f
o
r
o
p
tim
ized
p
a
r
am
eter
s
.
T
o
av
o
id
o
v
e
r
f
itti
n
g
,
f
iv
e
-
f
o
ld
c
r
o
s
s
-
v
alid
atio
n
is
p
e
r
f
o
r
m
ed
i
n
th
i
s
wo
r
k
.
T
h
e
tu
n
in
g
p
ar
am
eter
s
ar
e
n
u
m
b
e
r
o
f
f
u
ll
y
co
n
n
ec
ted
lay
er
s
,
a
n
d
th
e
r
e
g
u
lar
izatio
n
s
tr
en
g
th
(
ʎ
)
v
alu
e
.
T
h
e
h
ig
h
e
r
v
al
u
e
o
f
ʎ
will
r
esu
lt
in
u
n
d
er
f
itti
n
g
o
f
th
e
p
lo
t
an
d
th
e
lo
wer
v
alu
e
s
h
o
ws
o
v
er
f
itti
n
g
o
f
th
e
v
alu
e
s
.
W
ith
th
r
ee
f
u
lly
co
n
n
ec
ted
lay
er
s
,
t
h
e
f
ir
s
t
la
y
er
s
ize
,
is
1
0
0
with
th
e
s
ec
o
n
d
-
an
d
-
t
h
ir
d
-
la
y
er
s
ize
1
0
.
W
ith
th
ese
p
a
r
am
eter
s
ettin
g
s
,
th
e
r
esu
lts
o
b
tain
ed
a
r
e
s
h
o
wn
in
T
a
b
le
3
.
T
ab
le
3
.
Per
f
o
r
m
an
ce
o
f
W
NN
af
ter
p
ar
am
eter
tu
n
n
i
n
g
P
a
r
a
me
t
e
r
s
e
t
t
i
n
g
R
M
S
E
M
S
E
M
A
E
P
r
e
d
i
c
t
i
o
n
s
p
e
e
d
Tr
a
i
n
i
n
g
t
i
me
sec
N
o
.
o
f
l
a
y
e
r
s=
3
,
ʎ
=
0
0
.
0
1
0
6
6
0
.
0
0
0
1
1
3
6
3
0
.
0
0
3
1
8
7
3
0
0
0
0
2
4
2
8
.
6
N
o
.
o
f
l
a
y
e
r
s=
3
,
ʎ
=
0
.
1
0
.
1
1
3
9
3
0
.
0
1
2
9
8
1
0
.
0
7
8
0
0
3
3
1
0
0
0
2
5
6
.
8
4
5.
CO
NCLU
SI
O
N
E
lectr
icity
u
s
ag
e
is
in
f
lu
e
n
ce
d
b
y
a
r
an
g
e
o
f
f
ac
to
r
s
,
i
n
clu
d
in
g
wea
th
e
r
,
tim
e
o
f
d
ay
,
h
o
lid
ay
s
,
an
d
s
o
cial
ev
en
ts
.
ML
alg
o
r
ith
m
s
ca
n
co
m
p
r
eh
en
d
co
m
p
lex
r
elati
o
n
s
h
ip
s
an
d
c
r
ea
te
ad
a
p
tab
le
m
o
d
els
th
at
r
esp
o
n
d
to
alter
in
g
c
o
n
d
itio
n
s
.
I
n
t
h
is
p
ap
er
,
a
n
ef
f
ec
tiv
e
f
o
r
ec
asti
n
g
ap
p
r
o
ac
h
f
o
r
th
e
3
3
/1
1
k
V
s
u
b
s
tatio
n
at
W
av
i,
Nasik
,
I
n
d
ia.
1
4
r
e
g
r
ess
io
n
m
o
d
els
ar
e
ev
alu
ate
d
b
ased
o
n
d
if
f
er
en
t
p
er
f
o
r
m
an
ce
in
d
ices
in
itially
.
Am
o
n
g
t
h
e
1
4
m
o
d
els
s
tu
d
ied
,
a
w
id
e
n
e
u
r
al
n
etwo
r
k
m
o
d
el
is
r
ec
o
m
m
en
d
ed
b
ased
o
n
R
MSE
,
MSE
,
an
d
MA
E
.
Ma
n
y
r
esear
ch
er
s
h
av
e
ex
p
lo
r
ed
v
ar
io
u
s
r
eg
r
ess
io
n
m
o
d
els
f
o
r
f
o
r
ec
asti
n
g
.
B
u
t
th
e
p
er
f
o
r
m
an
ce
o
f
n
eu
r
al
n
etwo
r
k
m
o
d
els
in
r
eg
r
ess
io
n
r
em
ain
ed
u
n
ex
p
lo
r
ed
m
an
y
r
esear
c
h
er
s
h
av
e
u
tili
ze
d
SVM
an
d
GPR
tech
n
iq
u
es.
T
h
e
cu
r
r
e
n
t
s
tu
d
y
ev
alu
ates
th
e
p
r
ec
is
io
n
o
f
SVM,
d
ec
is
io
n
tr
ee
s
,
an
d
n
eu
r
al
n
etwo
r
k
s
.
T
h
e
r
esu
lts
s
h
o
w
th
at
wid
e
n
eu
r
al
n
etwo
r
k
s
h
ad
th
e
b
est
p
er
f
o
r
m
an
ce
,
with
a
r
eg
r
ess
io
n
er
r
o
r
o
f
0
.
0
1
0
6
6
,
a
n
d
an
M
SE
o
f
0
.
0
0
0
1
1
3
6
3
.
T
h
e
s
tu
d
y
in
v
esti
g
ates
th
e
ef
f
ec
tiv
en
ess
o
f
n
eu
r
al
n
etwo
r
k
m
o
d
els
u
s
in
g
th
e
r
eg
r
ess
io
n
m
et
h
o
d
.
T
h
is
im
p
lies
th
at
wid
e
n
eu
r
al
n
etwo
r
k
s
h
a
v
e
s
ig
n
if
ican
t
p
r
o
m
is
e
f
o
r
p
r
e
cise
elec
tr
ical
lo
ad
f
o
r
ec
asti
n
g
.
T
h
is
im
p
lies
th
at
wid
e
n
eu
r
al
n
etwo
r
k
s
h
av
e
s
ig
n
if
ican
t
p
r
o
m
is
e
f
o
r
p
r
ec
is
e
elec
tr
ical
lo
ad
f
o
r
ec
asti
n
g
.
As
r
esear
ch
ad
v
an
ce
s
,
we
ex
p
ec
t
to
s
ee
th
e
d
ev
el
o
p
m
en
t
o
f
e
v
en
m
o
r
e
s
o
p
h
i
s
ticated
ML
m
o
d
els.
Utilizin
g
s
m
ar
t
g
r
id
s
an
d
r
en
ewa
b
le
en
er
g
y
s
o
u
r
ce
s
will
en
h
an
ce
f
o
r
ec
asti
n
g
ac
cu
r
a
cy
an
d
d
r
iv
e
to
wa
r
d
s
a
m
o
r
e
r
esil
ien
t
an
d
ec
o
-
f
r
ien
d
ly
e
n
er
g
y
f
u
t
u
r
e.
W
h
ile
n
eu
r
al
n
etwo
r
k
s
o
f
f
er
s
ig
n
if
ican
t
a
d
v
an
tag
es
f
o
r
task
s
lik
e
elec
tr
ical
lo
ad
f
o
r
ec
asti
n
g
,
t
h
ey
also
co
m
e
with
s
ev
e
r
al
ch
allen
g
es.
T
h
ese
ch
allen
g
es
ca
n
im
p
ac
t
th
eir
p
er
f
o
r
m
a
n
ce
,
u
s
ab
ilit
y
,
a
n
d
in
te
g
r
atio
n
in
to
r
ea
l
-
wo
r
ld
ap
p
licatio
n
s
.
L
ar
g
e
an
d
g
o
o
d
q
u
ality
d
ata,
h
ig
h
co
m
p
u
tatio
n
al
co
s
t,
ar
e
s
o
m
e
o
f
th
e
c
h
allen
g
es.
Ov
er
co
m
in
g
ch
allen
g
es
lik
e
Mo
d
el
co
m
p
lex
ity
,
in
te
r
p
r
eta
b
ilit
y
,
an
d
tr
ain
in
g
tim
e
r
eq
u
i
r
es
a
co
m
b
in
atio
n
o
f
ad
v
an
ce
d
tec
h
n
iq
u
es,
r
o
b
u
s
t
in
f
r
astru
ctu
r
e,
a
n
d
in
ter
d
is
cip
lin
ar
y
e
x
p
er
tis
e.
C
o
n
tin
u
o
u
s
r
esear
ch
an
d
d
e
v
e
l
o
p
m
e
n
t
i
n
n
e
u
r
a
l
n
e
t
w
o
r
k
m
e
t
h
o
d
o
l
o
g
i
e
s
,
a
l
o
n
g
w
it
h
i
m
p
r
o
v
e
m
e
n
t
s
i
n
d
a
t
a
m
a
n
a
g
e
m
en
t
a
n
d
c
o
m
p
u
t
a
t
i
o
n
a
l
r
e
s
o
u
r
c
e
s
,
a
r
e
e
s
s
e
n
t
ia
l
t
o
m
it
i
g
a
t
e
t
h
es
e
c
h
a
l
le
n
g
e
s
a
n
d
f
u
l
l
y
h
a
r
n
e
s
s
t
h
e
p
o
t
e
n
t
i
al
o
f
n
e
u
r
a
l
n
e
t
w
o
r
k
s
.
ACK
NO
WL
E
DG
E
M
E
NT
S
Au
th
o
r
s
th
an
k
Mr
.
S.
S.
Kh
an
d
ar
e,
C
h
ief
E
n
g
in
ee
r
MSE
B
Nasik
Div
is
io
n
I
n
d
ia
f
o
r
h
is
h
elp
in
ac
q
u
ir
in
g
s
ite
d
ata
a
n
d
s
ite
v
is
its
f
o
r
th
is
r
esear
ch
wo
r
k
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
No
f
u
n
d
in
g
in
v
o
lv
e
d
in
th
is
p
r
o
ject.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ap
p
l Po
wer
E
n
g
I
SS
N:
2252
-
8
7
9
2
P
o
w
erin
g
th
e
fu
tu
r
e
o
f e
lectrica
l lo
a
d
f
o
r
ec
a
s
tin
g
u
s
in
g
a
r
eg
r
ess
io
n
lea
r
n
er
…
(
S
u
s
h
a
ma
D.
Wa
n
kh
a
d
e
)
273
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
T
h
is
jo
u
r
n
al
u
s
es
th
e
C
o
n
t
r
ib
u
to
r
R
o
les
T
a
x
o
n
o
m
y
(
C
R
ed
iT)
to
r
ec
o
g
n
ize
in
d
iv
i
d
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
th
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
llab
o
r
atio
n
.
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
Su
s
h
am
a
D.
W
an
k
h
ad
e
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
B
ab
asah
eb
R
.
Patil
✓
✓
✓
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
s
i
s
I
:
I
n
v
e
s
t
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
s
i
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
Au
th
o
r
s
s
tate
n
o
co
n
f
lict o
f
in
t
er
est.
DATA AV
AI
L
AB
I
L
I
T
Y
Der
iv
ed
d
ata
s
u
p
p
o
r
tin
g
th
e
f
i
n
d
in
g
s
o
f
th
is
s
tu
d
y
ar
e
av
ailab
le
f
r
o
m
th
e
co
r
r
esp
o
n
d
i
n
g
au
t
h
o
r
[
SDW
],
o
n
r
eq
u
est.
RE
F
E
R
E
NC
E
S
[
1
]
A
.
A
l
h
e
n
d
i
,
A
.
S
.
A
l
-
S
u
mai
t
i
,
M
.
M
a
r
z
b
a
n
d
,
R
.
K
u
m
a
r
,
a
n
d
A
.
A
.
Z.
D
i
a
b
,
“
S
h
o
r
t
-
t
e
r
m
l
o
a
d
a
n
d
p
r
i
c
e
f
o
r
e
c
a
st
i
n
g
u
si
n
g
a
r
t
i
f
i
c
i
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
w
i
t
h
e
n
h
a
n
c
e
d
M
a
r
k
o
v
c
h
a
i
n
f
o
r
I
S
O
N
e
w
En
g
l
a
n
d
,
”
En
e
rg
y
R
e
p
o
r
t
s
,
v
o
l
.
9
,
p
p
.
4
7
9
9
–
4
8
1
5
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
g
y
r
.
2
0
2
3
.
0
3
.
1
1
6
.
[
2
]
M
.
C
o
r
d
e
i
r
o
-
C
o
s
t
a
s
,
D
.
V
i
l
l
a
n
u
e
v
a
,
P
.
Eg
u
í
a
-
O
l
l
e
r
,
M
.
M
a
r
t
í
n
e
z
-
C
o
mes
a
ñ
a
,
a
n
d
S
.
R
a
m
o
s,
“
L
o
a
d
f
o
r
e
c
a
st
i
n
g
w
i
t
h
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
n
d
d
e
e
p
l
e
a
r
n
i
n
g
m
e
t
h
o
d
s
,
”
Ap
p
l
i
e
d
S
c
i
e
n
c
e
s
(
S
w
i
t
zer
l
a
n
d
)
,
v
o
l
.
1
3
,
n
o
.
1
3
,
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
a
p
p
1
3
1
3
7
9
3
3
.
[
3
]
J.
S
.
C
h
o
u
a
n
d
D
.
S
.
Tr
a
n
,
“
F
o
r
e
c
a
s
t
i
n
g
e
n
e
r
g
y
c
o
n
s
u
m
p
t
i
o
n
t
i
m
e
s
e
r
i
e
s
u
si
n
g
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s
b
a
s
e
d
o
n
u
sa
g
e
p
a
t
t
e
r
n
s
o
f
r
e
s
i
d
e
n
t
i
a
l
h
o
u
s
e
h
o
l
d
e
r
s,”
E
n
e
r
g
y
,
v
o
l
.
1
6
5
,
p
p
.
7
0
9
–
7
2
6
,
2
0
1
8
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
n
e
r
g
y
.
2
0
1
8
.
0
9
.
1
4
4
.
[
4
]
R
.
W
a
z
i
r
a
l
i
,
E.
Y
a
g
h
o
u
b
i
,
M
.
S
.
S
.
A
b
u
j
a
z
a
r
,
R
.
A
h
ma
d
,
a
n
d
A
.
H
.
V
a
k
i
l
i
,
“
S
t
a
t
e
-
of
-
t
h
e
-
a
r
t
r
e
v
i
e
w
o
n
e
n
e
r
g
y
a
n
d
l
o
a
d
f
o
r
e
c
a
st
i
n
g
i
n
m
i
c
r
o
g
r
i
d
s
u
s
i
n
g
a
r
t
i
f
i
c
i
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
s,
mac
h
i
n
e
l
e
a
r
n
i
n
g
,
a
n
d
d
e
e
p
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s
,
”
E
l
e
c
t
r
i
c
Po
w
e
r
S
y
st
e
m
s
Re
s
e
a
r
c
h
,
v
o
l
.
2
2
5
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
p
sr
.
2
0
2
3
.
1
0
9
7
9
2
.
[
5
]
X
.
C
h
e
n
,
J.
P
e
n
g
,
Y
.
H
e
,
B
.
Zh
a
n
g
,
D
.
Ji
a
n
g
,
a
n
d
P
.
L
i
n
,
“
P
o
w
e
r
l
o
a
d
f
o
r
e
c
a
s
t
i
n
g
a
l
g
o
r
i
t
h
m
b
a
s
e
d
o
n
r
e
g
r
e
ssi
o
n
s
u
p
p
o
r
t
v
e
c
t
o
r
mac
h
i
n
e
,
”
i
n
L
e
c
t
u
re
N
o
t
e
s i
n
El
e
c
t
r
i
c
a
l
E
n
g
i
n
e
e
r
i
n
g
,
v
o
l
.
8
9
5
,
2
0
2
2
,
p
p
.
1
2
6
8
–
1
2
7
4
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
9
8
1
-
19
-
4
7
7
5
-
9
_
1
6
3
.
[
6
]
M
.
W
.
N
u
r
A
z
a
m
i
,
I
.
W
a
h
y
u
d
i
F
a
r
i
d
,
C
.
W
.
P
r
i
a
n
a
n
d
a
,
a
n
d
A
.
M
u
s
t
h
o
f
a
,
“
El
e
c
t
r
i
c
l
o
a
d
f
o
r
e
c
a
st
i
n
g
o
n
sm
a
r
t
e
n
e
r
g
y
m
e
t
e
r
(
S
E
M
)
u
si
n
g
l
i
n
e
a
r
r
e
g
r
e
ss
i
o
n
,
”
i
n
2
0
2
3
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
A
d
v
a
n
c
e
d
Me
c
h
a
t
r
o
n
i
c
s
,
I
n
t
e
l
l
i
g
e
n
t
Ma
n
u
f
a
c
t
u
r
e
a
n
d
I
n
d
u
st
r
i
a
l
Au
t
o
m
a
t
i
o
n
,
I
C
A
MIMI
A
2
0
2
3
-
Pro
c
e
e
d
i
n
g
s
,
2
0
2
3
,
p
p
.
7
9
3
–
798
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
A
M
I
M
I
A
6
0
8
8
1
.
2
0
2
3
.
1
0
4
2
7
9
1
0
.
[
7
]
S
.
G
o
c
h
h
a
i
t
a
n
d
D
.
S
h
a
r
ma,
“
R
e
g
r
e
s
si
o
n
m
o
d
e
l
-
b
a
s
e
d
sh
o
r
t
-
t
e
r
m
l
o
a
d
f
o
r
e
c
a
s
t
i
n
g
f
o
r
l
o
a
d
d
e
s
p
a
t
c
h
c
e
n
t
r
e
,
”
J
o
u
r
n
a
l
o
f
A
p
p
l
i
e
d
En
g
i
n
e
e
ri
n
g
a
n
d
T
e
c
h
n
o
l
o
g
i
c
a
l
S
c
i
e
n
c
e
(
J
AETS
)
,
v
o
l
.
4
,
n
o
.
2
,
p
p
.
6
9
3
–
7
1
0
,
Ju
n
.
2
0
2
3
,
d
o
i
:
1
0
.
3
7
3
8
5
/
j
a
e
t
s.
v
4
i
2
.
1
6
8
2
.
[
8
]
K
.
H
o
p
f
,
H
.
H
a
r
t
s
t
a
n
g
,
a
n
d
T.
S
t
a
a
k
e
,
“
M
e
t
a
-
r
e
g
r
e
ssi
o
n
a
n
a
l
y
s
i
s
o
f
e
r
r
o
r
s
i
n
s
h
o
r
t
-
t
e
r
m
e
l
e
c
t
r
i
c
i
t
y
l
o
a
d
f
o
r
e
c
a
st
i
n
g
,
”
i
n
e
-
En
e
r
g
y
2
0
2
3
C
o
m
p
a
n
i
o
n
-
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
1
4
t
h
AC
M
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
F
u
t
u
re
E
n
e
r
g
y
S
y
s
t
e
m
s
,
2
0
2
3
,
p
p
.
3
2
–
3
9
,
d
o
i
:
1
0
.
1
1
4
5
/
3
5
9
9
7
3
3
.
3
6
0
0
2
4
8
.
[
9
]
K
.
A
k
a
sh
,
V
.
A
r
u
l
mo
z
h
i
,
A
.
B
a
l
a
k
r
i
s
h
n
a
n
,
B
.
H
a
r
i
s
a
n
k
a
r
,
a
n
d
J.
A
n
u
d
e
v
,
“
C
o
m
p
a
r
i
so
n
o
f
l
o
a
d
f
o
r
e
c
a
st
i
n
g
a
l
g
o
r
i
t
h
m
f
o
r
i
n
d
u
st
r
i
a
l
l
o
a
d
s
,
”
i
n
2
0
2
3
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
C
o
n
t
r
o
l
,
C
o
m
m
u
n
i
c
a
t
i
o
n
a
n
d
C
o
m
p
u
t
i
n
g
,
I
C
C
C
2
0
2
3
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
C
C
5
7
7
8
9
.
2
0
2
3
.
1
0
1
6
5
3
1
8
.
[
1
0
]
H
.
A
.
El
-
S
a
d
a
h
a
n
d
S
.
H
.
Z.
A
l
-
T
h
a
l
a
b
i
,
“
M
o
d
e
l
i
n
g
a
n
d
f
o
r
e
c
a
st
i
n
g
u
s
i
n
g
s
u
p
p
o
r
t
v
e
c
t
o
r
r
e
g
r
e
ss
i
o
n
a
n
d
c
h
a
o
t
i
c
a
l
g
o
r
i
t
h
ms
/
a
p
p
l
i
e
d
st
u
d
y
,
”
i
n
AI
P
C
o
n
f
e
re
n
c
e
Pr
o
c
e
e
d
i
n
g
s
,
2
0
2
3
,
p
.
0
5
0
0
1
5
,
d
o
i
:
1
0
.
1
0
6
3
/
5
.
0
1
1
9
5
7
5
.
[
1
1
]
A
.
Y
a
d
a
v
,
A
.
K
u
mar
,
R
.
P
.
S
.
R
a
n
a
,
M
.
C
h
a
n
d
r
a
k
a
r
,
M
.
P
a
z
o
k
i
,
a
n
d
R
.
A
.
El
S
e
h
i
e
m
y
,
“
A
n
e
f
f
i
c
i
e
n
t
m
o
n
t
h
l
y
l
o
a
d
f
o
r
e
c
a
st
i
n
g
mo
d
e
l
u
s
i
n
g
G
a
u
ssi
a
n
p
r
o
c
e
s
s reg
r
e
ss
i
o
n
,
”
i
n
2
0
2
1
I
E
EE
4
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
C
o
m
p
u
t
i
n
g
,
P
o
w
e
r
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
T
e
c
h
n
o
l
o
g
i
e
s
,
G
U
C
O
N
2
0
2
1
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
0
9
/
G
U
C
O
N
5
0
7
8
1
.
2
0
2
1
.
9
5
7
4
0
0
8
.
[
1
2
]
M
.
D
i
n
e
s
h
R
e
d
d
y
a
n
d
N
.
V
i
sh
a
l
i
,
“
Lo
a
d
f
o
r
e
c
a
s
t
i
n
g
u
s
i
n
g
l
i
n
e
a
r
r
e
g
r
e
ss
i
o
n
a
n
a
l
y
si
s
i
n
t
i
m
e
seri
e
s mo
d
e
l
f
o
r
R
G
U
K
T,
R
.
K
.
v
a
l
l
e
y
c
a
m
p
u
s
H
T
f
e
e
d
e
r
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
En
g
i
n
e
e
ri
n
g
Re
se
a
r
c
h
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
V
6
,
n
o
.
0
5
,
2
0
1
7
,
d
o
i
:
1
0
.
1
7
5
7
7
/
i
j
e
r
t
v
6
i
s
0
5
0
4
4
3
.
[
1
3
]
M
.
H
u
a
n
g
,
“
Th
e
o
r
y
a
n
d
i
mp
l
e
m
e
n
t
a
t
i
o
n
o
f
l
i
n
e
a
r
r
e
g
r
e
ss
i
o
n
,
”
i
n
P
ro
c
e
e
d
i
n
g
s
-
2
0
2
0
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
C
o
m
p
u
t
e
r
Vi
s
i
o
n
,
I
m
a
g
e
a
n
d
D
e
e
p
L
e
a
rn
i
n
g
,
C
VI
D
L
2
0
2
0
,
2
0
2
0
,
p
p
.
2
1
0
–
2
1
7
,
d
o
i
:
1
0
.
1
1
0
9
/
C
V
I
D
L5
1
2
3
3
.
2
0
2
0
.
0
0
-
9
9
.
[
1
4
]
D
.
M
a
u
l
u
d
a
n
d
A
.
M
.
A
b
d
u
l
a
z
e
e
z
,
“
A
r
e
v
i
e
w
o
n
l
i
n
e
a
r
r
e
g
r
e
ssi
o
n
c
o
mp
r
e
h
e
n
si
v
e
i
n
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
,
”
J
o
u
r
n
a
l
o
f
A
p
p
l
i
e
d
S
c
i
e
n
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
T
re
n
d
s
,
v
o
l
.
1
,
n
o
.
2
,
p
p
.
1
4
0
–
1
4
7
,
2
0
2
0
,
d
o
i
:
1
0
.
3
8
0
9
4
/
j
a
st
t
1
4
5
7
.
[
1
5
]
B
.
G
a
y
e
,
D
.
Z
h
a
n
g
,
a
n
d
A
.
W
u
l
a
m
u
,
“
I
mp
r
o
v
e
me
n
t
o
f
su
p
p
o
r
t
v
e
c
t
o
r
m
a
c
h
i
n
e
a
l
g
o
r
i
t
h
m
i
n
b
i
g
d
a
t
a
b
a
c
k
g
r
o
u
n
d
,
”
M
a
t
h
e
m
a
t
i
c
a
l
Pro
b
l
e
m
s i
n
En
g
i
n
e
e
ri
n
g
,
v
o
l
.
2
0
2
1
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
5
5
/
2
0
2
1
/
5
5
9
4
8
9
9
.
[
1
6
]
J.
C
e
r
v
a
n
t
e
s,
F
.
G
a
r
c
i
a
-
La
m
o
n
t
,
L.
R
o
d
r
í
g
u
e
z
-
M
a
z
a
h
u
a
,
a
n
d
A
.
L
o
p
e
z
,
“
A
c
o
mp
r
e
h
e
n
s
i
v
e
s
u
r
v
e
y
o
n
su
p
p
o
r
t
v
e
c
t
o
r
ma
c
h
i
n
e
c
l
a
ss
i
f
i
c
a
t
i
o
n
:
A
p
p
l
i
c
a
t
i
o
n
s
,
c
h
a
l
l
e
n
g
e
s
a
n
d
t
r
e
n
d
s
,
”
N
e
u
r
o
c
o
m
p
u
t
i
n
g
,
v
o
l
.
4
0
8
,
p
p
.
1
8
9
–
2
1
5
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
n
e
u
c
o
m.
2
0
1
9
.
1
0
.
1
1
8
.
[
1
7
]
P
.
Y
.
Ta
s
e
r
,
“
A
p
p
l
i
c
a
t
i
o
n
o
f
b
a
g
g
i
n
g
a
n
d
b
o
o
s
t
i
n
g
a
p
p
r
o
a
c
h
e
s
u
si
n
g
d
e
c
i
s
i
o
n
t
r
e
e
-
b
a
se
d
a
l
g
o
r
i
t
h
ms
i
n
d
i
a
b
e
t
e
s
r
i
sk
p
r
e
d
i
c
t
i
o
n
,
”
i
n
7
t
h
I
n
t
e
rn
a
t
i
o
n
a
l
M
a
n
a
g
e
m
e
n
t
I
n
f
o
rm
a
t
i
o
n
S
y
s
t
e
m
s
C
o
n
f
e
r
e
n
c
e
,
2
0
2
1
,
p
.
6
,
d
o
i
:
1
0
.
3
3
9
0
/
p
r
o
c
e
e
d
i
n
g
s
2
0
2
1
0
7
4
0
0
6
.
Evaluation Warning : The document was created with Spire.PDF for Python.