T
E
L
KO
M
NI
K
A
,
V
ol
.
14,
N
o.
3,
S
ept
em
ber
20
16,
pp.
9
4
8~
9
55
I
S
S
N
:
1
693
-
6
930
,
ac
c
r
edi
t
ed
A
b
y
D
IK
T
I,
D
e
c
r
e
e
N
o
:
58/
D
I
K
T
I
/
K
ep/
2013
D
O
I
:
10.
12928/
T
E
LK
O
M
N
I
K
A
.
v
1
4
i
3
.
3589
94
8
R
ec
ei
v
ed
M
a
y 2
,
2
01
6
;
R
e
v
i
s
ed
J
ul
6
,
2
01
6
;
A
c
c
ep
t
ed
J
ul
y
20
,
20
1
6
A
p
p
lic
at
i
o
n
o
f
N
o
n
lin
ear
D
y
n
am
ical M
et
h
o
d
s f
o
r
Ar
c
W
eld
in
g
Q
u
alit
y
M
oni
t
or
i
n
g
S
h
u
g
u
a
n
g
W
u
*
,
Y
i
q
i
n
g
Z
h
o
u
D
epar
t
m
ent
of
E
l
e
c
t
r
o
ni
c
s
and
I
nf
or
m
at
i
on T
ec
h
nol
ogy
,
J
i
an
gm
en
P
ol
y
t
e
c
hni
c
,
J
i
an
gm
e
n 5
2909
0,
G
uan
gdo
n
g,
C
hi
n
a
*
C
or
r
es
po
ndi
ng a
ut
hor
,
e
-
m
a
i
l
:
5683
4186
@
qq.
c
om
A
b
st
r
act
O
w
i
ng t
o
i
t
s
d
i
v
er
s
e
,
t
h
e
s
t
ab
i
l
i
t
y
o
f
ar
c
s
i
g
nal
s
i
n
hi
g
h
-
po
w
er
ed
s
ubm
er
g
ed
ar
c
w
el
d
i
n
g i
s
n
o
t
v
er
y
s
al
i
ent
,
and w
el
d d
ef
e
c
t
s
ar
e d
i
f
f
i
c
ul
t
t
o d
et
ec
t
aut
o
m
at
i
c
al
l
y
.
A
i
m
ed at
t
hi
s
pr
ob
l
em
,
t
hi
s
pap
er
pr
opo
s
es
a no
i
s
e r
ob
us
t
nes
s
al
gor
i
t
hm
f
or
c
al
i
br
at
i
ng
t
he
s
i
ngul
ar
i
t
y
poi
nt
s
and
den
ot
i
n
g
t
he k
i
net
i
c
s
a
nd
s
t
a
b
ilit
y
of
ar
c
.
F
i
r
s
t
l
y
,
r
ec
ons
t
r
uc
t
a
v
e
c
t
or
,
w
hi
c
h
i
s
t
h
e
c
al
c
ul
a
t
i
on
o
f
t
he
ap
pr
o
x
i
m
at
e
e
nt
r
op
y
i
n
ph
a
s
e
s
pa
c
e,
den
ot
e
s
t
he di
s
t
or
t
i
on
of
ar
c
.
T
he
n,
a al
gor
i
t
hm
f
or
c
al
c
u
l
at
i
on i
s
gi
v
e
n bas
ed on r
ec
on
s
t
r
u
c
t
i
on of
c
hao
t
i
c
t
i
m
e
s
er
i
e
s
i
n
pha
s
e
s
pac
e
.
F
i
n
al
l
y
,
w
e
appl
y
t
he
c
a
l
c
ul
at
i
on
of
appr
ox
i
m
at
e
ent
r
o
py
al
gor
i
t
hm
i
n
phas
e s
pac
e t
o
f
l
aw
d
et
e
c
t
i
o
n
f
or
ar
c
s
i
gna
l
s
,
w
hi
c
h i
s
ef
f
i
c
i
e
nt
pr
o
v
ed
by
ex
p
er
i
m
ent
al
r
e
s
ul
t
s
.
Ke
y
w
o
rd
s
:
w
el
d
d
ef
ec
t
s
,
s
i
n
g
ul
ar
i
t
y
p
oi
nt
s
,
p
ha
s
e
s
pa
c
e,
ap
pr
ox
i
m
at
e
e
nt
r
o
py
C
o
p
y
r
i
g
h
t
©
20
16 U
n
i
ver
si
t
a
s A
h
mad
D
ah
l
an
.
A
l
l
r
i
g
h
t
s
r
eser
ved
.
1
.
I
n
tr
o
d
u
c
ti
o
n
W
i
t
h t
he de
v
e
l
o
pm
ent
and a
ppl
i
c
at
i
on
of
aut
om
at
i
o
n t
ec
h
no
l
og
y
a
nd c
o
m
put
er
t
ec
hno
l
og
y
,
t
h
e di
g
i
t
i
z
at
i
on
and i
n
t
e
l
l
i
ge
nt
of
w
el
di
ng h
as
bec
om
e a
f
oc
us
of
t
he dev
e
l
o
pm
ent
o
f
w
el
d
i
n
g t
ec
h
nol
og
y
n
o
w
a
d
a
y
s
.
W
el
di
ng pr
oc
es
s
i
s
a
c
o
m
pl
ex
ph
y
s
i
c
a
l
an
d c
he
m
i
c
al
pr
oc
es
s
w
i
t
h
v
ar
i
ous
i
n
t
er
f
er
enc
e,
and t
h
e s
t
ab
i
l
i
t
y
of
w
e
l
d
i
n
g qua
l
i
t
y
c
ont
r
ol
i
s
a m
or
e c
om
pl
ex
and
di
f
f
i
c
ul
t
no
nl
i
n
ear
pr
obl
e
m
[1
-
3
].
E
l
ec
t
r
i
c
al
s
i
g
nal
s
,
s
uc
h as
w
e
l
d
i
ng
c
ur
r
e
nt
a
nd
w
el
di
ng
v
o
l
t
ag
e,
c
ont
ai
ns
ab
und
ant
i
nf
or
m
at
i
on of
w
e
l
d
i
n
g
pr
oc
es
s
.
I
f
w
e
c
an
ex
t
r
ac
t
t
he
f
eat
ur
e
i
nf
or
m
at
i
on
ac
c
ur
at
el
y
,
a
nd
t
hen
us
i
ng
t
h
e
s
c
i
ent
i
f
i
c
m
e
t
hod
t
o
u
nder
t
ak
e
s
t
at
i
s
t
i
c
a
l
ana
l
y
s
i
s
and
c
al
c
ul
a
t
i
o
n,
w
i
l
l
s
ur
el
y
c
an
be bet
t
er
s
i
g
ht
i
nt
o t
h
e c
h
ar
ac
t
er
i
s
t
i
c
s
of
t
he m
ec
hani
s
m
o
f
w
el
di
ng
pr
oc
es
s
[
4
,
5
].
Ma
n
y
s
c
ho
l
ar
s
hav
e
s
t
u
di
e
d
i
n
t
hi
s
r
e
s
pec
t
:
t
o
r
educ
e
t
h
e
s
ubj
ec
t
i
v
i
t
y
of
doub
l
e
w
i
r
e
pul
s
e
d
w
el
di
n
g
w
el
di
ng pr
oc
es
s
s
t
abi
l
i
t
y
e
v
al
u
at
i
o
n.
C
ons
i
der
i
ng
w
el
d s
t
r
engt
h
as
t
he
qua
l
i
t
y
c
h
ar
ac
t
er
i
s
t
i
c
i
n t
he
s
el
ec
t
i
on
of
pr
oc
es
s
par
a
m
et
er
s
,
f
u
m
e
f
or
m
at
i
on i
n a
p
ul
s
ed
g
as
m
et
al
ar
c
w
el
di
ng (
G
MA
W
)
pr
oc
es
s
i
s
i
nv
es
t
i
g
at
ed
b
y
c
oupl
i
ng
a t
i
m
e
-
depe
nde
nt
ax
i
-
s
ym
m
e
t
r
i
c
tw
o
-
di
m
ens
i
on
al
m
odel
,
w
hi
c
h t
ak
es
i
nt
o ac
c
ou
nt
bo
t
h dr
op
l
et
d
et
ac
hm
ent
and
pr
oduc
t
i
on of
m
et
al
v
apo
ur
,
w
i
t
h
a
m
odel
f
or
f
u
m
e
f
or
m
at
i
on
an
d
t
r
a
ns
por
t
bas
e
d
on
t
he
m
et
ho
d
of
m
o
m
ent
s
f
or
t
he s
ol
ut
i
on of
t
he aer
os
ol
gen
er
al
d
y
n
am
i
c
equat
i
o
n
[
6
]
.
T
hr
ee geom
et
r
y
c
hanges
t
o t
he
i
nn
er
bor
e of
a
w
e
l
di
ng
no
z
z
l
e
an
d t
hei
r
ef
f
ec
t
s
on
w
el
d
qu
al
i
t
y
d
ur
i
n
g g
as
m
et
al
ar
c
w
el
di
ng
(
G
MA
W
)
w
er
e
i
n
v
es
t
i
ga
t
ed
t
hr
ou
gh
t
he
us
e
o
f
c
om
put
at
i
ona
l
f
l
u
i
d
d
y
nam
i
c
(
C
F
D
)
m
odel
s
an
d
ex
per
i
m
ent
al
t
r
i
a
l
s
[7
].
A
n
ada
pt
i
v
e
t
ec
h
ni
q
ue
bas
e
d on
es
t
i
m
at
i
on
of
s
i
gn
al
par
am
et
er
s
v
i
a
r
ot
at
i
ona
l
i
nv
ar
i
anc
e t
ec
h
ni
q
ue (
E
S
P
R
I
T
)
i
s
pr
opos
ed t
hat
o
pt
i
m
i
z
es
t
h
e ac
c
ur
ac
y
and
c
o
m
put
at
i
on t
i
m
e f
or
har
m
oni
c
/
i
n
t
er
har
m
oni
c
es
t
i
m
at
i
on of
s
t
at
i
onar
y
as
w
e
l
l
as
nons
t
at
i
onar
y
p
ow
e
r
s
up
pl
y
s
i
gn
al
s
[8
].
T
he us
e
of
f
uz
z
y
r
u
l
e
b
as
ed s
y
s
t
em
s
t
o m
odel
t
he r
e
l
at
i
ons
hi
p
bet
w
e
en
w
e
l
d c
ont
r
o
l
par
am
et
er
s
and t
he w
e
l
d be
ad geom
et
r
y
f
eat
ur
es
i
s
ex
pl
or
ed i
n t
hi
s
paper
.
T
he
s
y
s
t
em
i
s
t
es
t
e
d
o
n
t
hr
ee
dat
as
et
s
a
nd
t
h
e
p
er
f
or
m
anc
e
i
s
f
ound
t
o
be
s
at
i
s
f
ac
t
or
y
c
o
m
par
ed t
o t
he m
ul
t
i
l
a
y
er
per
c
ept
r
on (
ML
P
)
and r
ad
i
al
b
as
i
s
f
unc
t
i
on (
R
B
F
)
ne
ur
al
net
w
or
k
s
bas
ed s
y
s
t
em
s
[
9]
.
S
i
m
ps
on S
W
el
uc
i
dat
es
t
he s
i
gnat
ur
e i
m
age ap
pr
oac
h t
o
w
el
di
ng f
aul
t
det
ec
t
i
on,
c
o
v
er
i
ng t
he c
a
l
c
ul
a
t
i
o
n of
s
i
gna
t
ur
e i
m
age dat
a obj
ec
t
s
f
r
o
m
bl
oc
k
s
of
w
el
d
i
ng
el
ec
t
r
i
c
a
l
dat
a (
v
ol
t
ag
e an
d c
ur
r
ent
)
,
t
he def
i
ni
t
i
o
n of
appr
opr
i
at
e v
ec
t
or
oper
a
t
i
ons
,
an
d t
he
m
ani
pul
at
i
o
n of
t
he s
i
gn
at
u
r
es
t
o per
m
i
t
det
ec
t
i
o
n of
w
el
d
i
ng
f
aul
t
s
[
10
].
T
he abov
e an
al
y
s
i
s
of
t
he
s
t
at
us
of
t
he s
y
s
t
em
i
s
not
c
om
pr
ehens
i
v
e,
t
h
e ex
t
r
ac
t
ed
c
har
ac
t
er
i
s
t
i
c
i
nf
or
m
at
i
on i
s
m
or
e
r
el
y
on
t
he
ex
per
i
e
nc
e of
t
he
peo
pl
e,
an
d
o
v
er
em
phas
i
z
e
on
t
he i
nde
pen
denc
e of
t
he s
t
at
e.
L
o
ss t
i
m
e
-
v
ar
y
i
ng c
h
ar
ac
t
er
i
s
t
i
c
s
of
t
he s
i
g
na
l
,
and
i
gn
or
e t
h
e
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
K
A
I
S
S
N
:
1
693
-
6
930
A
pp
l
i
c
at
i
o
n of
N
on
l
i
near
D
y
nam
i
c
al
Met
hods
f
or
A
r
c
W
el
d
i
ng
Q
u
a
lit
y
M
on
i
t
or
i
ng
(
S
hu
gua
ng
W
u
)
949
par
am
et
er
s
i
n
c
hr
o
no
l
og
i
c
al
as
s
oc
i
at
e
d
f
ac
t
or
s
,
s
uc
h
as
t
he
t
her
m
al
i
ner
t
i
a
of
ar
c
s
pac
e.
T
her
ef
or
e,
i
n
t
i
m
e
s
er
i
es
，
da
t
a
g
at
h
er
of
ev
er
y
m
om
e
nt
r
ef
l
ec
t
s
as
depe
nde
nc
e
r
el
at
i
ons
h
i
p.
I
n
t
he ear
l
y
196
0
,
i
n
or
der
t
o ov
er
c
om
e t
he di
f
f
i
c
ul
t
y
t
o s
o
l
v
e t
he
ent
r
op
y
i
n t
h
e c
haos
phen
om
enon,
P
i
nc
us
pr
op
os
ed a c
onc
ept
of
appr
ox
i
m
at
e ent
r
op
y
,
w
hi
c
h
i
s
a non
-
ne
gat
i
v
e
quan
t
i
t
at
i
v
e
des
c
r
i
pt
i
o
n
of
t
he
c
om
pl
ex
i
t
y
of
t
h
e
no
nl
i
n
ear
t
i
m
e
s
er
i
es
.
I
f
t
h
e
p
h
y
s
i
c
al
pr
oc
es
s
of
a n
onl
i
n
ear
c
om
pl
ex
d
egr
ee
i
s
h
i
gh
er
,
t
he appr
ox
i
m
at
e
ent
r
op
y
w
i
l
l
be
bi
gg
er
[
11
-
13
]
.
T
h
i
s
m
et
hod i
s
of
t
he a
dv
ant
a
ges
of
s
hor
t
dem
anded
d
at
a a
nd t
he a
dv
ant
ages
of
s
t
r
ong
ant
i
-
i
nt
er
f
er
enc
e c
a
pab
i
l
i
t
y
.
I
n
pr
ac
t
i
c
e,
i
t
of
t
en b
e us
e
d t
o a
di
agn
os
t
i
c
c
r
i
t
er
i
on
,
an
d a
l
r
ead
y
ha
v
e
t
r
i
ed an
d
i
s
a huge s
uc
c
es
s
i
n t
he f
i
el
ds
of
at
m
os
pher
i
c
r
es
ear
c
h
[1
4
]
,
m
ec
hani
c
al
equ
i
pm
ent
f
aul
t
d
i
ag
nos
i
s
[1
5
]
,
t
el
ec
o
m
m
uni
c
at
i
ons
[
16,
17]
,
po
w
er
t
r
a
n
sm
i
ssi
o
n
[1
8
].
A
l
t
hou
gh A
pE
n m
et
hod c
an
i
den
t
i
f
y
d
y
n
am
i
c
s
t
r
uc
t
ur
e
m
ut
at
i
on of
t
i
m
e s
er
i
es
,
it
w
il
l n
o
t
be
ab
l
e
t
o
gi
v
e
j
um
p
t
i
m
e
and
t
h
e
det
ec
t
e
d
m
ut
at
i
on
i
nt
er
v
a
l
t
hat
h
ea
v
i
l
y
r
e
l
y
o
n
t
he
c
hoi
c
e
of
t
he s
ubs
e
que
nc
e.
T
her
ef
or
e,
t
he a
ut
h
or
bui
l
t
c
a
l
c
ul
a
t
i
o
n of
ap
pr
ox
i
m
at
e ent
r
op
y
al
gor
i
t
hm
t
o
s
ol
v
e t
h
e d
y
n
am
i
c
ar
c
t
i
m
e s
er
i
es
of
a
ppr
ox
i
m
at
e en
t
r
op
y
v
al
u
e
.
T
he pr
es
e
nt
p
aper
i
s
m
ai
nl
y
dev
ot
e
d t
o t
hepr
o
bl
em
o
f
ex
pl
or
i
ng
ar
c
and pr
oc
es
s
s
t
abi
l
i
t
y
i
n h
i
gh
-
p
o
w
er
e
d
s
ubm
er
ged b
y
v
i
r
t
u
e
of
t
he c
h
aot
i
c
par
a
m
et
er
,
A
pE
n
.
I
n
p
ar
t
i
c
u
l
ar
,
t
he
go
al
of
t
he
pa
per
i
s
t
o
at
t
em
pt
t
o
pr
opos
e
a
ne
w
n
um
er
i
c
al
s
t
and
ar
d
t
o
ac
c
ur
at
el
y
qu
an
t
i
f
y
and
ev
al
uat
e
t
he
ar
c
s
t
abi
l
i
t
y
i
n
h
i
gh
-
po
w
er
e
d s
ubm
er
ged ar
c
w
el
d
i
ng
.
2.
R
e
sea
r
ch
M
et
h
o
d
2
.1
.
T
h
e
D
e
fi
n
i
ti
o
n
o
f
B
u
r
s
t D
e
te
c
ti
o
n
Ap
E
n
A
l
g
o
r
i
th
m
S
t
ep
1.
Mi
r
r
or
t
h
e d
at
a
(
)
fn
a
nd
ex
pan
d
t
o L,
t
hat
bui
l
d
-
u
p t
he n
e
w
s
eq
uenc
e
(
)
gn
;
(
)
(
2
)
(
)
,
1
,
2
,
...,
=
−+
+
=
g
n
f
n
L
f
n
n
L
(
1)
S
t
ep
2.
R
ec
ons
t
r
uc
t
(
)
Tn
as
a
L
-
di
m
ens
i
onal
of
ph
as
e s
pac
e;
(
)
{
}
(
)
1
,
(
2)
,
...,
(
)
,
1
,
2
,
...
=
+
+
+=
T
n
g
i
gi
gi
L
i
L
(2
)
S
t
ep
3.
C
al
c
u
l
at
e t
h
e A
pE
n
v
a
l
ue
of
d
(
)
,
1
,
2
,
...
=
Tn
n
L
b
y
f
as
t
appr
ox
i
m
at
e e
nt
r
op
y
c
a
lc
u
la
t
io
n
;
S
t
ep
4.
R
ec
ons
t
r
uc
t
(
)
,
1
,
2
,
...
=
Yn
n
L
as
a
L
-
di
m
ens
i
onal
of
phas
e s
p
ac
e
;
S
t
ep
5.
S
eq
uenc
e
t
he
(
)
Yn
,
i
f
t
he
r
ec
or
d
br
eak
do
w
n
t
hr
es
ho
l
d
p
oi
n
t
,
r
ec
or
d
t
he
br
eak
dow
n po
i
nt
c
oor
di
nat
es
x
0,
at
(
x
0,
x
0 +
L)
w
i
t
hi
n t
he s
c
op
e of
r
et
r
i
e
v
i
n
g
t
he m
ax
i
m
u
m
ent
r
op
y
p
oi
nt
of
s
i
n
gul
ar
i
t
y
.
2
.2
.
T
h
e
D
e
fi
n
i
ti
o
n
o
f
F
ast
A
p
E
n
I
n or
d
er
t
o
d
ef
i
ne
A
p
E
n(
r
,
m
,
N
)
f
or
an
N
-
d
i
m
ens
i
ona
l
t
i
m
e s
er
i
es
{
}
12
,
,
...,
,
εε
ε
N
g
i
v
en
t
he
par
am
et
er
s
m
,
r
,
t
he
m
-
di
m
ens
i
onal
em
bedded v
e
c
t
or
(
)
{
}
12
1
,
,
...,
,
εε
ε
+−
=
i
m
xi
hav
e b
een t
o
be c
ons
i
der
e
d.
T
hen,
t
he
A
pE
n
i
s
def
i
n
ed
as
[1
0
]:
(
)
(
)
1
(
,
,
)
li
m
φφ
+
→∞
=
−
mm
N
A
p
E
n
m
r
N
r
r
(3
)
W
h
er
e:
(
)
(
)
1
1
1
ln
1
φ
−+
=
=
−+
∑
Nm
mm
i
j
r
Cr
Nm
(
4)
(
)
(
)
(
)
1
1
1
,
1
θ
εε
−+
=
=
−
−+
∑
Nm
m
mm
i
ij
j
Cr
d
r
Nm
(
5)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
1
6
9
3
-
6
930
T
E
L
KO
M
NI
K
A
V
o
l.
14
,
N
o
.
3,
S
ept
em
ber
2016
:
9
48
–
9
55
950
(
)
1
0
0
θ
>
=
if
x
x
ot
he
r
w
i
s
e
(
6)
A
nd
,
(
)
(
)
ma
x
,
0
1
εε
=
+
−
+
≤≤
−
ij
d
ik
j
k
k
m
(
7)
(
)
(
)
(
)
2
11
1...
2
−−
=
∩
=
−
∑
i
ij
ij
Cr
d
d
j
N
(
8)
(
)
(
)
(
)
(
)
(
)
3
11
2
2
1...
2
−−
−
−
=
∩∩
=
−
∑
i
ij
ij
i
j
Cr
d
d
d
j
N
(
9)
O
bv
i
ous
l
y
,
t
he
v
al
ue
of
t
he
es
t
i
m
at
e dep
ends
on
m
and
r
.
A
s
s
ugg
es
t
ed
b
y
P
i
nc
us
,
m
c
an be t
ak
en as
2 and
r
as
(
0.
1
–
0.
25)
S
D
,
w
h
er
e S
D
i
s
t
he s
t
an
dar
d d
ev
i
at
i
on f
r
om
t
he or
i
g
i
n
al
dat
a
s
eq
uenc
e
.
A
s
a
r
ul
e,
i
n
en
gi
neer
i
ng
pr
ac
t
i
c
e,
m
or
e
t
han
1
00
da
t
a
ar
e
n
eed
ed
t
o
m
eet
t
he
r
equi
r
em
ent
s
f
or
es
t
i
m
at
i
n
g a r
obus
t
v
al
ue of
A
p
E
n.
C
ons
equ
ent
l
y
,
i
n t
hi
s
pa
p
er
t
he A
pE
n i
s
c
al
c
ul
a
t
ed
und
er
t
h
e f
ol
l
o
w
i
ng c
on
di
t
i
ons
:
100
,
0.15
,
2.
≥=
=
N
r
SD
m
2
.
3
.
T
h
e
C
a
l
c
u
l
a
ti
o
n
o
f
A
p
p
r
o
x
i
m
a
te
E
n
tr
o
y
A
l
g
o
r
i
t
h
m
P
r
o
p
e
r
ti
e
s
D
e
fi
n
i
ti
o
n
1
:
T
he A
p
E
n
v
a
l
ues
ha
v
e
j
us
t
as
m
uc
h
per
i
odi
c
t
o or
i
gi
nal
s
i
gn
al
.
P
r
o
o
f:
I
f
w
e n
o
w
d
ef
i
ne
(
)
fx
a
s
a per
i
od
i
c
f
unc
t
i
on
w
i
t
h a
per
i
o
d
T
,
(
)
(
)
+
=
fx
T
fx
(
10)
T
hen,
(
)
(
)
(
)
()
+=
A
p
E
n
f
x
T
A
p
E
n
f
x
(1
1
)
T
he
v
ec
t
or
(
)
()
+
A
p
E
n
f
x
T
s
ho
w
th
e
p
e
r
io
d
ic
i
nt
ens
i
t
y
m
odu
l
at
i
on
in
t
he
c
ha
os
dom
ai
n
.
A
l
s
o,
t
he
ma
x
i
mu
m
and
m
in
im
u
m
of
A
pE
n v
al
ues
ar
e
s
ho
w
n
i
n
a
s
am
e
per
i
od
i
c
p
r
oper
t
y
.
(
)
(
)
(
)
(
)
(
)
()
+
=
M
a
x
A
p
E
n
f
x
T
M
a
x
A
p
E
n
f
x
(1
2
)
(
)
(
)
(
)
(
)
(
)
()
+
=
M
i
n
A
p
E
n
f
x
T
M
i
n
A
p
E
n
f
x
(1
3
)
D
e
fi
n
i
ti
o
n
2
:
T
he ne
w
s
y
s
t
em
c
ons
i
s
t
s
of
a s
t
abl
e s
y
s
t
em
i
s
not
nec
es
s
ar
i
l
y
s
t
a
bl
e
.
S
y
s
t
em
A
and S
y
s
t
em
B
ar
e s
t
abl
e s
y
s
t
em
.
W
hen
t
he s
y
s
t
em
A
i
s
c
onv
er
t
e
d i
nt
o
s
y
s
t
em
B
,
s
y
s
t
em
A
w
i
l
l
no
t
al
w
a
y
s
i
n a
s
t
ab
l
e s
t
at
e.
P
r
o
o
f:
I
f
,
(
)
2
s
i
n(
0.2
)
1
1
500
=
+
≤
≤
yn
n
n
(
14)
(
)
1.5
s
i
n(
0.2
)
2
c
os
(
0.5
)
0.2
,
1
500
=
+
−
≤
≤
zn
n
n
n
(
15)
(
)
(
)
(
)
1
500
500
1000
≤
≤
=
<
≤
zn
n
fn
yn
n
(
16
)
(
(
))
=
A
S
A
p
E
n
y
n
(
17)
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
K
A
I
S
S
N
:
1
693
-
6
930
A
pp
l
i
c
at
i
o
n of
N
on
l
i
near
D
y
nam
i
c
al
Met
hods
f
or
A
r
c
W
el
d
i
ng
Q
u
a
lit
y
M
on
i
t
or
i
ng
(
S
hu
gua
ng
W
u
)
951
(
(
))
=
B
S
A
p
E
n
z
n
(
18)
(
(
))
=
f
S
A
p
E
n
f
n
(
19)
A
s se
e
i
ng
f
ro
m
F
i
g
ur
e
1b
a
nd F
i
g
ur
e
2
(
b
):
<
AB
SS
(
20
)
T
hen,
(
(
)
)
(
(
)
)
,
[
400
,
500]
<∈
A
p
E
n
f
n
A
p
E
n
z
n
n
(
21)
T
he ne
w
s
y
s
t
em
(
)
fn
i
s
not
al
w
a
y
s
i
n a s
t
abl
e s
t
at
e
,
s
ee i
n F
i
gur
e 3
.
F
i
gur
e 1
.
(
a
)
T
he
T
i
m
e S
er
i
es
o
f
Eq
u
at
i
on
1
4;
(
b
)
t
h
e A
pE
n
of
(
a)
F
i
gur
e 2.
(
a
)
T
he
T
i
m
e S
er
i
es
o
f
Eq
u
at
i
on
1
5;
(
b
)
t
h
e A
pE
n
of
(
a)
D
e
fi
n
i
ti
o
n
3
:
A
n
y
s
i
gn
al
c
a
n be r
ec
o
ns
t
r
uc
t
ed
b
y
s
i
nus
oi
d
al
f
unc
t
i
on.
P
r
o
o
f:
I
f
w
e
no
w
def
i
n
e
(
)
T
ft
as
a per
i
od
i
c
f
unc
t
i
o
n
w
i
t
h a
p
er
i
od
T
,
(
)
(
)
li
m
→∞
=
T
T
f
t
ft
(
22)
0
50
100
150
200
250
300
350
400
450
500
-2
0
2
4
y
(n
)
n
(
a)
n
0
50
100
150
200
250
300
350
400
450
500
0.
17
0.
175
0.
18
0.
185
Ap
En
(b
)
0
100
200
300
400
500
600
700
800
900
1000
-5
0
5
y
(n
)
n
(b
)
(
a)
0
50
100
150
200
250
300
350
400
450
500
0
0
.5
1
Ap
En
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
1
6
9
3
-
6
930
T
E
L
KO
M
NI
K
A
V
o
l.
14
,
N
o
.
3,
S
ept
em
ber
2016
:
9
48
–
9
55
952
T
he s
i
ngl
e s
i
n
us
oi
d
(
)
gt
i
s
def
i
n
ed as
:
(
)
s
i
n(
2
)
πϖ
ϕ
=
+
gt
A
t
(
23)
T
he par
am
et
er
s
i
n E
q
uat
i
o
n
2
3
ar
e t
h
e am
pl
i
t
u
de
A
,
t
he
f
r
equenc
y
ω
and phas
e
φ
.
It
w
as
F
our
i
er
’
s
ho
t
i
dea
t
o c
o
ns
i
der
a
s
um
of
s
i
nus
oi
ds
a
s
a m
odel
f
or
(
)
ft
di
s
t
r
i
b
ut
i
on.
(
)
0
1
s
i
n(
)
ϖϕ
∞
=
=
++
∑
n
n
ft
A
A
n
t
(
24
)
A
n
y
s
i
g
na
l
c
an
be r
ec
o
ns
t
r
uc
t
ed b
y
s
i
nus
o
i
da
l
f
unc
t
i
o
n.
F
i
gur
e
3
.
(
a
)
T
he
T
i
m
e S
er
i
es
o
f
Eq
u
at
i
on
1
6
;
(
b
)
t
h
e A
pE
n
of
(
a)
D
e
fi
n
i
ti
o
n
4
:
A
br
e
ak
dow
n of
t
he s
y
s
t
em
poi
nt
m
ut
at
i
o
n i
nt
o t
h
e s
y
s
t
em
B
,
t
h
e bur
s
t
poi
nt
f
l
ag
A
p
E
n
v
a
l
u
e i
s
def
i
ne
d as
:
0.2
ψ
=
(
25)
F
i
gur
e
4
. E
x
tr
a
c
t
W
av
ef
or
m of
ar
c
W
i
t
hw
a
v
el
et
s
A
l
g
or
i
t
hm
.
(
a
)
O
r
ig
in
a
l A
r
c
;
(
b
)
t
he
D
eno
i
s
e
w
i
t
h W
av
el
et
s
A
l
gor
i
t
hm
of
(
a)
0
100
200
300
400
500
600
700
800
900
1000
-5
0
5
y
(n
)
0
100
200
300
400
500
600
700
800
900
1000
0
0
.5
1
Ap
En
n
(b
)
(
a)
n
0
100
200
300
400
500
600
700
800
900
1000
30
40
50
60
U
/V
(
a)
n
(b
)
0
100
200
300
400
500
600
700
800
900
1000
30
40
50
60
U
/V
n
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
K
A
I
S
S
N
:
1
693
-
6
930
A
pp
l
i
c
at
i
o
n of
N
on
l
i
near
D
y
nam
i
c
al
Met
hods
f
or
A
r
c
W
el
d
i
ng
Q
u
a
lit
y
M
on
i
t
or
i
ng
(
S
hu
gua
ng
W
u
)
953
F
i
gur
e
5.
G
et
t
he
Bu
r
s
t
Po
i
n
t
. (
a
O
r
i
g
in
a
l A
r
c
;
(
b
)
t
he
A
p
E
n of
(
a)
;
(
c
)
t
he b
ur
s
t
po
i
nt
of
(
a)
3.
R
e
su
l
t
s
a
n
d
A
n
a
l
y
s
i
s
3.
1.
C
o
m
p
ar
at
i
v
e A
n
al
y
s
i
s
o
f A
l
g
o
r
i
th
m
s
W
av
el
et
al
gor
i
t
hm
,
MC
-
A
p
E
n a
l
gor
i
t
hm
(
m
ov
i
ng c
ut
dat
a ap
pr
ox
i
m
at
e ent
r
op
y
)
,
B
D
-
A
p
E
n (
B
ur
s
t
de
t
ec
t
i
on
dat
a ap
pr
ox
i
m
at
e
ent
r
op
y
)
al
g
or
i
t
hm
ar
e hi
gh
-
p
o
w
er
w
el
di
n
g ar
c
s
i
g
na
l
F
ig
ur
e 4
a
,
F
ig
ur
e 5
a
an
al
y
s
i
s
.
W
av
el
et
anal
y
s
i
s
m
et
hod t
o el
i
m
i
nat
e s
i
gna
l
no
i
s
e,
ac
c
ur
at
e w
av
ef
or
m
f
eat
ur
e
ex
t
r
ac
t
i
on
as
i
l
l
us
t
r
at
ed
i
n F
i
g
ur
e
4
b.
A
v
ec
t
or
t
hat
d
en
ot
es
t
he
m
agni
t
ude a
nd d
i
r
ec
t
i
on of
w
e
l
d ar
c
,
i
s
r
ec
ons
t
r
uc
t
ed
MC
-
A
p
E
n a
l
g
or
i
t
hm
,
as
i
l
l
us
t
r
at
e
d i
n
F
i
g
ur
e
5
b.
BD
-
A
p
E
n
a
l
gor
i
t
hm
not
o
nl
y
r
ec
ons
t
r
uc
t
e
A
v
ec
t
or
t
hat
den
ot
es
t
he
m
agni
t
ud
e an
d
di
r
ec
t
i
on
of
w
e
l
d
ar
c
,
but
al
s
o ex
t
r
ac
t
bur
es
t
po
i
nt
,
as
i
l
l
u
s
t
r
at
ed i
n F
i
g
ur
e
5c
.
3.
2.
T
h
e
i
n
fl
u
e
n
c
e
o
f
t
h
e p
h
ase sp
ace
si
z
e f
o
r
t
h
e cl
assi
f
i
c
at
i
o
n
H
er
e,
w
e
s
e
l
ec
t
a
f
au
l
t
ar
c
dat
a,
s
h
o
w
n
i
n
F
i
gur
e
6
a,
and
t
hen
f
i
x
e
t
h
e
d
i
f
f
er
ent
s
i
z
es
of
phas
e s
pac
e D
,
r
e
sp
e
ct
i
v
e
l
y
D
=
50
,
D
=
150,
D
=
2
50,
D
=
35
0,
D
=
4
5
0,
c
al
c
ul
at
e i
t
s
appr
ox
i
m
at
e
ent
r
op
y
i
n p
has
e s
p
ac
e.
F
ix
ed
D
=
50,
appr
ox
i
m
at
e ent
r
op
y
dr
i
l
l
-
do
w
n t
o
z
er
o,
and t
h
e ap
pr
ox
i
m
at
e ent
r
o
p
y
g
ap
d
is
lo
c
a
t
io
n
,
s
e
e i
n F
i
gu
r
e
6b
.
T
he s
i
z
e of
phas
e s
pac
e
i
s
i
nc
r
eas
ed s
uf
f
i
c
i
ent
l
y
t
o
150
and
t
her
e
i
s
a c
l
ear
ap
pr
ox
i
m
at
e ent
r
op
y
c
onc
a
v
e g
ap
,
s
ee
i
n
F
i
gur
e 6c
.
T
hen,
t
h
e s
c
an
ni
n
g
r
ange i
s
i
nc
r
eas
ed s
uf
f
i
c
i
ent
l
y
t
o 2
50,
a
nd t
h
e
g
ap d
i
s
l
oc
at
i
o
n
w
ou
l
d
c
om
e
ev
en
s
oo
ner
i
n F
i
gur
e
6
c.
F
i
na
l
l
y
,
t
he s
i
z
e of
p
has
e
s
pac
e
i
s
i
nc
r
eas
ed
ag
ai
n
t
o
D
=
3
50,
t
h
e
f
o
rm
of
appr
ox
i
m
at
e
ent
r
op
y
b
ec
om
e
f
ur
t
her
s
m
ot
her
,
s
e
e i
n F
i
gur
e
6(
e)
.
0
100
200
300
400
500
600
700
800
900
1000
20
40
60
U
/V
n
(
c
)
(
b
)
(
a
)
0
100
200
300
400
500
600
700
800
900
1000
0
1
2
Ap
En
n
0
100
200
300
400
500
600
700
800
900
1000
0
50
U
/V
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
1
6
9
3
-
6
930
T
E
L
KO
M
NI
K
A
V
o
l.
14
,
N
o
.
3,
S
ept
em
ber
2016
:
9
48
–
9
55
954
F
i
gur
e 6.
(
a
)
o
r
ig
in
a
l a
r
c
;
(
b
)
t
he
A
pE
n of
(
a)
w
her
e D
=
50;
(
c
)
t
he
A
p
E
n of
(
a)
w
h
er
e D
=
150
;
(
d
)
t
he
A
p
E
n of
(
a)
w
h
er
e D
=
2
50;
(
e
)
t
he
A
p
E
n
of
(
a)
w
h
er
e D
=
35
0
;
(
e
)
t
h
e A
pE
n of
(
a)
w
her
e
D=
40
0
F
i
gur
e 7.
(
a
)
o
r
ig
in
a
l a
r
c
;
(
b
)
t
he bur
s
t
poi
nt
of
(
a)
w
h
er
e D
=
150
;
(
c
)
t
he b
ur
s
t
po
i
nt
of
(
a)
w
her
e D
=
2
50;
(
d
)
t
he
bur
s
t
poi
nt
of
(
a)
w
her
e D
=
35
0
E
x
per
i
m
ent
al
r
es
u
l
t
s
s
ho
w
al
gor
i
t
hm
i
s
r
obus
t
t
o c
al
i
br
at
e t
he s
i
n
gul
ar
i
t
y
po
i
nt
s
and
deno
t
e t
he k
i
net
i
c
s
a
nd
s
t
a
b
i
lit
y
of
ar
c
,
s
ee F
i
g
ur
e 7
.
T
he r
es
ol
ut
i
o
n an
d s
epar
at
i
o
n ef
f
i
c
i
enc
y
w
er
e
i
m
pr
ov
ed
gr
e
at
l
y
b
y
r
egu
l
at
i
ng
t
he
s
i
z
es
of
phas
e
s
pac
e.
T
he
l
ar
ger
of
pha
s
e
s
pac
e
,
T
he
hol
e of
ar
c
i
s
m
or
e ob
v
i
o
us
and
v
i
c
e
v
er
s
a.
I
n
pr
a
c
t
i
c
e,
t
h
e ef
f
ec
t
of
phas
e
s
pac
e
w
i
l
l
be
c
ons
i
der
e
d.
4
.
C
o
n
c
l
u
s
i
o
n
I
n
es
s
enc
e,
t
he
m
et
hod
of
di
g
i
t
a
l
f
i
l
t
er
i
ng
i
s
us
e
d
f
i
r
s
t
,
and
t
hen
r
es
t
or
ed
t
o
i
t
s
or
i
gi
na
l
f
eat
ur
e.
H
o
w
e
v
er
,
i
t
i
s
u
na
bl
e
t
o
de
not
es
t
he
no
n
-
l
i
ne
ar
d
y
nam
i
c
al
f
eat
ur
es
.
T
he
c
al
c
ul
at
i
on
of
t
he
a
ppr
ox
i
m
at
e
ent
r
o
p
y
i
n
phas
e
s
p
ac
e,
w
hi
c
h
i
s
di
f
f
er
ent
f
r
o
m
w
av
e
l
et
s
i
gna
l
det
ec
t
i
o
n
m
et
hod,
w
i
l
l
not
o
nl
y
c
a
l
i
br
a
t
e
t
he
s
i
ng
ul
ar
i
t
y
poi
nt
s
but
al
s
o
deno
t
e
t
he
k
i
net
i
c
s
a
nd
s
t
a
b
i
lit
y
of
ar
c
.
A
v
ec
t
or
,
w
hi
c
h
i
s
w
i
t
h
t
he
c
al
c
ul
a
t
i
o
n
of
t
he
appr
ox
i
m
at
e
ent
r
op
y
i
n
phas
e
s
pac
e,
d
enot
es
t
he d
i
s
t
or
t
i
o
n of
ar
c
.
T
he
c
al
c
ul
a
t
i
o
n of
t
he ap
pr
ox
i
m
at
e ent
r
op
y
i
n ph
as
e s
pac
e,
w
hi
c
h
i
s
a
c
o
m
pl
ex
i
t
y
m
eas
ur
e s
ui
t
a
bl
e f
or
s
hor
t
dat
a,
e
v
o
l
v
e as
a pr
ob
l
em
-
s
t
at
e m
oni
t
or
i
ng
s
y
s
t
em
.
0
200
400
600
800
1000
30
40
50
60
n
(f
)
(e
)
(c
)
(d
)
(b
)
(
a)
0
200
400
600
800
1000
0
0
.5
1
Ap
En
n
0
200
400
600
800
1000
0
0
.5
1
1
.5
Ap
En
n
U
/V
0
200
400
600
800
1000
0
0
.5
1
1
.5
Ap
En
n
0
200
400
600
800
1000
0
0
.5
1
1
.5
Ap
En
n
0
200
400
600
800
1000
0
0
.5
1
1
.5
Ap
En
n
0
200
400
600
800
1000
30
40
50
60
U
/V
n
(d
)
(
a)
(c
)
n
0
200
400
600
800
1000
0
20
40
60
U
/V
(b
)
0
200
400
600
800
1000
0
20
40
60
U
/V
n
0
200
400
600
800
1000
0
20
40
60
U
/V
n
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
K
A
I
S
S
N
:
1
693
-
6
930
A
pp
l
i
c
at
i
o
n of
N
on
l
i
near
D
y
nam
i
c
al
Met
hods
f
or
A
r
c
W
el
d
i
ng
Q
u
a
lit
y
M
on
i
t
or
i
ng
(
S
hu
gua
ng
W
u
)
955
A
c
k
n
o
w
l
e
d
g
e
m
e
n
ts
T
hi
s
w
or
k
w
as
s
up
por
t
e
d b
y
p
dj
h201
6b0
75
2,
20
14Z
J
0
02
.
R
ef
er
en
ces
[1
]
Z
hang W
e
i
c
hao
,
Y
ang
Li
j
un,
L
ü
X
i
ao
q
i
ng
.
A
ppr
ox
i
m
at
e ent
r
o
py
bas
ed m
e
s
o
-
s
pr
ay
t
r
an
s
f
er
anal
y
s
i
s
o
f
Al
-
al
l
oy
pul
s
ed m
et
al
i
ner
t
-
g
as
w
el
di
ng u
nder
s
el
f
-
ada
pt
i
ng c
o
nt
r
ol
.
A
c
t
a
Ph
y
s
.
Si
n
.
2011
;
6
0(
2)
:
02060
1
.
[2
]
N
i
l
s
i
am
Y
,
H
as
el
h
uhn A
,
W
i
j
n
en B
,
et
al
.
I
nt
egr
at
ed V
o
l
t
ag
e
-
C
ur
r
ent
M
oni
t
or
i
ng and C
o
n
t
r
ol
o
f
G
as
M
et
al
A
r
c
W
e
l
d M
agnet
i
c
B
al
l
-
J
oi
n
t
ed O
p
en S
o
ur
c
e
3
-
D
Pri
n
t
e
r.
M
ac
hi
ne
s
.
2
015
;
3(
4)
:
339
-
351.
[3
]
Y
ao
P
,
X
ue
J
X
,
Z
hu
Q
i
ang.
Q
uant
i
t
at
i
v
e
ev
a
l
uat
i
on
of
do
ubl
e
w
i
r
e
pul
s
ed
w
el
di
n
g
s
t
abi
l
i
t
y
bas
e
d
o
n
pr
obab
i
l
i
t
y
den
s
i
t
y
di
s
t
r
i
but
i
on.
T
r
ans
ac
t
i
on
s
of
t
he C
hi
n
a W
el
d
i
ng
I
ns
t
i
t
u
t
i
on
.
2
014;
35(
7)
:
5
1
-
54.
[4
]
Z
ade F
,
T
al
ent
a A
.
E
d
i
t
o
rs
.
A
dv
anc
ed
F
uz
z
y
C
ont
r
ol
S
y
s
t
e
m
.
Y
ogy
ak
ar
t
a
:
U
A
D
P
r
es
s
.
2
0
10.
[5
]
W
u
S
G
,
C
he
n
J
X
.
D
e
s
ig
n
o
f
D
is
t
r
i
bu
t
e
d P
ul
s
e D
a
t
a A
c
q
u
is
i
t
i
o
n S
y
s
t
em
.
A
dv
a
nc
e
d
M
at
er
i
al
s
R
es
ear
c
h.
2013
:
7
56
-
75
9.
[6
]
B
os
el
l
i
M
,
C
ol
om
b
o V
,
G
hed
i
n
i
E
,
et
a
l
.
T
w
o
-
di
m
en
s
i
on
al
t
i
m
e
-
depe
nde
nt
m
ode
l
l
i
n
g of
f
um
e f
or
m
at
i
o
n
i
n a pul
s
ed ga
s
m
et
a
l
ar
c
w
el
di
ng pr
oc
es
s
.
J
o
ur
nal
o
f
P
hy
s
i
c
s
D
:
A
ppl
i
e
d P
hy
s
i
c
s
.
2
013;
46(
2
2)
:
22400
6.
[7
]
C
am
pbe
l
l
S
W
,
G
al
l
ow
ay
A
M
.
R
am
s
ey
G
M
,
et
al
.
A
c
om
put
a
t
i
ona
l
f
l
ui
d
dy
nam
i
c
ana
l
y
s
i
s
o
f
t
he
ef
f
ec
t
of
w
el
d noz
z
l
e g
eo
m
et
r
y
c
han
ges
on
s
hi
el
d
i
ng
gas
c
ov
er
ag
e dur
i
ng
gas
m
e
t
al
ar
c
w
el
di
n
g.
J
ou
r
n
al
of
M
anuf
ac
t
ur
i
n
g S
c
i
en
c
e a
nd E
n
gi
neer
i
ng
.
20
13;
1
35(
5)
:
05
10
16.
[8
]
J
ai
n S
K
,
S
i
ng
h S
N
,
S
i
ng
h
J
G
.
A
n ada
pt
i
v
e t
i
m
e
-
e
ffi
c
i
e
n
t te
c
hni
que f
or
har
m
on
i
c
e
s
t
i
m
at
i
on o
f
nons
t
at
i
onar
y
s
i
gn
al
s
.
I
nd
us
t
r
i
al
E
l
e
c
t
r
o
ni
c
s
.
I
E
E
E
T
r
an
s
a
c
t
i
ons
on
.
2
013;
60(
8)
:
32
95
-
33
03
.
[9
]
G
hant
y
P
,
P
aul
S
M
,
R
oy
A
B
,
et
al
.
F
uz
z
y
r
ul
e ba
s
ed
appr
o
ac
h f
or
pr
e
di
c
t
i
n
g w
el
d be
ad
geom
et
r
y
i
n
gas
t
ung
s
t
en
ar
c
w
el
di
ng.
S
c
ien
c
e and
T
ec
h
nol
ogy
of
W
el
d
i
n
g
and J
o
i
ni
n
g
.
2
008
;
13(
2)
:
1
67
-
175.
[1
0
]
S
i
m
p
s
on S
W
.
S
i
gnat
ur
e
i
m
ag
es
f
or
ar
c
w
el
di
n
g f
a
ul
t
det
ec
t
i
on.
S
c
i
e
nc
e
an
d
T
e
c
h
nol
o
gy
of
W
e
l
di
n
g
and J
oi
n
i
ng
.
20
07
;
1
2
(6
):
481
-
486
[1
1
]
P
i
n
cu
s S
M
.
A
ppr
o
x
i
m
at
e ent
r
opy
as
a m
e
as
ur
e of
s
y
s
t
em
c
om
p
l
ex
i
t
y
.
P
r
oc
eedi
ngs
o
f
t
he N
at
i
o
nal
A
c
adem
y
of
S
c
i
en
c
e
s
of
t
he U
ni
t
ed
S
t
at
e
s
of
A
m
er
i
c
a
.
19
91
;
88(
6)
:
229
7
-
230
1.
[1
2
]
P
i
nc
u
s
S
M
.
A
s
s
es
s
i
ng
S
er
i
a
l
I
r
r
egul
ar
i
t
y
and
I
t
s
I
m
pl
i
c
at
i
ons
f
or
H
ea
l
t
h.
A
nn
al
s
of
t
h
e
N
ew
Y
o
rk
A
c
ade
m
y
of
S
c
i
en
c
e
s
.
2
001
;
9
54(
1)
:
245
–
267.
[1
3
]
P
i
n
cu
s S
M
,
G
ol
dber
ger
A
L.
P
hy
s
i
ol
o
gi
c
al
t
i
m
e
-
s
er
i
es
a
nal
y
s
i
s
:
W
h
a
t
doe
s
r
eg
ul
ar
i
t
y
quant
i
f
y
?
.
A
m
er
i
c
an
J
our
nal
of
P
h
y
s
i
ol
og
y
.
19
94
;
1643
-
56.
[1
4
]
W
a
n
g Li
u,
H
e
W
en
-
P
i
ng
,
W
a
n
S
hi
q
uan.
E
v
ol
ut
i
onar
y
m
od
el
i
n
g
f
or
par
a
m
et
er
e
s
t
i
m
at
i
on
f
or
c
hao
t
i
c
s
y
s
te
m
.
A
ct
a
P
h
ys.
S
i
n
.
2014
;
63(
1)
:
0192
03.
[1
5
]
K
oy
unc
u
I
,
O
z
c
er
i
t
A
T
,
P
ehl
i
v
an
I
.
I
m
pl
em
ent
at
i
on
of
F
P
G
A
-
bas
e
d
r
eal
t
i
m
e
nov
el
c
hao
t
i
c
os
c
i
l
l
at
or
.
N
onl
i
n
ear
D
y
nam
i
c
s
.
2
014
;
7
7(
1
-
2
)
:
49
-
59.
[1
6
]
Li
L,
K
e
qi
L
,
G
en H
.
R
e
c
og
ni
t
i
on of
F
i
s
s
i
o
n S
i
g
nal
s
B
a
s
ed o
n
W
av
el
et
A
nal
y
s
i
s
and N
e
ur
al
N
et
w
or
k
.
T
EL
KO
M
N
I
KA
T
el
ec
om
m
uni
c
at
i
on
C
om
put
i
ng E
l
ec
t
r
oni
c
s
a
nd C
ont
r
ol
.
2016
;
14(
3)
.
[1
7
]
Y
un H
,
Li
u R
,
S
han
ggu
an L
.
T
r
ans
f
or
m
er
F
a
ul
t
D
i
agno
s
i
s
M
et
hod B
as
e
d on D
y
na
m
i
c
W
e
i
ght
e
d
Co
m
bi
nat
i
on M
ode
.
T
EL
KO
M
N
I
KA
T
el
e
c
om
m
uni
c
at
i
o
n C
o
m
put
i
ng E
l
ec
t
r
oni
c
s
and C
on
t
r
ol
. 2
016
;
14(
3)
.
D
O
I
:
h
t
t
p:
/
/
dx
.
doi
.
or
g/
10
.
1292
8/
t
e
l
k
o
m
ni
k
a.
v
14
i
3.
3545
.
[1
8
]
Z
hang
X
ue
q
i
n
g,
Li
ang
J
u
n.
C
haot
i
c
t
i
m
e
s
er
i
e
s
pr
edi
c
t
i
o
n
m
ode
l
of
w
i
nd pow
er
ba
s
ed
o
n en
s
em
bl
e
em
pi
r
i
c
al
m
ode
d
ec
om
p
os
i
t
ion
-
appr
ox
i
m
at
e ent
r
opy
an
d r
es
er
v
oi
r
.
Ac
t
a
Ph
y
s
.
S
i
n
.
20
13
;
6
2
(5
):
05050
5.
Evaluation Warning : The document was created with Spire.PDF for Python.