T
E
L
K
O
M
NIKA
T
elec
o
mm
un
ica
t
io
n Co
m
pu
t
i
ng
E
lect
ro
nics
a
nd
Co
ntr
o
l
Vo
l.
2
4
,
No
.
5
,
Octo
b
er
2
0
2
6
,
p
p
.
1
6
4
4
~
1
6
5
5
I
SS
N:
1
6
9
3
-
6
9
3
0
,
DOI
: 1
0
.
1
2
9
2
8
/
T
E
L
KOM
NI
K
A
.
v
2
4
i5
.
27967
1644
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//telko
mn
ika
.
u
a
d
.
a
c.
i
d
Foreca
sting
parti
cula
te
m
a
t
ter
co
n
centration
using
deep
lea
rning
and w
ea
t
her data
Ah
m
a
d F
urqo
n
1
,
Sa
nto
s
o
So
ek
ir
no
1
,
Na
ufa
l A
na
nd
a
2
,
H
a
ry
a
s
Su
by
a
nta
ra
Wica
k
s
a
na
3
,
Ah
m
a
d
M
eij
l
a
n Ya
s
ir
4
,
T
o
nn
y
Wa
h
y
u Aj
i
4
,
Av
iv
M
a
g
hridl
o
4
,
F
a
dh
li
L
uthf
ia
ns
y
a
h
4
,
M
a
rz
uk
i
Sin
a
m
bela
4
,
Su
k
o
P
ra
y
it
no
Adi
4
1
D
e
p
a
r
t
me
n
t
o
f
P
h
y
si
c
s,
F
a
c
u
l
t
y
o
f
M
a
t
h
e
mat
i
c
s
a
n
d
S
c
i
e
n
c
e
,
U
n
i
v
e
r
si
t
y
o
f
I
n
d
o
n
e
si
a
,
D
e
p
o
k
,
W
e
st
Jav
a
,
I
n
d
o
n
e
si
a
2
D
e
p
a
r
t
me
n
t
o
f
I
n
st
r
u
me
n
t
a
t
i
o
n
a
n
d
C
a
l
i
b
r
a
t
i
o
n
,
R
e
g
i
o
n
I
I
o
f
M
e
t
e
o
r
o
l
o
g
i
c
a
l
C
l
i
mat
o
l
o
g
i
c
a
l
a
n
d
G
e
o
p
h
y
si
c
a
l
A
g
e
n
c
y
,
B
a
n
t
e
n
,
I
n
d
o
n
e
si
a
3
D
i
r
e
c
t
o
r
a
t
e
o
f
I
n
st
r
u
me
n
t
a
t
i
o
n
C
a
l
i
b
r
a
t
i
o
n
,
A
g
e
n
c
y
f
o
r
M
e
t
e
o
r
o
l
o
g
y
C
l
i
mat
o
l
o
g
y
a
n
d
G
e
o
p
h
y
si
c
s,
Ja
k
a
r
t
a
,
I
n
d
o
n
e
si
a
4
D
e
p
a
r
t
me
n
t
o
f
I
n
st
r
u
me
n
t
a
t
i
o
n
,
S
t
a
t
e
C
o
l
l
e
g
e
o
f
M
e
t
e
o
r
o
l
o
g
y
C
l
i
mat
o
l
o
g
y
a
n
d
G
e
o
p
h
y
si
c
s,
T
a
n
g
e
r
a
n
g
,
B
a
n
t
e
n
,
I
n
d
o
n
e
si
a
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Ma
r
3
1
,
2
0
2
6
R
ev
i
s
ed
J
u
l 1
3
,
2
0
2
6
A
cc
ep
ted
A
u
g
2
8
,
2
0
2
6
P
a
rti
c
u
late
m
a
tt
e
r
(P
M
2
.
5
)
c
o
n
c
e
n
tratio
n
is
a
c
rit
ica
l
a
ir
q
u
a
li
ty
in
d
ica
to
r
w
it
h
se
v
e
re
h
e
a
lt
h
im
p
a
c
ts.
P
re
d
ictin
g
i
ts
d
y
n
a
m
ics
is
c
h
a
ll
e
n
g
in
g
d
u
e
t
o
c
o
m
p
lex
in
tera
c
ti
o
n
s
w
it
h
m
e
te
o
ro
l
o
g
ica
l
v
a
riab
les
.
T
h
is
st
u
d
y
e
v
a
lu
a
tes
lo
n
g
s
h
o
rt
-
term
m
e
m
o
r
y
(
L
S
T
M
)
a
n
d
m
u
lt
il
a
y
e
r
p
e
rc
e
p
tro
n
(M
L
P
)
n
e
tw
o
rk
s
to
i
d
e
n
ti
f
y
th
e
m
o
st
a
c
c
u
ra
te
f
o
re
c
a
stin
g
a
rc
h
it
e
c
tu
re
.
W
e
in
teg
ra
ted
c
o
n
ti
n
u
o
u
s
d
a
ta
f
ro
m
a
u
to
m
a
ti
c
w
e
a
th
e
r
sta
ti
o
n
s
(AW
S
)
a
n
d
a
ir
q
u
a
li
ty
se
n
so
rs.
Da
ta
p
re
p
ro
c
e
ss
in
g
i
n
v
o
lv
e
d
tem
p
o
ra
l
sy
n
c
h
ro
n
iza
ti
o
n
,
o
u
t
li
e
r
re
m
o
v
a
l,
a
n
d
m
u
lt
iv
a
riate
i
m
p
u
tatio
n
b
y
c
h
a
in
e
d
e
q
u
a
ti
o
n
s
(M
I
CE).
Ke
y
in
p
u
t
f
e
a
tu
re
s
p
a
st
p
a
rti
c
u
late
m
a
tt
e
r,
te
m
p
e
ra
tu
re
,
a
n
d
re
lativ
e
h
u
m
id
it
y
w
e
r
e
se
le
c
ted
v
ia
P
e
a
rso
n
c
o
rre
latio
n
a
n
d
train
e
d
u
sin
g
a
2
4
-
h
o
u
r
slid
i
n
g
w
in
d
o
w
.
Re
su
lt
s
d
e
m
o
n
stra
te
th
a
t
th
e
sin
g
le
L
S
T
M
m
o
d
e
l
o
u
tp
e
rf
o
rm
s
b
o
th
th
e
M
L
P
a
n
d
h
y
b
rid
a
rc
h
it
e
c
tu
re
s.
T
h
e
L
S
T
M
a
c
h
iev
e
d
th
e
lo
w
e
st
ro
o
t
m
e
a
n
sq
u
a
re
e
rro
r
(RM
S
E
)
o
f
6
.
8
6
2
µ
g
/
m
³
a
n
d
a
m
e
a
n
a
b
so
lu
te
p
e
rc
e
n
tag
e
e
rro
r
(M
A
P
E)
o
f
3
7
.
3
8
%
.
A
lt
h
o
u
g
h
th
e
M
L
P
y
ield
e
d
a
slig
h
tl
y
h
ig
h
e
r
c
o
e
ff
icie
n
t
o
f
d
e
term
in
a
ti
o
n
(0
.
8
1
0
),
it
e
x
h
i
b
it
e
d
sig
n
if
ica
n
t
m
a
g
n
it
u
d
e
b
ias
b
y
tre
a
ti
n
g
se
q
u
e
n
ti
a
l
d
a
ta
a
s
sta
ti
c
f
e
a
tu
re
s.
Ulti
m
a
tel
y
,
th
i
s
re
se
a
rc
h
p
r
o
v
id
e
s
a
ro
b
u
st,
d
a
ta
-
d
riv
e
n
f
ra
m
e
w
o
rk
f
o
r
re
a
l
-
ti
m
e
a
ir
q
u
a
li
ty
e
a
rl
y
wa
rn
in
g
sy
ste
m
s
in
tro
p
ica
l
re
g
io
n
s
.
K
ey
w
o
r
d
s
:
A
ir
q
u
al
it
y
L
o
n
g
s
h
o
r
t
-
ter
m
m
e
m
o
r
y
Mu
ltil
a
y
er
p
er
ce
p
tr
o
n
P
ar
ticu
late
m
atter
W
ea
th
er
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
:
San
to
s
o
So
ek
ir
n
o
Dep
ar
t
m
en
t o
f
P
h
y
s
ics
,
Fac
u
lt
y
o
f
Ma
t
h
e
m
a
tics
a
n
d
Scien
ce
,
Un
iv
er
s
it
y
o
f
I
n
d
o
n
es
ia
Dep
o
k
,
W
est J
av
a,
I
n
d
o
n
esia
E
m
ail: sa
n
to
s
o
.
s
@
s
c
i.u
i.a
c.
id
1.
I
NT
RO
D
UCT
I
O
N
A
ir
q
u
a
lit
y
s
i
g
n
if
ican
t
l
y
i
m
p
ac
ts
g
lo
b
al
p
u
b
lic
h
ea
lth
an
d
ec
o
lo
g
ical
b
alan
ce
[
1
]
.
Am
o
n
g
at
m
o
s
p
h
er
ic
p
o
llu
tan
t
s
,
p
ar
ticu
late
m
atter
≤
2
.
5
µm
(
P
M2
.
5
)
is
p
ar
ticu
lar
ly
cr
itica
l,
co
n
tr
ib
u
tin
g
to
ap
p
r
o
x
im
a
tel
y
o
n
e
-
ei
g
h
th
o
f
g
lo
b
al
p
o
llu
tan
t
-
r
elate
d
d
ea
th
s
d
u
e
t
o
s
ev
er
e
r
esp
ir
ato
r
y
an
d
ca
r
d
io
v
ascu
lar
r
is
k
s
[
2
]
,
[
3
]
.
T
h
e
ac
cu
m
u
latio
n
a
n
d
d
is
p
er
s
io
n
o
f
P
M2
.
5
ar
e
s
tr
o
n
g
l
y
d
r
iv
e
n
b
y
co
m
p
lex
,
n
o
n
lin
ea
r
m
eteo
r
o
lo
g
ical
f
ac
to
r
s
s
u
c
h
as
te
m
p
er
at
u
r
e,
r
elativ
e
h
u
m
id
it
y
,
a
n
d
w
i
n
d
s
p
ee
d
[
4
]
.
C
o
n
v
en
t
io
n
al
s
tati
s
tic
al
m
o
d
els,
s
u
c
h
as
au
to
r
eg
r
es
s
iv
e
in
te
g
r
ated
m
o
v
in
g
av
er
a
g
e
(
A
R
I
M
A
)
a
n
d
li
n
ea
r
r
eg
r
ess
io
n
,
s
tr
u
g
g
le
to
ca
p
tu
r
e
t
h
ese
d
y
n
a
m
ic
in
ter
ac
tio
n
s
[
5
]
.
Si
m
ilar
l
y
,
tr
a
d
itio
n
al
en
g
i
n
ee
r
in
g
s
o
lu
tio
n
s
lik
e
Kal
m
a
n
f
ilter
s
o
r
s
tan
d
ar
d
m
ac
h
i
n
e
lear
n
i
n
g
alg
o
r
ith
m
s
o
f
te
n
r
eq
u
ir
e
r
i
g
o
r
o
u
s
m
an
u
al
f
ea
t
u
r
e
e
n
g
in
ee
r
i
n
g
a
n
d
ass
u
m
e
Ga
u
s
s
ia
n
n
o
i
s
e
d
is
tr
ib
u
tio
n
s
.
T
h
ese
ass
u
m
p
tio
n
s
r
ar
el
y
h
o
ld
tr
u
e
in
h
ig
h
l
y
v
o
lati
le
r
ea
l
-
w
o
r
ld
at
m
o
s
p
h
er
ic
s
en
s
o
r
s
tr
ea
m
s
.
C
o
n
s
eq
u
en
tl
y
,
d
ee
p
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
F
o
r
ec
a
s
tin
g
p
a
r
ticu
la
te
ma
tte
r
co
n
ce
n
tr
a
tio
n
u
s
in
g
d
ee
p
le
a
r
n
in
g
a
n
d
w
ea
th
er d
a
ta
(
S
a
n
to
s
o
S
o
ek
ir
n
o
)
1645
s
eq
u
en
tial
n
e
u
r
al
n
et
w
o
r
k
s
o
f
f
er
a
s
u
p
er
io
r
en
g
i
n
ee
r
i
n
g
alt
er
n
ativ
e
b
y
a
u
to
n
o
m
o
u
s
l
y
lea
r
n
in
g
h
ier
ar
ch
ica
l
te
m
p
o
r
al
f
ea
t
u
r
es d
ir
ec
tl
y
f
r
o
m
r
a
w
m
u
lti
-
s
en
s
o
r
d
ata.
I
n
r
ec
en
t
y
ea
r
s
,
th
e
d
ev
elo
p
m
en
t
o
f
m
ac
h
in
e
lear
n
in
g
m
et
h
o
d
s
h
as
p
r
o
v
id
ed
n
e
w
o
p
p
o
r
tu
n
itie
s
in
air
q
u
alit
y
m
o
d
eli
n
g
an
d
p
r
ed
ict
io
n
.
T
h
e
l
o
n
g
s
h
o
r
t
-
ter
m
m
e
m
o
r
y
(
L
ST
M)
m
o
d
el
is
co
n
s
id
er
ed
o
p
ti
m
al
in
p
r
ed
ictin
g
P
M2
.
5
c
o
n
ce
n
tr
ati
o
n
s
b
ec
au
s
e
it
is
d
esi
g
n
ed
to
o
v
er
co
m
e
th
e
p
r
o
b
lem
o
f
v
a
n
is
h
i
n
g
g
r
ad
ien
t
s
an
d
ca
n
r
etai
n
i
m
p
o
r
tan
t
i
n
f
o
r
m
at
io
n
i
n
lo
n
g
t
i
m
e
s
er
ies
[
6
]
.
I
n
p
r
ed
ictin
g
P
M2
.
5
,
h
o
u
r
l
y
i
n
ter
v
al
s
h
av
e
a
r
o
o
t
m
ea
n
s
q
u
ar
e
e
r
r
o
r
(
R
MSE
)
i
n
th
e
r
a
n
g
e
o
f
6
.
3
-
1
3
.
1
μ
g
/
m
³
[
7
]
.
On
th
e
o
t
h
er
h
an
d
,
t
h
e
m
u
ltil
a
y
er
p
er
ce
p
tr
o
n
(
ML
P
)
is
a
n
eu
r
al
n
et
w
o
r
k
m
o
d
el
th
at
is
also
ef
f
ec
tiv
e
in
h
an
d
lin
g
n
o
n
l
in
ea
r
r
elatio
n
s
h
ip
s
b
et
w
ee
n
v
ar
iab
le
s
,
alth
o
u
g
h
w
it
h
a
s
i
m
p
ler
ap
p
r
o
ac
h
th
a
n
L
ST
M.
T
h
is
m
o
d
e
l
ca
n
p
r
ed
ict
P
M2
.
5
co
n
ce
n
tr
atio
n
s
w
it
h
e
v
alu
a
tio
n
m
etr
ics i
n
t
h
e
f
o
r
m
o
f
an
R
MS
E
o
f
3
.
4
1
-
8
.
1
2
μ
g
/
m
³
an
d
an
MA
E
o
f
2
.
2
1
-
4
.
3
6
μ
g
/
m
³
[
7
]
.
W
h
ile
p
r
ev
io
u
s
s
tu
d
ies
h
av
e
e
x
p
lo
r
ed
L
ST
M
an
d
ML
P
ar
ch
itectu
r
es
f
o
r
air
q
u
alit
y
f
o
r
ec
a
s
tin
g
[
8
]
,
m
an
y
s
i
m
p
l
y
f
ee
d
r
a
w
m
ete
o
r
o
lo
g
ical
v
ar
iab
les
in
to
n
eu
r
al
n
et
w
o
r
k
s
w
i
th
o
u
t
r
ig
o
r
o
u
s
l
y
e
v
al
u
ati
n
g
t
h
e
n
ec
es
s
it
y
o
f
ea
c
h
f
ea
t
u
r
e
o
r
ad
d
r
ess
in
g
t
h
e
te
m
p
o
r
al
ir
r
e
g
u
lar
ities
i
n
h
er
e
n
t
i
n
s
en
s
o
r
d
ata.
Mo
s
t
e
x
is
ti
n
g
L
ST
M
-
b
ased
w
o
r
k
s
r
el
y
o
n
s
tan
d
ar
d
ized
,
clea
n
d
atasets
f
r
o
m
te
m
p
er
ate
r
eg
io
n
s
an
d
o
v
er
lo
o
k
th
e
co
m
p
le
x
th
er
m
o
h
y
g
r
o
m
etr
ic
i
n
ter
ac
tio
n
s
t
y
p
ica
l
o
f
tr
o
p
ical
cli
m
at
es.
Fu
r
t
h
er
m
o
r
e,
th
er
e
is
a
lack
o
f
s
y
s
te
m
atic
co
m
p
ar
ati
v
e
ev
a
lu
at
io
n
s
th
at
i
s
o
late
th
e
e
x
ac
t
p
er
f
o
r
m
a
n
ce
t
r
ad
e
-
o
f
f
s
b
et
w
ee
n
s
eq
u
e
n
tial
L
ST
M,
s
tatic
f
ee
d
-
f
o
r
w
ar
d
ML
P
,
an
d
h
y
b
r
id
ar
ch
itect
u
r
es
u
n
d
er
id
en
tical
d
ata
p
r
e
-
p
r
o
ce
s
s
in
g
p
ip
elin
es.
T
o
b
r
id
g
e
th
is
cr
itical
g
ap
,
o
u
r
s
tu
d
y
i
n
tr
o
d
u
ce
s
a
u
n
i
f
ied
f
r
a
m
e
w
o
r
k
t
h
at
co
m
b
in
e
s
te
m
p
o
r
al
s
y
n
c
h
r
o
n
iz
atio
n
,
m
u
lti
v
ar
iate
i
m
p
u
ta
tio
n
b
y
ch
ai
n
ed
eq
u
atio
n
s
(
MI
C
E
)
,
an
d
r
ig
o
r
o
u
s
P
e
ar
s
o
n
co
r
r
elatio
n
f
ea
tu
r
e
s
elec
tio
n
.
W
e
ex
p
licitl
y
ev
alu
a
te
h
o
w
i
n
teg
r
ati
n
g
lo
c
alize
d
au
to
m
atic
w
ea
t
h
er
s
ta
tio
n
(
A
W
S)
d
ata
o
p
ti
m
izes
P
M2
.
5
p
r
ed
ictio
n
ac
cu
r
ac
y
,
p
r
o
v
id
in
g
a
r
ep
r
o
d
u
cib
le
b
en
ch
m
ar
k
f
o
r
tr
o
p
ical
a
ir
q
u
alit
y
m
o
d
elin
g
.
T
h
e
s
p
ec
if
ic
co
n
tr
ib
u
tio
n
s
o
f
t
h
i
s
s
t
u
d
y
ar
e
th
r
ee
f
o
ld
:
(
i)
a
co
m
p
ar
ativ
e
e
v
al
u
atio
n
o
f
L
ST
M
an
d
ML
P
ar
ch
itect
u
r
es
f
o
r
s
eq
u
e
n
tial
air
q
u
alit
y
d
ata;
(
ii)
th
e
in
teg
r
atio
n
o
f
co
n
ti
n
u
o
u
s
AW
S
p
ar
am
eter
s
a
s
p
r
ed
icto
r
s
;
an
d
(
iii)
th
e
i
m
p
le
m
en
tatio
n
o
f
a
2
4
-
h
o
u
r
s
lid
i
n
g
w
i
n
d
o
w
ap
p
r
o
ac
h
to
ca
p
tu
r
e
d
ail
y
p
er
io
d
icit
y
.
T
h
e
s
ig
n
i
f
ica
n
ce
o
f
th
is
r
esear
ch
lies
i
n
p
r
o
v
id
in
g
a
h
ig
h
l
y
r
eliab
le,
d
ata
-
d
r
iv
en
f
r
a
m
e
w
o
r
k
f
o
r
ea
r
l
y
w
ar
n
i
n
g
s
y
s
te
m
s
,
p
ar
ticu
lar
l
y
i
n
tr
o
p
ical
r
eg
io
n
s
w
h
er
e
co
m
p
lex
m
e
teo
r
o
lo
g
ical
d
y
n
a
m
ics
s
tr
o
n
g
l
y
d
icta
te
p
o
llu
tan
t
ac
cu
m
u
lat
io
n
.
2.
M
E
T
H
O
D
T
h
is
s
ec
tio
n
d
escr
ib
es
t
h
e
r
es
ea
r
ch
s
ta
g
es
co
n
d
u
cted
to
ev
a
lu
ate
t
h
e
L
ST
M
an
d
ML
P
alg
o
r
ith
m
s
in
p
r
ed
ictin
g
P
M2
.
5
co
n
ce
n
tr
atio
n
u
s
i
n
g
m
e
teo
r
o
lo
g
ical
d
ata
as
in
p
u
t
f
ea
t
u
r
es
[
9
]
.
T
h
e
s
tag
es
o
f
t
h
i
s
r
esear
ch
m
et
h
o
d
in
clu
d
e
liter
atu
r
e
s
t
u
d
y
,
d
ata
co
llec
tio
n
,
d
ata
p
r
e
-
p
r
o
ce
s
s
in
g
,
d
ata
i
m
p
u
tat
io
n
,
d
eter
m
in
in
g
p
r
ed
icto
r
s
,
d
esig
n
in
g
i
n
p
u
t
f
ea
t
u
r
es
an
d
m
o
d
el
h
y
p
er
p
ar
a
m
eter
s
,
a
n
d
ev
alu
a
tin
g
t
h
e
L
ST
M
an
d
ML
P
m
o
d
els.
T
h
e
f
ir
s
t
s
tag
e
o
f
t
h
i
s
r
esear
c
h
m
et
h
o
d
is
a
l
iter
atu
r
e
s
t
u
d
y
,
t
h
is
s
ta
g
e
is
ca
r
r
ied
o
u
t
b
y
e
x
p
l
o
r
in
g
,
r
e
ad
in
g
,
t
h
e
r
e
s
ea
r
ch
b
eg
an
w
it
h
a
co
m
p
r
eh
e
n
s
iv
e
liter
atu
r
e
r
ev
ie
w
o
f
s
cien
tific
j
o
u
r
n
als
an
d
ar
ticles.
T
h
i
s
s
tep
f
o
cu
s
ed
o
n
estab
lis
h
in
g
t
h
e
t
h
eo
r
etica
l
f
o
u
n
d
atio
n
o
f
P
M2
.
5
d
y
n
a
m
ics,
th
e
u
n
d
er
l
y
in
g
m
ec
h
an
ic
s
o
f
L
ST
M
an
d
M
L
P
alg
o
r
ith
m
s
,
an
d
id
e
n
ti
f
y
in
g
m
et
h
o
d
o
lo
g
ical
g
ap
s
i
n
p
r
io
r
s
t
u
d
ies
u
tili
z
in
g
i
n
te
g
r
ated
A
W
S
an
d
air
q
u
alit
y
s
en
s
o
r
d
ata.
T
h
e
d
ata
co
llectio
n
p
h
ase
o
f
t
h
is
s
t
u
d
y
f
o
cu
s
ed
o
n
ca
p
tu
r
i
n
g
r
a
w
d
ata
f
r
o
m
v
ar
io
u
s
s
o
u
r
ce
s
to
o
b
tain
co
m
p
r
e
h
en
s
iv
e
i
n
f
o
r
m
atio
n
a
b
o
u
t
en
v
ir
o
n
m
en
tal
co
n
d
itio
n
s
.
I
n
t
h
i
s
co
n
te
x
t,
d
ata
w
a
s
c
o
llected
u
s
i
n
g
t
w
o
p
r
im
ar
y
i
n
s
tr
u
m
e
n
ts
:
A
W
S
an
d
an
air
q
u
alit
y
s
en
s
o
r
.
T
h
e
c
alib
r
ated
A
W
S
in
s
tr
u
m
e
n
t
s
er
v
ed
as
a
s
o
u
r
ce
o
f
m
eteo
r
o
lo
g
ical
d
ata,
co
v
er
in
g
v
ar
iab
les
s
u
c
h
as
air
te
m
p
er
atu
r
e
an
d
r
elativ
e
h
u
m
id
it
y
.
Me
an
w
h
ile,
th
e
air
qu
alit
y
s
e
n
s
o
r
s
er
v
ed
as
a
s
o
u
r
ce
o
f
h
i
s
to
r
ical
d
ata
o
n
P
M2
.
5
an
d
P
M1
0
p
ar
ticu
late
m
a
tter
co
n
ce
n
tr
at
io
n
s
.
T
h
e
d
ata
p
r
o
ce
s
s
in
g
p
h
a
s
e
th
en
tr
an
s
f
o
r
m
ed
t
h
e
r
a
w
d
ata
i
n
to
a
f
o
r
m
at
r
ea
d
y
f
o
r
p
r
o
ce
s
s
in
g
b
y
a
m
ac
h
i
n
e
lear
n
in
g
m
o
d
el.
T
h
ese
s
tep
s
in
cl
u
d
ed
d
ata
clea
n
i
n
g
to
r
e
m
o
v
e
co
r
r
u
p
ted
o
r
an
o
m
al
o
u
s
d
ata
an
d
ti
m
e
s
y
n
ch
r
o
n
izatio
n
to
en
s
u
r
e
th
a
t
th
e
te
m
p
o
r
al
r
eso
lu
tio
n
o
f
th
e
m
eteo
r
o
lo
g
ical
d
ata
an
d
th
e
air
q
u
alit
y
s
e
n
s
o
r
d
ata
w
as a
li
g
n
ed
[
1
0
]
.
T
h
e
f
o
u
r
th
s
ta
g
e
is
d
ata
i
m
p
u
tatio
n
,
a
s
p
ec
ial
s
tep
in
p
r
e
-
p
r
o
ce
s
s
i
n
g
to
h
a
n
d
le
m
i
s
s
in
g
v
al
u
es
.
Sen
s
o
r
s
o
f
te
n
ex
p
er
ien
ce
tech
n
ical
f
ail
u
r
es.
Mis
s
in
g
d
ata
is
f
illed
u
s
in
g
s
p
ec
if
ic
s
tat
is
tica
l
tech
n
iq
u
es
(
s
u
ch
as
lin
ea
r
i
n
ter
p
o
latio
n
o
r
m
ea
n
i
m
p
u
tat
io
n
)
to
m
a
in
ta
in
th
e
co
n
ti
n
u
i
t
y
o
f
t
h
e
ti
m
e
s
er
ie
s
d
at
a
w
i
th
o
u
t
h
a
v
i
n
g
to
d
elete
m
u
c
h
i
m
p
o
r
tan
t
in
f
o
r
m
atio
n
[
1
1
]
.
T
h
e
f
if
t
h
s
ta
g
e
is
d
eter
m
i
n
in
g
p
r
ed
icto
r
s
.
T
h
is
s
ta
g
e
is
t
h
e
p
r
o
ce
s
s
o
f
s
elec
ti
n
g
i
n
p
u
t
v
ar
iab
les
(
f
ea
tu
r
es)
th
at
h
a
v
e
th
e
m
o
s
t
i
n
f
lu
en
ce
o
n
th
e
p
r
ed
ictio
n
tar
g
et
(
P
M2
.
5
)
.
I
n
p
u
t
f
ea
t
u
r
e
s
elec
tio
n
u
s
es
P
ea
r
s
o
n
co
r
r
elatio
n
an
al
y
s
is
to
i
d
en
tify
m
eteo
r
o
lo
g
ical
p
ar
am
eter
s
t
h
at
h
a
v
e
a
s
ig
n
i
f
ica
n
t i
m
p
ac
t o
n
f
l
u
ct
u
ati
o
n
s
in
air
p
o
llu
t
io
n
co
n
ce
n
tr
ati
o
n
s
[
1
2
]
.
T
h
e
s
ix
th
s
tag
e,
i
n
p
u
t
f
ea
t
u
r
e
an
d
h
y
p
er
p
ar
a
m
eter
d
esi
g
n
,
is
a
cr
u
cial
p
h
ase
in
t
h
e
tech
n
ic
al
s
tr
u
ct
u
r
e
d
esig
n
b
ef
o
r
e
th
e
m
o
d
el
ev
a
lu
atio
n
p
r
o
ce
s
s
b
eg
i
n
s
.
A
t
th
is
s
ta
g
e,
i
n
p
u
t
f
ea
t
u
r
e
d
esig
n
is
ca
r
r
ied
o
u
t
b
y
d
eter
m
in
i
n
g
th
e
ti
m
e
w
i
n
d
o
w
to
d
eter
m
i
n
e
th
e
d
u
r
atio
n
o
f
h
is
to
r
ical
d
ata,
o
r
th
e
n
u
m
b
er
o
f
p
ast
h
o
u
r
s
th
at
w
il
l
b
e
u
s
ed
as
th
e
b
asis
f
o
r
f
u
tu
r
e
p
r
ed
ictio
n
s
.
Si
m
u
lta
n
e
o
u
s
l
y
,
h
y
p
er
p
ar
a
m
eter
s
et
tin
g
s
ar
e
ca
r
r
ied
o
u
t
to
o
p
tim
ize
m
o
d
el
p
er
f
o
r
m
a
n
ce
b
y
d
eter
m
in
in
g
k
e
y
p
ar
a
m
e
ter
s
,
s
u
ch
as
th
e
n
u
m
b
er
o
f
h
id
d
en
lay
er
s
,
th
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
2
4
,
No
.
5
,
Octo
b
e
r
2
0
2
6
:
1
6
4
4
-
1655
1646
n
u
m
b
er
o
f
n
e
u
r
o
n
s
,
th
e
lear
n
i
n
g
r
ate,
t
h
e
b
atch
s
ize,
th
e
ac
ti
v
atio
n
f
u
n
ctio
n
,
th
e
s
e
g
m
en
ta
t
io
n
p
o
r
tio
n
,
an
d
th
e
t
y
p
e
o
f
o
p
ti
m
izer
.
T
h
is
ai
m
s
t
o
o
b
tain
th
e
m
o
s
t
o
p
ti
m
al
L
S
T
M
an
d
ML
P
m
o
d
el
co
n
f
i
g
u
r
atio
n
s
to
p
r
o
d
u
ce
a
h
ig
h
le
v
el
o
f
p
r
ed
ictio
n
ac
cu
r
a
c
y
.
T
h
e
f
in
al
s
tep
is
to
ev
a
lu
ate
t
h
e
L
ST
M
an
d
ML
P
m
o
d
els
t
o
m
ea
s
u
r
e
t
h
e
ac
cu
r
ac
y
p
er
f
o
r
m
an
ce
o
f
b
o
th
alg
o
r
it
h
m
s
.
T
h
e
m
o
d
els
ar
e
test
ed
u
s
in
g
p
r
ev
io
u
s
l
y
u
n
s
ee
n
d
ata.
P
r
ed
icted
r
esu
lts
ar
e
co
m
p
ar
ed
w
it
h
ac
tu
al
v
al
u
es
u
s
in
g
e
v
al
u
atio
n
m
etr
ic
s
.
Fi
g
u
r
e
1
s
h
o
w
s
a
f
lo
w
c
h
ar
t
o
f
th
e
ev
al
u
atio
n
o
f
th
e
L
ST
M
an
d
M
L
P
m
o
d
el
s
in
p
r
ed
ictin
g
P
M2
.
5
in
teg
r
ated
w
ith
A
W
S
m
eteo
r
o
lo
g
ical
d
ata.
F
ig
u
r
e
1
.
Flo
w
c
h
ar
t r
esear
ch
B
ased
o
n
Fig
u
r
e
1
,
d
ata
co
llectio
n
w
as c
ar
r
ied
o
u
t b
y
d
o
w
n
l
o
ad
in
g
w
ea
t
h
er
an
d
air
q
u
a
lit
y
d
ata
f
r
o
m
A
W
S
lo
g
g
er
s
an
d
eq
u
ip
m
e
n
t
at
th
e
C
ite
k
o
Me
teo
r
o
lo
g
ical
Statio
n
.
T
h
e
d
ataset
u
tili
ze
d
i
n
t
h
is
s
tu
d
y
s
p
a
n
s
a
n
o
b
s
er
v
atio
n
p
er
io
d
f
r
o
m
J
an
u
ar
y
1
to
J
u
l
y
2
0
,
2
0
2
3
.
Data
p
r
ep
r
o
ce
s
s
in
g
w
as
ca
r
r
ied
o
u
t
b
y
s
y
n
c
h
r
o
n
izi
n
g
d
ata
an
d
r
em
o
v
i
n
g
o
u
tlier
s
.
Data
s
y
n
ch
r
o
n
izatio
n
w
as
ca
r
r
ied
o
u
t
b
y
m
atc
h
in
g
th
e
s
a
m
p
lin
g
i
n
ter
v
al
s
o
f
m
eteo
r
o
lo
g
ical
d
ata,
P
M
1
0
,
an
d
P
M2
.
5
ev
er
y
3
h
o
u
r
s
to
en
s
u
r
e
a
co
n
s
is
te
n
t
te
m
p
o
r
al
r
eso
lu
tio
n
.
Ou
tlier
ch
ec
k
i
n
g
w
a
s
ca
r
r
ied
o
u
t
u
s
in
g
t
h
e
r
a
n
g
e
ch
ec
k
an
d
s
tep
ch
ec
k
m
et
h
o
d
s
[
1
3
]
.
E
m
p
t
y
d
ata
an
d
o
u
tlier
s
w
er
e
i
m
p
u
ted
u
s
i
n
g
MI
C
E
[
1
4
]
.
MI
C
E
is
co
n
s
id
er
ed
to
h
av
e
ad
v
a
n
tag
e
s
i
n
r
ep
r
esen
ti
n
g
m
u
lti
v
a
r
iate
d
ata,
r
ed
u
cin
g
b
ias an
d
p
r
o
v
id
in
g
m
o
r
e
s
tab
l
e
d
ata
d
is
tr
ib
u
tio
n
af
ter
i
m
p
u
t
atio
n
[
1
5
]
.
Af
ter
t
h
e
d
ata
w
a
s
f
illed
in
t
e
m
p
o
r
all
y
,
th
e
P
M2
.
5
p
r
ed
ict
o
r
w
as
d
eter
m
i
n
ed
an
d
t
h
e
d
ataset
w
a
s
s
p
lit,
allo
ca
tin
g
8
0
%
f
o
r
tr
ain
in
g
d
ata
an
d
2
0
%
f
o
r
test
i
n
g
d
ata.
T
h
e
m
o
d
el
co
n
f
i
g
u
r
atio
n
s
w
er
e
d
etailed
to
en
s
u
r
e
r
ep
r
o
d
u
cib
ilit
y
.
Sp
ec
if
icall
y
,
t
h
e
L
ST
M
ar
ch
itectu
r
e
co
n
s
is
t
s
o
f
t
w
o
h
id
d
en
la
y
er
s
w
it
h
6
4
an
d
3
2
n
eu
r
o
n
s
,
r
esp
ec
tiv
el
y
.
T
h
e
ML
P
m
o
d
el
w
a
s
d
esig
n
ed
with
a
s
i
m
ilar
f
u
ll
y
co
n
n
ec
ted
s
tr
u
ct
u
r
e
f
o
r
a
f
air
co
m
p
ar
is
o
n
.
B
o
th
m
o
d
els
u
til
ized
th
e
A
d
a
m
o
p
ti
m
izer
w
it
h
a
lear
n
i
n
g
r
ate
o
f
0
.
0
0
1
,
a
b
a
tch
s
ize
o
f
3
2
,
an
d
w
er
e
tr
ain
ed
f
o
r
a
m
ax
i
m
u
m
o
f
1
0
0
ep
o
ch
s
.
T
h
e
L
ST
M
m
o
d
el
h
y
p
er
p
ar
am
eter
s
w
er
e
f
i
n
e
-
tu
n
ed
u
s
in
g
t
h
e
ea
r
l
y
s
to
p
p
in
g
m
et
h
o
d
to
p
r
ev
en
t
o
v
er
f
itti
n
g
.
T
o
en
s
u
r
e
r
o
b
u
s
tn
es
s
a
n
d
s
tr
ict
r
ep
r
o
d
u
c
ib
ilit
y
,
t
h
e
m
o
d
el
tr
ain
i
n
g
p
r
o
ce
s
s
w
as
ex
ec
u
ted
ac
r
o
s
s
f
iv
e
in
d
ep
en
d
e
n
t
r
u
n
s
u
s
i
n
g
f
i
x
ed
r
an
d
o
m
s
ee
d
s
in
b
o
th
th
e
Nu
m
P
y
an
d
T
en
s
o
r
Flo
w
lib
r
ar
ies.
T
h
e
r
ep
o
r
ted
ev
alu
atio
n
m
etr
ics
r
ep
r
esen
t
th
e
av
er
ag
e
p
er
f
o
r
m
an
ce
ac
r
o
s
s
th
e
s
e
m
u
ltip
le
r
u
n
s
,
th
er
eb
y
v
al
id
at
in
g
t
h
e
s
tab
ilit
y
o
f
th
e
m
o
d
els
u
n
d
er
v
ar
y
i
n
g
in
it
ializatio
n
s
.
Af
ter
o
b
tain
in
g
th
e
o
p
tim
a
l h
y
p
er
p
ar
a
m
e
ter
s
,
th
e
m
o
d
el
s
w
er
e
p
r
o
ce
s
s
ed
f
o
r
ev
alu
atio
n
.
Mo
d
el
p
er
f
o
r
m
an
ce
w
a
s
q
u
an
tif
ied
u
s
in
g
t
h
r
ee
s
tan
d
ar
d
s
ta
tis
tical
m
etr
ics:
R
MSE
,
m
ea
n
ab
s
o
lu
te
p
er
ce
n
tag
e
er
r
o
r
(
MA
P
E
)
,
an
d
th
e
co
e
f
f
ic
ien
t
o
f
d
eter
m
in
at
i
o
n
(
R
²)
.
R
M
SE
m
ea
s
u
r
es
t
h
e
av
er
ag
e
m
a
g
n
itu
d
e
o
f
p
r
ed
ictio
n
er
r
o
r
s
,
g
iv
in
g
h
ig
h
er
p
en
alt
y
to
lar
g
e
d
ev
iati
o
n
s
.
MA
P
E
ass
es
s
es
f
o
r
ec
ast
in
g
ac
c
u
r
ac
y
as
an
in
t
u
iti
v
e
p
er
ce
n
ta
g
e
er
r
o
r
.
Fin
all
y
,
R
²
ev
a
lu
ate
s
t
h
e
p
r
o
p
o
r
ti
o
n
o
f
v
ar
ian
ce
in
P
M2
.
5
co
n
ce
n
tr
atio
n
s
ex
p
lai
n
ed
b
y
t
h
e
i
n
p
u
t
f
ea
t
u
r
es,
r
ef
lecti
n
g
th
e
m
o
d
el’
s
ab
ilit
y
to
ca
p
tu
r
e
g
en
er
al
d
y
n
a
m
ic
tr
en
d
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
F
o
r
ec
a
s
tin
g
p
a
r
ticu
la
te
ma
tte
r
co
n
ce
n
tr
a
tio
n
u
s
in
g
d
ee
p
le
a
r
n
in
g
a
n
d
w
ea
th
er d
a
ta
(
S
a
n
to
s
o
S
o
ek
ir
n
o
)
1647
2
=
1
−
∑
(
1
−
̂
=
1
)
2
∑
(
̅
̅
̅
−
̂
=
1
)
2
(
1
)
=
√
∑
(
̂
−
)
2
=
1
(
2
)
=
1
∑
(
−
̂
)
2
=
1
(
3
)
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
is
s
ec
tio
n
p
r
ese
n
ts
t
h
e
r
es
u
lts
o
f
t
h
e
P
M2
.
5
co
n
ce
n
tr
atio
n
p
r
ed
ictio
n
s
y
s
te
m
i
m
p
le
m
e
n
tatio
n
a
n
d
an
i
n
-
d
ep
th
an
al
y
s
is
o
f
t
h
e
p
er
f
o
r
m
an
ce
o
f
t
h
e
d
ev
el
o
p
ed
m
o
d
el.
T
h
e
d
is
cu
s
s
io
n
b
eg
i
n
s
w
it
h
t
h
e
tr
an
s
f
o
r
m
atio
n
o
f
r
a
w
d
a
ta
in
to
m
ac
h
i
n
e
lear
n
i
n
g
-
r
ea
d
y
f
e
atu
r
es,
f
o
llo
w
ed
b
y
a
co
m
p
ar
ativ
e
ev
al
u
atio
n
o
f
L
ST
M,
ML
P
,
an
d
h
y
b
r
id
L
S
T
M
-
ML
P
m
o
d
els,
an
d
co
n
c
lu
d
es
w
i
th
a
d
i
s
cu
s
s
io
n
o
f
t
h
e
i
m
p
licatio
n
s
o
f
th
e
f
i
n
d
in
g
s
f
o
r
u
n
d
er
s
ta
n
d
i
n
g
air
p
o
llu
tio
n
d
y
n
a
m
ics b
ased
o
n
t
h
e
r
esear
ch
co
n
d
u
cted
.
3
.
1
.
P
re
pro
ce
s
s
ing
da
t
a
Me
teo
r
o
lo
g
ical
d
ata
s
o
u
r
ce
d
f
r
o
m
A
W
S
co
n
ta
in
s
n
o
is
e,
o
u
t
lier
s
,
an
d
te
m
p
o
r
al
in
co
n
s
is
te
n
cies
t
h
a
t
ca
n
i
n
ter
f
er
e
w
i
th
th
e
m
o
d
el’
s
g
e
n
er
aliza
tio
n
p
r
o
ce
s
s
.
T
h
er
ef
o
r
e,
d
ata
p
r
ep
r
o
ce
s
s
in
g
is
ess
e
n
tial,
i
n
cl
u
d
in
g
d
ata
clea
n
in
g
,
h
an
d
li
n
g
m
i
s
s
i
n
g
d
ata,
co
r
r
elatio
n
an
al
y
s
is
f
o
r
f
ea
tu
r
e
s
elec
tio
n
,
an
d
d
ata
s
ca
le
n
o
r
m
aliza
tio
n
[
1
6
]
.
T
h
e
f
ir
s
t
s
tep
is
d
ata
in
teg
r
atio
n
an
d
clea
n
i
n
g
.
T
h
e
r
aw
d
ataset
co
n
s
is
ts
o
f
w
ea
t
h
er
p
ar
a
m
eter
s
at
h
o
u
r
l
y
r
eso
lu
tio
n
a
n
d
air
p
ar
ticle
co
n
ce
n
tr
atio
n
s
at
th
r
ee
-
h
o
u
r
r
eso
lu
tio
n
.
T
h
e
w
ea
th
er
d
ata
is
t
h
e
n
alig
n
ed
to
th
r
ee
-
h
o
u
r
l
y
r
eso
lu
tio
n
to
en
s
u
r
e
co
n
s
i
s
te
n
c
y
w
i
th
t
h
e
p
ar
ticu
la
te
d
ata.
A
f
ter
w
ar
d
,
q
u
alit
y
co
n
tr
o
l
(
QC
)
p
r
o
ce
d
u
r
es
ar
e
p
er
f
o
r
m
ed
a
s
d
escr
ib
ed
in
th
e
m
et
h
o
d
s
s
ec
tio
n
.
T
h
e
o
u
tl
ier
d
etec
tio
n
u
tili
ze
d
s
tr
i
ct
p
h
y
s
ical
th
r
e
s
h
o
ld
s
to
en
s
u
r
e
d
ata
v
alid
it
y
.
Fo
r
m
et
eo
r
o
lo
g
ical
p
ar
am
eter
s
,
d
ata
p
o
in
ts
o
u
t
s
id
e
th
e
b
o
u
n
d
s
o
f
1
5
°C
to
4
5
°C
f
o
r
te
m
p
er
atu
r
e
an
d
3
0
%
to
1
0
0
%
f
o
r
r
elativ
e
h
u
m
id
it
y
w
er
e
f
la
g
g
ed
as
a
n
o
m
alie
s
an
d
r
e
m
o
v
ed
.
Fo
llo
w
i
n
g
t
h
e
r
e
m
o
v
al,
t
h
e
MI
C
E
i
m
p
u
tati
o
n
v
alid
atio
n
w
as
co
n
d
u
cted
b
y
co
m
p
ar
i
n
g
th
e
s
tatis
tical
d
is
tr
ib
u
tio
n
o
f
th
e
d
ataset
b
ef
o
r
e
an
d
af
ter
im
p
u
tatio
n
to
en
s
u
r
e
th
e
s
tr
u
ct
u
r
al
in
teg
r
it
y
o
f
th
e
ti
m
e
-
s
er
i
es
w
as
m
ain
tain
e
d
w
it
h
o
u
t
i
n
tr
o
d
u
cin
g
ar
ti
f
icial
b
ias.
T
h
e
v
is
u
aliza
t
io
n
i
n
F
ig
u
r
e
2
s
h
o
w
s
t
h
e
p
lo
t
o
f
te
m
p
e
r
atu
r
e
an
d
h
u
m
id
it
y
af
ter
QC
,
t
h
e
th
r
ee
-
h
o
u
r
av
er
a
g
e
te
m
p
er
at
u
r
e,
an
d
P
M2
.
5
c
o
n
ce
n
tr
at
io
n
t
h
r
o
u
g
h
o
u
t t
h
e
o
b
s
er
v
atio
n
p
er
io
d
.
Fig
u
r
e
2
.
P
lo
ttin
g
d
ata
h
u
m
id
it
y
,
te
m
p
er
at
u
r
e,
P
M1
0
,
an
d
PM2
.
5
B
ased
o
n
th
is
v
i
s
u
a
lizatio
n
,
t
h
e
d
ail
y
f
lu
ct
u
atio
n
d
y
n
a
m
ic
s
ar
e
v
is
ib
le
w
i
th
s
e
v
er
al
g
ap
s
in
m
is
s
in
g
v
alu
e
s
af
ter
d
ata
clea
n
in
g
an
d
QC
.
T
o
h
an
d
le
th
is
m
i
s
s
i
n
g
d
ata,
th
e
im
p
u
tatio
n
tec
h
n
i
q
u
e
u
s
i
n
g
t
h
e
MI
C
E
alg
o
r
ith
m
w
a
s
i
m
p
le
m
en
ted
d
u
e
to
it
s
ab
ilit
y
to
u
til
ize
th
e
m
u
lti
v
ar
iate
co
r
r
elatio
n
s
tr
u
ct
u
r
e
b
et
w
ee
n
at
m
o
s
p
h
er
ic
p
ar
a
m
eter
s
to
esti
m
ate
t
h
e
clo
s
est
v
al
u
es,
th
u
s
m
ai
n
tai
n
in
g
s
tati
s
tical
d
is
tr
ib
u
tio
n
a
n
d
m
i
n
i
m
izi
n
g
b
ias
i
n
th
e
i
m
p
u
t
atio
n
p
r
o
ce
s
s
[
1
7
]
.
T
h
e
n
ex
t
s
tag
e,
f
ea
tu
r
e
s
elec
t
io
n
,
is
ca
r
r
ied
o
u
t
to
id
en
tify
in
p
u
t
p
ar
a
m
eter
s
t
h
at
h
a
v
e
a
p
o
s
i
tiv
e
co
n
tr
ib
u
tio
n
to
th
e
p
ar
a
m
eter
s
to
b
e
p
r
ed
icted
.
T
h
i
s
is
cr
u
cial
b
ec
au
s
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
2
4
,
No
.
5
,
Octo
b
e
r
2
0
2
6
:
1
6
4
4
-
1655
1648
ir
r
elev
an
t
p
ar
a
m
eter
s
ca
n
a
ctu
all
y
i
n
cr
ea
s
e
t
h
e
co
m
p
u
t
atio
n
al
b
u
r
d
en
a
n
d
t
h
e
r
is
k
o
f
o
v
er
f
itti
n
g
o
r
u
n
d
er
f
itti
n
g
.
P
ea
r
s
o
n
co
r
r
elati
o
n
m
atr
i
x
a
n
al
y
s
is
w
as
p
er
f
o
r
m
ed
to
d
eter
m
i
n
e
th
e
o
p
ti
m
al
in
p
u
t
p
ar
a
m
eter
s
,
w
it
h
t
h
e
r
esu
lts
s
h
o
w
n
in
Fig
u
r
e
3
.
Fig
u
r
e
3
.
P
ea
r
s
o
n
co
r
r
elatio
n
m
atr
i
x
B
ased
o
n
Fig
u
r
e
3
,
th
e
co
r
r
elatio
n
an
al
y
s
i
s
s
h
o
w
s
a
s
tr
o
n
g
s
tatis
tical
r
elatio
n
s
h
ip
b
et
w
ee
n
th
e
tar
g
et
v
ar
iab
le
(
P
M2
.
5
)
an
d
ce
r
tain
p
ar
am
eter
s
.
B
ased
o
n
th
e
co
r
r
elatio
n
co
ef
f
icie
n
t
(
r
)
,
it
w
a
s
d
ec
id
ed
to
u
s
e
o
n
l
y
th
r
ee
m
ai
n
f
ea
t
u
r
es
as
p
r
ed
icto
r
s
:
P
M1
0
,
te
m
p
er
atu
r
e,
an
d
h
u
m
id
it
y
.
P
M1
0
s
h
o
w
s
a
s
tr
o
n
g
p
o
s
itiv
e
co
r
r
elatio
n
w
i
th
P
M2
.
5
,
in
li
n
e
w
it
h
t
h
e
f
i
n
d
in
g
s
o
f
Z
h
an
g
et
a
l
.
[
1
7
]
w
h
o
co
n
d
u
cted
a
s
h
ap
le
y
ad
d
itiv
e
ex
p
lan
atio
n
(
SH
A
P
)
an
al
y
s
i
s
o
f
P
M1
0
an
d
P
M2
.
5
,
w
h
ich
f
o
u
n
d
th
at
P
M1
0
is
th
e
d
o
m
i
n
an
t
p
r
ed
icto
r
f
o
r
P
M2
.
5
,
co
n
s
id
er
in
g
th
a
t
P
M2
.
5
is
a
p
h
y
s
ical
s
u
b
s
et
o
f
P
M1
0
w
it
h
s
i
m
i
lar
e
m
i
s
s
io
n
s
o
u
r
ce
s
an
d
d
is
p
er
s
io
n
p
atter
n
s
.
Me
an
w
h
ile,
te
m
p
er
a
tu
r
e
an
d
h
u
m
id
it
y
al
s
o
s
h
o
w
a
s
ig
n
i
f
ica
n
t
co
r
r
elatio
n
.
T
h
is
is
in
ac
co
r
d
an
ce
w
it
h
r
esear
c
h
b
y
I
s
tia
n
a
et
a
l
.
[
4
]
w
h
ich
h
i
g
h
l
ig
h
t
s
th
e
i
m
p
o
r
ta
n
ce
o
f
t
h
e
in
ter
ac
tio
n
b
et
w
ee
n
th
e
s
e
t
w
o
v
ar
iab
les
in
tr
o
p
ical
r
eg
io
n
s
.
T
h
e
s
tu
d
y
s
h
o
w
ed
th
at
te
m
p
er
atu
r
e
an
d
h
u
m
id
it
y
h
a
v
e
a
g
r
ea
ter
im
p
ac
t
o
n
p
o
llu
tan
t
ac
c
u
m
u
la
tio
n
t
h
an
r
ain
f
all.
C
o
n
v
er
s
el
y
,
o
th
er
v
ar
i
ab
les
w
er
e
eli
m
i
n
ated
f
r
o
m
t
h
e
in
p
u
t
lis
t
d
u
e
to
th
eir
v
er
y
w
ea
k
co
r
r
elatio
n
s
in
th
e
P
M2
.
5
d
ataset
in
th
i
s
s
tu
d
y
.
T
h
is
eli
m
i
n
atio
n
o
f
f
ea
tu
r
es
w
it
h
lo
w
co
r
r
elatio
n
s
ai
m
s
to
r
ed
u
ce
n
o
is
e
th
at
co
u
ld
co
n
f
u
s
e
t
h
e
m
o
d
el
in
lear
n
i
n
g
d
eter
m
i
n
is
t
ic
p
atter
n
s
[
1
8
]
.
Af
ter
f
ea
t
u
r
e
s
elec
tio
n
,
d
ata
n
o
r
m
al
izatio
n
w
a
s
p
er
f
o
r
m
ed
u
s
in
g
t
h
e
m
i
n
-
m
a
x
s
ca
lin
g
tec
h
n
iq
u
e.
T
h
e
r
a
w
d
ata
h
as
a
h
i
g
h
l
y
h
eter
o
g
en
eo
u
s
s
ca
le
r
an
g
e,
w
h
er
e
p
ar
ticu
late
co
n
ce
n
tr
at
io
n
s
ca
n
r
ea
ch
h
u
n
d
r
ed
s
o
f
µg
/
m
³,
w
h
ile
air
te
m
p
er
atu
r
es
r
an
g
e
in
th
e
ten
s
o
f
d
eg
r
ee
s
C
elsi
u
s
.
T
h
is
d
if
f
er
e
n
ce
in
s
ca
le
ca
u
s
es
th
e
m
o
d
el
g
r
ad
ien
t
to
o
s
cilla
te
an
d
s
lo
ws
co
n
v
er
g
e
n
ce
.
A
tr
an
s
f
o
r
m
a
t
io
n
w
as
ap
p
lied
to
m
ap
all
f
e
atu
r
es
to
t
h
e
[
0
-
1
]
r
an
g
e
to
en
s
u
r
e
th
at
ea
ch
f
ea
t
u
r
e
h
as
a
p
r
o
p
o
r
tio
n
al
w
ei
g
h
t
co
n
tr
ib
u
tio
n
in
t
h
e
m
o
d
el’
s
l
o
s
s
f
u
n
ctio
n
[
1
9
]
.
Fu
r
t
h
er
m
o
r
e,
th
is
n
o
r
m
al
izati
o
n
tech
n
iq
u
e
also
ai
m
s
to
m
ak
e
t
h
e
co
m
p
u
tatio
n
a
l
p
r
o
ce
s
s
a
n
d
u
s
e
o
f
th
e
m
o
d
el
’
s
m
e
m
o
r
y
ca
p
ac
it
y
m
o
r
e
ef
f
ic
ien
t.
T
h
e
m
o
d
el
co
n
f
i
g
u
r
atio
n
w
as
d
esig
n
ed
to
ca
p
tu
r
e
te
m
p
o
r
al
p
atter
n
s
in
a
m
u
lti
-
le
v
el
m
a
n
n
er
.
T
h
e
f
ir
s
t
la
y
er
u
s
ed
L
ST
M
an
d
M
L
P
w
ith
a
t
w
o
-
h
id
d
en
la
y
er
s
t
r
u
ctu
r
e.
B
ased
o
n
e
m
p
ir
ical
e
x
p
er
i
m
e
n
ts
,
t
h
e
b
est
h
y
p
er
p
ar
a
m
eter
s
w
er
e
o
b
tain
e
d
in
Fi
g
u
r
e
4
.
T
h
e
f
i
n
al
s
tep
i
n
d
ata
p
r
ep
r
o
ce
s
s
in
g
w
a
s
tr
an
s
f
o
r
m
in
g
t
h
e
ti
m
e
s
er
ies
in
to
a
f
o
r
m
at
s
u
itab
le
f
o
r
s
u
p
er
v
is
ed
lear
n
i
n
g
,
w
h
ich
w
a
s
p
er
f
o
r
m
ed
u
s
i
n
g
a
s
lid
i
n
g
w
i
n
d
o
w
tech
n
iq
u
e.
T
h
is
tech
n
iq
u
e
u
t
ilizes
h
i
s
to
r
ical
d
ata
o
v
er
a
s
p
ec
i
f
ic
t
i
m
e
p
er
io
d
(
lo
o
k
b
ac
k
)
to
co
n
s
tr
u
c
t
an
i
n
p
u
t
s
eq
u
en
ce
(
X)
th
at
is
u
s
ed
to
p
r
ed
ict
th
e
tar
g
et
v
al
u
e
(
y
)
at
t
h
e
n
ex
t
t
i
m
e
s
tep
[
2
0
]
.
I
n
th
is
s
t
u
d
y
,
th
e
s
lid
in
g
w
in
d
o
w
w
a
s
s
et
to
2
4
h
o
u
r
s
,
allo
w
in
g
t
h
e
m
o
d
el
to
u
tili
ze
in
f
o
r
m
atio
n
f
r
o
m
th
e
p
r
ev
io
u
s
2
4
h
o
u
r
s
to
p
r
o
j
ec
t
P
M
2
.
5
co
n
ce
n
tr
atio
n
s
o
n
e
h
o
u
r
in
to
th
e
f
u
t
u
r
e.
T
h
e
lo
o
k
b
ac
k
p
er
i
o
d
tak
es
in
to
ac
co
u
n
t
th
e
p
er
io
d
icit
y
o
f
P
M2
.
5
co
n
ce
n
tr
atio
n
lev
el
s
th
a
t r
ep
ea
t d
ail
y
.
T
h
e
s
lid
in
g
w
i
n
d
o
w
p
r
o
ce
s
s
ef
f
ec
tiv
e
l
y
tr
a
n
s
f
o
r
m
s
tab
u
lar
d
ata
in
to
3
D
ten
s
o
r
s
w
it
h
d
i
m
e
n
s
io
n
s
(
s
a
m
p
les,
ti
m
e
s
tep
s
,
f
ea
tu
r
es)
,
w
h
ic
h
ar
e
th
e
n
p
r
o
ce
s
s
ed
b
y
th
e
L
ST
M
ar
ch
itect
u
r
e
to
lear
n
h
ier
ar
c
h
ical
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
F
o
r
ec
a
s
tin
g
p
a
r
ticu
la
te
ma
tte
r
co
n
ce
n
tr
a
tio
n
u
s
in
g
d
ee
p
le
a
r
n
in
g
a
n
d
w
ea
th
er d
a
ta
(
S
a
n
to
s
o
S
o
ek
ir
n
o
)
1649
te
m
p
o
r
al
d
ep
en
d
en
cies.
Fo
r
ML
P
m
o
d
els,
th
e
in
p
u
t
te
n
s
o
r
s
ar
e
f
la
tten
ed
(
r
es
h
ap
ed
)
in
t
o
a
2
D
f
o
r
m
a
t
w
it
h
d
i
m
en
s
io
n
s
(
s
a
m
p
les,
t
i
m
e
s
te
p
s
x
f
ea
t
u
r
es)
to
b
e
co
m
p
ati
b
le
w
i
th
f
u
ll
y
co
n
n
ec
ted
la
y
er
s
.
T
h
e
u
s
e
o
f
t
h
e
s
lid
in
g
w
i
n
d
o
w
tec
h
n
iq
u
e
h
as
th
e
ad
v
a
n
tag
e
o
f
p
r
eser
v
i
n
g
t
h
e
te
m
p
o
r
al
r
elatio
n
s
h
ip
s
p
r
ese
n
t
i
n
p
o
llu
ta
n
t
d
ata
an
d
co
m
p
r
eh
e
n
s
i
v
el
y
u
tili
zi
n
g
h
is
to
r
ical
d
ata.
F
u
r
th
er
m
o
r
e,
th
is
ap
p
r
o
ac
h
also
p
r
ev
e
n
ts
d
ata
leak
a
g
e
b
y
en
s
u
r
in
g
a
clea
r
s
ep
ar
atio
n
b
et
w
ee
n
tr
ain
in
g
an
d
tes
t d
ata
b
ased
o
n
ch
r
o
n
o
lo
g
ical
o
r
d
er
[
2
1
]
.
Fig
u
r
e
4
.
Mo
d
el
ar
ch
itectu
r
e
3
.
2
.
E
v
a
lua
t
i
o
n L
ST
M
a
nd
M
L
P
m
o
de
l
Mo
d
el
p
er
f
o
r
m
a
n
ce
e
v
al
u
atio
n
w
a
s
co
n
d
u
c
ted
b
y
co
m
p
ar
i
n
g
th
e
p
r
ed
ictio
n
r
es
u
lts
to
a
test
s
et
r
ep
r
esen
tin
g
air
p
o
llu
tio
n
co
n
d
itio
n
s
th
e
m
o
d
el
h
ad
n
ev
er
e
n
co
u
n
ter
ed
b
ef
o
r
e.
Vis
u
al
an
a
l
y
s
i
s
w
as
p
er
f
o
r
m
ed
s
eq
u
en
tiall
y
,
s
tar
ti
n
g
w
i
th
th
e
h
y
b
r
id
,
L
ST
M,
a
n
d
M
L
P
m
o
d
els,
to
id
en
ti
f
y
p
atter
n
s
i
n
th
e
r
esp
o
n
s
i
v
en
e
s
s
o
f
ea
ch
ar
ch
itect
u
r
e
to
ac
tu
al
d
at
a
f
lu
c
tu
at
io
n
s
.
T
h
e
v
is
u
a
lizatio
n
i
n
Fi
g
u
r
e
5
d
em
o
n
s
tr
ates
t
h
e
v
er
y
s
o
l
id
p
er
f
o
r
m
a
n
ce
o
f
L
ST
M
m
o
d
el.
As
illu
s
tr
ated
b
y
th
e
ali
g
n
m
e
n
t
o
f
th
e
p
r
ed
ictio
n
an
d
ac
tu
al
cu
r
v
es
in
Fi
g
u
r
e
5
,
th
is
m
o
d
el
d
e
m
o
n
s
tr
ates
a
h
ig
h
ab
ilit
y
to
m
in
i
m
ize
th
e
d
i
f
f
er
en
ce
b
et
w
ee
n
p
r
ed
icted
an
d
ac
tu
al
v
alu
e
s
,
esp
ec
iall
y
at
f
ast
ti
m
e
s
tep
s
.
T
h
e
s
h
o
r
t
-
an
d
lo
n
g
-
ter
m
m
e
m
o
r
y
m
ec
h
an
i
s
m
s
in
th
e
L
ST
M
allo
w
it
to
m
e
m
o
r
ize
h
i
s
to
r
ical
p
atter
n
s
m
o
r
e
ef
f
ec
tiv
e
l
y
t
h
a
n
o
th
er
m
o
d
els,
r
esu
lti
n
g
in
a
p
r
ed
ictio
n
cu
r
v
e
th
at
is
q
u
ite
p
r
ec
is
e
w
it
h
th
e
a
ctu
al
d
ata.
I
n
co
n
tr
ast,
t
h
e
M
L
P
m
o
d
el
ex
h
ib
it
s
a
d
if
f
er
e
n
t
tr
en
d
,
as
s
h
o
w
n
in
Fig
u
r
e
6
.
Alth
o
u
g
h
it
s
p
r
ed
ictio
n
lin
e
f
o
llo
w
s
t
h
e
u
p
-
an
d
-
d
o
w
n
tr
en
d
v
er
y
w
ell,
th
er
e
is
a
n
o
ti
ce
ab
le
m
a
g
n
it
u
d
e
b
ias
in
s
o
m
e
s
eg
m
en
t
s
,
w
h
er
e
th
e
m
o
d
el
s
y
s
te
m
atica
ll
y
p
r
ed
icts
s
l
ig
h
tl
y
h
ig
h
er
o
r
lo
w
er
v
alu
es
(
o
f
f
s
et)
.
T
h
is
is
u
n
d
er
s
ta
n
d
ab
le
co
n
s
id
er
in
g
th
at
M
L
P
s
lac
k
a
r
ec
u
r
r
en
t
m
ec
h
a
n
i
s
m
to
ca
p
tu
r
e
s
eq
u
e
n
tial
d
ep
en
d
e
n
cies
in
tr
i
n
s
ical
l
y
,
i
n
s
tead
tr
ea
ti
n
g
s
lid
in
g
w
i
n
d
o
w
i
n
p
u
t
s
as s
tati
c
f
ea
tu
r
e
s
[
1
7
]
.
Fig
u
r
e
5
.
L
ST
M
p
r
ed
ictio
n
v
s
ac
tu
al
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
2
4
,
No
.
5
,
Octo
b
e
r
2
0
2
6
:
1
6
4
4
-
1655
1650
Fig
u
r
e
6
.
ML
P
p
r
ed
ictio
n
v
s
a
ctu
al
Fu
r
t
h
er
m
o
r
e,
t
h
e
h
y
b
r
id
m
o
d
el
ev
al
u
atio
n
is
v
is
u
alize
d
i
n
Fig
u
r
e
7
.
B
ased
o
n
Fi
g
u
r
e
7
,
th
e
h
y
b
r
id
m
o
d
el
d
e
m
o
n
s
tr
ates
g
o
o
d
g
en
er
aliza
tio
n
,
w
it
h
th
e
p
r
ed
icted
lin
e
f
o
llo
w
i
n
g
th
e
g
e
n
er
al
p
a
tter
n
o
f
th
e
ac
tu
al
lin
e.
Ho
w
e
v
er
,
th
er
e
ar
e
s
li
g
h
t
d
ev
iat
io
n
s
a
t
s
o
m
e
ex
tr
e
m
e
p
ea
k
s
,
w
h
er
e
t
h
e
m
o
d
el
ap
p
ea
r
s
to
s
tr
u
g
g
le
to
ca
p
tu
r
e
s
u
d
d
en
f
l
u
ct
u
atio
n
s
w
it
h
f
u
ll
p
r
ec
is
io
n
.
T
h
is
in
d
icate
s
th
at
co
m
b
i
n
i
n
g
t
h
e
L
ST
M
an
d
ML
P
ar
ch
itect
u
r
es
as
a
h
y
b
r
id
m
o
d
el
f
ac
es
co
m
p
le
x
p
ar
a
m
eter
o
p
tim
izatio
n
i
s
s
u
es,
r
es
u
lti
n
g
in
its
p
er
f
o
r
m
an
ce
b
ein
g
lo
w
er
th
a
n
t
h
at
o
f
t
h
e
s
in
g
le
L
ST
M
m
o
d
el
i
n
t
h
i
s
s
t
u
d
y
.
T
o
ev
al
u
ate
t
h
e
o
v
er
all
m
o
d
el
p
er
f
o
r
m
a
n
ce
q
u
an
tita
tiv
e
l
y
,
t
h
e
ev
al
u
atio
n
m
atr
i
x
as s
h
o
w
n
i
n
T
ab
le
1
w
a
s
u
til
ized
.
Fig
u
r
e
7
.
H
y
b
r
id
L
ST
M
-
M
L
P
p
r
ed
ictio
n
v
s
ac
tu
a
l
T
ab
le
1
.
Mo
d
el’
s
m
etr
ic
e
v
al
u
atio
n
M
o
d
e
l
M
e
t
r
i
c
e
v
a
l
u
a
t
i
o
n
R
M
S
E
M
A
P
E
R
2
L
S
T
M
6
.
8
6
2
3
7
.
3
8
4
0
.
8
0
4
M
L
P
7
.
0
0
4
3
3
.
8
8
1
0
.
8
1
0
H
y
b
r
i
d
6
.
9
2
1
3
6
.
3
2
8
0
.
7
9
8
I
n
ter
m
s
o
f
R
M
SE,
T
ab
le
1
s
h
o
w
s
t
h
at
L
ST
M
ac
h
ie
v
ed
th
e
b
est
p
er
f
o
r
m
an
ce
w
i
th
a
v
al
u
e
o
f
6
.
8
6
2
,
w
h
ic
h
i
s
t
h
e
lo
w
es
t
a
m
o
n
g
t
h
e
t
h
r
ee
m
o
d
el
s
s
t
u
d
ied
.
T
h
i
s
i
n
d
icate
s
t
h
at
L
ST
M
h
as
th
e
s
m
allest
av
er
a
g
e
ab
s
o
lu
te
er
r
o
r
d
ev
iatio
n
in
th
e
o
r
ig
in
al
u
n
its
o
f
p
o
llu
ta
n
t
co
n
ce
n
tr
atio
n
,
w
h
ic
h
m
ea
n
s
t
h
is
m
o
d
el
p
r
o
d
u
ce
s
th
e
m
o
s
t
ac
c
u
r
ate
p
r
ed
ictio
n
s
a
n
d
w
i
th
m
i
n
i
m
al
er
r
o
r
s
co
m
p
ar
e
d
to
th
e
h
y
b
r
id
m
o
d
el
(
6
.
9
2
1
)
an
d
M
L
P
(
7
.
0
0
4
)
.
T
h
is
lo
w
R
M
SE
v
a
lu
e
in
d
icat
es
th
a
t
L
ST
M
is
m
o
r
e
ef
f
ec
ti
v
e
i
n
h
an
d
li
n
g
d
ata
v
ar
iab
ili
t
y
an
d
p
r
o
v
id
es
m
o
r
e
co
n
s
is
ten
t
r
e
s
u
l
ts
i
n
p
r
ed
ictin
g
P
M2
.
5
co
n
ce
n
tr
atio
n
s
.
Fu
r
t
h
er
m
o
r
e,
in
ter
m
s
o
f
M
SE,
L
ST
M
ag
ain
e
x
ce
l
s
w
it
h
a
v
al
u
e
o
f
4
7
.
0
8
1
,
w
h
ic
h
is
lo
w
er
th
a
n
h
y
b
r
id
(
4
7
.
9
0
1
)
an
d
ML
P
(
4
9
.
0
5
0
)
.
MSE
g
iv
es
g
r
ea
ter
w
ei
g
h
t
to
lar
g
e
p
r
ed
ict
io
n
er
r
o
r
s
,
s
o
a
lo
w
er
MSE
v
al
u
e
f
o
r
L
ST
M
in
d
icate
s
th
at
th
i
s
m
o
d
el
p
r
o
d
u
ce
s
ex
tr
e
m
e
p
r
ed
ictio
n
er
r
o
r
s
less
f
r
eq
u
en
t
l
y
.
L
ST
M
m
o
d
els
ar
e
b
etter
at
m
i
n
i
m
izi
n
g
lar
g
e
er
r
o
r
s
,
w
h
ic
h
in
tu
r
n
in
cr
ea
s
e
s
p
r
ed
ictio
n
r
eliab
ilit
y
i
n
s
i
tu
ati
o
n
s
w
it
h
h
ig
h
d
ata
v
ar
iatio
n
.
M
L
P
a
n
d
h
y
b
r
i
d
,
a
l
th
o
u
g
h
p
r
o
d
u
c
in
g
g
o
o
d
r
e
s
u
lt
s
,
s
h
o
w
l
a
r
g
e
r
e
r
r
o
r
s
i
n
s
o
m
e
ex
t
r
em
e
p
r
e
d
ic
t
i
o
n
s
,
w
h
i
ch
is
r
ef
l
ec
t
e
d
in
th
e
h
ig
h
e
r
MS
E
v
a
lu
es
i
n
T
a
b
l
e
1
.
Fin
all
y
,
f
o
r
th
e
co
ef
f
icien
t
o
f
d
eter
m
i
n
atio
n
(
R
2
)
in
T
ab
le
1
,
ML
P
h
as
a
s
lig
h
t
ad
v
an
ta
g
e
w
i
th
a
v
alu
e
o
f
0
.
8
1
0
,
h
ig
h
er
th
a
n
L
S
T
M
(
0
.
8
0
4
)
an
d
h
y
b
r
id
(
0
.
7
9
8
)
.
T
h
is
R
2
v
a
lu
e
i
n
d
icate
s
t
h
at
ML
P
is
b
etter
ab
le
to
f
o
llo
w
t
h
e
u
p
s
a
n
d
d
o
w
n
s
o
f
th
e
d
ata,
r
ef
lectin
g
its
ab
ilit
y
to
ca
p
tu
r
e
d
ata
v
ar
iab
ili
t
y
o
r
f
lu
ct
u
atio
n
s
.
Ho
w
e
v
er
,
th
e
s
li
g
h
tl
y
h
ig
h
er
R
2
v
al
u
e
o
f
M
L
P
is
n
o
t
en
o
u
g
h
to
o
v
er
co
m
e
L
ST
M’
s
s
u
p
er
io
r
ity
i
n
ter
m
s
o
f
ab
s
o
lu
te
p
r
ed
ictio
n
ac
c
u
r
ac
y
,
as
r
ef
lecte
d
i
n
t
h
e
lo
w
er
R
M
SE
an
d
MSE
v
al
u
es.
T
h
er
ef
o
r
e,
alth
o
u
g
h
M
L
P
is
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
F
o
r
ec
a
s
tin
g
p
a
r
ticu
la
te
ma
tte
r
co
n
ce
n
tr
a
tio
n
u
s
in
g
d
ee
p
le
a
r
n
in
g
a
n
d
w
ea
th
er d
a
ta
(
S
a
n
to
s
o
S
o
ek
ir
n
o
)
1651
b
etter
at
m
atc
h
i
n
g
d
ata
p
atter
n
s
,
L
ST
M
is
s
u
p
er
io
r
in
p
r
ed
ictin
g
m
o
r
e
ac
c
u
r
ate
v
alu
e
s
o
v
er
al
l.
Ov
er
all,
L
ST
M
w
a
s
th
e
b
est
m
o
d
el
in
th
is
s
t
u
d
y
.
A
lt
h
o
u
g
h
M
L
P
h
ad
a
m
ar
g
in
a
l
ad
v
an
ta
g
e
in
th
e
co
r
r
elatio
n
co
ef
f
icie
n
t
(
R
2
)
,
th
e
d
if
f
er
en
ce
w
a
s
n
o
t
p
r
ac
ticall
y
s
ig
n
i
f
ica
n
t.
I
n
th
e
co
n
tex
t
o
f
an
air
q
u
alit
y
ea
r
l
y
w
ar
n
i
n
g
s
y
s
te
m
,
m
i
n
i
m
izi
n
g
t
h
e
m
a
g
n
itu
d
e
o
f
p
r
ed
ictio
n
er
r
o
r
s
(
R
MSE
an
d
MSE
)
is
f
ar
m
o
r
e
cr
itical
th
an
s
i
m
p
l
y
tr
en
d
co
r
r
elatio
n
.
T
h
e
s
u
p
er
io
r
ity
o
f
L
ST
M
in
m
i
n
i
m
izi
n
g
th
e
s
e
er
r
o
r
s
alig
n
s
w
ith
t
h
e
s
tu
d
y
b
y
Ki
m
et
a
l
.
[
1
9
]
,
w
h
ic
h
e
m
p
h
as
ized
th
at
r
ec
u
r
r
en
t
n
e
u
r
al
n
et
w
o
r
k
ar
c
h
itec
tu
r
es
ar
e
b
etter
ab
le
to
ca
p
t
u
r
e
th
e
s
to
c
h
asti
c
d
y
n
a
m
ics o
f
air
p
o
llu
tan
ts
.
3
.
3
.
Dis
cus
s
io
n
T
h
is
s
tu
d
y
f
o
u
n
d
th
e
L
ST
M
ar
ch
itectu
r
e
to
b
e
s
u
p
er
io
r
in
m
i
n
i
m
izi
n
g
P
M2
.
5
p
r
ed
ict
io
n
er
r
o
r
s
co
m
p
ar
ed
to
ML
P
an
d
h
y
b
r
i
d
m
o
d
els.
T
h
eo
r
etica
ll
y
,
t
h
is
s
u
p
er
io
r
it
y
i
s
b
ased
o
n
th
e
L
ST
M’
s
in
ter
n
al
ce
ll
s
tate
m
ec
h
a
n
is
m
,
w
h
ich
ca
n
s
t
o
r
e
r
elev
an
t
in
f
o
r
m
atio
n
lo
n
g
-
ter
m
an
d
f
o
r
g
e
tti
n
g
u
n
n
ec
es
s
a
r
y
n
o
is
e
t
h
r
o
u
g
h
a
f
o
r
g
et
g
ate.
T
h
is
ca
p
ab
ilit
y
i
s
cr
u
cial
i
n
th
e
co
n
tex
t
o
f
ai
r
p
o
llu
tio
n
d
ata,
w
h
ic
h
is
ti
m
e
-
s
er
ies
w
i
th
h
i
g
h
au
to
co
r
r
elatio
n
.
R
ec
u
r
r
en
ce
-
b
ased
d
ee
p
lear
n
in
g
ap
p
r
o
ac
h
es
s
u
ch
a
s
L
ST
M
ar
e
in
h
er
e
n
t
l
y
b
etter
s
u
ited
to
ca
p
tu
r
in
g
n
o
n
-
li
n
ea
r
te
m
p
o
r
al
d
y
n
a
m
ics
t
h
an
M
L
P
[
2
1
]
,
w
h
ich
tr
ea
t
s
l
id
in
g
w
in
d
o
w
in
p
u
ts
as
s
tatic
f
ea
t
u
r
e
v
ec
to
r
s
w
it
h
o
u
t
a
n
ex
p
licit
ti
m
e
d
i
m
e
n
s
io
n
[
2
2
]
.
T
h
is
p
h
en
o
m
e
n
o
n
e
x
p
lai
n
s
w
h
y
M
L
P
s
,
d
esp
ite
h
av
i
n
g
h
i
g
h
co
r
r
elatio
n
,
f
ail
to
ac
h
ie
v
e
th
e
s
a
m
e
le
v
el
o
f
ab
s
o
l
u
te
p
r
ec
is
i
o
n
as L
ST
Ms.
T
h
e
o
b
s
er
v
atio
n
t
h
at
t
h
e
s
i
n
g
le
L
ST
M
ar
ch
itec
tu
r
e
o
u
tp
er
f
o
r
m
ed
t
h
e
h
y
b
r
id
L
ST
M
-
M
L
P
m
o
d
el
p
r
o
v
id
es
v
alu
ab
le
in
s
i
g
h
ts
in
to
n
eu
r
al
n
et
w
o
r
k
d
esi
g
n
f
o
r
lo
ca
lized
en
v
ir
o
n
m
e
n
tal
ti
m
e
s
er
ies.
A
lt
h
o
u
g
h
h
y
b
r
id
m
o
d
els
co
n
ce
p
t
u
all
y
a
tte
m
p
t
to
lev
er
a
g
e
th
e
s
eq
u
e
n
tial
m
e
m
o
r
y
o
f
L
ST
Ms
alo
n
g
s
id
e
th
e
n
o
n
-
li
n
ea
r
m
ap
p
in
g
ca
p
ab
ilit
ies
o
f
f
u
ll
y
co
n
n
ec
ted
la
y
er
s
,
o
u
r
r
es
u
lt
s
i
n
d
icate
a
s
tab
ilit
y
an
d
o
p
ti
m
iz
atio
n
tr
ad
e
-
o
f
f
.
W
e
attr
ib
u
te
th
e
h
y
b
r
id
m
o
d
el’
s
l
o
w
er
p
er
f
o
r
m
an
ce
to
ar
ch
itec
t
u
r
al
o
v
er
-
p
ar
a
m
e
ter
izatio
n
.
I
n
a
lo
ca
lized
d
ataset
w
it
h
a
2
4
-
h
o
u
r
lo
o
k
b
ac
k
w
in
d
o
w
,
ad
d
in
g
d
en
s
e
f
ee
d
-
f
o
r
w
ar
d
lay
er
s
af
ter
r
ec
u
r
r
en
t
lay
er
s
i
n
tr
o
d
u
ce
s
r
ed
u
n
d
an
t
w
ei
g
h
t
m
atr
ice
s
th
at
d
is
to
r
t
th
e
te
m
p
o
r
al
g
r
ad
ien
ts
lear
n
ed
b
y
t
h
e
L
ST
M
ce
lls
.
T
h
is
co
m
p
le
x
it
y
in
cr
ea
s
es
th
e
m
o
d
el’
s
s
u
s
ce
p
tib
ilit
y
to
o
v
er
f
itt
in
g
o
n
tr
ai
n
in
g
n
o
is
e
a
n
d
cr
ea
tes
s
u
b
o
p
tim
al
lo
ca
l
m
i
n
i
m
a
d
u
r
in
g
b
ac
k
p
r
o
p
ag
atio
n
.
C
o
n
s
eq
u
en
tl
y
,
th
e
s
i
m
p
ler
,
d
ed
icate
d
L
ST
M
ar
ch
itectu
r
e
m
ai
n
tai
n
s
s
u
p
er
io
r
w
ei
g
h
t
co
n
v
er
g
e
n
ce
a
n
d
b
etter
g
en
er
a
lizatio
n
w
h
en
p
r
ed
ictin
g
u
n
s
ee
n
ex
tr
e
m
e
p
o
llu
tio
n
p
ea
k
s
[
2
3
]
.
I
n
ad
d
itio
n
to
ar
ch
itectu
r
al
d
if
f
er
en
ce
s
,
o
u
r
f
i
n
d
in
g
s
e
m
p
h
a
s
ize
th
e
cr
itical
r
o
le
o
f
s
p
ec
if
ic
m
eteo
r
o
lo
g
ical
d
r
iv
er
s
i
n
p
r
ed
ictin
g
P
M2
.
5
d
y
n
a
m
ic
s
.
T
h
e
em
p
ir
ical
s
elec
tio
n
o
f
air
te
m
p
er
atu
r
e
an
d
r
elativ
e
h
u
m
id
it
y
w
h
ile
eli
m
i
n
atin
g
w
ea
k
l
y
co
r
r
elate
d
v
ar
iab
les
l
ik
e
w
i
n
d
s
p
ee
d
d
em
o
n
s
tr
ates
th
at
th
er
m
al
an
d
m
o
is
t
u
r
e
in
ter
ac
tio
n
s
d
o
m
i
n
at
e
lo
ca
l
p
ar
ticu
late
b
eh
av
io
r
in
th
i
s
s
t
u
d
y
ar
ea
.
Fro
m
a
n
atm
o
s
p
h
er
ic
p
h
y
s
ic
s
p
er
s
p
ec
tiv
e,
h
i
g
h
r
elati
v
e
h
u
m
id
it
y
ac
ce
ler
ates
t
h
e
h
y
g
r
o
s
co
p
ic
g
r
o
w
t
h
o
f
ae
r
o
s
o
ls
an
d
p
r
o
m
o
te
s
t
h
e
f
o
r
m
atio
n
o
f
s
ec
o
n
d
ar
y
in
o
r
g
an
ic
p
ar
ticu
lates.
Si
m
u
lta
n
eo
u
s
l
y
,
te
m
p
er
at
u
r
e
v
ar
iatio
n
s
d
ir
ec
tl
y
in
f
l
u
en
ce
t
h
e
h
eig
h
t
o
f
t
h
e
at
m
o
s
p
h
er
ic
b
o
u
n
d
ar
y
la
y
e
r
an
d
th
er
m
al
i
n
v
er
s
io
n
ev
e
n
ts
,
w
h
ic
h
tr
ap
p
o
llu
tan
t
s
n
ea
r
th
e
s
u
r
f
ac
e.
B
y
s
u
p
p
l
y
i
n
g
t
h
e
L
ST
M
w
ith
th
e
s
e
co
n
ti
n
u
o
u
s
t
h
er
m
o
h
y
g
r
o
m
e
tr
ic
p
ar
am
eter
s
,
t
h
e
n
e
t
w
o
r
k
ef
f
ec
tiv
e
l
y
lear
n
s
t
h
e
n
o
n
-
li
n
ea
r
th
er
m
o
d
y
n
a
m
ic
t
h
r
es
h
o
ld
s
th
at
tr
ig
g
er
s
u
d
d
en
p
o
llu
tan
t
ac
cu
m
u
lat
io
n
,
s
ig
n
i
f
ica
n
tl
y
i
m
p
r
o
v
i
n
g
ac
cu
r
a
c
y
o
v
er
p
u
r
e
au
to
r
eg
r
es
s
i
v
e
(
P
M
-
o
n
l
y
)
h
i
s
to
r
ical
in
p
u
ts
[
2
4
]
.
Fro
m
an
en
g
i
n
ee
r
in
g
an
d
p
r
ac
tical
d
ep
lo
y
m
en
t
p
er
s
p
ec
ti
v
e,
th
e
ar
ch
itect
u
r
al
s
u
p
er
io
r
it
y
o
f
th
e
s
in
g
le
L
ST
M
m
o
d
el
e
x
te
n
d
s
b
e
y
o
n
d
p
r
ed
ictiv
e
ac
cu
r
ac
y
to
co
m
p
u
tat
io
n
al
e
f
f
icie
n
c
y
.
I
n
r
e
al
-
w
o
r
ld
air
q
u
alit
y
m
o
n
ito
r
i
n
g
,
f
o
r
ec
asti
n
g
m
o
d
els
m
u
s
t
b
e
i
n
te
g
r
ated
i
n
to
i
n
t
er
n
et
o
f
th
in
g
s
(
I
o
T
)
ed
g
e
d
e
v
ices
o
r
ce
n
tr
alize
d
clo
u
d
p
latf
o
r
m
s
t
h
at
p
r
o
ce
s
s
co
n
tin
u
o
u
s
,
h
i
g
h
-
f
r
eq
u
e
n
c
y
s
en
s
o
r
s
tr
ea
m
s
.
W
h
ile
m
as
s
i
v
e
h
y
b
r
id
ar
ch
itect
u
r
e
s
d
em
a
n
d
s
u
b
s
ta
n
tial
m
e
m
o
r
y
f
o
o
tp
r
in
ts
an
d
h
i
g
h
co
m
p
u
tatio
n
al
late
n
c
y
,
o
u
r
o
p
ti
m
i
ze
d
L
ST
M
m
o
d
el
m
ai
n
tai
n
s
a
co
m
p
ac
t
p
ar
a
m
e
te
r
s
ize
a
f
ter
co
n
v
er
g
e
n
ce
.
T
h
is
lig
h
t
w
ei
g
h
t
p
r
o
f
ile
e
n
ab
les
lo
w
-
late
n
c
y
r
ea
l
-
ti
m
e
in
f
er
en
ce
,
m
a
k
i
n
g
it
ex
ce
p
ti
o
n
all
y
w
el
l
-
s
u
ited
f
o
r
d
ep
lo
y
m
e
n
t
o
n
ed
g
e
ar
tif
ic
ial
in
t
ellig
e
n
ce
(
E
d
g
e
-
A
I
)
m
icr
o
co
n
tr
o
ller
s
e
m
b
ed
d
ed
w
i
th
i
n
A
W
S.
Desp
ite
it
s
r
o
b
u
s
t
p
er
f
o
r
m
an
ce
,
t
h
is
s
t
u
d
y
h
as
li
m
it
atio
n
s
th
at
p
r
esen
t
o
p
p
o
r
tu
n
itie
s
f
o
r
m
et
h
o
d
o
lo
g
ical
s
ca
li
n
g
.
First,
w
h
ile
MI
C
E
i
m
p
u
ta
tio
n
e
f
f
ec
tiv
el
y
m
ai
n
tai
n
s
s
tatis
tical
d
is
tr
ib
u
tio
n
d
u
r
in
g
m
is
s
i
n
g
d
ata
in
ter
v
a
ls
[
2
5
]
,
ar
tif
icial
i
m
p
u
tatio
n
in
h
er
en
tl
y
ca
r
r
ies
esti
m
atio
n
u
n
ce
r
tai
n
t
y
d
u
r
in
g
u
n
p
r
ec
ed
en
ted
ex
tr
e
m
e
p
o
llu
ti
o
n
ev
e
n
ts
.
Seco
n
d
,
th
e
cu
r
r
e
n
t
i
m
p
le
m
en
tat
io
n
v
alid
ates
t
h
e
ar
ch
itect
u
r
e
u
s
i
n
g
a
s
in
g
le
-
s
ta
tio
n
d
ataset,
w
h
i
ch
d
o
es
n
o
t
ac
co
u
n
t
f
o
r
s
p
atial
p
o
llu
ta
n
t
tr
an
s
p
o
r
t.
Ho
w
ev
er
,
o
u
r
m
o
d
u
lar
p
r
ep
r
o
ce
s
s
in
g
,
i
m
p
u
tatio
n
,
an
d
s
lid
in
g
-
w
i
n
d
o
w
f
r
a
m
e
w
o
r
k
i
s
in
h
er
en
t
l
y
s
ca
lab
le
to
m
u
lti
-
s
tatio
n
a
n
d
r
eg
io
n
a
l
n
et
w
o
r
k
s
.
T
o
g
en
er
alize
th
i
s
m
et
h
o
d
o
lo
g
y
ac
r
o
s
s
b
r
o
ad
er
g
eo
g
r
ap
h
ic
ar
ea
s
,
t
h
e
p
ip
eli
n
e
ca
n
b
e
r
ea
d
il
y
ex
ten
d
ed
i
n
to
s
p
atio
te
m
p
o
r
al
g
r
ap
h
n
e
u
r
al
n
et
w
o
r
k
s
(
GNN)
o
r
co
n
v
o
lu
tio
n
al
L
ST
Ms
(
C
o
n
v
L
ST
M)
.
I
n
s
u
c
h
s
ca
led
ar
ch
itectu
r
e
s
,
m
u
ltip
le
A
W
S
an
d
air
q
u
alit
y
s
tatio
n
s
w
o
u
ld
s
er
v
e
as
s
p
atial
n
o
d
es,
allo
w
in
g
th
e
n
et
w
o
r
k
to
s
i
m
u
ltan
eo
u
s
l
y
ca
p
tu
r
e
lo
ca
lized
tem
p
o
r
al
d
ep
en
d
en
cie
s
an
d
cr
o
s
s
-
r
e
g
io
n
al
p
o
llu
tan
t
d
is
p
er
s
io
n
p
atter
n
s
.
B
ased
o
n
th
e
ab
o
v
e
an
al
y
s
is
,
f
u
r
th
er
r
esear
ch
is
r
ec
o
m
m
en
d
ed
to
ex
p
lo
r
e
th
e
in
teg
r
atio
n
o
f
atten
tio
n
m
ec
h
a
n
i
s
m
s
o
r
tr
an
s
f
o
r
m
er
s
t
h
at
ca
n
p
r
o
v
id
e
d
y
n
a
m
ic
w
ei
g
h
ts
at
s
p
ec
i
f
ic
ti
m
e
-
s
tep
s
t
h
at
ar
e
m
o
s
t
in
f
l
u
en
t
ial,
r
ath
er
th
an
s
i
m
p
l
y
a
s
s
i
g
n
in
g
u
n
i
f
o
r
m
w
e
ig
h
t
s
to
th
e
en
tire
s
lid
in
g
w
i
n
d
o
w
.
F
u
r
th
er
m
o
r
e,
th
e
d
ev
elo
p
m
en
t
o
f
a
s
p
atio
te
m
p
o
r
al
m
o
d
el
th
at
in
te
g
r
ates
m
u
lti
-
s
tat
io
n
d
at
a
u
s
i
n
g
a
g
r
id
-
e
m
b
ed
d
ed
ap
p
r
o
ac
h
is
h
ig
h
l
y
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
2
4
,
No
.
5
,
Octo
b
e
r
2
0
2
6
:
1
6
4
4
-
1655
1652
r
ec
o
m
m
e
n
d
ed
to
ca
p
tu
r
e
cr
o
s
s
-
r
e
g
io
n
al
p
o
llu
tio
n
d
is
tr
ib
u
ti
o
n
p
atter
n
s
i
n
co
m
p
le
x
u
r
b
an
ar
ea
s
[2
6]
.
Ov
er
all,
th
is
s
t
u
d
y
s
u
cc
ess
f
u
ll
y
d
e
m
o
n
s
tr
ate
s
t
h
at
t
h
e
L
ST
M
ar
ch
i
tectu
r
e
w
it
h
ap
p
r
o
p
r
iate
d
ata
p
r
e
-
p
r
o
ce
s
s
in
g
ca
n
p
r
o
v
id
e
r
eliab
le
P
M2
.
5
p
r
ed
ictio
n
ac
c
u
r
ac
y
f
o
r
ea
r
l
y
w
a
r
n
in
g
s
y
s
te
m
s
.
T
h
ese
r
es
u
lt
s
p
r
o
v
id
e
a
s
tr
o
n
g
tech
n
ical
f
o
u
n
d
atio
n
f
o
r
p
o
lic
y
m
a
k
er
s
to
ad
o
p
t
A
I
-
b
ased
m
o
d
els
to
m
iti
g
ate
th
e
h
ea
lth
i
m
p
ac
t
s
o
f
air
p
o
llu
tio
n
.
4.
CO
NCLU
SI
O
N
T
h
is
s
tu
d
y
p
r
o
v
id
ed
a
co
m
p
r
eh
en
s
iv
e
co
m
p
ar
ati
v
e
ev
al
u
atio
n
o
f
d
ee
p
lear
n
in
g
ar
ch
i
t
ec
tu
r
es
f
o
r
f
o
r
ec
asti
n
g
h
o
u
r
l
y
P
M2
.
5
co
n
ce
n
tr
at
io
n
s
b
y
i
n
te
g
r
ati
n
g
l
o
ca
lized
A
W
S
an
d
air
q
u
alit
y
s
e
n
s
o
r
d
ata.
T
h
e
ev
alu
a
tio
n
d
e
m
o
n
s
tr
ates
th
at
t
h
e
s
in
g
le
L
ST
M
m
o
d
el
is
th
e
m
o
s
t
r
o
b
u
s
t
a
n
d
r
eliab
le
ar
c
h
itectu
r
e,
ac
h
iev
in
g
th
e
lo
w
est er
r
o
r
r
ates
w
it
h
a
n
R
MSE
o
f
6
.
8
6
2
µg
/
m
³
an
d
a
n
MSE
o
f
4
7
.
0
8
1
.
T
h
e
s
u
p
er
io
r
it
y
o
f
t
h
e
L
ST
M
lies
in
its
i
n
ter
n
a
l
g
ati
n
g
a
n
d
ce
ll
s
tate
m
ec
h
a
n
is
m
s
,
w
h
ic
h
ef
f
e
ctiv
el
y
ca
p
t
u
r
e
co
m
p
lex
te
m
p
o
r
al
d
y
n
a
m
ic
s
an
d
lo
n
g
-
ter
m
s
eq
u
e
n
tia
l
d
ep
en
d
en
cies.
C
o
n
v
er
s
el
y
,
w
h
ile
t
h
e
s
tan
d
ar
d
ML
P
m
o
d
el
ac
h
iev
ed
a
s
lig
h
tl
y
h
i
g
h
er
R
²
(
0
.
8
1
0
)
,
it
ex
h
ib
ited
a
n
o
ticea
b
le
m
a
g
n
itu
d
e
b
ias
b
y
tr
ea
t
in
g
ti
m
e
-
s
er
ie
s
s
eq
u
e
n
ce
s
as
s
tatic
f
ea
t
u
r
es.
Fu
r
t
h
er
m
o
r
e,
co
m
b
i
n
i
n
g
ar
ch
i
tectu
r
es
in
to
a
h
y
b
r
id
L
ST
M
-
ML
P
m
o
d
el
d
e
g
r
ad
ed
p
r
ed
ictio
n
ac
cu
r
ac
y
d
u
e
to
in
cr
ea
s
ed
s
tr
u
ct
u
r
al
co
m
p
lex
i
t
y
an
d
o
v
er
f
itti
n
g
r
is
k
s
.
O
u
r
f
in
d
in
g
s
also
co
n
f
ir
m
t
h
at
s
elec
tin
g
P
M1
0
,
air
te
m
p
er
atu
r
e,
an
d
r
elativ
e
h
u
m
id
it
y
a
s
co
r
e
p
r
e
d
icto
r
s
ef
f
ec
tiv
el
y
m
i
n
i
m
ize
s
in
p
u
t
n
o
i
s
e
w
h
ile
ca
p
tu
r
i
n
g
ess
e
n
tial a
t
m
o
s
p
h
er
ic
i
n
ter
ac
ti
o
n
s
.
T
o
b
u
ild
u
p
o
n
th
ese
f
i
n
d
i
n
g
s
,
f
u
t
u
r
e
r
esear
ch
s
h
o
u
ld
ex
p
lo
r
e
th
e
in
te
g
r
atio
n
o
f
A
tte
n
tio
n
m
ec
h
a
n
i
s
m
s
o
r
T
r
an
s
f
o
r
m
er
-
ba
s
ed
ar
ch
itec
tu
r
es
to
d
y
n
a
m
icall
y
w
ei
g
h
t
cr
itica
l
s
eq
u
en
t
ial
ti
m
e
s
tep
s
.
A
d
d
itio
n
al
l
y
,
ex
p
a
n
d
in
g
th
e
m
et
h
o
d
o
lo
g
y
to
in
co
r
p
o
r
ate
m
u
lti
-
y
ea
r
d
atasets
an
d
m
u
l
ti
-
s
t
atio
n
s
p
atio
te
m
p
o
r
al
n
et
w
o
r
k
s
w
il
l
f
u
r
t
h
er
en
h
an
ce
m
o
d
el
g
e
n
er
aliza
tio
n
.
I
m
p
le
m
en
tin
g
s
u
ch
s
p
atia
l
-
te
m
p
o
r
al
f
r
a
m
e
w
o
r
k
s
w
il
l
b
e
cr
u
cial
f
o
r
d
ev
elo
p
in
g
s
ca
lab
l
e,
r
ea
l
-
ti
m
e
air
q
u
al
it
y
ea
r
l
y
war
n
in
g
s
y
s
te
m
s
in
co
m
p
le
x
u
r
b
an
en
v
ir
o
n
m
e
n
ts
.
ACK
NO
WL
E
D
G
M
E
NT
S
W
e
w
o
u
ld
li
k
e
to
ex
p
r
ess
o
u
r
g
r
atit
u
d
e
to
B
MK
G
an
d
C
ite
k
o
Me
teo
r
o
lo
g
ical
Statio
n
f
o
r
th
ei
r
s
u
p
p
o
r
t
in
p
r
o
v
id
in
g
d
ata
an
d
in
s
tr
u
m
en
t t
h
at
h
av
e
b
ee
n
ess
e
n
tial
f
o
r
th
is
r
esear
c
h
p
r
o
j
ec
t.
F
UNDIN
G
I
NF
O
RM
AT
I
O
N
T
h
is
r
esear
ch
w
a
s
f
u
n
d
ed
b
y
Facu
lt
y
Ma
th
e
m
atic
s
a
n
d
N
atu
r
al
Scie
n
ce
U
n
i
v
er
s
ita
s
I
n
d
o
n
esia
f
o
r
th
is
w
o
r
k
,
u
n
d
er
p
r
o
j
ec
t
s
ch
e
m
e
o
f
Hib
ah
P
en
elitia
n
P
a
s
ca
s
ar
j
an
a
w
it
h
n
u
m
b
er
P
KS
-
2
7
/UN2
.
F3
.
D/PP
M.
0
0
.
0
2
/2
0
2
5
.
AUTHO
R
CO
NT
RIB
UT
I
O
NS ST
A
T
E
M
E
NT
T
h
is
j
o
u
r
n
al
u
s
e
s
th
e
C
o
n
tr
ib
u
to
r
R
o
les
T
ax
o
n
o
m
y
(
C
R
ed
iT
)
to
r
ec
o
g
n
ize
in
d
i
v
id
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
t
h
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
lla
b
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
Ah
m
ad
Fu
r
q
o
n
✓
✓
✓
✓
✓
✓
✓
San
t
s
o
So
er
k
ir
n
o
✓
✓
✓
✓
✓
✓
✓
✓
Nau
f
al
A
n
a
n
d
a
✓
✓
✓
✓
✓
✓
✓
Har
y
as S
u
b
y
a
n
tar
a
W
icak
s
a
n
a
✓
✓
✓
✓
✓
✓
✓
Ah
m
ad
Me
ij
lan
Yasir
✓
✓
✓
✓
✓
T
o
n
n
y
W
ah
y
u
A
j
i
✓
✓
✓
✓
✓
✓
Av
i
v
Ma
g
h
r
id
lo
✓
✓
✓
✓
✓
Fad
h
li
L
u
th
f
ia
n
s
y
a
h
✓
✓
✓
✓
✓
Ma
r
zu
k
i
Sin
a
m
b
ela
✓
✓
✓
✓
✓
✓
Su
k
o
P
r
ay
it
n
o
A
d
i
✓
✓
✓
✓
✓
✓
✓
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
si
s
I
:
I
n
v
e
st
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
si
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
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
F
o
r
ec
a
s
tin
g
p
a
r
ticu
la
te
ma
tte
r
co
n
ce
n
tr
a
tio
n
u
s
in
g
d
ee
p
le
a
r
n
in
g
a
n
d
w
ea
th
er d
a
ta
(
S
a
n
to
s
o
S
o
ek
ir
n
o
)
1653
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
Au
t
h
o
r
s
s
tate
n
o
co
n
f
lic
t o
f
i
n
t
er
est.
DATA AV
AI
L
AB
I
L
I
T
Y
T
h
e
d
ata
th
at
s
u
p
p
o
r
t
th
e
f
i
n
d
in
g
s
o
f
th
i
s
s
tu
d
y
ar
e
av
ai
la
b
le
f
r
o
m
t
h
e
co
r
r
esp
o
n
d
in
g
a
u
th
o
r
[
SS
]
,
u
p
o
n
r
ea
s
o
n
ab
le
r
eq
u
est
R
E
F
E
R
E
NC
E
S
[
1
]
M
.
T
.
U
d
r
i
s
t
i
o
i
u
,
Y
.
EL
M
g
h
o
u
c
h
i
,
a
n
d
H
.
Y
i
l
d
i
z
h
a
n
,
“
P
r
e
d
i
c
t
i
o
n
,
m
o
d
e
l
l
i
n
g
,
a
n
d
f
o
r
e
c
a
s
t
i
n
g
o
f
P
M
a
n
d
A
Q
I
u
s
i
n
g
h
y
b
r
i
d
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
,
”
J
o
u
r
n
a
l
o
f
C
l
e
a
n
e
r
P
r
o
d
u
c
t
i
o
n
,
v
o
l
.
4
2
1
,
p
.
1
3
8
4
9
6
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
c
l
e
p
r
o
.
2
0
2
3
.
1
3
8
4
9
6
.
[
2
]
I
.
N
.
G
.
S
u
y
a
s
a
,
N
.
M
.
M
a
r
w
a
t
i
,
a
n
d
N
.
K
.
R
u
s
m
i
n
i
n
g
s
i
h
,
“
M
e
a
s
u
r
i
n
g
A
i
r
Q
u
a
l
i
t
y
O
v
e
r
D
e
n
p
a
s
a
r
C
i
t
y
I
n
d
o
n
e
s
i
a
i
n
2
0
2
1
,
”
J
o
u
r
n
a
l
o
f
H
e
a
l
t
h
a
n
d
M
e
d
i
c
a
l
S
c
i
e
n
c
e
s
,
v
o
l
.
4
,
n
o
.
4
,
D
e
c
.
2
0
2
1
,
d
o
i
:
1
0
.
3
1
0
1
4
/
a
i
o
r
.
1
9
9
4
.
0
4
.
0
4
.
1
9
1
.
[
3
]
G
.
M
a
u
l
e
,
“
A
i
r
P
o
l
l
u
t
i
o
n
a
n
d
I
n
t
e
r
s
t
i
t
i
a
l
L
u
n
g
D
i
s
e
a
s
e
:
I
n
s
i
g
h
t
s
f
r
o
m
t
h
e
G
l
o
b
a
l
B
u
r
d
e
n
o
f
D
i
s
e
a
se
D
a
t
a
b
a
s
e
,
”
E
u
r
o
p
e
a
n
M
e
d
i
c
a
l
J
o
u
r
n
a
l
,
M
a
r
.
2
0
2
6
,
d
o
i
:
1
0
.
3
3
5
9
0
/
e
m
j
/
d
k
v
u
1
7
1
1
.
[
4
]
T
.
I
s
t
i
a
n
a
e
t
a
l
.
,
“
C
a
u
s
a
l
i
t
y
A
n
a
l
y
s
i
s
o
f
A
i
r
Q
u
a
l
i
t
y
a
n
d
M
e
t
e
o
r
o
l
o
g
i
c
a
l
P
a
r
a
m
e
t
e
r
s
f
o
r
P
M
2
.
5
C
h
a
r
a
c
t
e
r
i
s
t
i
c
s
D
e
t
e
r
m
i
n
a
t
i
o
n
:
E
v
i
d
e
n
c
e
f
r
o
m
J
a
k
a
r
t
a
,
”
A
e
r
o
s
o
l
a
n
d
A
i
r
Q
u
a
l
i
t
y
R
e
s
e
a
r
c
h
,
v
o
l
.
2
3
,
n
o
.
9
,
p
.
2
3
0
0
1
4
,
2
0
2
3
,
d
o
i
:
1
0
.
4
2
0
9
/
a
a
q
r
.
2
3
0
0
1
4
.
[
5
]
Y
.
W
a
n
g
e
t
a
l
.
,
“
A
c
o
m
p
a
r
a
t
i
v
e
s
t
u
d
y
o
f
t
h
r
e
e
m
o
d
e
l
s
t
o
a
n
a
l
y
z
e
t
h
e
i
m
p
a
c
t
o
f
a
i
r
p
o
l
l
u
t
a
n
t
s
o
n
t
h
e
n
u
m
b
e
r
o
f
p
u
l
m
o
n
a
r
y
t
u
b
e
r
c
u
l
o
s
i
s
c
a
s
e
s
i
n
U
r
u
m
q
i
,
X
i
n
j
i
a
n
g
,
”
P
L
o
S
O
N
E
,
v
o
l
.
1
8
,
n
o
.
1
,
p
.
e
0
2
7
7
3
1
4
,
J
a
n
.
2
0
2
3
,
d
o
i
:
1
0
.
1
3
7
1
/
j
o
u
r
n
a
l
.
p
o
n
e
.
0
2
7
7
3
1
4
.
[
6
]
S
.
E
.
H
a
sy
i
m
,
N
u
r
a
z
i
z
a
h
,
M
.
Y
.
P
r
a
t
a
m
a
,
U
.
R
.
G
u
r
n
i
n
g
,
B
.
K
h
a
i
r
u
n
n
i
s
a
,
“
P
M
2
.
5
P
r
e
d
i
c
t
i
o
n
U
s
i
n
g
t
h
e
L
o
n
g
S
h
o
r
t
-
T
e
r
m
M
e
m
o
r
y
A
l
g
o
r
i
t
h
m
,”
I
J
A
T
I
S
:
I
n
d
o
n
e
s
i
a
n
J
o
u
r
n
a
l
o
f
A
p
p
l
i
e
d
T
e
c
h
n
o
l
o
g
y
a
n
d
I
n
n
o
v
a
t
i
o
n
S
c
i
e
n
c
e
,
v
o
l
.
2
,
n
o
,
2
,
pp
.
63
-
69
,
2
0
2
5
.
[
7
]
S
.
H
o
f
f
m
a
n
a
n
d
R
.
J
a
s
i
ń
s
k
i
,
“
T
h
e
U
s
e
o
f
M
u
l
t
i
l
a
y
e
r
P
e
r
c
e
p
t
r
o
n
s
t
o
M
o
d
e
l
P
M
2
.
5
C
o
n
c
e
n
t
r
a
t
i
o
n
s
a
t
A
i
r
M
o
n
i
t
o
r
i
n
g
S
t
a
t
i
o
n
s
i
n
P
o
l
a
n
d
,
”
A
t
m
o
s
p
h
e
r
e
,
v
o
l
.
1
4
,
n
o
.
1
,
p
.
9
6
,
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
a
t
m
o
s
1
4
0
1
0
0
9
6
.
[
8
]
F
.
M
o
h
a
m
m
a
d
i
,
H
.
T
e
i
r
i
,
Y
.
H
a
j
i
z
a
d
e
h
,
A
.
A
b
d
o
l
a
h
n
e
j
a
d
,
a
n
d
A
.
E
b
r
a
h
i
m
i
,
“
P
r
e
d
i
c
t
i
o
n
o
f
a
t
m
o
s
p
h
e
r
i
c
P
M
2
.
5
l
e
v
e
l
b
y
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
i
n
I
s
f
a
h
a
n
,
I
r
a
n
,
”
S
c
i
e
n
t
i
f
i
c
R
e
p
o
r
t
s
,
v
o
l
.
1
4
,
n
o
.
1
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
3
8
/
s
4
1
5
9
8
-
0
2
4
-
5
2
6
1
7
-
z.
[
9
]
S
.
S
o
e
k
i
r
n
o
e
t
a
l
.
,
“
D
e
s
i
g
n
o
f
d
r
o
u
g
h
t
e
a
r
l
y
w
a
r
n
i
n
g
sy
s
t
e
m
b
a
s
e
d
o
n
s
t
a
n
d
a
r
d
i
z
e
d
p
r
e
c
i
p
i
t
a
t
i
o
n
i
n
d
e
x
p
r
e
d
i
c
t
i
o
n
u
s
i
n
g
h
y
b
r
i
d
A
R
I
M
A
-
M
L
P
i
n
B
a
n
t
e
n
p
r
o
v
i
n
c
e
,
”
I
A
E
S
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
A
r
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
(
I
J
-
A
I
)
,
v
o
l
.
1
3
,
n
o
.
2
,
p
.
1
8
7
8
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
5
9
1
/
i
j
a
i
.
v
1
3
.
i
2
.
p
p
1
8
7
8
-
1
8
8
7
.
[
1
0
]
H
a
r
t
a
n
t
o
,
S
.
H
u
m
a
i
d
i
,
E
.
F
r
i
d
a
,
N
.
A
n
a
n
d
a
,
a
n
d
M
.
S
i
n
a
m
b
e
l
a
,
“
S
p
a
t
i
a
l
e
v
a
l
u
a
t
i
o
n
r
a
i
n
f
a
l
l
e
s
t
i
m
a
t
i
o
n
o
n
w
e
a
t
h
e
r
r
a
d
a
r
u
s
i
n
g
M
a
r
s
h
a
l
l
–
P
a
l
m
e
r
r
e
f
l
e
c
t
i
v
i
t
y
–
r
a
i
n
f
a
l
l
r
a
t
e
i
n
B
a
n
t
e
n
,
”
J
o
u
r
n
a
l
o
f
W
a
t
e
r
a
n
d
L
a
n
d
D
e
v
e
l
o
p
m
e
n
t
,
n
o
.
6
2
,
p
p
.
1
9
3
–
2
0
0
,
2
0
2
4
,
d
o
i
:
1
0
.
2
4
4
2
5
/
j
w
l
d
.
2
0
2
4
.
1
5
1
5
6
7
.
[
1
1
]
M.
P
e
r
i
n
i
a
n
d
M
.
N
i
k
o
l
i
c
,
“
I
n
-
D
a
t
a
b
a
s
e
D
a
t
a
I
m
p
u
t
a
t
i
o
n
,
”
P
r
o
c
e
e
d
i
n
g
s
o
f
t
h
e
A
C
M
o
n
M
a
n
a
g
e
m
e
n
t
o
f
D
a
t
a
,
v
o
l
.
2
,
n
o
.
1
,
p
p
.
1
–
2
7
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
4
5
/
3
6
3
9
3
2
6
.
[
1
2
]
T
.
I
s
t
i
a
n
a
e
t
a
l
.
,
“
F
i
n
e
p
a
r
t
i
c
u
l
a
t
e
m
a
t
t
e
r
c
o
n
c
e
n
t
r
a
t
i
o
n
f
o
r
e
c
a
s
t
i
n
g
u
s
i
n
g
l
o
n
g
s
h
o
r
t
-
t
e
r
m
me
m
o
r
y
n
e
t
w
o
r
k
a
n
d
m
e
t
e
o
r
o
l
o
g
i
c
a
l
i
n
p
u
t
s
,
”
G
l
o
b
a
l
J
o
u
r
n
a
l
o
f
E
n
v
i
r
o
n
m
e
n
t
a
l
S
c
i
e
n
c
e
a
n
d
M
a
n
a
g
e
m
e
n
t
,
v
o
l
.
1
0
,
n
o
.
4
,
p
p
.
1
7
5
9
–
1
7
7
4
,
2
0
2
4
,
d
o
i
:
1
0
.
2
2
0
3
4
/
g
j
e
s
m
.
2
0
2
4
.
0
4
.
1
6
.
[
1
3
]
C
.
A
.
F
i
e
b
r
i
c
h
,
Y
.
R
.
M
o
r
g
a
n
,
A
.
G
.
M
c
c
o
m
b
s
,
P
.
K
.
H
a
l
l
,
a
n
d
R
.
A
.
M
c
p
h
e
r
s
o
n
,
“
Q
u
a
l
i
t
y
a
s
s
u
r
a
n
c
e
p
r
o
c
e
d
u
r
e
s
f
o
r
me
so
s
c
a
l
e
m
e
t
e
o
r
o
l
o
g
i
c
a
l
d
a
t
a
,
”
J
o
u
r
n
a
l
o
f
A
t
m
o
s
p
h
e
r
i
c
a
n
d
O
c
e
a
n
i
c
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
2
7
,
n
o
.
1
0
,
p
p
.
1
5
6
5
–
1
5
8
2
,
2
0
1
0
,
d
o
i
:
1
0
.
1
1
7
5
/
2
0
1
0
J
T
E
C
H
A
1
4
3
3
.
1
.
[
1
4
]
G
.
A
k
b
a
r
,
P
.
P
r
a
j
i
t
n
o
,
A
r
i
f
f
u
d
i
n
,
a
n
d
N
.
A
n
a
n
d
a
,
“
M
u
l
t
i
v
a
r
i
a
t
e
I
m
p
u
t
a
t
i
o
n
C
h
a
i
n
e
d
E
q
u
a
t
i
o
n
o
n
S
o
l
a
r
R
a
d
i
a
t
i
o
n
i
n
A
u
t
o
m
a
t
i
c
W
e
a
t
h
e
r
S
t
a
t
i
o
n
,
”
J
u
r
n
a
l
P
e
n
e
l
i
t
i
a
n
P
e
n
d
i
d
i
k
a
n
I
P
A
,
v
o
l
.
1
0
,
n
o
.
7
,
p
p
.
3
6
3
3
–
3
6
3
9
,
2
0
2
4
,
d
o
i
:
1
0
.
2
9
3
0
3
/
j
p
p
i
p
a
.
v
1
0
i
7
.
7
6
7
9
.
[
1
5
]
R
.
P
a
h
l
e
p
i
,
S
.
S
o
e
k
i
r
n
o
,
a
n
d
H
.
S
.
W
i
c
a
k
s
a
n
a
,
“
S
o
l
a
r
R
a
d
i
a
t
i
o
n
I
n
t
e
n
s
i
t
y
I
mp
u
t
a
t
i
o
n
i
n
P
y
r
a
n
o
m
e
t
e
r
o
f
A
u
t
o
m
a
t
i
c
W
e
a
t
h
e
r
S
t
a
t
i
o
n
B
a
s
e
d
o
n
L
o
n
g
S
h
o
r
t
T
e
r
m
M
e
m
o
r
y
,
”
U
l
t
i
m
a
C
o
m
p
u
t
i
n
g
:
J
u
r
n
a
l
S
i
s
t
e
m
K
o
m
p
u
t
e
r
,
p
p
.
3
5
–
4
0
,
2
0
2
3
,
d
o
i
:
1
0
.
3
1
9
3
7
/
s
k
.
v
1
5
i
2
.
3
3
4
8
.
[
1
6
]
K
.
S
.
H
e
r
n
á
n
d
e
z
,
D
.
N
i
e
v
e
s
,
J
.
S
.
P
é
r
e
z
-
C
a
r
r
a
s
q
u
i
l
l
a
,
P
.
M
o
n
t
o
y
a
,
M
.
D
.
Z
u
l
u
a
g
a
,
a
n
d
M
.
R
a
m
í
r
e
z
,
“
F
o
r
e
c
a
s
t
i
n
g
P
M
2
.
5
c
o
n
c
e
n
t
r
a
t
i
o
n
s
u
s
i
n
g
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
e
s
:
a
d
d
e
d
v
a
l
u
e
o
f
l
o
w
-
c
o
s
t
m
o
n
i
t
o
r
i
n
g
a
n
d
r
e
g
i
o
n
a
l
mo
d
e
l
i
n
g
,
”
2
0
2
4
,
C
o
p
e
r
n
i
c
u
s
G
m
b
H
.
d
o
i
:
1
0
.
5
1
9
4
/
e
g
u
s
p
h
e
r
e
-
e
g
u
2
4
-
1
2
1
4
.
[
1
7
]
Y
.
Z
h
a
n
g
,
Q
.
S
u
n
,
J
.
L
i
u
,
a
n
d
O
.
P
e
t
r
o
s
i
a
n
,
“
L
o
n
g
-
T
e
r
m
F
o
r
e
c
a
s
t
i
n
g
o
f
A
i
r
P
o
l
l
u
t
i
o
n
P
a
r
t
i
c
u
l
a
t
e
M
a
t
t
e
r
(
P
M
2
.
5
)
a
n
d
A
n
a
l
y
s
i
s
o
f
I
n
f
l
u
e
n
c
i
n
g
F
a
c
t
o
r
s
,
”
S
u
s
t
a
i
n
a
b
i
l
i
t
y
(
S
w
i
t
z
e
r
l
a
n
d
)
,
v
o
l
.
1
6
,
n
o
.
1
,
p
.
1
9
,
2
0
2
4
,
d
o
i
:
1
0
.
3
3
9
0
/
s
u
1
6
0
1
0
0
1
9
.
[
1
8
]
D
.
M
.
P
.
W
.
D
i
s
s
a
n
a
y
a
k
e
,
R
.
M
.
K
.
T
.
R
a
t
h
n
a
y
a
k
e
,
a
n
d
L
.
L
.
G
.
C
h
a
t
h
u
r
a
n
g
a
,
“
C
r
o
p
Y
i
e
l
d
F
o
r
e
c
a
s
t
i
n
g
u
s
i
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
-
A
S
y
s
t
e
m
a
t
i
c
L
i
t
e
r
a
t
u
r
e
R
e
v
i
e
w
,
”
K
D
U
J
o
u
r
n
a
l
o
f
M
u
l
t
i
d
i
s
c
i
p
l
i
n
a
r
y
S
t
u
d
i
e
s
,
v
o
l
.
5
,
n
o
.
1
,
p
p
.
5
4
–
6
5
,
2
0
2
3
,
d
o
i
:
1
0
.
4
0
3
8
/
k
j
m
s
.
v
5
i
1
.
6
2
.
[
1
9
]
T
.
K
i
m
,
W
.
K
o
,
a
n
d
J
.
K
i
m
,
“
A
n
a
l
y
s
i
s
a
n
d
i
m
p
a
c
t
e
v
a
l
u
a
t
i
o
n
o
f
m
i
s
s
i
n
g
d
a
t
a
i
m
p
u
t
a
t
i
o
n
i
n
d
a
y
-
a
h
e
a
d
P
V
g
e
n
e
r
a
t
i
o
n
f
o
r
e
c
a
s
t
i
n
g
,
”
A
p
p
l
i
e
d
S
c
i
e
n
c
e
s
(
S
w
i
t
z
e
r
l
a
n
d
)
,
v
o
l
.
9
,
n
o
.
1
,
p
.
2
0
4
,
2
0
1
9
,
d
o
i
:
1
0
.
3
3
9
0
/
a
p
p
9
0
1
0
2
0
4
.
[
2
0
]
D
.
M
.
A
s
a
d
,
M
.
S
h
a
h
r
i
a
r
,
a
n
d
A
.
H
.
C
h
o
w
d
h
u
r
y
,
“
L
o
a
d
F
o
r
e
c
a
s
t
i
n
g
f
o
r
D
h
a
k
a
C
i
t
y
U
s
i
n
g
R
N
N
,
L
S
T
M
,
a
n
d
G
R
U
A
r
c
h
i
t
e
c
t
u
r
e
s
w
i
t
h
M
e
t
e
o
r
o
l
o
g
i
c
a
l
a
n
d
T
e
m
p
o
r
a
l
D
a
t
a
,
”
2
0
2
5
,
I
E
E
E
.
d
o
i
:
1
0
.
1
1
0
9
/
E
C
C
E
6
4
5
7
4
.
2
0
2
5
.
1
1
0
1
3
7
7
9
.
[
2
1
]
B
.
E
r
e
n
,
İ
.
A
k
s
a
n
g
ü
r
,
a
n
d
C
.
E
r
d
e
n
,
“
P
r
e
d
i
c
t
i
n
g
n
e
x
t
h
o
u
r
f
i
n
e
p
a
r
t
i
c
u
l
a
t
e
m
a
t
t
e
r
(
P
M
2
.
5
)
i
n
t
h
e
I
s
t
a
n
b
u
l
M
e
t
r
o
p
o
l
i
t
a
n
C
i
t
y
u
s
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
a
l
g
o
r
i
t
h
m
s
w
i
t
h
t
i
m
e
w
i
n
d
o
w
i
n
g
s
t
r
a
t
e
g
y
,
”
U
r
b
a
n
C
l
i
m
a
t
e
,
v
o
l
.
4
8
,
p
.
1
0
1
4
1
8
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
u
c
l
i
m
.
2
0
2
3
.
1
0
1
4
1
8
.
[
2
2
]
Y
.
W
u
,
C
.
Q
i
a
n
,
a
n
d
H
.
H
u
a
n
g
,
“
E
n
h
a
n
c
e
d
A
i
r
Q
u
a
l
i
t
y
P
r
e
d
i
c
t
i
o
n
U
s
i
n
g
a
C
o
u
p
l
e
d
D
V
M
D
I
n
f
o
r
me
r
-
C
N
N
-
L
S
T
M
M
o
d
e
l
O
p
t
i
m
i
z
e
d
w
i
t
h
D
u
n
g
B
e
e
t
l
e
A
l
g
o
r
i
t
h
m
,
”
E
n
t
r
o
p
y
,
v
o
l
.
2
6
,
n
o
.
7
,
p
.
5
3
4
,
2
0
2
4
,
d
o
i
:
1
0
.
3
3
9
0
/
e
2
6
0
7
0
5
3
4
.
[
2
3
]
Z
.
L
.
T
h
a
k
k
e
r
a
n
d
S
.
H
.
B
u
c
h
,
“
E
f
f
e
c
t
o
f
F
e
a
t
u
r
e
S
c
a
l
i
n
g
P
r
e
-
p
r
o
c
e
s
s
i
n
g
T
e
c
h
n
i
q
u
e
s
o
n
M
a
c
h
i
n
e
L
e
a
r
n
i
n
g
A
l
g
o
r
i
t
h
m
s
t
o
P
r
e
d
i
c
t
P
a
r
t
i
c
u
l
a
t
e
M
a
t
t
e
r
C
o
n
c
e
n
t
r
a
t
i
o
n
f
o
r
G
a
n
d
h
i
n
a
g
a
r
,
G
u
j
a
r
a
t
,
I
n
d
i
a
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
S
c
i
e
n
t
i
f
i
c
R
e
s
e
a
r
c
h
i
n
S
c
i
e
n
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
,
p
p
.
4
1
0
–
4
1
9
,
2
0
2
4
,
d
o
i
:
1
0
.
3
2
6
2
8
/
i
j
s
r
s
t
5
2
4
1
1
1
5
0
.
[
2
4
]
M
.
A
.
B
o
u
c
h
e
r
,
J
.
Q
u
i
l
t
y
,
a
n
d
J
.
A
d
a
m
o
w
s
k
i
,
“
D
a
t
a
A
s
s
i
m
i
l
a
t
i
o
n
f
o
r
S
t
r
e
a
mf
l
o
w
F
o
r
e
c
a
s
t
i
n
g
U
s
i
n
g
E
x
t
r
e
me
L
e
a
r
n
i
n
g
M
a
c
h
i
n
e
s
a
n
d
M
u
l
t
i
l
a
y
e
r
P
e
r
c
e
p
t
r
o
n
s
,
”
W
a
t
e
r
R
e
s
o
u
r
c
e
s
R
e
s
e
a
r
c
h
,
v
o
l
.
5
6
,
n
o
.
6
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
2
9
/
2
0
1
9
W
R
0
2
6
2
2
6
.
[
2
5
]
D
.
Y
u
l
i
z
a
r
,
D
.
A
o
f
a
n
y
,
S
.
S
o
e
k
i
r
n
o
,
N
.
A
n
a
n
d
a
,
M
.
A
.
P
r
a
b
o
w
o
,
a
n
d
S
.
P
.
A
d
i
,
“
V
i
s
u
a
l
i
z
a
t
i
o
n
o
f
R
a
i
n
f
a
l
l
C
l
a
s
s
i
f
i
c
a
t
i
o
n
u
s
i
n
g
R
a
i
n
G
a
u
g
e
b
a
s
e
d
o
n
W
e
b
s
i
t
e
,
”
2
0
2
4
,
I
E
E
E
.
d
o
i
:
1
0
.
1
1
0
9
/
A
I
M
S
6
1
8
1
2
.
2
0
2
4
.
1
0
5
1
2
6
0
0
.
[
2
6
]
L
.
K
.
K
a
l
y
a
n
a
m
a
n
d
S
.
K
a
t
k
o
o
r
i
,
“
S
i
g
m
o
i
d
-
b
a
s
e
d
N
e
u
r
o
n
P
r
u
n
i
n
g
T
e
c
h
n
i
q
u
e
f
o
r
M
L
P
s
o
n
I
o
T
E
d
g
e
D
e
v
i
c
e
s
,
”
2
0
2
3
,
I
E
E
E
.
d
o
i
:
1
0
.
1
1
0
9
/
I
C
P
E
E
V
5
8
6
5
0
.
2
0
2
3
.
1
0
3
9
1
8
7
5
.
Evaluation Warning : The document was created with Spire.PDF for Python.