I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
p
ute
r
E
ng
in
ee
ring
(
I
J
E
CE
)
Vo
l.
8
,
No
.
4
,
A
u
g
u
s
t
201
8
,
p
p
.
2
3
1
0
~2
3
1
8
I
SS
N:
2
0
8
8
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v8
i
4
.
p
p
2
3
1
0
-
2318
2310
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
co
r
e
.
co
m/
jo
u
r
n
a
ls
/in
d
ex
.
p
h
p
/
I
JE
C
E
Applica
tion o
f
M
ultiple
K
er
nel S
u
ppo
rt
Ve
ctor R
eg
ress
io
n f
o
r
Weld
Bea
d
G
eo
metry
P
r
ediction
in
Ro
bo
tic
G
M
AW
Process
Na
der
M
o
lla
y
i
1
,
M
o
ha
m
m
a
d
J
a
v
a
d
E
idi
2
1
F
a
c
u
lt
y
o
f
Co
m
p
u
ter an
d
I
n
d
u
str
ial
En
g
in
e
e
rin
g
,
Bir
jan
d
U
n
iv
e
rsity
o
f
Tec
h
n
o
lo
g
y
,
Ira
n
2
F
a
c
u
lt
y
o
f
M
e
c
h
a
n
ica
lan
d
M
a
teria
ls
En
g
in
e
e
rin
g
,
Birj
a
n
d
U
n
iv
e
rsity
o
f
T
e
c
h
n
o
l
o
g
y
,
Ira
n
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
No
v
1
2
,
2
0
1
7
R
ev
i
s
ed
J
an
4
,
2
0
1
8
A
cc
ep
ted
J
an
11
,
2
0
1
8
M
o
d
e
ll
i
n
g
a
n
d
p
re
d
ictio
n
o
f
w
e
ld
b
e
a
d
g
e
o
m
e
tr
y
is
a
n
i
m
p
o
rtan
t
issu
e
in
ro
b
o
ti
c
G
M
AW
p
ro
c
e
ss
.
T
h
is
p
ro
c
e
ss
is
h
ig
h
ly
n
o
n
-
li
n
e
a
r
a
n
d
c
o
u
p
led
m
u
lt
iv
a
riab
le
s
y
ste
m
a
n
d
th
e
re
latio
n
sh
i
p
b
e
tw
e
e
n
p
ro
c
e
ss
p
a
ra
m
e
ters
a
n
d
w
e
ld
b
e
a
d
g
e
o
m
e
tr
y
c
a
n
n
o
t
b
e
d
e
f
in
e
d
b
y
a
n
e
x
p
li
c
it
m
a
th
e
m
a
ti
c
a
l
e
x
p
re
ss
io
n
.
T
h
e
re
f
o
re
,
a
p
p
li
c
a
ti
o
n
o
f
su
p
e
rv
ise
d
le
a
rn
in
g
a
lg
o
rit
h
m
s
c
a
n
b
e
u
se
f
u
l
f
o
r
th
is
p
u
r
p
o
se
.
S
u
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
is
a
v
e
r
y
su
c
c
e
ss
f
u
l
a
p
p
r
o
a
c
h
to
su
p
e
rv
ise
d
lea
rn
in
g
.
In
t
h
is
a
p
p
r
o
a
c
h
,
a
h
i
g
h
e
r
d
e
g
re
e
o
f
a
c
c
u
ra
c
y
a
n
d
g
e
n
e
ra
li
z
a
ti
o
n
c
a
p
a
b
il
it
y
c
a
n
b
e
o
b
tain
e
d
b
y
u
sin
g
t
h
e
m
u
lt
ip
le
k
e
rn
e
l
lea
rn
in
g
f
ra
m
e
w
o
r
k
,
w
h
ich
is
c
o
n
sid
e
re
d
a
s
a
g
re
a
t
a
d
v
a
n
tag
e
in
p
re
d
ictio
n
o
f
w
e
ld
b
e
a
d
g
e
o
m
e
tr
y
d
u
e
to
th
e
h
ig
h
d
e
g
re
e
o
f
p
re
d
ictio
n
a
c
c
u
ra
c
y
re
q
u
ired
.
In
t
h
is
p
a
p
e
r,
a
n
o
v
e
l
a
p
p
r
o
a
c
h
f
o
r
m
o
d
e
ll
in
g
a
n
d
p
re
d
ic
ti
o
n
o
f
th
e
w
e
ld
b
e
a
d
g
e
o
m
e
tr
y
,
b
a
se
d
o
n
m
u
lt
ip
le
k
e
rn
e
l
su
p
p
o
rt
v
e
c
to
r
re
g
re
ss
io
n
a
n
a
ly
sis
h
a
s
b
e
e
n
p
r
o
p
o
se
d
,
w
h
ich
b
e
n
e
f
it
s
f
ro
m
a
h
ig
h
d
e
g
re
e
o
f
a
c
c
u
ra
c
y
a
n
d
g
e
n
e
ra
li
z
a
ti
o
n
c
a
p
a
b
il
it
y
.
T
h
is
m
o
d
e
l
c
a
n
b
e
u
se
d
f
o
r
p
ro
p
e
r
s
e
lec
ti
o
n
o
f
w
e
ld
in
g
p
a
ra
m
e
ters
in
o
rd
e
r
to
o
b
tain
a
d
e
sire
d
w
e
ld
b
e
a
d
g
e
o
m
e
tr
y
in
ro
b
o
ti
c
G
M
AW
p
ro
c
e
ss
.
K
ey
w
o
r
d
:
Mu
ltip
le
k
er
n
el
lear
n
in
g
R
eg
r
es
s
io
n
a
n
al
y
s
i
s
R
o
b
o
tic
g
as
m
e
tal
ar
c
w
e
ld
in
g
Su
p
p
o
r
t v
ec
to
r
m
ac
h
i
n
e
W
eld
b
ea
d
g
eo
m
etr
y
p
r
ed
ictio
n
Co
p
y
rig
h
t
©
2
0
1
8
In
stit
u
te o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
rin
g
a
n
d
S
c
ien
c
e
.
Al
l
rig
h
ts
re
se
rv
e
d
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Nad
er
Mo
lla
y
i,
Facu
lt
y
o
f
C
o
m
p
u
ter
an
d
I
n
d
u
s
tr
ial
E
n
g
i
n
ee
r
in
g
,
B
ir
j
an
d
Un
iv
er
s
it
y
o
f
T
ec
h
n
o
l
o
g
y
,
B
ir
j
an
d
,
So
u
th
Kh
o
r
asan
P
r
o
v
in
ce
,
I
r
an
E
m
ail:
m
o
lla
y
i@
b
ir
j
an
d
u
t.a
c.
i
r
1.
I
NT
RO
D
UCT
I
O
N
I
n
to
d
ay
's
in
d
u
s
tr
ial
w
o
r
ld
,
i
n
cr
ea
s
ed
co
m
p
etitio
n
in
t
h
e
g
l
o
b
al
ec
o
n
o
m
y
h
as
led
to
in
cr
e
asin
g
n
ee
d
f
o
r
q
u
ick
d
eliv
er
y
o
f
c
u
s
to
m
iz
ed
p
r
o
d
u
cts
o
f
m
a
n
u
f
ac
t
u
r
er
s
t
o
m
ee
tc
u
s
to
m
er
d
e
m
a
n
d
s
[
1
]
.
T
o
th
is
en
d
,
t
h
e
er
a
o
f
r
ap
id
p
r
o
to
ty
p
i
n
g
e
m
er
g
ed
in
1
9
8
8
an
d
q
u
ic
k
l
y
d
e
v
elo
p
ed
b
ased
o
n
r
ap
id
ad
v
a
n
ce
s
o
f
d
ig
ital
co
m
p
u
t
i
n
g
s
y
s
te
m
s
a
n
d
ad
v
a
n
ce
d
tech
n
o
lo
g
ies
s
u
c
h
as
laser
.
T
h
is
n
o
v
el
in
n
o
v
at
iv
e
tech
n
o
lo
g
y
w
a
s
s
u
cc
e
s
s
f
u
l
to
o
v
er
co
m
e
th
e
s
h
o
r
tag
e
s
o
f
tr
a
d
it
io
n
al
p
r
o
to
ty
p
i
n
g
m
et
h
o
d
s
a
n
d
r
esu
lted
i
n
r
e
m
ar
k
ab
le
r
ed
u
ctio
n
o
f
p
r
o
to
ty
p
es
p
r
o
d
u
ctio
n
ti
m
e
[
2
]
.
I
n
r
ap
i
d
p
r
o
to
ty
p
in
g
,
c
o
n
s
tr
u
ctio
n
o
r
ass
e
m
b
l
y
o
f
t
h
e
co
n
s
is
tin
g
p
ar
t
s
is
u
s
u
al
l
y
p
er
f
o
r
m
ed
b
y
ap
p
licatio
n
o
f
a
d
d
itiv
e
m
a
n
u
f
ac
t
u
r
i
n
g
tech
n
o
lo
g
y
,
i
n
w
h
ic
h
a
t
h
r
ee
-
d
i
m
en
s
io
n
a
l
o
b
j
ec
t
is
cr
ea
ted
b
y
ad
d
i
tio
n
o
f
la
y
er
-
u
p
o
n
-
la
y
er
o
f
m
ater
ials
u
n
d
er
co
m
p
u
ter
co
n
tr
o
l
.
I
n
r
ec
en
t
y
ea
r
s
,
p
len
t
y
o
f
s
tu
d
ies
h
a
v
e
b
ee
n
p
r
esen
ted
w
it
h
r
eg
ar
d
to
r
ap
id
p
r
o
to
t
y
p
in
g
o
f
m
etalli
c
p
ar
ts
b
ased
o
n
g
as
m
etal
ar
c
w
eld
i
n
g
(
GM
A
W
)
[
3
]
.
I
n
t
h
i
s
p
r
o
ce
s
s
,
an
elec
tr
ic
ar
c
b
et
w
ee
n
a
co
n
s
u
m
ab
le
w
ir
e
elec
tr
o
d
e
an
d
th
e
m
eta
lli
c
w
o
r
k
p
iece
is
g
e
n
er
ated
,
w
h
ich
h
ea
t
s
t
h
e
w
o
r
k
p
iece
an
d
ca
u
s
e
s
th
e
m
to
m
elt
an
d
j
o
in
.
A
r
o
b
o
tic
s
y
s
te
m
is
ap
p
lied
in
s
o
m
e
o
f
th
e
GM
A
W
p
r
o
ce
s
s
es
to
co
n
tr
o
l
th
e
d
ep
o
s
it
o
f
w
eld
ed
m
ater
ial
,
p
o
s
itio
n
o
f
w
eld
i
n
g
to
r
ch
an
d
s
o
m
eo
th
er
p
ar
a
m
eter
s
s
u
c
h
a
s
s
lo
p
e
an
d
r
o
tati
o
n
o
f
t
h
e
to
r
ch
.
A
s
ch
e
m
atic
v
ie
w
o
f
t
h
e
r
o
b
o
tic
GM
A
W
p
r
o
ce
s
s
is
d
ep
icted
in
Fig
u
r
e
1
[
4
]
.
I
n
w
eld
i
n
g
p
r
o
ce
s
s
es,
w
eld
b
ea
d
g
eo
m
etr
y
,
n
a
m
e
l
y
w
eld
w
id
t
h
a
n
d
w
eld
b
ea
d
h
ei
g
h
t
,
as
s
h
o
w
n
i
n
Fig
u
r
e
2
is
o
n
e
o
f
th
e
i
m
p
o
r
tan
t
ch
ar
ac
ter
is
tics
o
f
t
h
e
w
el
d
in
g
li
n
e.
E
s
p
ec
iall
y
,
i
n
t
h
e
GM
A
W
p
r
o
ce
s
s
,
th
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
A
p
p
lica
tio
n
o
f Mu
ltip
le
K
ern
el
S
u
p
p
o
r
t V
ec
to
r
R
eg
r
ess
io
n
fo
r
W
e
ld
B
ea
d
Geo
metry
…
(
N
a
d
er Mo
lla
yi
)
2311
b
ea
d
g
eo
m
e
tr
y
h
a
s
a
s
i
g
n
i
f
ica
n
t
ef
f
ec
t
o
n
t
h
e
la
y
er
th
ic
k
n
es
s
,
s
u
r
f
ac
e
q
u
alit
y
,
a
n
d
d
i
m
e
n
s
io
n
al
p
r
ec
is
io
n
.
T
h
e
m
o
s
t
i
m
p
o
r
tan
t
p
ar
a
m
eter
s
w
h
ic
h
a
f
f
ec
t
th
eb
ea
d
g
eo
m
e
tr
y
ar
e
w
eld
in
g
c
u
r
r
en
t
a
n
d
v
o
ltag
e,
t
y
p
e
a
n
d
p
er
ce
n
tag
e
o
f
i
n
er
t g
a
s
,
an
d
th
e
d
is
ta
n
ce
o
f
n
o
zz
le
to
th
e
w
o
r
k
p
iece
.
As
ap
p
r
o
p
r
iate
w
eld
b
ea
d
g
eo
m
e
tr
y
r
es
u
lt
s
i
n
h
i
g
h
w
el
d
q
u
alit
y
,
p
r
o
p
er
s
elec
tio
n
o
f
w
eld
i
n
g
p
ar
am
eter
s
s
o
as
to
o
b
tain
a
d
esire
d
w
eld
g
eo
m
etr
y
i
s
o
f
g
r
ea
t
atten
tio
n
in
GM
A
W
p
r
o
ce
s
s
.
T
o
th
is
en
d
,
it
i
s
g
r
ea
tl
y
i
m
p
o
r
tan
t
to
d
ev
elo
p
a
g
lo
b
al
m
o
d
el
o
f
w
eld
g
eo
m
etr
y
b
ased
o
n
t
h
e
p
r
o
ce
s
s
p
ar
a
m
eter
s
.
D
u
e
to
t
h
e
h
ig
h
l
y
n
o
n
-
l
in
ea
r
an
d
co
u
p
led
m
u
lt
iv
ar
iab
le
ef
f
ec
t
o
f
t
h
e
in
p
u
t
p
ar
am
eter
s
o
n
th
e
w
eld
g
e
o
m
e
tr
y
,
th
i
s
m
o
d
el
ca
n
n
o
t
b
e
d
ef
in
ed
th
r
o
u
g
h
an
ex
p
licit
m
ath
e
m
at
ical
ex
p
r
ess
io
n
a
n
d
ad
v
an
ce
d
m
o
d
elli
n
g
tec
h
n
iq
u
es
ar
e
in
v
e
s
ti
g
ated
f
o
r
th
is
p
u
r
p
o
s
e.
Fig
u
r
e
1
.
Sch
e
m
atic
Vie
w
o
f
t
h
e
r
o
b
o
tic
GM
A
W
P
r
o
ce
s
s
[
4
]
Fig
u
r
e
2
.
W
eld
b
ea
d
g
eo
m
e
tr
y
Ma
ch
i
n
e
lear
n
in
g
i
s
a
s
u
b
f
i
eld
o
f
co
m
p
u
ter
s
cie
n
ce
,
i
n
w
h
ic
h
t
h
e
s
tu
d
y
a
n
d
co
n
s
t
r
u
ctio
n
o
f
alg
o
r
ith
m
s
,
ca
p
ab
le
o
f
lear
n
i
n
g
f
r
o
m
a
n
d
m
ak
i
n
g
p
r
ed
icti
o
n
s
b
ased
o
n
a
l
i
m
ited
s
et
o
f
o
b
s
er
v
ed
d
ata
is
ex
p
lo
r
ed
.
I
n
s
u
ch
al
g
o
r
ith
m
s
a
m
o
d
el
i
s
b
u
ilt
f
r
o
m
ex
a
m
p
le
i
n
p
u
t
s
i
n
o
r
d
er
to
m
ak
e
d
ata
-
d
r
iv
en
p
r
ed
ictio
n
s
o
r
d
ec
is
io
n
s
.
Su
p
er
v
i
s
ed
lear
n
i
n
g
is
th
e
m
ac
h
i
n
e
lear
n
i
n
g
tas
k
o
f
in
f
er
r
in
g
a
f
u
n
ctio
n
f
r
o
m
a
s
et
o
f
lab
eled
tr
ain
i
n
g
d
ata
[
5
]
.
T
h
e
alg
o
r
ith
m
s
i
n
t
h
i
s
f
ield
ca
n
b
e
u
s
ed
to
estab
lis
h
a
m
o
d
el
b
ased
o
n
a
li
m
ited
s
et
o
f
o
b
s
er
v
atio
n
s
f
o
r
m
ak
i
n
g
p
r
ed
ictio
n
s
in
ca
s
es
w
h
ic
h
h
av
e
n
o
t
o
b
s
er
v
ed
.
T
h
er
ef
o
r
e,
th
ese
alg
o
r
ith
m
s
ca
n
b
e
u
s
ed
f
o
r
m
o
d
elli
n
g
an
d
p
r
ed
ictio
n
o
f
w
eld
g
eo
m
etr
y
in
G
MA
W
p
r
o
ce
s
s
an
d
s
e
v
er
al
r
e
s
ea
r
ch
es
h
a
v
e
b
ee
n
p
er
f
o
r
m
ed
i
n
th
is
f
ield
w
h
ic
h
ar
e
m
ai
n
l
y
b
ased
o
n
n
e
u
r
al
n
et
w
o
r
k
s
an
d
f
u
zz
y
s
y
s
te
m
s
[
6
]
.
I
n
f
ield
r
o
b
o
tic
GM
A
W
p
r
o
ce
s
s
,
a
g
lo
b
al
d
atab
ase
o
f
p
r
o
ce
s
s
p
ar
am
eter
s
an
d
th
e
co
r
r
esp
o
n
d
in
g
w
eld
g
eo
m
etr
y
h
as
b
ee
n
p
r
o
v
id
ed
b
y
[
7
]
an
d
p
r
ed
ictiv
e
m
o
d
elli
n
g
h
a
s
b
ee
n
p
er
f
o
r
m
ed
b
y
b
o
th
t
h
e
n
eu
r
al
n
et
w
o
r
k
a
n
d
s
ec
o
n
d
o
r
d
er
r
eg
r
ess
io
n
an
al
y
s
i
s
m
eth
o
d
s
,
w
h
ic
h
p
r
o
v
es th
e
h
i
g
h
er
ac
c
u
r
ac
y
o
f
t
h
e
n
e
u
r
al
n
et
w
o
r
k
ap
p
r
o
ac
h
o
v
er
th
e
s
ec
o
n
d
o
r
d
er
r
eg
r
ess
io
n
.
Su
p
p
o
r
t
v
ec
to
r
m
ac
h
i
n
e
(
SV
M
)
is
a
s
tate
-
of
-
th
e
-
ar
t
ap
p
r
o
ac
h
to
s
u
p
er
v
is
ed
lear
n
in
g
,
u
s
ed
f
o
r
class
i
f
icatio
n
an
d
r
eg
r
es
s
io
n
an
al
y
s
is
,
w
h
ic
h
h
a
s
b
ee
n
p
r
o
v
en
as
a
p
o
w
er
f
u
l
m
e
th
o
d
in
m
a
n
y
p
r
ac
tical
ap
p
licatio
n
s
[
8
]
.
S
tr
u
ct
u
r
al
r
is
k
m
in
i
m
izat
io
n
alo
n
g
s
id
e
w
it
h
e
m
p
ir
ical
r
i
s
k
m
in
i
m
izatio
n
i
s
t
h
e
m
ai
n
ad
v
an
ta
g
e
o
f
th
e
SVM
s
o
v
er
th
e
n
eu
r
al
n
et
w
o
r
k
s
r
es
u
lt
in
g
i
n
a
b
etter
g
en
er
aliza
t
io
n
c
ap
ab
ilit
y
i
n
m
an
y
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2088
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
4
,
A
u
g
u
s
t 2
0
1
8
:
2
3
1
0
–
2
3
1
8
2312
p
r
o
b
lem
s
[
9
]
.
T
h
e
ac
cu
r
ac
y
o
f
SVM
-
b
ased
m
o
d
elli
n
g
ca
n
b
e
en
h
a
n
ce
d
b
y
a
n
e
w
er
a
p
p
r
o
ac
h
k
n
o
w
n
a
s
m
u
ltip
le
k
er
n
el
lear
n
i
n
g
,
w
h
i
ch
i
s
i
n
tr
o
d
u
ce
d
in
S
ec
tio
n
3
[
1
0
]
.
R
eg
ar
d
in
g
to
t
h
e
h
ig
h
d
eg
r
ee
o
f
ac
c
u
r
ac
y
r
eq
u
ir
ed
in
p
r
ed
ictio
n
o
f
w
el
d
b
ea
d
g
eo
m
etr
y
in
r
o
b
o
tic
GM
A
W
p
r
o
ce
s
s
,
ap
p
licatio
n
o
f
m
u
lt
ip
le
k
er
n
el
s
u
p
p
o
r
t v
ec
to
r
m
ac
h
i
n
e
f
o
r
t
h
i
s
p
r
ed
ictio
n
h
as b
ee
n
d
is
c
u
s
s
e
d
in
t
h
is
p
ap
er
an
d
t
h
is
ap
p
r
o
ac
h
h
as b
ee
n
p
r
o
v
en
to
p
r
o
v
id
e
m
o
r
e
ac
cu
r
ac
y
a
n
d
g
en
er
aliza
tio
n
ca
p
ab
ilit
y
.
2.
SUPP
O
RT
V
E
CT
O
R
M
ACH
I
NE
Su
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
s
(
SV
Ms)
ar
e
s
u
p
er
v
i
s
ed
lear
n
i
n
g
m
o
d
els
w
it
h
as
s
o
ciate
d
lear
n
in
g
alg
o
r
it
h
m
s
w
h
ic
h
a
n
al
y
ze
d
ata
an
d
r
ec
o
g
n
ize
p
atter
n
s
,
u
s
ed
f
o
r
class
if
ica
tio
n
a
n
d
r
eg
r
ess
io
n
an
al
y
s
i
s
[
1
1
]
.
A
lin
ea
r
SVM
-
b
ased
clas
s
i
f
ier
s
y
s
te
m
f
in
d
s
th
e
h
y
p
er
-
p
la
n
e
w
h
ic
h
le
ad
s
to
t
h
e
m
a
x
i
m
u
m
m
ar
g
i
n
b
et
w
ee
n
th
e
s
a
m
p
les
o
f
t
h
e
t
w
o
cla
s
s
e
s
i
n
t
h
e
tr
ai
n
in
g
d
ataset,
w
h
ile
m
in
i
m
iz
in
g
th
e
c
lass
if
ica
tio
n
er
r
o
r
.
Su
c
h
a
clas
s
if
ier
ca
n
b
e
d
escr
ib
ed
b
y
E
q
u
atio
n
(
1
)
,
in
w
h
ich
x
is
t
h
e
i
n
p
u
t
v
ec
to
r
an
d
w
an
d
b
ar
e
th
e
w
eig
h
t
s
an
d
b
ias
v
ec
to
r
s
,
re
s
p
ec
tiv
el
y
[
1
2
]
.
T
h
e
o
p
tim
u
m
v
al
u
e
s
o
f
w
a
n
d
b
ar
e
o
b
ta
in
ed
b
y
m
in
i
m
izat
io
n
o
f
t
h
e
r
is
k
f
u
n
ctio
n
R
(
w
)
ex
p
r
ess
ed
i
n
E
q
u
atio
n
(
2
)
,
s
u
b
j
ec
ted
to
th
e
co
n
s
tr
ai
n
ts
o
f
E
q
u
atio
n
(
3
)
,
f
o
r
t
h
e
N
s
a
m
p
le
s
o
f
th
e
(
x
i
,
y
i
)
i
n
t
h
e
tr
ain
i
n
g
d
ataset
[
1
3
]
.
(
)
(
)
(
1
)
(
)
‖
‖
∑
(
2
)
(
)
(
3
)
I
n
th
e
r
is
k
f
u
n
ctio
n
o
f
E
q
u
a
t
io
n
(
2
)
,
th
e
f
ir
s
t
ter
m
s
tan
d
s
f
o
r
th
e
s
tr
u
ct
u
r
al
r
is
k
,
i.e
.
th
e
m
ar
g
i
n
b
et
w
ee
n
th
e
t
w
o
clas
s
es
a
n
d
th
e
s
ec
o
n
d
ter
m
s
tan
d
s
f
o
r
th
e
e
m
p
ir
ical
r
is
k
,
i.e
.
th
e
tr
ain
i
n
g
er
r
o
r
.
T
h
e
p
ar
am
eter
C
,
i
s
th
e
r
eg
u
lar
izat
io
n
f
ac
to
r
an
d
it
tr
ad
es
o
f
f
th
e
r
elativ
e
i
m
p
o
r
tan
ce
o
f
m
a
x
i
m
izin
g
th
e
s
tr
u
ct
u
r
al
an
d
e
m
p
ir
ical
er
r
o
r
s
.
Fig
u
r
e
3
s
h
o
w
s
t
h
e
SV
M
-
b
ased
clas
s
i
f
i
ca
tio
n
.
Fig
u
r
e
3
.
SVM
-
b
ased
clas
s
i
f
ic
atio
n
I
n
ca
s
e
o
f
d
ata
w
ith
n
o
n
li
n
e
ar
b
o
r
d
er
b
etw
ee
n
th
e
t
w
o
cl
ass
es
t
h
e
o
r
ig
in
al
f
ea
t
u
r
e
s
p
ac
e
ca
n
b
e
m
ap
p
ed
to
s
o
m
e
h
i
g
h
er
-
d
i
m
e
n
s
io
n
al
f
ea
tu
r
e
s
p
ac
e
w
h
er
e
t
h
e
tr
ain
in
g
s
et
is
s
ep
ar
ab
le,
t
h
r
o
u
g
h
a
n
o
n
li
n
ea
r
f
u
n
ctio
n
k
n
o
w
n
a
s
th
e
k
er
n
el
f
u
n
ct
io
n
,
as d
ep
icted
in
Fi
g
u
r
e
4
[
1
4
]
.
Fig
u
r
e
4
.
Ma
p
p
in
g
o
f
t
h
e
f
ea
t
u
r
e
s
p
ac
e
b
ased
o
n
a
k
er
n
el
f
u
n
ctio
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
A
p
p
lica
tio
n
o
f Mu
ltip
le
K
ern
el
S
u
p
p
o
r
t V
ec
to
r
R
eg
r
ess
io
n
fo
r
W
e
ld
B
ea
d
Geo
metry
…
(
N
a
d
er Mo
lla
yi
)
2313
T
h
er
ef
o
r
e
SVM
-
b
ased
class
i
f
i
ca
tio
n
ca
n
b
e
ex
p
r
ess
ed
as
:
(
)
(
(
)
)
(
4
)
I
n
w
h
ic
h
w
a
n
d
b
ar
e
o
b
tain
ed
b
y
m
in
i
m
iz
in
g
th
e
r
is
k
f
u
n
cti
o
n
R
(
w
)
i
n
E
q
u
atio
n
(
5
)
s
u
b
j
ec
ted
to
th
e
co
n
s
tr
ain
ts
o
f
:
(
(
)
)
(
5
)
T
h
e
co
n
ce
p
t
o
f
SVM
class
if
ic
atio
n
ca
n
b
e
g
e
n
er
alize
d
f
o
r
t
h
e
p
u
r
p
o
s
e
o
f
r
eg
r
es
s
io
n
b
y
i
n
tr
o
d
u
cin
g
th
e
m
ar
g
i
n
o
f
to
ler
an
ce
f
o
r
th
e
f
u
n
ctio
n
to
b
e
esti
m
ated
,
b
ased
o
n
th
e
p
er
m
itted
esti
m
atio
n
er
r
o
r
.
Giv
en
a
li
m
ited
n
u
m
b
er
o
f
o
b
s
er
v
atio
n
s
f
r
o
m
t
h
e
f
u
n
ctio
n
f
(
x
)
w
it
h
th
e
p
er
m
itted
m
ar
g
i
n
o
f
to
ler
an
ce
,
SVM
-
b
ased
class
i
f
icatio
n
b
et
w
ee
n
f
(
x
)
+
an
d
f
(
x
)
-
ca
n
b
e
co
n
s
id
er
ed
as
esti
m
ati
n
g
f
(
x
)
in
th
e
p
er
m
itted
m
ar
g
i
n
o
f
to
ler
an
ce
,
as
d
ep
icted
in
F
ig
u
r
e
5
[
1
5
]
.
I
n
o
th
er
w
o
r
d
s
,
i
n
SVM
-
b
ased
r
e
g
r
ess
io
n
(
SV
R
)
,
th
e
i
n
p
u
t
s
p
ac
e
is
m
ap
p
ed
in
to
a
h
ig
h
d
i
m
e
n
s
io
n
al
f
ea
tu
r
e
s
p
ac
e
v
ia
t
h
e
k
er
n
el
f
u
n
ctio
n
a
n
d
th
e
n
a
lin
ea
r
o
p
ti
m
al
r
eg
r
e
s
s
io
n
i
s
p
er
f
o
r
m
ed
i
n
th
i
s
s
p
ac
e.
T
h
er
ef
o
r
e,
th
e
f
o
r
m
u
latio
n
o
f
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
i
n
es
ca
n
b
e
g
en
er
alize
d
f
o
r
th
e
p
u
r
p
o
s
e
o
f
r
eg
r
ess
io
n
a
s
:
(
)
∑
(
)
(
)
(
6
)
Fig
u
r
e
5
.
Gen
er
aliza
tio
n
o
f
S
VM
-
b
ased
clas
s
i
f
icatio
n
to
S
VM
-
b
ased
r
eg
r
es
s
io
n
[
1
5
]
T
h
e
o
p
tim
al
r
eg
r
ess
io
n
is
o
b
tain
ed
b
y
m
a
x
i
m
izi
n
g
t
h
e
(
)
f
u
n
ctio
n
in
E
q
u
atio
n
(
7
)
s
u
b
j
ec
te
d
to
th
e
co
n
s
tr
ai
n
ts
g
i
v
en
b
y
E
q
u
atio
n
(
8
)
[
1
5
].
(
)
∑
∑
{
(
)
(
)
〈
(
)
(
)
〉
}
∑
(
)
∑
(
)
(7)
{
∑
[
]
(
8
)
A
cc
o
r
d
in
g
to
t
h
e
Me
r
ce
r
’
s
t
h
eo
r
em
[
1
4
]
,
th
e
i
n
n
er
p
r
o
d
u
ct
〈
(
)
(
)
〉
ca
n
b
e
d
ef
i
n
ed
t
h
r
o
u
g
h
a
k
er
n
el
f
u
n
ctio
n
as
(
)
〈
(
)
(
)
〉
.
T
h
er
ef
o
r
e,
th
e
(
)
f
u
n
c
tio
n
in
E
q
u
atio
n
(
7
)
ca
n
b
e
ex
p
r
ess
ed
as
:
(
)
∑
∑
{
(
)
(
)
(
)
}
∑
(
)
∑
(
)
(9)
T
h
e
o
p
tim
izatio
n
p
r
o
b
le
m
ca
n
b
e
s
o
lv
ed
v
ia
q
u
ad
r
atic
p
r
o
g
r
a
m
m
i
n
g
o
p
ti
m
izatio
n
an
d
t
h
e
esti
m
ated
f
u
n
ctio
n
is
e
x
p
r
ess
ed
b
ased
o
n
th
e
o
p
ti
m
al
v
alu
e
s
as
E
q
u
atio
n
(
1
0
)
[
1
6
]
.
(
)
∑
(
)
(
)
(
1
0
)
T
h
e
m
o
s
t c
o
m
m
o
n
f
o
r
m
u
latio
n
s
f
o
r
th
e
k
er
n
el
f
u
n
ctio
n
ar
e
l
is
ted
in
T
ab
le
1
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2088
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
4
,
A
u
g
u
s
t 2
0
1
8
:
2
3
1
0
–
2
3
1
8
2314
T
ab
le
1
.
Mo
s
t
C
o
m
m
o
n
Fo
r
m
u
latio
n
s
f
o
r
t
h
e
Ker
n
el
F
u
n
cti
o
n
K
e
r
n
e
l
t
y
p
e
F
o
r
mu
l
a
t
i
o
n
G
a
u
ssi
a
n
r
a
d
i
a
l
b
a
si
s (
R
B
F
)
(
)
(
‖
‖
)
P
o
l
y
n
o
mi
a
l
o
f
d
e
g
r
e
e
d
(
)
(
〈
〉
)
M
u
l
t
i
-
L
a
y
e
r
P
e
r
c
e
p
t
r
o
n
(
M
L
P
)
(
)
(
〈
〉
)
3.
SUPP
O
RT
V
E
CT
O
R
R
E
G
RE
SS
I
O
N
B
ASE
D
O
N
M
U
L
T
I
P
L
E
K
E
RN
E
L
L
E
ARNI
NG
I
n
SVM
-
b
ased
r
eg
r
es
s
io
n
,
t
h
e
p
er
f
o
r
m
a
n
ce
o
f
th
e
lear
n
i
n
g
al
g
o
r
ith
m
h
ig
h
l
y
d
ep
en
d
s
o
n
th
e
d
ata
r
ep
r
esen
tatio
n
,
w
h
ic
h
i
s
c
h
o
s
en
th
r
o
u
g
h
t
h
e
k
er
n
el
f
u
n
c
tio
n
.
Ker
n
e
l
f
u
n
ctio
n
m
ea
s
u
r
es
th
e
n
o
n
li
n
ea
r
s
i
m
ilar
it
y
b
et
w
ee
n
s
a
m
p
les,
s
o
an
ef
f
icie
n
t
k
er
n
el
s
h
o
u
ld
r
ep
r
esen
t
d
ata
ad
ap
tiv
el
y
.
I
n
ad
d
itio
n
,
an
ap
p
r
o
p
r
iate
r
eg
u
lar
izatio
n
te
r
m
is
d
ef
in
ed
f
o
r
th
e
lear
n
i
n
g
p
r
o
b
le
m
i
n
ter
m
s
o
f
t
h
e
k
er
n
el
f
u
n
ctio
n
's
p
ar
am
eter
s
.
I
n
m
o
s
t
ca
s
es,
t
h
e
p
ar
a
m
eter
s
of
a
s
in
g
le
k
e
r
n
el
f
u
n
ctio
n
i
s
tu
n
ed
f
o
r
th
e
w
h
o
le
d
ata
s
ets.
A
lt
h
o
u
g
h
th
e
k
er
n
el
p
ar
a
m
et
er
ca
n
b
e
o
p
ti
m
all
y
c
h
o
s
e
n
to
en
h
an
ce
t
h
e
g
e
n
er
aliza
tio
n
ca
p
ab
ilit
y
,
lear
n
in
g
w
it
h
s
i
n
g
le
k
er
n
el
is
n
o
t
v
er
y
d
ata
-
ad
ap
ted
o
r
d
is
cr
i
m
in
a
t
iv
e.
Mu
ltip
le
k
e
r
n
el
lear
n
i
n
g
(
MK
L
)
p
r
o
v
id
es
a
m
o
r
e
f
lex
ib
le
f
r
a
m
e
w
o
r
k
th
a
n
s
in
g
le
k
er
n
el
an
d
m
i
n
es
d
ata
in
f
o
r
m
atio
n
m
o
r
e
ad
ap
tiv
el
y
a
n
d
m
o
r
e
ef
f
ec
ti
v
el
y
[
1
7
]
.
I
n
th
e
MK
L
f
r
a
m
e
w
o
r
k
,
th
e
k
er
n
el
f
u
n
ctio
n
is
f
o
r
m
ed
b
ased
a
li
n
ea
r
co
n
v
e
x
co
m
b
i
n
atio
n
o
f
M
f
u
n
ctio
n
s
w
h
ic
h
s
ati
s
f
y
t
h
e
M
er
ce
r
'
s
co
n
d
itio
n
s
,
f
o
r
m
u
la
ted
as
:
(
)
∑
(
)
(
1
1
)
w
h
er
e
is
t
h
e
w
eig
h
t o
f
t
h
e
m
-
th
b
asis
k
er
n
el
f
u
n
ctio
n
an
d
m
u
s
t
s
atis
f
y
th
e
co
n
d
it
io
n
s
o
f
:
∑
(
1
2
)
T
h
e
co
m
b
in
in
g
w
ei
g
h
ts
ar
e
co
n
s
id
er
ed
as a
v
ec
to
r
o
f
w
ei
g
h
t
s
,
n
a
m
el
y
[
]
.
T
h
e
m
u
ltip
le
k
er
n
el
lear
n
i
n
g
(
MK
L
)
p
r
o
b
le
m
ca
n
b
e
d
escr
ib
ed
as
lear
n
i
n
g
t
h
e
co
m
b
i
n
i
n
g
w
ei
g
h
ts
an
d
th
e
s
o
lu
tio
n
s
o
f
th
e
o
r
ig
in
al
p
r
o
b
lem
,
f
o
r
ex
a
m
p
le,
th
e
s
o
lu
tio
n
s
o
f
an
d
f
o
r
SVR
p
r
o
b
lem
i
n
E
q
u
atio
n
(
15)
,
in
a
s
in
g
le
o
p
ti
m
izat
i
o
n
p
r
o
b
lem
.
B
y
s
u
b
s
t
itu
tio
n
o
f
E
q
u
atio
n
(
1
1
)
in
to
E
q
u
atio
n
(9
)
,
th
e
o
p
tim
izatio
n
p
r
o
b
le
m
o
f
M
KL
-
b
ased
SV
R
is
o
b
tai
n
ed
as
m
a
x
i
m
izatio
n
o
f
(
)
in
E
q
u
atio
n
(
1
3
)
s
u
b
j
ec
ted
to
th
e
co
n
s
tr
ain
ts
o
f
E
q
u
atio
n
(
1
4
)
[
1
7
]
.
(
)
∑
∑
{
(
)
(
)
∑
(
)
}
∑
(
)
∑
(
)
(13
)
{
∑
[
]
∑
(
1
4
)
R
ec
en
t
l
y
,
a
s
i
m
p
le
an
d
ef
f
icie
n
t
alg
o
r
it
h
m
f
o
r
m
u
ltip
le
k
er
n
el
lear
n
in
g
h
a
s
b
ee
n
p
r
o
p
o
s
ed
b
y
w
h
ic
h
s
o
lv
es
t
h
e
o
p
ti
m
izatio
n
p
r
o
b
le
m
b
y
ap
p
licatio
n
o
f
th
e
g
r
ad
ien
t
d
escen
t
m
et
h
o
d
[
1
8
]
.
T
h
is
ap
p
r
o
ac
h
,
k
n
o
w
n
as
Si
m
p
le
MK
L
,
i
s
b
ased
o
n
th
e
f
ac
t
t
h
at
t
h
e
o
b
j
ec
tiv
e
f
u
n
ctio
n
L
in
E
q
u
a
tio
n
(
1
3
)
is
co
n
v
e
x
an
d
d
if
f
er
e
n
tiab
le.
T
h
er
ef
o
r
e,
th
e
o
p
ti
m
u
m
v
ec
to
r
o
f
w
eig
h
t
s
d
ca
n
b
e
o
b
tain
e
d
b
y
m
ea
n
s
o
f
u
p
d
ati
n
g
it
o
n
t
h
e
g
r
ad
ie
n
t
d
esce
n
t
d
ir
ec
tio
n
o
f
L
.
I
n
t
h
i
s
m
et
h
o
d
,
th
e
g
r
ad
ie
n
t o
f
o
b
j
ec
tiv
e
f
u
n
ct
io
n
is
co
m
p
u
ted
b
y
th
e
d
er
iv
at
iv
es o
f
L
a
s
∑
∑
(
)
(
)
(
)
(
1
5
)
I
n
p
r
o
g
r
ess
,
th
e
d
esce
n
t d
ir
ec
ti
o
n
D
o
f
g
r
ad
ien
ts
is
f
o
u
n
d
an
d
d
is
u
p
d
ated
as
:
(
1
6
)
w
h
er
e
is
th
e
s
tep
len
g
t
h
.
T
h
e
g
r
ad
ien
t
o
f
th
e
o
b
j
ec
tiv
e
f
u
n
ctio
n
is
o
n
l
y
u
p
d
ated
w
h
e
n
th
e
o
b
j
ec
tiv
e
v
alu
e
d
ec
r
ea
s
es.
T
h
is
u
p
d
ate
p
r
o
ce
d
u
r
e
is
r
ep
ea
ted
u
n
ti
l th
e
s
to
p
p
in
g
cr
iter
io
n
i
s
m
et
[
1
8
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
A
p
p
lica
tio
n
o
f Mu
ltip
le
K
ern
el
S
u
p
p
o
r
t V
ec
to
r
R
eg
r
ess
io
n
fo
r
W
e
ld
B
ea
d
Geo
metry
…
(
N
a
d
er Mo
lla
yi
)
2315
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
I
n
o
r
d
er
to
estab
lis
h
t
h
e
M
K
-
SVR
p
r
ed
icti
v
e
m
o
d
els,
a
d
at
ab
ase
o
f
m
ea
s
u
r
ed
v
alu
e
s
o
f
w
eld
b
ea
d
g
eo
m
etr
y
to
g
e
th
er
w
i
th
th
e
co
r
r
esp
o
n
d
in
g
p
r
o
ce
s
s
p
ar
a
m
ete
r
s
,
p
r
o
v
id
ed
b
y
X
io
n
g
,
et
a
l
.
was u
t
ilized
,
w
h
ich
is
s
h
o
w
n
in
T
ab
le
2
[
7
]
.
Fr
o
m
t
h
is
d
atab
ase,
th
e
f
ir
s
t
t
h
ir
t
y
o
n
e
s
a
m
p
les
w
er
e
u
s
ed
to
tr
ain
th
e
MK
-
S
VR
m
o
d
els
an
d
th
e
p
r
ed
ictab
le
ac
cu
r
ac
y
o
f
t
h
e
estab
li
s
h
ed
m
o
d
els
was
ev
al
u
ated
b
ased
o
n
th
e
n
e
x
t
t
w
el
v
e
s
a
m
p
le
s
,
m
ar
k
ed
i
n
b
o
ld
.
T
o
im
p
r
o
v
e
t
h
e
ac
cu
r
ac
y
,
all
t
h
e
i
n
p
u
t
a
n
d
tar
g
et
v
a
lu
e
s
w
er
e
n
o
r
m
alize
d
b
et
w
ee
n
−1
an
d
+1
as
:
(
)
(
)
(
1
7
)
w
h
er
e,
ma
x
an
d
min
ar
e
r
esp
e
ctiv
el
y
th
e
m
ax
i
m
u
m
o
r
m
in
i
m
u
m
v
alu
e
o
f
th
e
in
p
u
t
o
r
th
e
o
u
tp
u
t
a
m
o
n
g
t
h
e
w
h
o
le
d
ataset,
is
th
e
in
p
u
t
o
r
o
u
tp
u
t
an
d
is
th
e
co
r
r
esp
o
n
d
i
n
g
n
o
r
m
alize
d
v
al
u
e.
B
ased
o
n
th
e
n
o
r
m
alize
d
d
ataset,
th
e
s
i
n
g
le
k
er
n
el
a
n
d
m
u
ltip
le
k
er
n
el
SVM
m
o
d
els
w
er
e
i
m
p
le
m
en
ted
b
y
t
h
e
SV
M
-
KM
[
2
0
]
an
d
t
h
e
Si
m
p
leM
K
L
Ma
tlab
to
o
lb
o
x
es
,
r
esp
ec
tiv
el
y
.
T
r
ain
in
g
th
e
m
o
d
el
s
a
n
d
ca
lc
u
lati
n
g
t
h
e
p
r
ed
icted
n
o
r
m
alize
d
o
u
tp
u
ts
,
t
h
e
y
w
er
e
s
ca
led
to
th
eir
o
r
ig
in
al
r
an
g
e,
as
:
̂
(
)
(
)
(
1
8
)
w
h
er
e
̂
is
th
e
p
r
ed
icted
o
u
tp
u
t
in
t
h
e
o
r
ig
i
n
al
r
an
g
e
a
n
d
is
th
e
n
o
r
m
alize
d
p
r
ed
icted
o
u
tp
u
t
.
Fo
r
p
r
ed
ictiv
e
m
o
d
ell
in
g
o
f
t
h
e
b
ea
d
w
id
t
h
,
th
e
k
er
n
el
f
u
n
ctio
n
w
a
s
s
elec
t
ed
as
co
m
b
i
n
atio
n
o
f
4
0
1
Gau
s
s
ian
b
asis
k
er
n
el
f
u
n
ctio
n
s
w
it
h
p
ar
a
m
eter
s
v
ar
y
in
g
f
r
o
m
8
w
it
h
i
n
cr
e
m
e
n
t
o
f
0
.
0
1
to
1
2
an
d
in
ca
s
e
o
f
th
e
b
ea
d
h
eig
h
t
it
w
a
s
s
elec
ted
as
co
m
b
i
n
atio
n
o
f
t
h
r
ee
p
o
ly
n
o
m
ia
l
b
asis
f
u
n
ct
io
n
s
w
it
h
p
ar
a
m
eter
s
o
f
1
,
2
,
3
an
d
8
1
Gau
s
s
ia
n
b
asi
s
k
er
n
el
f
u
n
ctio
n
s
w
it
h
p
ar
a
m
e
ter
s
v
ar
y
i
n
g
f
r
o
m
0
.
2
w
it
h
i
n
cr
e
m
en
t
o
f
0
.
0
1
to
1
.
T
h
e
s
i
n
g
le
k
er
n
el
m
o
d
els
w
er
e
i
m
p
le
m
en
ted
b
ased
o
n
th
e
Ga
u
s
s
ia
n
k
er
n
el
f
u
n
c
tio
n
.
T
h
e
p
ar
am
eter
s
o
f
s
i
n
g
le
k
er
n
el
an
d
m
u
ltip
le
ke
r
n
el
m
o
d
els ar
e
lis
ted
in
T
ab
le
3
.
T
ab
le
2
.
T
h
e
GM
A
W
P
r
o
ce
s
s
P
ar
am
eter
s
a
n
d
th
e
C
o
r
r
esp
o
n
d
in
g
Val
u
es o
f
W
eld
Geo
m
e
tr
y
[
7
]
Ex
p
e
r
i
me
n
t
No
W
i
r
e
f
e
e
d
r
a
t
e
(
m/
mi
n)
W
e
l
d
i
n
g
sp
e
e
d
(
c
m/
mi
n
)
A
r
c
v
o
l
t
a
g
e
(
V
)
N
o
z
z
l
e
t
o
p
l
a
t
e
d
i
s
t
a
n
c
e
(
mm
)
B
e
a
d
w
i
d
t
h
(
mm
)
B
e
a
d
h
e
i
g
h
t
(
mm
)
1
5
.
2
2
2
.
5
1
7
.
5
9
8
.
9
5
2
.
8
8
2
3
.
6
2
2
.
5
1
7
.
5
9
1
0
.
7
2
3
.
3
5
3
5
.
2
3
7
.
5
1
7
.
5
9
7
.
1
9
2
.
4
5
4
3
.
6
3
7
.
5
1
7
.
5
9
8
.
2
9
2
.
7
5
5
5
.
2
2
2
.
5
2
0
.
5
9
1
0
.
2
5
2
.
6
6
6
3
.
6
2
2
.
5
2
0
.
5
9
1
1
.
5
3
.
2
6
7
5
.
2
3
7
.
5
2
0
.
5
9
8
.
3
6
2
.
1
7
8
3
.
6
3
7
.
5
2
0
.
5
9
9
.
3
5
2
.
5
8
9
5
.
2
2
2
.
5
1
7
.
5
15
8
.
3
6
3
10
3
.
6
2
2
.
5
1
7
.
5
15
9
.
5
2
3
.
5
6
11
5
.
2
3
7
.
5
1
7
.
5
15
6
.
8
3
2
.
4
5
12
3
.
6
3
7
.
5
1
7
.
5
15
7
.
9
8
2
.
9
13
5
.
2
2
2
.
5
2
0
.
5
15
9
.
9
2
2
.
7
9
14
3
.
6
2
2
.
5
2
0
.
5
15
1
1
.
1
2
3
.
3
5
15
5
.
2
3
7
.
5
2
0
.
5
15
7
.
9
1
2
.
2
6
16
2
.
8
3
7
.
5
2
0
.
5
15
9
.
2
5
2
.
7
17
6
30
19
12
7
.
3
9
2
.
3
2
18
4
.
4
30
19
12
9
.
9
3
.
2
8
19
4
.
4
15
19
12
1
1
.
7
6
3
.
8
20
4
.
4
45
19
12
7
.
5
4
2
.
3
4
21
4
.
4
30
16
12
8
.
0
8
2
.
9
4
22
4
.
4
30
22
12
9
.
9
2
.
4
5
23
4
.
4
30
19
6
9
.
5
1
2
.
7
7
24
4
.
4
30
19
18
8
.
5
8
2
.
8
3
25
4
.
4
30
19
12
8
.
8
8
2
.
7
5
26
4
.
4
30
19
12
9
.
0
9
2
.
8
3
27
4
.
4
30
19
12
8
.
9
2
2
.
7
9
28
4
.
4
30
19
12
8
.
9
1
2
.
7
5
29
4
.
4
30
19
12
8
.
9
2
2
.
8
3
30
4
.
4
30
19
12
9
.
0
2
2
.
8
1
31
4
.
4
30
19
12
8
.
8
2
.
8
32
4
21
17
12
8
.
7
9
8
3
.
3
4
6
33
4
27
17
12
8
.
8
9
9
2
.
9
6
1
34
4
30
17
12
7
.
9
5
4
2
.
8
5
4
35
4
36
17
12
7
.
2
4
9
2
.
6
6
2
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2088
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
4
,
A
u
g
u
s
t 2
0
1
8
:
2
3
1
0
–
2
3
1
8
2316
Ex
p
e
r
i
me
n
t
No
W
i
r
e
f
e
e
d
r
a
t
e
(
m/
mi
n
)
W
e
l
d
i
n
g
sp
e
e
d
(
c
m/
mi
n
)
A
r
c
v
o
l
t
a
g
e
(
V
)
N
o
z
z
l
e
t
o
p
l
a
t
e
d
i
s
t
a
n
c
e
(
mm
)
B
e
a
d
w
i
d
t
h
(
mm
)
B
e
a
d
h
e
i
g
h
t
(
mm
)
36
4
39
17
12
7
.
1
9
3
2
.
5
3
3
37
4
27
1
8
.
9
12
1
0
.
0
0
2
3
.
2
1
8
38
5
.
2
30
1
8
.
9
12
9
.
1
1
6
3
.
0
4
7
39
6
27
2
0
.
3
12
1
1
.
2
3
3
3
.
3
0
4
40
5
.
2
2
2
.
5
19
12
1
0
.
6
1
3
.
4
1
1
41
5
.
2
3
7
.
5
19
12
8
.
4
9
4
2
.
7
9
42
4
.
4
3
7
.
5
1
7
.
5
12
7
.
7
8
8
2
.
8
1
1
43
6
3
7
.
5
2
0
.
5
12
9
.
8
4
9
2
.
8
7
6
T
ab
le
3
.
T
h
e
Sin
g
le
Ker
n
el
an
d
Mu
ltip
le
Ker
n
el
S
VM
P
ar
am
eter
s
P
a
r
a
me
t
e
r
M
e
t
h
o
d
B
e
a
d
w
i
d
t
h
B
e
a
d
h
e
i
g
h
t
k
e
r
n
e
l
p
a
r
a
me
t
e
r
(
)
si
n
g
l
e
k
e
r
n
e
l
1
0
.
2
r
e
g
u
l
a
r
i
z
a
t
i
o
n
f
a
c
t
o
r
(
C
)
mu
l
t
i
p
l
e
k
e
r
n
e
l
si
n
g
l
e
k
e
r
n
e
l
7
1
0
0
10
1
0
0
2
5
0
0
0
i
n
se
n
s
i
t
i
v
i
t
y
p
a
r
a
me
t
e
r
(
)
mu
l
t
i
p
l
e
k
e
r
n
e
l
si
n
g
l
e
k
e
r
n
e
l
0
.
0
0
9
10
-
7
0
.
1
10
-
7
A
cc
u
r
ac
y
o
f
t
h
e
f
i
n
al
m
o
d
e
ls
w
a
s
ev
al
u
ated
b
ased
o
n
th
e
r
o
o
t
m
ea
n
s
s
q
u
ar
e
er
r
o
r
(
R
MSE
)
,
n
o
r
m
alize
d
r
o
o
t
m
ea
n
s
s
q
u
a
r
e
er
r
o
r
(
NR
MSE
)
an
d
m
ea
n
ab
s
o
lu
te
p
er
ce
n
ta
g
e
er
r
o
r
(
MA
P
E
)
s
tatis
t
ical
in
d
ices,
d
ef
i
n
ed
as
:
√
∑
(
̂
)
(
1
9
)
(
2
0
)
∑
|
̂
|
(
2
1
)
I
n
th
e
s
e
eq
u
atio
n
s
,
an
d
̂
ar
e
th
e
co
r
r
esp
o
n
d
in
g
m
ea
s
u
r
ed
an
d
th
e
p
r
ed
icted
o
u
tp
u
ts
,
r
esp
ec
t
iv
el
y
,
N
is
t
h
e
co
r
r
esp
o
n
d
in
g
n
u
m
b
er
o
f
tr
ain
i
n
g
o
r
te
s
tin
g
s
a
m
p
l
es
an
d
̅
is
th
e
m
ea
n
v
alu
e
o
f
t
h
e
to
tal
m
ea
s
u
r
ed
o
u
tp
u
ts
.
T
h
e
ca
lcu
la
ted
v
al
u
es
o
f
th
e
i
n
d
ices
ar
e
lis
ted
in
T
ab
le
4
.
B
esid
es
th
e
s
u
p
er
io
r
p
e
r
f
o
r
m
an
ce
o
f
MK
-
SVR
o
v
er
th
e
SK
-
S
VR
,
th
i
s
m
et
h
o
d
h
as
a
b
etter
test
in
g
m
ea
n
ab
s
o
l
u
te
p
er
ce
n
tag
e
er
r
o
r
(
MA
P
E
)
th
an
th
e
A
N
N
-
b
ased
ap
p
r
o
ac
h
p
r
o
p
o
s
e
d
b
y
[
7
]
w
h
ich
h
as r
ep
o
r
ted
a
MA
P
E
o
f
2
.
0
1
3
% f
o
r
th
e
test
d
ata.
T
h
e
SK
-
SV
R
te
s
ti
n
g
r
o
o
t
m
ea
n
s
s
q
u
ar
e
er
r
o
r
ca
n
b
e
f
u
r
th
er
r
ed
u
ce
d
to
0
.
1
4
9
3
b
y
ch
a
n
g
i
n
g
t
h
e
SVM
k
er
n
el
a
n
d
m
o
d
el
p
ar
am
e
ter
s
,
b
u
t
th
is
v
al
u
e
f
o
r
k
er
n
e
l
p
ar
a
m
eter
ca
n
n
o
t
b
e
o
b
tain
ed
f
r
o
m
th
e
tr
ai
n
i
n
g
d
atab
ase.
I
n
o
th
er
w
o
r
d
s
,
t
h
e
SK
-
SV
R
m
o
d
el
i
s
o
v
er
lear
n
ed
in
tr
ain
i
n
g
p
r
o
ce
s
s
a
n
d
ca
n
n
o
t
b
e
tr
ain
ed
to
m
ak
e
th
e
b
est
p
r
ed
ictio
n
s
f
o
r
t
h
e
tes
t
d
ata
b
esid
es
th
e
tr
ain
i
n
g
d
ata.
T
h
er
ef
o
r
e
,
th
e
MK
-
SVR
m
eth
o
d
b
en
ef
i
ts
f
r
o
m
b
etter
g
en
er
aliza
tio
n
ca
p
ab
ilit
y
a
s
w
e
ll
as
h
i
g
h
er
p
r
ec
is
io
n
f
o
r
th
e
test
d
ata.
T
h
e
m
ea
s
u
r
ed
o
u
tp
u
ts
to
g
et
h
er
w
it
h
t
h
e
o
u
tp
u
t
s
p
r
ed
icted
b
y
th
e
MK
-
S
VM
m
eth
o
d
ar
e
d
ep
icted
in
Fi
g
u
r
e
6
a
n
d
a
g
o
o
d
ag
r
ee
m
e
n
t
ca
n
b
e
o
b
s
er
v
ed
b
et
w
ee
n
t
h
e
m
.
T
ab
le
4
.
C
alcu
lated
Valu
es o
f
th
e
Stati
s
tical
I
n
d
ices f
o
r
th
e
T
r
ain
in
g
a
n
d
T
est Da
ta
D
a
t
a
b
a
se
M
e
t
h
o
d
R
M
S
E
N
R
M
S
E
M
A
P
E
T
r
a
i
n
i
n
g
MK
-
S
V
M
0
.
1
4
7
8
0
.
0
2
4
9
0
.
9
2
%
SK
-
S
V
M
0
.
0
3
1
6
0
.
0
0
5
3
0
.
1
8
%
T
e
st
i
n
g
MK
-
S
V
M
0
.
1
5
0
7
0
.
0
2
5
4
1
.
7
7
%
SK
-
S
V
M
0
.
3
2
9
1
0
.
0
5
5
5
5
.
5
2
%
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
A
p
p
lica
tio
n
o
f Mu
ltip
le
K
ern
el
S
u
p
p
o
r
t V
ec
to
r
R
eg
r
ess
io
n
fo
r
W
e
ld
B
ea
d
Geo
metry
…
(
N
a
d
er Mo
lla
yi
)
2317
Fig
u
r
e
6
.
P
r
ed
icte
d
an
d
m
ea
s
u
r
ed
v
alu
es
f
o
r
A
)
T
h
e
b
ea
d
h
eig
h
t B
)
T
h
e
b
ea
d
w
id
t
h
5.
CO
NCLU
SI
O
N
I
n
t
h
is
p
ap
er
,
ap
p
licatio
n
o
f
m
u
ltip
le
k
er
n
el
SVM
r
e
g
r
e
s
s
io
n
an
a
l
y
s
is
h
a
s
b
ee
n
p
r
o
p
o
s
ed
f
o
r
m
o
d
ell
in
g
a
n
d
p
r
ed
ictio
n
o
f
t
h
e
w
eld
g
eo
m
e
tr
y
in
r
o
b
o
tic
GM
A
W
-
b
ased
r
ap
id
m
a
n
u
f
ac
t
u
r
in
g
p
r
o
ce
s
s
b
ased
o
n
th
e
i
n
p
u
t
p
ar
a
m
eter
s
o
f
wir
e
f
ee
d
r
ate,
w
e
ld
in
g
s
p
ee
d
,
ar
c
v
o
ltag
e
a
n
d
n
o
zz
le
to
p
la
te
d
is
tan
ce
.
I
n
th
is
an
al
y
s
is
,
t
h
e
k
er
n
el
f
u
n
ctio
n
i
s
f
o
r
m
ed
b
ased
o
n
a
l
in
ea
r
co
m
b
in
at
io
n
o
f
b
asi
s
k
er
n
e
l
f
u
n
ctio
n
s
a
n
d
u
s
i
n
g
t
h
e
Si
m
p
le
MK
L
al
g
o
r
ith
m
,
t
h
e
o
p
tim
ized
co
m
b
i
n
atio
n
o
f
th
e
k
er
n
e
l
f
u
n
ctio
n
a
n
d
t
h
e
s
o
l
u
tio
n
s
o
f
t
h
e
S
V
R
p
r
o
b
lem
ar
e
o
b
tain
ed
.
B
ased
o
n
th
e
r
esu
lts
,
it
h
a
s
b
ee
n
co
n
clu
d
ed
th
at
th
e
b
est
p
r
ed
ictio
n
r
esu
lt
s
ca
n
n
o
t
b
e
o
b
tain
ed
in
s
i
n
g
le
k
er
n
el
SV
R
f
o
r
b
o
th
t
h
e
tr
ai
n
i
n
g
a
n
d
te
s
t
d
ata,
w
h
ile
ap
p
licatio
n
o
f
m
u
ltip
le
k
er
n
el
SV
R
r
esu
lt
s
i
n
th
e
b
est
r
e
s
u
l
ts
f
o
r
b
o
th
o
f
th
e
s
e
d
atab
ase
s
.
P
r
ed
ictio
n
r
es
u
lt
s
a
ls
o
p
r
o
v
e
h
i
g
h
er
ac
cu
r
ac
y
o
f
t
h
e
m
u
ltip
le
k
er
n
el
SV
R
b
esid
es
its
e
n
h
a
n
ce
d
g
e
n
er
aliza
tio
n
c
ap
ab
ilit
y
o
v
er
t
h
e
s
i
n
g
le
k
er
n
el
SVM
a
n
d
A
NN
r
eg
r
ess
io
n
ap
p
r
o
ac
h
es.
B
ased
o
n
th
e
m
u
ltip
le
k
er
n
el
SV
M
m
o
d
els,
th
e
i
n
p
u
t
p
ar
a
m
et
er
s
ca
n
b
e
tu
n
ed
to
o
b
tain
a
d
esire
d
w
eld
g
eo
m
e
tr
y
in
t
h
i
s
m
a
n
u
f
ac
t
u
r
i
n
g
p
r
o
ce
s
s
w
ith
a
h
i
g
h
er
d
eg
r
ee
o
f
ac
c
u
r
ac
y
.
RE
F
E
R
E
NC
E
S
[1
]
S
.
H
.
M
a
so
o
d
,
"
In
tr
o
d
u
c
ti
o
n
t
o
a
d
v
a
n
c
e
s
in
a
d
d
it
iv
e
m
a
n
u
f
a
c
tu
rin
g
a
n
d
to
o
li
n
g
"
,
Co
m
p
re
h
e
n
siv
e
M
a
ter
ia
ls
Pro
c
e
ss
in
g
,
v
o
l
.
10
,
p
p
.
1
-
2
,
F
e
b
2
0
1
4
.
[2
]
G
.
Co
lo
m
b
o
,
e
t
a
l
.
,
"
Re
v
e
rse
e
n
g
i
n
e
e
rin
g
a
n
d
ra
p
id
p
ro
t
o
ty
p
in
g
tec
h
n
i
q
u
e
s to
in
n
o
v
a
te p
ro
sth
e
sis so
c
k
e
t
d
e
si
g
n
"
,
i
n
T
h
re
e
-
Dime
n
sio
n
a
l
Ima
g
e
Ca
p
tu
r
e
a
n
d
Ap
p
li
c
a
ti
o
n
s
,
p
.
6
0
5
6
0
P
,
2
0
06
.
[3
]
D.
Zh
a
n
g
,
e
t
a
l
.
,
"
Ra
p
i
d
p
r
o
to
ty
p
in
g
o
f
m
e
tal
p
a
rts
b
y
th
re
e
-
d
i
m
e
n
s
io
n
a
l
w
e
ld
in
g
"
,
Pro
c
e
e
d
in
g
s
o
f
th
e
In
stit
u
t
io
n
o
f
M
e
c
h
a
n
ica
l
E
n
g
in
e
e
rs
,
v
o
l.
2
1
2
,
n
o
.
3
,
p
p
.
1
7
5
-
1
8
2
,
1
9
9
8
.
[4
]
J
.
Xio
n
g
,
e
t
a
l
.
,
"
Clo
se
d
-
lo
o
p
c
o
n
tr
o
l
o
f
v
a
riab
le
lay
e
r
w
id
th
f
o
r
t
h
in
-
w
a
ll
e
d
p
a
rts
in
w
ire
a
n
d
a
rc
a
d
d
it
iv
e
m
a
n
u
f
a
c
tu
rin
g
"
,
J
o
u
rn
a
l
o
f
M
a
ter
i
a
ls
Pro
c
e
ss
in
g
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
2
3
3
,
p
p
.
1
0
0
-
1
0
6
,
2
0
1
6
.
[5
]
T
.
Ha
stie,
e
t
a
l
.
,
"
Ov
e
rv
ie
w
o
f
su
p
e
rv
ise
d
lea
rn
in
g
,
"
T
h
e
e
lem
e
n
ts
o
f
sta
ti
stica
l
lea
rn
i
n
g
,
p
p
.
9
-
41
,
2
0
09
.
[6
]
K.
P
a
l
a
n
d
S
.
K.
P
a
l,
"
S
o
f
t
c
o
m
p
u
ti
n
g
m
e
th
o
d
s
u
se
d
f
o
r
t
h
e
m
o
d
e
ll
in
g
a
n
d
o
p
ti
m
isa
ti
o
n
o
f
Ga
s
M
e
t
a
l
A
r
c
Weld
in
g
:
a
re
v
ie
w"
,
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
M
a
n
u
f
a
c
tu
rin
g
Res
e
a
rc
h
,
v
o
l.
6
,
No
.
1
,
p
p
.
1
5
-
2
9
,
2
0
1
1
.
[7
]
J.
X
io
n
g
,
e
t
a
l
.
,
"
Be
a
d
g
e
o
m
e
tr
y
p
re
d
ictio
n
f
o
r
ro
b
o
ti
c
G
M
AW
-
b
a
se
d
ra
p
id
m
a
n
u
f
a
c
tu
rin
g
t
h
ro
u
g
h
a
n
e
u
ra
l
n
e
tw
o
rk
a
n
d
a
s
e
c
o
n
d
-
o
rd
e
r
re
g
re
s
sio
n
a
n
a
ly
sis
"
,
J
o
u
rn
a
l
o
f
In
telli
g
e
n
t
M
a
n
u
fa
c
t
u
rin
g
,
v
o
l.
2
5
,
n
o
.
1
,
p
p
.
1
5
7
-
1
6
3
,
2
0
1
4
.
[8
]
L
.
W
a
n
g
,
"
S
u
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
s: t
h
e
o
ry
a
n
d
a
p
p
li
c
a
ti
o
n
s
"
,
S
p
r
in
g
e
r S
c
ien
c
e
&
Bu
sin
e
ss
M
e
d
ia
,
v
o
l.
1
7
7
,
2
0
05
.
[9
]
D
.
M
e
y
e
r,
e
t
a
l
.
, "
T
h
e
su
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
u
n
d
e
r
tes
t
"
,
Ne
u
r
o
c
o
mp
u
t
in
g
,
v
o
l
.
55
,
n
o
.
1
,
p
p
.
1
6
9
-
186
,
2
0
0
3
.
[1
0
]
S
.
S
o
n
n
e
n
b
u
rg
,
e
t
a
l
.
,
"
L
a
rg
e
s
c
a
le
m
u
lt
ip
le
k
e
rn
e
l
lea
rn
in
g
"
,
J
o
u
rn
a
l
o
f
M
a
c
h
i
n
e
L
e
a
r
n
in
g
Re
se
a
rc
h
,
v
o
l.
7
,
p
p
.
1
5
3
1
-
1
5
6
5
,
2
0
0
6
.
[1
1
]
C
.
Co
rtes
,
e
t
a
l
.
"
S
u
p
p
o
rt
-
v
e
c
to
r
n
e
tw
o
rk
s
"
,
M
a
c
h
in
e
lea
rn
in
g
,
v
o
l
.
20
, n
o
.
3
,
p
p
.
2
7
3
-
2
9
7
,
1
9
9
5
.
[1
2
]
S
.
P
ra
sa
d
,
e
t
a
l
.
,
"
Co
m
p
a
riso
n
o
f
A
c
c
u
ra
c
y
M
e
a
su
re
s
f
o
r
R
S
Im
a
g
e
Cla
ss
i
f
ica
ti
o
n
u
sin
g
S
V
M
a
n
d
A
NN
Clas
sif
ier
s
"
,
In
ter
n
a
ti
o
n
a
l
J
o
u
r
n
a
l
o
f
E
lec
trica
l
a
n
d
Co
mp
u
ter
En
g
i
n
e
e
rin
g
,
v
o
l.
7
,
n
o
.
3
,
p
p
.
1
1
8
0
-
1
1
8
7
,
2
0
17
.
[1
3
]
G
.
T
.
N
g
o
,
e
t
a
l
.
, "
I
m
a
g
e
R
e
tri
e
v
a
l
w
it
h
Re
le
v
a
n
c
e
F
e
e
d
b
a
c
k
u
sin
g
S
V
M
A
c
ti
v
e
L
e
a
rn
in
g
"
,
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
El
e
c
trica
l
a
n
d
C
o
mp
u
ter
En
g
in
e
e
rin
g
,
v
o
l.
6
,
n
o
.
6
,
p
.
3
2
3
8
,
2
0
16
.
[1
4
]
N.
Cristi
a
n
in
i
a
n
d
J.
S
h
a
w
e
-
T
a
y
l
o
r,
"
A
n
in
tro
d
u
c
ti
o
n
to
s
u
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
s
a
n
d
o
th
e
r
k
e
rn
e
l
-
b
a
se
d
lea
rn
in
g
m
e
th
o
d
s
"
,
Ca
m
b
rid
g
e
u
n
iv
e
rsity
p
re
ss
,
2
0
0
0
.
[1
5
]
F
.
P
a
rre
ll
a
,
"
On
li
n
e
su
p
p
o
rt
v
e
c
to
r
re
g
re
ss
io
n
"
,
M
a
ste
r'
s
T
h
e
sis
,
De
p
a
rtme
n
t
o
f
In
f
o
r
m
a
ti
o
n
S
c
ien
c
e
,
Un
iv
e
rsit
y
o
f
G
e
n
o
a
,
Ital
y
,
2
0
0
7
.
[1
6
]
L
.
G
e
n
g
,
e
t
a
l
.
,
"
F
o
re
c
a
stin
g
Ra
n
g
e
V
o
latil
i
ty
U
sin
g
S
u
p
p
o
rt
V
e
c
to
r
M
a
c
h
in
e
s
w
it
h
Im
p
ro
v
e
d
P
S
O
A
lg
o
rit
h
m
s
"
,
T
EL
KOM
NIKA
(
T
e
lec
o
mm
u
n
ica
t
io
n
C
o
mp
u
ti
n
g
E
lec
tro
n
ics
a
n
d
C
o
n
tro
l)
,
v
o
l
.
1
3
,
n
o
.
3
A
,
p
p
.
2
0
8
-
2
1
6
,
2
0
1
6
.
[1
7
]
W
.
Bu
d
ih
a
rto
,
A
.
S
a
n
to
so
,
D.
P
u
rw
a
n
to
,
A
.
Ja
z
id
e
,
“
M
u
lt
ip
le
m
o
v
in
g
o
b
sta
c
les
a
v
o
id
a
n
c
e
o
f
se
r
v
ice
ro
b
o
t
u
si
n
g
ste
re
o
v
isio
n
”
,
T
EL
KOM
NIKA
(
T
e
lec
o
mm
u
n
ica
ti
o
n
C
o
mp
u
ti
n
g
El
e
c
tro
n
ics
a
n
d
Co
n
tro
l)
,
2
0
1
1
,
v
o
l.
9
,
n
o
.
3
,
p
p
.
4
3
3
-
4
4
4
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2088
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
4
,
A
u
g
u
s
t 2
0
1
8
:
2
3
1
0
–
2
3
1
8
2318
[1
8
]
C
.
Y
.
Ye
h
,
e
t
a
l
.
,
"
A
m
u
lt
ip
le
-
k
e
r
n
e
l
su
p
p
o
rt
v
e
c
to
r
re
g
re
ss
io
n
a
p
p
ro
a
c
h
f
o
r
sto
c
k
m
a
rk
e
t
p
rice
f
o
re
c
a
stin
g
"
,
Exp
e
rt
S
y
ste
ms
wit
h
Ap
p
li
c
a
ti
o
n
s
,
v
o
l
.
38
,
n
o
.
3
,
p
p
.
2
1
7
7
-
2
1
8
6
,
2
0
1
1
.
[1
9
]
A
.
Ra
k
o
to
m
a
m
o
n
j
y
,
e
t
a
l
.
,
"
S
im
p
le M
KL
"
,
J
o
u
rn
a
l
o
f
M
a
c
h
i
n
e
L
e
a
rn
in
g
Res
e
a
rc
h
,
v
o
l
.
9
,
p
p
.
2
4
9
1
-
2
5
2
1
,
2
0
0
8
.
[2
0
]
S
.
Ca
n
u
,
"
S
v
m
a
n
d
k
e
rn
e
l
m
e
th
o
d
s
m
a
tl
a
b
to
o
lb
o
x
"
,
h
tt
p
://
a
si.
in
s
a
-
ro
u
e
n
.
fr/en
se
ig
n
a
n
ts/~
a
ra
k
o
to
/t
o
o
l
b
o
x
/
in
d
e
x
.
h
tml
,
2
0
0
5.
B
I
O
G
RAP
H
I
E
S
O
F
AUTH
O
RS
Na
d
e
r
M
o
ll
a
y
i
w
a
s
b
o
rn
in
M
a
sh
h
a
d
,
Ira
n
in
1
9
8
5
.
He
re
c
e
i
v
e
d
h
is
B.
S
c
.
d
e
g
re
e
in
e
lec
tri
c
a
l
e
n
g
in
e
e
rin
g
f
ro
m
S
h
a
h
ro
o
d
Un
iv
e
rsit
y
o
f
Tec
h
n
o
lo
g
y
,
S
h
a
h
ro
o
d
,
I
ra
n
a
n
d
h
is
M
.
S
c
.
d
e
g
re
e
in
p
o
w
e
r
e
lec
tro
n
ics
f
ro
m
S
h
a
rif
Un
iv
e
rsit
y
o
f
T
e
c
h
n
o
lo
g
y
,
T
e
h
ra
n
,
Ira
n
in
2
0
0
7
a
n
d
2
0
1
0
,
re
sp
e
c
ti
v
e
ly
.
S
in
c
e
S
e
p
tem
b
e
r
2
0
1
0
,
h
e
h
a
s
b
e
e
n
a
l
ec
tu
r
er
in
Birj
a
n
d
Un
iv
e
rsity
o
f
T
e
c
h
n
o
lo
g
y
,
Birj
a
n
d
,
Ira
n
.
His
re
se
a
rc
h
in
tere
sts
in
c
l
u
d
e
p
o
w
e
r
e
lec
tro
n
ics
,
p
o
w
e
r
q
u
a
li
ty
,
d
ig
it
a
l
sig
n
a
l
p
ro
c
e
ss
in
g
,
p
a
tt
e
r
n
re
c
o
g
n
it
io
n
a
n
d
m
a
c
h
in
e
lea
rn
in
g
.
M
o
h
a
m
m
a
d
J
a
v
a
d
Eid
i
w
a
s
b
o
rn
i
n
Ira
n
in
1
9
9
2
in
F
a
rs,
Ira
n
.
He
re
c
e
iv
e
d
h
is
B.
S
c
.
d
e
g
re
e
in
M
e
c
h
a
n
ica
l
En
g
in
e
e
rin
g
i
n
2
0
1
4
.
S
in
c
e
S
e
p
tem
b
e
r
2
0
1
5
h
e
is
a
M
.
S
c
stu
d
e
n
t
in
Bi
rjan
d
u
n
iv
e
rsity
o
f
tec
h
n
o
l
o
g
y
.
His
f
ield
s o
f
in
tere
st
a
re
in
d
u
strial
d
e
sig
n
,
a
d
d
it
iv
e
m
a
n
u
f
a
c
tu
rin
g
a
n
d
ra
p
id
p
ro
to
ty
p
in
g
.
Evaluation Warning : The document was created with Spire.PDF for Python.