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
6
1
4
~
2
6
2
3
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
6
1
4
-
262
3
2614
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
Neura
l Ne
tw
o
rk
Mo
del Develo
p
ment w
ith
So
ft
Co
m
pu
ting
Techniqu
es
for M
e
m
b
ra
ne
Fi
ltratio
n P
ro
cess
Z
a
k
a
ria
h Y
us
uf
,
No
rha
liza
Abdu
l
Wa
ha
b
,
Sh
a
f
is
h
uh
a
za
Sa
hla
n
Co
n
tr
o
l
&
M
e
c
h
a
tr
o
n
ics
E
n
g
in
e
e
rin
g
,
F
a
c
u
lt
y
o
f
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
,
Un
iv
e
rsiti
T
e
k
n
o
lo
g
i
M
a
lay
s
ia,
8
1
3
1
0
S
k
u
d
a
i
Jo
h
o
r
Ba
h
r
u
,
M
a
lay
sia
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Feb
1
0
,
2
0
1
8
R
ev
i
s
ed
Ma
y
2
8
,
201
8
A
cc
ep
ted
Ma
y
7
,
2
0
1
8
M
e
m
b
ra
n
e
b
io
re
a
c
to
r
e
m
p
lo
y
s
a
n
e
ff
i
c
ien
t
f
il
tratio
n
tec
h
n
o
l
o
g
y
f
o
r
so
li
d
a
n
d
li
q
u
i
d
se
p
a
ra
ti
o
n
in
w
a
ste
wa
t
e
r
trea
tme
n
t
p
ro
c
e
ss
.
De
v
e
lo
p
m
e
n
t
o
f
m
e
m
b
ra
n
e
f
il
tratio
n
m
o
d
e
l
is
sig
n
if
ica
n
t
a
s
th
is
m
o
d
e
l
c
a
n
b
e
u
se
d
to
p
re
d
ict
f
il
tratio
n
d
y
n
a
m
ic
w
h
ich
is
lat
e
r
u
ti
li
z
e
d
in
c
o
n
tro
l
d
e
v
e
lo
p
m
e
n
t.
M
o
st
o
f
th
e
a
v
a
il
a
b
le
m
o
d
e
ls
o
n
ly
su
it
a
b
le
f
o
r
m
o
n
it
o
rin
g
p
u
rp
o
se
,
w
h
ich
a
re
to
o
c
o
m
p
lex
,
re
q
u
ired
m
a
n
y
v
a
riab
les
a
n
d
n
o
t
s
u
it
a
b
le
f
o
r
c
o
n
tr
o
l
sy
st
e
m
d
e
sig
n
.
T
h
is
w
o
rk
f
o
c
u
sin
g
o
n
t
h
e
sim
p
le
ti
m
e
s
e
ris
m
o
d
e
l
f
o
r
m
e
m
b
ra
n
e
f
il
tratio
n
p
ro
c
e
ss
u
sin
g
n
e
u
ra
l
n
e
tw
o
rk
tec
h
n
i
q
u
e
.
I
n
th
is
p
a
p
e
r,
su
b
m
e
rg
e
d
m
e
m
b
ra
n
e
f
il
tratio
n
m
o
d
e
l
d
e
v
e
lo
p
e
d
u
sin
g
re
c
u
rre
n
t
n
e
u
ra
l
n
e
tw
o
rk
(RNN
)
train
u
sin
g
g
e
n
e
ti
c
a
lg
o
rit
h
m
(
GA
),
in
e
rti
a
w
e
ig
h
t
p
a
rti
c
le
sw
a
r
m
o
p
ti
m
iz
a
ti
o
n
(IW
-
P
S
O)
a
n
d
g
ra
v
it
a
ti
o
n
a
l
se
a
rc
h
a
lg
o
rit
h
m
(
G
S
A
).
T
h
e
se
o
p
ti
m
iza
ti
o
n
a
lg
o
rit
h
m
s
a
re
c
o
m
p
a
re
d
in
term
o
f
it
s
a
c
c
u
ra
c
y
a
n
d
c
o
n
v
e
rg
e
n
t
sp
e
e
d
i
n
u
p
d
a
t
in
g
th
e
w
e
ig
h
ts
a
n
d
b
ias
e
s
o
f
th
e
RNN
f
o
r
o
p
ti
m
a
l
f
il
tratio
n
m
o
d
e
l.
T
h
e
e
v
a
lu
a
ti
o
n
o
f
th
e
m
o
d
e
ls
is
m
e
a
su
re
d
u
si
n
g
th
re
e
p
e
rf
o
rm
a
n
c
e
e
v
a
lu
a
ti
o
n
s,
w
h
ich
a
re
m
e
a
n
sq
u
a
re
e
rro
r
(M
S
E),
m
e
a
n
a
b
so
lu
te
d
e
v
iatio
n
(
M
A
D)
a
n
d
c
o
e
ff
icie
n
t
o
f
d
e
term
in
a
ti
o
n
(R2
).
F
r
o
m
th
e
re
su
lt
s
o
b
tai
n
e
d
,
a
ll
m
e
th
o
d
s
y
i
e
ld
sa
ti
sf
a
c
to
r
y
re
su
lt
f
o
r
th
e
m
o
d
e
l,
w
it
h
th
e
b
e
st
re
su
lt
s
g
iv
e
n
b
y
IW
-
P
S
O
.
K
ey
w
o
r
d
:
A
N
N
m
o
d
eli
n
g
GA
GS
A
P
SO
SMB
R
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
:
No
r
h
aliza
A
b
d
u
l W
ah
ab
,
C
o
n
tr
o
l &
Me
c
h
atr
o
n
ic
s
E
n
g
i
n
ee
r
in
g
Dep
ar
t
m
e
n
t,
Facu
lt
y
o
f
E
lectr
ical
E
n
g
in
ee
r
in
g
,
Un
i
v
er
s
iti T
ek
n
o
lo
g
i M
ala
y
s
ia
,
8
1
3
1
0
Sk
u
d
ai,
J
o
h
o
r
,
Ma
lay
s
i
a.
E
m
ail:aliza
@
f
k
e.
u
t
m
.
m
y
1.
I
NT
RO
D
UCT
I
O
N
Me
m
b
r
an
e
f
il
tr
atio
n
i
s
a
p
r
o
m
is
in
g
tec
h
n
o
lo
g
y
i
n
m
a
n
y
s
ep
ar
atio
n
an
d
p
u
r
i
f
icatio
n
p
r
o
ce
s
s
.
T
h
is
tech
n
o
lo
g
y
is
ab
le
to
ac
h
iev
e
h
ig
h
p
er
f
o
r
m
an
ce
i
n
ter
m
o
f
p
er
m
ea
te
q
u
alit
y
.
T
h
e
ap
p
lic
atio
n
o
f
m
e
m
b
r
a
n
e
f
iltra
tio
n
is
w
id
el
y
ap
p
lied
in
m
an
y
i
n
d
u
s
tr
ial
ap
p
licatio
n
s
s
u
ch
as
w
ater
tr
ea
t
m
e
n
t
an
d
waste
w
ater
tr
ea
t
m
e
n
t
p
r
o
ce
s
s
,
f
o
o
d
p
r
o
ce
s
s
in
g
,
d
esalin
a
tio
n
,
m
ed
ical
ap
p
licatio
n
a
n
d
m
a
n
y
m
o
r
e.
De
s
p
ite
t
h
e
ad
v
a
n
ce
i
n
t
h
e
cu
r
r
en
t
tec
h
n
o
lo
g
y
,
m
e
m
b
r
an
e
f
iltra
tio
n
s
till
e
x
p
er
ien
ci
n
g
n
u
m
er
o
u
s
d
r
a
w
b
ac
k
s
in
ter
m
s
o
f
f
o
u
li
n
g
,
h
ig
h
en
er
g
y
co
n
s
u
m
p
tio
n
a
n
d
a
s
ig
n
i
f
ican
tl
y
h
i
g
h
er
o
p
er
atio
n
al
c
o
s
t c
o
m
p
ar
ed
to
co
n
v
en
tio
n
a
l te
ch
n
iq
u
es [
1
]
.
T
o
im
p
r
o
v
e
th
e
m
e
m
b
r
a
n
e
f
i
lt
r
atio
n
p
r
o
ce
s
s
,
o
n
e
o
f
th
e
o
p
tio
n
s
is
to
o
b
tain
ac
cu
r
ate
m
o
d
elin
g
o
f
th
e
f
iltra
tio
n
p
r
o
ce
s
s
.
W
ith
t
h
e
o
b
tain
ed
m
o
d
el,
an
ac
c
u
r
ate
p
r
ed
ictio
n
ca
n
b
e
m
ad
e
f
o
r
f
u
tu
r
e
ac
tio
n
o
f
th
e
p
r
o
ce
s
s
in
o
r
d
er
to
i
m
p
r
o
v
e
th
e
f
i
ltra
tio
n
p
er
f
o
r
m
a
n
ce
o
f
h
y
d
r
a
u
lic
clea
n
i
n
g
i
n
v
o
l
v
i
n
g
ae
r
atio
n
air
f
lo
w
,
b
ac
k
w
a
s
h
,
r
elax
at
io
n
a
n
d
ch
e
m
ical
c
lean
i
n
g
.
Fro
m
th
e
o
b
tain
ed
m
o
d
el,
t
h
e
b
est
s
etti
n
g
f
o
r
t
h
e
f
iltra
tio
n
p
er
m
ea
te
f
l
u
x
s
e
t
p
o
in
t
an
d
th
e
tr
an
s
m
e
m
b
r
an
e
p
r
ess
u
r
e
(
T
M
P
)
d
y
n
a
m
ic
b
eh
av
io
r
ca
n
also
b
e
d
eter
m
in
ed
.
Ho
w
e
v
er
,
d
ev
elo
p
in
g
a
w
o
r
k
in
g
m
o
d
el
f
o
r
m
e
m
b
r
an
e
f
ilt
r
atio
n
p
r
o
ce
s
s
is
n
o
t
an
ea
s
y
tas
k
,
d
u
e
to
th
e
n
o
n
li
n
ea
r
c
h
ao
tic
b
eh
a
v
io
r
o
f
f
o
u
li
n
g
.
I
n
l
iter
atu
r
e,
s
ev
er
al
m
ec
h
an
i
s
m
s
w
er
e
d
ev
el
o
p
ed
to
o
b
tain
an
o
p
tim
u
m
f
il
tr
atio
n
m
o
d
el.
Dar
c
y
’
s
la
w
s
f
o
r
e
x
a
m
p
le,
s
e
v
er
al
ter
m
s
o
f
f
o
u
li
n
g
r
es
is
ta
n
ce
ar
e
ad
d
ed
in
s
er
ies
to
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
N
eu
r
a
l Net
w
o
r
k
Mo
d
el
Dev
el
o
p
men
t wi
th
S
o
ft C
o
mp
u
tin
g
Tech
n
iq
u
es fo
r
…
(
N
o
r
h
a
liz
a
A
b
d
u
l
)
2615
r
ep
r
esen
t
th
e
f
o
u
li
n
g
la
y
er
i
n
t
h
e
m
e
m
b
r
a
n
e
f
iltra
tio
n
p
r
o
ce
s
s
to
d
ev
elo
p
th
e
f
iltra
tio
n
m
o
d
el.
Si
m
ilar
to
o
th
er
m
ec
h
a
n
i
s
tic
o
r
f
ir
s
t p
r
in
c
ip
le
m
o
d
el,
ted
io
u
s
ca
lib
r
atio
n
p
r
o
ce
s
s
i
n
v
o
l
v
in
g
m
an
y
p
ar
a
m
ete
r
s
is
r
eq
u
ir
ed
.
An
alter
n
ati
v
e
tec
h
n
iq
u
e
to
r
e
p
r
esen
t
a
d
y
n
a
m
ic
b
e
h
av
io
r
o
f
t
h
e
f
iltra
tio
n
p
r
o
ce
s
s
i
s
u
s
i
n
g
ar
ti
f
icial
n
eu
r
al
n
et
w
o
r
k
(
A
NN)
,
w
h
ic
h
h
as
h
i
g
h
ac
cu
r
ac
y
o
f
p
r
ed
icti
o
n
ca
p
ab
ilit
y
.
T
h
e
A
NN
tech
n
iq
u
e
is
e
m
p
lo
y
ed
i
n
n
u
m
er
o
u
s
ar
ea
s
i
n
cl
u
d
in
g
c
h
e
m
ical
p
r
o
ce
s
s
,
m
ec
h
an
ical
ap
p
licatio
n
s
,
f
in
a
n
cial
f
o
r
ec
a
s
tin
g
a
n
d
tr
en
d
i
n
g
,
en
v
ir
o
n
m
e
n
tal
p
r
ed
ictio
n
a
n
d
o
th
er
s
.
O
n
th
e
ar
ea
o
f
m
e
m
b
r
an
e
f
i
ltra
tio
n
m
o
d
eli
n
g
,
Ge
is
s
ler
et
a
l
[
2
]
h
a
s
d
ev
elo
p
ed
m
o
d
els
w
h
ic
h
ar
e
s
e
m
i
e
m
p
ir
ical
m
o
d
el
a
n
d
ANN
b
a
s
ed
m
o
d
el
f
o
r
p
er
m
ea
t
e
f
lu
x
m
o
d
eli
n
g
i
n
s
u
b
m
er
g
ed
ca
p
illar
y
MB
R
.
T
h
e
A
NN
m
o
d
el
is
b
ased
o
n
E
l
m
a
n
n
e
u
r
al
n
et
w
o
r
k
s
tr
u
ctu
r
e
to
p
r
ed
ict
th
e
p
er
m
ea
te
f
lu
x
.
T
h
e
tec
h
n
iq
u
es
y
ield
v
er
y
g
o
o
d
r
es
u
lts
,
w
h
er
e
th
e
s
e
m
i
e
m
p
ir
ical
m
o
d
el
r
e
q
u
ir
ed
s
m
all
i
n
p
u
t
v
ar
iab
les co
m
p
ar
ed
to
A
NN.
Ho
w
e
v
er
,
th
e
A
NN
m
o
d
el
g
a
v
e
h
ig
h
ac
cu
r
ac
y
w
it
h
t
h
e
av
er
a
g
e
er
r
o
r
o
f
2
.
7
%.
I
n
[
3
]
,
a
f
lat
s
h
ee
t
s
u
b
m
er
g
e
d
m
e
m
b
r
an
e
b
io
r
ea
cto
r
(
SMB
R
)
f
iltra
tio
n
f
o
r
w
aste
w
a
ter
f
iltra
tio
n
i
s
m
o
d
eled
u
s
i
n
g
A
NN
m
o
d
el
t
o
r
ep
r
esen
t
th
e
b
ac
k
w
a
s
h
e
f
f
e
ct
to
th
e
p
er
m
ea
te
f
lu
x
.
Mu
lt
i
la
y
er
f
ee
d
f
o
r
w
ar
d
n
eu
r
al
n
et
w
o
r
k
is
u
s
ed
to
m
o
d
el
th
e
s
y
s
te
m
.
Var
io
u
s
b
ac
k
wash
ed
a
n
d
f
il
tr
atio
n
i
n
ter
v
a
l
ar
e
u
s
ed
as
th
e
in
p
u
t
to
th
e
m
o
d
el,
m
ea
n
w
h
ile
f
lu
x
is
u
s
ed
as
t
h
e
o
u
tp
u
t.
W
it
h
t
h
e
g
i
v
en
b
ac
k
w
a
s
h
ed
to
p
er
m
e
ate
r
atio
,
th
e
m
o
d
el
is
ab
le
to
p
r
ed
ict
th
e
p
er
m
ea
t
e
f
lu
x
p
er
f
o
r
m
a
n
ce
.
Nev
er
t
h
e
less
,
t
h
e
ap
p
licatio
n
o
f
A
NN
u
s
i
n
g
co
n
v
e
n
tio
n
al
b
ac
k
p
r
o
p
ag
atio
n
(
B
P
)
alg
o
r
it
h
m
f
o
r
tr
ain
in
g
h
a
s
f
ac
ed
s
ev
er
al
p
r
o
b
lem
s
s
u
c
h
as
s
lo
w
c
o
n
v
er
g
e
n
t
a
n
d
th
e
alg
o
r
ith
m
h
as
te
n
d
en
c
y
to
tar
p
in
th
e
lo
ca
l
m
in
i
m
a.
T
h
u
s
,
th
e
ap
p
licatio
n
o
f
h
eu
r
i
s
tic
s
ea
r
ch
o
p
ti
m
izatio
n
tech
n
iq
u
e
is
o
n
e
o
f
th
e
s
o
lu
t
io
n
s
.
Sev
er
al
w
o
r
k
s
h
as
f
o
u
n
d
in
lit
er
atu
r
e
to
f
in
d
an
o
p
ti
m
al
w
ei
g
h
t
a
n
d
b
ias
v
al
u
e
in
t
h
e
tr
ai
n
in
g
o
f
th
e
n
eu
r
al
n
e
t
w
o
r
k
.
Am
o
n
g
t
h
e
w
id
el
y
u
s
ed
h
eu
r
is
tic
s
ea
r
ch
alg
o
r
ith
m
f
o
r
A
NN
m
o
d
el
is
g
en
etic
al
g
o
r
ith
m
(
GA
)
.
t
h
is
a
lg
o
r
it
h
m
is
a
m
o
n
g
th
e
p
io
n
ee
r
in
h
e
u
r
is
t
ic
o
p
tim
izatio
n
w
h
ic
h
is
in
s
p
ir
ed
f
r
o
m
t
h
e
p
o
p
u
latio
n
ev
o
lu
tio
n
ar
y
t
h
eo
r
y
.
G
A
w
as
s
u
cc
e
s
s
f
u
l
u
s
ed
i
n
[
4
]
,
[
5
]
an
d
[
6
]
t
o
tr
ain
th
e
n
eu
r
al
n
et
w
o
r
k
m
o
d
el
in
v
ar
io
u
s
ap
p
licatio
n
s
.
T
h
is
alg
o
r
ith
m
i
s
ab
le
to
p
r
o
d
u
ce
a
g
lo
b
al
s
o
l
u
tio
n
f
o
r
a
g
iv
en
o
p
ti
m
izatio
n
p
r
o
b
lem
s
.
P
ar
ticle
s
w
ar
m
s
o
p
ti
m
iza
tio
n
s
(
P
SO)
in
p
ar
ticu
lar
,
h
a
v
e
b
ec
o
m
e
a
v
er
y
e
f
f
ec
ti
v
e
o
p
ti
m
iza
tio
n
te
ch
n
iq
u
e
i
n
v
ar
io
u
s
f
ield
s
.
T
h
e
alg
o
r
it
h
m
is
f
a
s
t
an
d
v
er
y
r
eliab
le
in
s
ea
r
ch
in
g
f
o
r
m
i
n
i
m
izatio
n
.
Se
v
er
al
w
ell
k
n
o
w
n
ad
v
an
ce
m
en
t
o
f
P
SO
al
g
o
r
ith
m
s
u
c
h
a
s
r
an
d
o
m
P
SO
(
R
P
S
O)
,
co
n
s
tr
ictio
n
f
ac
to
r
P
SO
(
C
P
SO)
an
d
i
n
er
tia
w
ei
g
h
t
P
SO
(
I
W
P
SO)
h
a
v
e
b
ee
n
d
e
v
elo
p
ed
th
r
o
u
g
h
o
u
t
th
e
y
ea
r
s
.
T
h
ese
al
g
o
r
ith
m
s
h
av
e
b
ee
n
test
ed
in
p
r
ev
io
u
s
n
e
u
r
al
n
et
w
o
r
k
s
ap
p
licatio
n
a
s
r
ep
o
r
ted
in
[
7
]
,
[
8
]
an
d
[
9
]
.
Gr
av
ita
tio
n
a
l
s
ea
r
c
h
al
g
o
r
ith
m
(
GS
A
)
[
1
0
]
is
an
o
th
er
ty
p
e
o
f
o
p
ti
m
i
za
tio
n
tech
n
iq
u
e.
T
h
is
m
eth
o
d
is
in
s
p
ir
ed
b
y
t
h
e
la
w
o
f
g
r
av
it
y
a
n
d
th
e
m
as
s
in
ter
ac
tio
n
i
n
u
n
iv
er
s
e.
T
h
e
GS
A
w
a
s
r
ep
o
r
ted
to
y
ield
f
aster
co
n
v
er
g
e
n
t
r
ate
an
d
is
r
eliab
le
to
m
ee
t
t
h
e
o
b
j
ec
tiv
e
f
u
n
c
tio
n
i
n
m
a
n
y
ap
p
licatio
n
s
as r
ep
o
r
ted
in
[
1
0
]
-
[
1
2
]
.
T
h
is
w
o
r
k
f
o
c
u
s
es
o
n
m
o
d
el
d
ev
elo
p
m
en
t
o
f
m
e
m
b
r
an
e
f
iltra
tio
n
p
r
o
ce
s
s
u
s
i
n
g
r
ec
u
r
r
en
t
n
eu
r
al
n
et
w
o
r
k
(
R
NN)
m
o
d
el
tr
ai
n
b
y
t
h
r
ee
t
y
p
e
’
s
o
f
s
o
f
t
co
m
p
u
ti
n
g
o
p
ti
m
izatio
n
tec
h
n
iq
u
es
wh
ich
ar
e
G
A
,
P
SO
an
d
GS
A
.
T
h
ese
th
r
ee
al
g
o
r
ith
m
s
w
i
ll b
e
u
s
ed
to
s
ea
r
ch
f
o
r
th
e
b
est
w
ei
g
h
t
s
an
d
b
iase
s
o
f
t
h
e
r
ec
u
r
r
en
t
n
eu
r
a
l
n
et
w
o
r
k
m
o
d
el.
T
h
e
tr
ain
ed
m
o
d
e
is
t
h
en
co
m
p
ar
ed
in
ter
m
o
f
its
ac
cu
r
ac
y
o
n
th
e
tr
ai
n
in
g
an
d
test
i
n
g
o
f
m
e
m
b
r
a
n
e
f
i
ltra
tio
n
d
ata
s
et.
2.
E
XP
E
R
I
M
E
NT
S
E
T
UP
T
h
e
d
ata
s
et
is
co
ll
ec
ted
f
r
o
m
th
e
d
ev
elo
p
ed
m
e
m
b
r
an
e
b
io
r
ea
cto
r
p
ilo
t
p
lan
t
lo
ca
ted
in
P
r
o
ce
s
s
C
o
n
tr
o
l
L
ab
,
Fac
u
lt
y
o
f
E
lect
r
ical
E
n
g
i
n
ee
r
in
g
,
U
n
i
v
er
s
iti
T
ek
n
o
lo
g
i
Ma
la
y
s
ia
(
UT
M)
.
R
an
d
o
m
m
ag
n
it
u
d
e
s
tep
s
i
n
p
u
t
w
er
e
g
iv
e
n
to
th
e
s
u
ctio
n
p
u
m
p
to
s
ti
m
u
late
t
h
e
d
y
n
a
m
ic
b
e
h
av
io
r
o
f
t
h
e
p
r
o
ce
s
s
.
Fi
g
u
r
e
1
s
h
o
ws
th
e
p
la
n
t
s
c
h
e
m
atic
d
ia
g
r
a
m
w
h
ile
F
i
g
u
r
e
2
s
h
o
w
s
t
h
e
d
at
a
co
llected
f
r
o
m
th
e
p
ilo
t
p
lan
t.
T
h
e
ex
p
er
i
m
en
ts
w
er
e
ca
r
r
ied
o
u
t
in
s
in
g
le
tan
k
s
u
b
m
er
g
ed
m
e
m
b
r
an
e
b
io
r
ea
cto
r
s
,
w
it
h
w
o
r
k
in
g
v
o
l
u
m
e
o
f
2
0
L
P
alm
Oil
Mill
E
f
f
l
u
e
n
t
(
P
OM
E
)
tak
e
n
f
r
o
m
Sed
en
a
k
P
al
m
Oil
Mill
S
d
n
.
B
h
d
.
in
J
o
h
o
r
,
Ma
la
y
s
ia.
T
h
e
ae
r
atio
n
d
u
r
in
g
f
iltra
tio
n
is
s
et
ar
o
u
n
d
6
to
8
litt
er
p
er
m
in
u
te
(
L
P
M)
to
m
ai
n
tai
n
t
h
e
d
is
s
o
l
v
ed
o
x
y
g
e
n
le
v
el
to
b
e
m
o
r
e
t
h
a
n
2
p
p
m
d
u
r
i
n
g
t
h
e
e
x
p
er
i
m
e
n
t.
I
n
th
i
s
w
o
r
k
,
v
o
lta
g
e
i
n
p
u
t
o
f
t
h
e
p
er
m
ea
te
p
u
m
p
is
u
s
ed
f
o
r
th
e
in
p
u
t,
m
ea
n
w
h
ile
t
h
e
f
lu
x
an
d
T
MP
ar
e
th
e
m
ea
s
u
r
e
m
e
n
t o
u
tp
u
ts
.
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
6
1
4
–
2
6
2
3
2616
Fig
u
r
e
1
.
P
lan
t s
ch
a
m
atic
Fig
u
r
e
2
.
E
x
p
er
i
m
en
tal
d
ata
I
n
th
i
s
w
o
r
k
,
P
o
ly
et
h
er
s
u
l
f
o
n
e
(
P
E
S)
m
ater
ial
w
it
h
ap
p
r
o
x
i
m
atel
y
8
0
-
1
0
0
k
d
a
p
o
r
e
s
ize
an
d
a
s
u
r
f
ac
e
ar
ea
o
f
ap
p
r
o
x
im
ate
l
y
0
.
3
5
m
2
is
u
s
ed
in
t
h
e
f
iltra
tio
n
s
y
s
te
m
3.
M
O
DE
L
DE
VE
L
O
P
M
E
NT
R
ec
u
r
r
en
t
n
e
u
r
al
n
et
w
o
r
k
(
R
NN)
m
o
d
el
i
s
a
m
at
h
e
m
a
tical
m
o
d
el
d
ev
elo
p
ed
b
ased
o
n
t
h
e
p
ast
i
n
p
u
t
an
d
p
ast
o
u
tp
u
t
o
f
t
h
e
s
y
s
te
m
.
T
h
e
R
N
N
is
d
ev
elo
p
ed
b
ased
o
n
t
h
e
b
asic
f
ee
d
f
o
r
w
ar
d
n
eu
r
al
n
et
w
o
r
k
(
FF
NN)
.
T
h
e
tr
ain
i
n
g
alg
o
r
it
h
m
i
s
e
m
p
lo
y
ed
to
o
b
tain
s
u
ita
b
le
w
ei
g
h
ts
a
n
d
b
iases
o
f
t
h
e
n
et
w
o
r
k
i
n
o
r
d
er
to
m
i
n
i
m
ize
th
e
er
r
o
r
in
th
e
tr
ai
n
in
g
p
r
o
ce
d
u
r
e.
Fig
u
r
e
3
s
h
o
w
s
th
e
b
asic str
u
ctu
r
e
o
f
t
h
e
n
eu
r
al
n
et
w
o
r
k
m
o
d
el
.
Fig
u
r
e
3
.
Neu
r
al
n
et
w
o
r
k
s
tr
u
ctu
r
e
Fig
u
r
e
4
.
R
ec
u
r
r
en
t
n
eu
r
al
n
et
w
o
r
k
s
tr
u
ct
u
r
e
T
h
e
n
eu
r
al
n
et
w
o
r
k
eq
u
atio
n
c
an
b
e
r
ep
r
esen
ted
as
:
̂
(
)
[
∑
(
∑
)
]
(
1
)
w
h
er
e
̂
(
)
is
th
e
p
r
ed
ictio
n
o
u
tp
u
t.
F
i
is
t
h
e
f
u
n
ct
io
n
o
f
th
e
n
et
w
o
r
k
,
φ
i
s
t
h
e
in
p
u
t
v
ec
t
o
r
,
W
ij
an
d
w
ij
r
ep
r
esen
t
th
e
n
et
w
o
r
k
co
n
n
ec
tio
n
la
y
er
w
ei
g
h
t
s
a
n
d
b
iases
r
esp
ec
tiv
el
y
.
I
n
th
is
w
o
r
k
t
h
r
ee
an
d
f
iv
e
h
id
d
e
n
n
eu
r
o
n
s
ar
e
test
ed
w
it
h
b
asic
th
r
ee
la
y
er
s
n
e
u
r
al
n
e
t
w
o
r
k
s
w
h
ic
h
ar
e
th
e
i
n
p
u
t
la
y
er
,
s
i
n
g
le
h
id
d
en
la
y
e
r
a
n
d
o
u
tp
u
t
la
y
er
.
T
h
i
s
w
o
r
k
e
m
p
l
o
y
ed
n
o
n
lin
ea
r
au
to
r
e
g
r
ess
i
v
e
w
it
h
e
x
o
g
e
n
o
u
s
i
n
p
u
t
(
N
AR
X)
A
NN
s
tr
u
c
tu
r
e
w
h
ic
h
s
h
o
w
s
i
n
Fi
g
u
r
e
4
.
T
h
e
o
p
tim
izatio
n
tech
n
iq
u
e
s
ap
p
lied
in
th
e
A
NN
m
o
d
el
tr
ain
in
g
is
o
p
ti
m
ize
t
h
e
w
ei
g
h
ts
a
n
d
b
iases
o
f
t
h
e
A
N
N
s
tr
u
ct
u
r
e
b
y
m
in
i
m
iz
in
g
t
h
e
m
ea
n
s
q
u
ar
e
er
r
o
r
(
MSE
)
o
f
t
h
e
m
o
d
el
3
.
1
.
G
enet
ic
a
lg
o
rit
h
m
(
G
A)
Op
ti
m
izatio
n
al
g
o
r
ith
m
i
s
a
d
ev
elo
p
in
g
ar
ea
i
n
t
h
e
s
o
f
t
co
m
p
u
tin
g
s
tu
d
y
.
G
A
a
lg
o
r
it
h
m
is
o
n
e
o
f
th
e
p
io
n
ee
r
s
i
n
e
v
o
lu
t
io
n
b
ased
o
p
ti
m
izatio
n
tec
h
n
iq
u
e
s
a
n
d
is
a
r
eliab
le
tec
h
n
iq
u
e
i
n
o
p
ti
m
izatio
n
.
T
h
is
0
500
1000
1500
2000
2500
3000
3500
4000
0
20
40
T
i
m
e
(
s
)
Fl
u
x
(
L
/
m
2
h
)
0
500
1000
1500
2000
2500
3000
3500
4000
0
100
200
300
T
i
m
e
(
s
)
TM
P
(
m
b
a
r
)
0
500
1000
1500
2000
2500
3000
3500
4000
0
10
20
T
i
m
e
(
s
)
A
i
r
f
l
o
w
(
S
L
P
M
)
0
500
1000
1500
2000
2500
3000
3500
4000
0
2
4
T
i
m
e
(
s
)
P
u
m
p
(
V
o
l
t
)
−
−
2
−
−
2
−
2
(
)
2
(
)
−
A
NN
(
)
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
N
eu
r
a
l Net
w
o
r
k
Mo
d
el
Dev
el
o
p
men
t wi
th
S
o
ft C
o
mp
u
tin
g
Tech
n
iq
u
es fo
r
…
(
N
o
r
h
a
liz
a
A
b
d
u
l
)
2617
o
p
tim
izatio
n
tec
h
n
iq
u
e
is
w
id
el
y
ap
p
lied
in
m
a
n
y
in
d
u
s
tr
ial
ap
p
licatio
n
s
s
u
c
h
as
i
n
[
1
3
]
an
d
[
1
4
]
.
T
h
e
f
lo
w
ch
ar
t
o
f
th
e
al
g
o
r
it
h
m
ca
n
b
e
r
ep
r
esen
ted
in
F
i
g
u
r
e
5
.
Fig
u
r
e
5
.
GA
Op
ti
m
izatio
n
f
lo
w
c
h
ar
t
3
.
2
.
P
a
rt
icle
s
w
a
rm
o
pti
m
iza
t
io
n
P
ar
ticle
s
w
ar
m
o
p
ti
m
izat
io
n
(
P
SO)
is
in
s
p
ir
ed
b
y
b
eh
a
v
io
r
o
f
a
g
r
o
u
p
o
f
an
i
m
a
ls
h
u
n
t
in
g
.
T
h
i
s
h
eu
r
i
s
tic
s
ea
r
ch
o
p
ti
m
iza
tio
n
is
v
er
y
ef
f
ec
ti
v
e
in
f
i
n
d
i
n
g
o
p
tim
al
s
o
l
u
tio
n
f
o
r
m
a
n
y
p
r
o
b
lem
s
.
I
n
P
SO
alg
o
r
ith
m
,
n
u
m
b
er
o
f
s
w
ar
m
m
u
s
t
b
e
s
elec
ted
to
s
ea
r
ch
f
o
r
th
e
s
o
lu
t
io
n
.
E
ac
h
s
w
ar
m
co
n
tai
n
s
a
n
in
d
i
v
id
u
al
ca
lls
a
p
ar
ticle.
T
h
e
P
SO
m
ai
n
alg
o
r
ith
m
is
to
u
p
d
ate
th
e
p
o
s
itio
n
o
f
ea
ch
p
ar
ticle
w
it
h
an
esti
m
ated
v
elo
cit
y
.
E
ac
h
o
f
t
h
e
co
m
p
o
n
e
n
t
s
o
f
th
e
v
elo
cit
y
eq
u
atio
n
r
ep
r
esen
t
s
t
h
e
ex
p
lo
r
atio
n
ab
ilit
y
an
d
ca
p
ab
ilit
y
o
f
in
d
iv
id
u
al
lear
n
in
g
as
w
ell
a
s
s
o
cial
lear
n
i
n
g
.
T
h
e
P
SO a
lg
o
r
ith
m
e
x
ec
u
tio
n
f
lo
w
ch
ar
t i
s
s
h
o
w
n
i
n
F
i
g
u
r
e
6
.
Fig
u
r
e
6
.
P
SO
f
lo
w
c
h
ar
t
3
.
2
.
G
ra
v
it
a
t
io
na
l
s
ea
rc
h a
lg
o
rit
h
m
(
G
SA)
T
h
e
GS
A
al
g
o
r
ith
m
i
s
a
n
e
f
f
ec
tiv
e
o
p
ti
m
izatio
n
al
g
o
r
ith
m
d
ev
elo
p
ed
in
[
1
0
]
.
I
t
is
in
s
p
ir
ed
f
r
o
m
g
r
av
it
y
la
w
an
d
m
a
s
s
i
n
ter
ac
t
io
n
i
n
t
h
e
u
n
i
v
er
s
e.
Fig
u
r
e
7
p
r
esen
ts
t
h
e
f
lo
w
c
h
ar
t
o
f
t
h
e
GS
A
o
p
ti
m
izatio
n
alg
o
r
ith
m
.
T
h
e
d
etails d
ev
elo
p
m
en
t o
f
t
h
e
alg
o
r
it
h
m
i
s
p
r
ese
n
ted
in
[
1
0
]
.
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
6
1
4
–
2
6
2
3
2618
Fig
u
r
e
7
.
GSA
f
lo
w
c
h
ar
t
4.
RE
SU
L
T
S
A
ND
AN
AL
Y
SI
S
Gen
er
all
y
,
all
t
h
e
tr
ain
i
n
g
m
e
th
o
d
s
u
s
i
n
g
s
o
f
t
co
m
p
u
ti
n
g
te
ch
n
iq
u
es
ar
e
ab
le
to
p
r
o
d
u
ce
ac
ce
p
tab
le
r
esu
lt
f
o
r
t
h
e
m
o
d
el
o
b
tain
ed
.
Fro
m
t
h
e
tr
ain
i
n
g
r
es
u
lt
it
s
h
o
w
s
th
at
P
SO
al
g
o
r
ith
m
p
er
f
o
r
m
s
m
u
c
h
b
etter
w
h
e
n
co
m
p
ar
ed
w
ith
G
A
a
n
d
GS
A
.
Fro
m
F
i
g
u
r
e
8
(
a)
it
ca
n
b
e
o
b
s
er
v
ed
th
at
G
A
h
as
a
m
u
ch
f
a
s
ter
s
ea
r
c
h
i
n
g
f
o
r
th
e
r
ig
h
t
w
ei
g
h
ts
an
d
b
ias
es
o
f
th
e
n
eu
r
al
n
et
w
o
r
k
co
m
p
ar
ed
w
it
h
t
h
e
o
th
er
.
Ho
w
e
v
e
r
,
w
h
e
n
ap
p
r
o
ac
h
in
g
th
e
f
ir
s
t
5
0
iter
atio
n
s
,
P
SO
s
h
o
w
s
s
u
p
er
io
r
it
y
in
f
i
n
d
in
g
g
lo
b
al
m
i
n
i
m
u
m
f
r
o
m
t
h
e
g
iv
e
n
co
s
t
f
u
n
c
tio
n
.
Me
an
w
h
i
le,
F
i
g
u
r
e
8
(
b
)
s
h
o
w
s
th
e
zo
o
m
i
n
g
co
n
v
er
g
e
n
t
ar
ea
o
f
th
e
tr
ain
i
n
g
al
g
o
r
ith
m
s
.
I
n
ter
m
o
f
m
o
d
eli
n
g
p
er
f
o
r
m
a
n
ce
cr
iter
ia,
t
h
e
P
SO
alg
o
r
ith
m
p
er
f
o
r
m
s
b
etter
co
m
p
ar
ed
w
ith
t
h
e
o
th
er
m
et
h
o
d
s
.
T
ab
le
1
(
a)
p
r
esen
t
th
e
p
er
m
ea
te
f
lu
x
tr
ai
n
in
g
r
es
u
lt.
T
ab
le
1
(
b
)
s
h
o
w
s
th
e
r
es
u
lt
f
o
r
th
e
T
MP
m
o
d
elin
g
.
(
a)
(
b
)
Fig
u
r
e
8
.
(
a)
T
h
e
p
er
f
o
r
m
a
n
c
e
o
f
th
e
tr
ain
in
g
al
g
o
r
ith
m
(
b
)
Z
o
o
m
at
t
h
e
co
n
v
er
g
e
n
t a
r
ea
o
f
th
e
tr
ai
n
i
n
g
alg
o
r
ith
m
T
ab
le
1
.
(
a
)
P
e
r
m
ea
t
e
F
lu
x
M
o
d
elin
g
T
r
ain
in
g
R
e
s
u
l
t
M
e
t
h
o
d
R
2
M
S
E
M
A
D
GA
-
R
N
N
9
3
.
4
8
0
.
0
0
66
0
.
0
4
7
3
PSO
-
R
N
N
9
4
.
3
8
0
.
0
0
5
7
0
.
0
4
1
8
G
S
A
-
R
N
N
9
3
.
1
2
0
.
0
0
7
0
0
.
0
4
9
3
T
ab
le
1
.
(
b
)
T
M
P
Mo
d
elin
g
T
r
ain
in
g
R
e
s
u
lt
M
e
t
h
o
d
R
2
M
S
E
M
A
D
GA
-
R
N
N
9
7
.
9
0
0
.
0
0
1
4
0
.
0
2
8
7
PSO
-
R
N
N
9
7
.
8
2
0
.
0
0
1
5
0
.
0
2
9
6
G
S
A
-
R
N
N
8
3
.
5
9
0
.
0
1
1
3
0
.
0
8
4
8
Fig
u
r
e
9
(
a)
an
d
Fig
u
r
e
9
(
b
)
s
h
o
w
s
t
h
e
co
m
p
ar
i
s
o
n
b
et
w
ee
n
ac
tu
al
an
d
p
r
ed
icted
tr
ain
i
n
g
m
o
d
el
u
s
i
n
g
P
SO
-
R
NN
f
o
r
th
e
p
er
m
ea
te
f
lu
x
an
d
T
MP
.
0
100
200
300
400
500
0
0
.
1
0
.
2
0
.
3
0
.
4
0
.
5
0
.
6
0
.
7
0
.
8
I
t
e
r
a
t
i
o
n
s
Fi
t
n
e
s
s
GA
PSO
G
S
A
0
100
200
300
400
500
0
.
0
1
0
.
0
2
0
.
0
3
0
.
0
4
0
.
0
5
0
.
0
6
0
.
0
7
0
.
0
8
0
.
0
9
0
.
1
I
t
e
r
a
t
i
o
n
s
Fi
t
n
e
s
s
GA
PSO
G
S
A
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
N
eu
r
a
l Net
w
o
r
k
Mo
d
el
Dev
el
o
p
men
t wi
th
S
o
ft C
o
mp
u
tin
g
Tech
n
iq
u
es fo
r
…
(
N
o
r
h
a
liz
a
A
b
d
u
l
)
2619
(
a)
(
b
)
Fig
u
r
e
9
.
(
a)
P
er
m
ea
te
Fl
u
x
m
o
d
elin
g
tr
ai
n
in
g
r
es
u
lt
u
s
i
n
g
R
NN
-
P
SO t
h
r
ee
h
id
d
en
n
eu
r
o
n
an
d
th
e
r
esid
u
a
l
s
b
et
w
ee
n
ac
tu
al
a
n
d
m
o
d
el
(
b
)
T
M
P
m
o
d
elin
g
t
r
ai
n
in
g
r
es
u
lt
u
s
i
n
g
R
NN
-
P
SO t
h
r
ee
h
id
d
en
n
eu
r
o
n
a
n
d
th
e
r
esid
u
als b
et
w
ee
n
ac
tu
a
l a
n
d
m
o
d
el
T
h
e
v
alid
atio
n
o
f
th
e
m
o
d
el
is
d
o
n
e
b
y
u
s
in
g
d
ata
n
o
t
u
tili
ze
d
f
o
r
th
e
m
o
d
el
d
ev
elo
p
m
e
n
t.
Fro
m
th
e
r
esu
lt
g
i
v
en
i
n
t
h
e
T
ab
le
2
(
a
)
an
d
(
b
)
,
it
is
p
r
o
v
en
th
at
t
h
e
t
r
ain
i
n
g
p
er
f
o
r
m
an
ce
s
o
f
all
t
h
e
alg
o
r
ith
m
s
ar
e
ab
le
to
m
o
d
el
th
e
s
y
s
te
m
.
Ge
n
er
all
y
,
t
h
e
test
in
g
r
es
u
lt
is
le
s
s
ac
c
u
r
ate
co
m
p
ar
ed
w
it
h
t
h
e
tr
ai
n
i
n
g
.
Ho
w
e
v
er
,
f
r
o
m
th
e
te
s
ti
n
g
p
er
f
o
r
m
a
n
ce
it
s
h
o
w
ed
th
at
t
h
e
m
o
d
el
s
ar
e
ab
le
to
r
ep
licate
th
e
d
y
n
a
m
i
c
o
f
t
h
e
m
e
m
b
r
a
n
e
f
iltra
tio
n
s
y
s
te
m
as s
h
o
w
n
in
Fig
u
r
e
1
0
.
T
ab
le
2
.
(
a
)
P
e
r
m
ea
te
F
lu
x
M
o
d
elin
g
T
esti
n
g
R
e
s
u
l
t
M
e
t
h
o
d
%R
2
M
S
E
M
A
D
GA
-
R
N
N
9
3
.
6
2
0
.
0
0
6
2
0
.
0
4
5
2
PSO
-
R
N
N
9
4
.
1
9
0
.
0
0
5
6
0
.
0
4
1
8
G
S
A
-
R
N
N
9
2
.
7
6
0
.
0
0
7
0
0
.
0
4
4
8
T
ab
le
2
.
(
b
)
T
M
P
Mo
d
elin
g
T
esti
n
g
R
e
s
u
lt
M
e
t
h
o
d
%R
2
M
S
E
M
A
D
GA
-
R
N
N
9
7
.
85
0
.
0
0
1
3
0
.
0
2
7
5
PSO
-
R
N
N
9
7
.
8
7
0
.
0
0
1
2
0
.
0
2
8
3
G
S
A
-
R
N
N
8
1
.
3
3
0
.
0
1
1
5
0
.
0
8
3
9
(
a)
(
b
)
Fig
u
r
e
1
0
.
(
a)
P
er
m
ea
te
Fl
u
x
m
o
d
eli
n
g
test
in
g
r
es
u
lt u
s
i
n
g
R
NN
-
P
SO th
r
ee
h
id
d
en
n
e
u
r
o
n
an
d
th
e
r
es
id
u
al
s
b
et
w
ee
n
ac
tu
al
a
n
d
m
o
d
el
(
b
)
T
M
P
m
o
d
elin
g
tes
tin
g
r
es
u
lt
u
s
in
g
R
NN
-
P
SO t
h
r
ee
h
id
d
en
n
eu
r
o
n
an
d
t
h
e
r
esid
u
als b
et
w
ee
n
ac
tu
a
l a
n
d
m
o
d
el
0
500
1000
1500
2000
0
0
.
2
0
.
4
0
.
6
0
.
8
1
T
i
m
e
(
s
)
Fl
u
x
(
L
/
m
2
h
)
A
c
t
u
a
l
P
S
O
-
R
N
N
0
500
1000
1500
2000
-1
-
0
.
5
0
0
.
5
1
T
i
m
e
(
s
)
E
r
r
o
r
0
500
1000
1500
2000
0
0
.
2
0
.
4
0
.
6
0
.
8
1
T
i
m
e
(
s
)
TM
P
(
m
b
a
r
)
A
c
t
u
a
l
P
S
O
-
R
N
N
0
500
1000
1500
2000
-1
-
0
.
5
0
0
.
5
1
T
i
m
e
(
s
)
E
r
r
o
r
0
500
1000
1500
2000
0
0
.
2
0
.
4
0
.
6
0
.
8
1
T
i
m
e
(
s
)
Fl
u
x
(
L
/
m
2
h
)
A
c
t
u
a
l
P
r
e
d
i
c
t
e
d
0
500
1000
1500
2000
-1
-
0
.
5
0
0
.
5
1
T
i
m
e
(
s
)
E
r
r
o
r
0
500
1000
1500
2000
0
0
.
2
0
.
4
0
.
6
0
.
8
1
T
i
m
e
(
s
)
TM
P
(
m
b
a
r
)
A
c
t
u
a
l
P
r
e
d
i
c
t
e
d
0
500
1000
1500
2000
-1
-
0
.
5
0
0
.
5
1
T
i
m
e
(
s
)
E
r
r
o
r
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
6
1
4
–
2
6
2
3
2620
T
h
e
r
esu
lt
o
f
th
e
in
cr
e
m
e
n
t
o
f
th
e
h
id
d
en
n
e
u
r
o
n
s
o
f
t
h
e
n
eu
r
al
n
et
w
o
r
k
h
as
s
h
o
w
n
s
ig
n
i
f
ica
n
t
d
if
f
er
e
n
ce
f
r
o
m
t
h
e
in
itial
(
t
h
r
ee
h
id
d
en
n
e
u
r
o
n
s
)
tr
ai
n
i
n
g
.
I
t
is
an
i
n
ter
est
in
g
f
i
n
d
in
g
w
h
er
e
s
o
m
e
o
f
t
h
e
tr
ain
i
n
g
alg
o
r
it
h
m
s
h
o
w
s
i
m
p
r
o
v
e
m
e
n
t
o
f
t
h
e
m
o
d
el
w
h
ile
th
e
o
t
h
er
i
s
n
o
t
ab
le
to
f
i
n
d
s
u
itab
le
w
e
ig
h
ts
a
n
d
b
iases
o
f
t
h
e
n
eu
r
al
n
et
w
o
r
k
m
o
d
el
u
n
til
t
h
e
e
n
d
o
f
t
h
e
ite
r
atio
n
s
.
Fi
g
u
r
e
1
1
(
a)
s
h
o
w
s
t
h
e
p
er
f
o
r
m
a
n
ce
f
o
r
ea
ch
o
f
t
h
e
tr
ai
n
i
n
g
m
et
h
o
d
s
i
n
ter
m
o
f
it
f
i
t
n
es
s
a
n
d
iter
ati
o
n
s
.
Fi
g
u
r
e
1
1
(
b
)
s
h
o
w
s
t
h
e
z
o
o
m
ed
i
m
ag
e
at
t
h
e
co
n
v
er
g
e
n
t a
r
ea
.
(
a)
(
b
)
Fig
u
r
e
1
1
.
(
a)
T
h
e
p
er
f
o
r
m
an
c
e
o
f
th
e
tr
ain
in
g
al
g
o
r
ith
m
f
o
r
f
i
v
e
h
id
d
en
n
eu
r
o
n
(
b
)
Z
o
o
m
at
th
e
co
n
v
er
g
e
n
t
ar
ea
o
f
th
e
tr
ain
i
n
g
alg
o
r
it
h
m
f
o
r
f
i
v
e
h
id
d
en
n
eu
r
o
n
T
h
e
o
v
er
all
o
f
t
h
e
m
o
d
eli
n
g
p
er
f
o
r
m
a
n
ce
f
o
r
th
e
tr
ain
i
n
g
r
esu
lt
is
s
h
o
w
s
i
n
T
ab
le
3
(
a)
an
d
T
ab
l
e
3
(
b
)
.
P
er
m
ea
te
f
l
u
x
m
o
d
eli
n
g
h
ad
s
h
o
w
n
th
e
P
SO
an
d
G
A
ca
n
b
e
u
s
ed
f
o
r
th
e
R
NN
tr
ain
i
n
g
.
Fig
u
r
e
1
2
(
a)
an
d
Fig
u
r
e
12
(
b
)
r
ep
r
esen
ts
th
e
tr
a
i
n
i
n
g
r
es
u
lt
f
o
r
th
e
p
er
m
ea
te
f
lu
x
an
d
T
MP
u
s
in
g
th
e
P
SO
-
R
NN
m
o
d
eli
n
g
tech
n
iq
u
e
.
Gen
er
all
y
,
th
e
test
in
g
r
es
u
lt
s
h
av
e
s
h
o
w
n
th
at
th
e
r
es
u
lt
i
s
m
u
c
h
i
m
p
r
o
v
ed
co
m
p
ar
e
d
w
it
h
t
h
e
t
h
r
ee
h
id
d
en
n
e
u
r
o
n
.
T
h
e
P
SO
an
d
GA
tr
ain
i
n
g
alg
o
r
it
h
m
ar
e
ab
le
to
m
ain
tain
t
h
e
ac
cu
r
ac
y
o
f
th
e
m
o
d
el
w
it
h
m
i
n
i
m
u
m
er
r
o
r
in
cr
e
m
en
t
co
m
p
ar
ed
w
it
h
th
e
tr
ain
i
n
g
d
ataset.
T
h
e
test
in
g
r
es
u
lt
r
ev
ea
l
s
i
m
ilar
tr
en
d
o
f
ac
cu
r
ac
y
w
it
h
t
h
e
tr
ain
in
g
r
e
s
u
lt
w
h
er
e
t
h
e
P
SO
alg
o
r
it
h
m
o
b
tain
s
h
i
g
h
er
ac
cu
r
ac
y
o
f
th
e
m
o
d
el
f
o
llo
w
ed
w
it
h
G
A
an
d
GS
A
.
T
ab
le
4
(
a
)
an
d
T
ab
le
4
(
b
)
p
r
esen
t
th
e
t
esti
n
g
p
er
f
o
r
m
a
n
ce
o
f
th
e
m
o
d
els
w
it
h
th
e
%
R
2
,
MSE
an
d
M
AD
p
er
f
o
r
m
a
n
ce
.
T
h
e
test
i
n
g
r
e
s
u
l
t
f
o
r
th
e
p
er
m
e
ate
f
l
u
x
an
d
T
MP
w
er
e
s
h
o
w
n
i
n
F
ig
u
r
e
1
3
(
a)
an
d
Fig
u
r
e
13
(
b
)
.
T
ab
le
3
.
(
a
)
P
e
r
m
ea
te
F
lu
x
M
o
d
elin
g
f
iv
e
h
id
d
en
Ne
u
r
o
n
T
r
ain
i
n
g
R
es
u
lt
M
e
t
h
o
d
%R
2
M
S
E
M
A
D
GA
-
R
N
N
9
3
.
0
0
0
.
0
0
7
1
0
.
0
4
8
9
PSO
-
R
N
N
9
5
.
2
6
0
.
0
0
4
8
0
.
0
3
2
1
G
S
A
-
R
N
N
7
6
.
1
0
0
.
0
2
4
2
0
.
0
9
74
T
ab
le
3
.
(
b
)
T
M
P
Mo
d
elin
g
T
r
a
in
i
n
g
f
iv
e
h
id
d
en
Ne
u
r
o
n
R
e
s
u
lt
M
e
t
h
o
d
%R
2
M
S
E
M
A
D
GA
-
R
N
N
9
6
.
1
0
0
.
0
0
2
7
0
.
0
3
4
9
PSO
-
R
N
N
9
8
.
4
7
0
.
0
0
1
1
0
.
0
2
5
9
G
S
A
-
R
N
N
6
5
.
1
2
0
.
0
5
8
6
0
.
0
8
3
7
0
100
200
300
400
500
0
0
.
5
1
1
.
5
2
2
.
5
3
I
t
e
r
a
t
i
o
n
s
Fi
t
n
e
s
s
GA
PSO
G
S
A
0
100
200
300
400
500
0
0
.
0
5
0
.
1
0
.
1
5
0
.
2
0
.
2
5
I
t
e
r
a
t
i
o
n
s
Fi
t
n
e
s
s
GA
PSO
G
S
A
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
N
eu
r
a
l Net
w
o
r
k
Mo
d
el
Dev
el
o
p
men
t wi
th
S
o
ft C
o
mp
u
tin
g
Tech
n
iq
u
es fo
r
…
(
N
o
r
h
a
liz
a
A
b
d
u
l
)
2621
(
a)
(
b
)
Fig
u
r
e
1
2
.
(
a)
P
er
m
ea
te
Fl
u
x
m
o
d
eli
n
g
tr
ain
in
g
r
es
u
lt u
s
i
n
g
R
NN
-
PS
O
f
i
v
e
h
id
d
en
n
e
u
r
o
n
an
d
th
e
r
esid
u
a
l
s
b
et
w
ee
n
ac
tu
al
a
n
d
m
o
d
el
(
b
)
T
M
P
m
o
d
elin
g
tr
ai
n
in
g
r
es
u
lt
u
s
i
n
g
R
NN
-
P
SO
f
iv
e
h
id
d
en
n
eu
r
o
n
an
d
t
h
e
r
esid
u
als b
et
w
ee
n
ac
tu
a
l a
n
d
m
o
d
el
T
ab
le
4.
(
a
)
P
e
r
m
ea
te
F
lu
x
M
o
d
elin
g
T
esti
n
g
R
e
s
u
l
t
M
e
t
h
o
d
%R
2
M
S
E
M
A
D
GA
-
R
N
N
9
3
.
2
2
0
.
0
0
6
6
0
.
0
4
6
8
PSO
-
R
N
N
9
4
.
9
1
0
.
0
0
4
9
0
.
0
3
4
1
G
S
A
-
R
N
N
7
4
.
4
7
0
.
0
2
4
8
0
.
0
9
9
5
T
ab
le
4
.
(
b
)
T
M
P
Mo
d
elin
g
T
esti
n
g
R
e
s
u
lt
M
e
t
h
o
d
%R
2
M
S
E
M
A
D
GA
-
R
N
N
9
6
.
6
8
0
.
0
0
2
0
0
.
0
3
2
2
PSO
-
R
N
N
9
8
.
4
6
9
.
4
6
4
0
e
-
0
0
4
0
.
0
2
4
3
G
S
A
-
R
N
N
6
0
.
4
3
0
.
0
6
2
0
0
.
0
8
1
8
(
a)
(
b
)
Fig
u
r
e
1
3
.
(
a)
P
er
m
ea
te
Fl
u
x
m
o
d
eli
n
g
test
in
g
r
es
u
lt u
s
i
n
g
R
NN
-
P
SO th
r
ee
h
id
d
en
n
e
u
r
o
n
an
d
th
e
r
es
id
u
al
s
b
et
w
ee
n
ac
tu
al
a
n
d
m
o
d
el
(
b
)
: T
M
P
m
o
d
elin
g
test
in
g
r
es
u
lt
u
s
i
n
g
R
NN
-
P
SO t
h
r
ee
h
id
d
en
n
eu
r
o
n
a
n
d
th
e
r
esid
u
als b
et
w
ee
n
ac
tu
a
l a
n
d
m
o
d
el
5.
CO
NCLU
SI
O
N
T
h
is
p
ap
er
p
r
esen
ts
t
h
e
d
y
n
a
m
ic
m
o
d
eli
n
g
tech
n
iq
u
e
f
o
r
s
u
b
m
er
g
ed
m
e
m
b
r
an
e
f
iltra
ti
o
n
p
r
o
ce
s
s
u
s
i
n
g
R
NN
w
it
h
v
ar
io
u
s
s
o
f
t
co
m
p
u
tin
g
tec
h
n
iq
u
es.
Fro
m
th
e
r
esu
lt
it
s
h
o
w
ed
th
at
t
h
e
tr
ain
in
g
alg
o
r
it
h
m
u
s
i
n
g
s
o
f
t
co
m
p
u
t
in
g
tec
h
n
i
q
u
e
is
ab
le
to
p
r
o
d
u
ce
h
ig
h
ac
cu
r
ac
y
o
f
R
N
N
m
o
d
el.
T
h
is
w
o
r
k
co
m
p
ar
e
s
o
p
tim
izatio
n
a
lg
o
r
it
h
m
s
o
f
G
A
,
P
SO
a
n
d
GS
A
m
et
h
o
d
to
tr
ain
t
h
e
R
NN
m
o
d
el.
I
t
is
o
b
s
er
v
ed
th
at
th
e
P
SO
alg
o
r
ith
m
i
s
ab
le
to
g
i
v
e
s
h
i
g
h
er
ac
c
u
r
ac
y
co
m
p
ar
ed
w
it
h
t
h
e
o
th
er
s
f
o
llo
w
ed
w
it
h
GA
a
n
d
GS
A
.
T
h
e
0
500
1000
1500
2000
0
0
.
2
0
.
4
0
.
6
0
.
8
1
T
i
m
e
(
s
)
Fl
u
x
(
L
/
m
2
h
)
A
c
t
u
a
l
P
S
O
-
R
N
N
0
500
1000
1500
2000
-1
-
0
.
5
0
0
.
5
1
T
i
m
e
(
s
)
E
r
r
o
r
0
500
1000
1500
2000
0
0
.
2
0
.
4
0
.
6
0
.
8
1
T
i
m
e
(
s
)
TM
P
(
m
b
a
r
)
A
c
t
u
a
l
P
S
O
-
R
N
N
0
500
1000
1500
2000
-1
-
0
.
5
0
0
.
5
1
T
i
m
e
(
s
)
E
r
r
o
r
0
500
1000
1500
2000
0
0
.
2
0
.
4
0
.
6
0
.
8
1
T
i
m
e
(
s
)
Fl
u
x
(
L
/
m
2
h
)
A
c
t
u
a
l
P
r
e
d
i
c
t
e
d
0
500
1000
1500
2000
-1
-
0
.
5
0
0
.
5
1
T
i
m
e
(
s
)
E
r
r
o
r
0
500
1000
1500
2000
0
0
.
2
0
.
4
0
.
6
0
.
8
1
T
i
m
e
(
s
)
TM
P
(
m
b
a
r
)
A
c
t
u
a
l
P
S
O
-
R
N
N
0
500
1000
1500
2000
-1
-
0
.
5
0
0
.
5
1
T
i
m
e
(
s
)
E
r
r
o
r
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
6
1
4
–
2
6
2
3
2622
p
er
f
o
r
m
a
n
ce
o
f
t
h
e
P
SO
an
d
GA
al
g
o
r
ith
m
ar
e
al
m
o
s
t
s
i
m
i
lar
esp
ec
iall
y
f
o
r
th
e
T
MP
m
o
d
el.
I
n
cr
em
e
n
t
f
r
o
m
th
r
ee
to
f
iv
e
h
id
d
en
n
e
u
r
o
n
ca
n
in
cr
ea
s
e
th
e
ac
c
u
r
ac
y
o
f
th
e
m
o
d
el,
h
o
w
e
v
er
,
i
t
is
ca
n
also
b
e
tr
o
u
b
leso
m
e
f
o
r
GS
A
alg
o
r
ith
m
to
f
i
n
d
an
o
p
tim
u
m
w
ei
g
h
ts
an
d
b
iases
v
a
lu
e
f
o
r
th
e
m
o
d
el.
Fin
a
ll
y
t
h
is
m
o
d
el
w
il
l
b
e
u
s
ef
u
l
in
o
r
d
er
to
d
o
p
r
e
d
ictio
n
p
er
f
o
r
m
an
ce
a
n
d
co
n
tr
o
ller
d
esig
n
s
tr
ateg
y
i
n
t
h
e
f
u
t
u
r
e
.
ACK
NO
WL
E
D
G
E
M
E
NT
S
T
h
e
au
th
o
r
s
w
o
u
ld
li
k
e
to
t
h
an
k
t
h
e
R
esear
ch
Un
iv
er
s
it
y
Gr
an
t
(
GUP
)
v
o
te
1
3
H7
0
,
Un
i
v
er
s
iti
T
ek
n
o
lo
g
i
Ma
la
y
s
ia
f
o
r
th
e
f
i
n
an
cia
l
s
u
p
p
o
r
t.
T
h
e
f
ir
s
t
au
th
o
r
w
an
t
s
to
th
an
k
th
e
U
n
iv
er
s
iti
T
ek
n
o
lo
g
i
MA
R
A
(
UiT
M)
an
d
th
e
MO
H
E
f
o
r
th
e
T
P
M
-
SLA
I
s
c
h
o
lar
s
h
ip
RE
F
E
R
E
NC
E
S
[1
]
S.
Ju
d
d
,
T
h
e
M
BR
Bo
o
k
P
ri
n
c
ip
les
a
n
d
A
p
p
li
c
a
ti
o
n
s
o
f
M
e
m
b
ra
n
e
Bio
re
a
c
to
rs
in
W
a
ter
a
n
d
W
a
ste
w
a
ter
T
re
a
t
m
e
n
t
,
S
e
c
o
n
d
Ed
i.
El
se
v
ier,
2
0
1
0
.
[2
]
S
.
G
e
issle
r,
T
.
W
in
tg
e
n
s,
T
.
M
e
li
n
,
K.
V
o
ss
e
n
k
a
u
l,
a
n
d
C
.
Ku
ll
m
a
n
n
,
“
M
o
d
e
ll
in
g
a
p
p
ro
a
c
h
e
s
f
o
r
f
il
tratio
n
p
ro
c
e
ss
e
s
w
it
h
n
o
v
e
l
su
b
m
e
rg
e
d
c
a
p
il
lary
m
o
d
u
les
in
m
e
m
b
ra
n
e
b
i
o
re
a
c
to
rs
f
o
r
wa
ste
wa
ter
trea
t
m
e
n
t,
”
De
sa
li
n
a
ti
o
n
,
v
o
l.
1
7
8
,
p
p
.
1
2
5
-
1
3
4
,
2
0
0
5
.
[3
]
A
.
A
id
a
n
,
N.
A
b
d
e
l
-
Ja
b
b
a
r,
T
.
H.
Ib
ra
h
im
,
V
.
Ne
n
o
v
,
a
n
d
F
.
M
jall
i
,
“
Ne
u
ra
l
n
e
tw
o
rk
m
o
d
e
li
n
g
a
n
d
o
p
ti
m
iza
ti
o
n
o
f
sc
h
e
d
u
li
n
g
b
a
c
k
wa
sh
f
o
r
m
e
m
b
r
a
n
e
b
io
re
a
c
to
r,
”
Cle
a
n
T
e
c
h
n
o
l.
En
v
iro
n
.
P
o
li
c
y
,
v
o
l.
1
0
,
n
o
.
4
,
p
p
.
3
8
9
-
3
9
5
,
De
c
.
2
0
0
7
.
[4
]
T
.
T
.
Ch
o
w
,
G
.
Q.
Zh
a
n
g
,
Z.
L
in
,
a
n
d
C.
L
.
S
o
n
g
,
“
G
lo
b
a
l
o
p
ti
m
iz
a
ti
o
n
o
f
a
b
so
rp
t
io
n
c
h
il
ler
sy
ste
m
b
y
g
e
n
e
ti
c
a
lg
o
rit
h
m
a
n
d
n
e
u
ra
l
n
e
tw
o
rk
,
”
En
e
rg
y
Bu
il
d
.
,
v
o
l.
3
4
,
p
p
.
1
0
3
-
1
0
9
,
2
0
0
2
.
[5
]
A
.
R.
P
e
n
d
a
sh
te
h
,
a
F
a
k
h
ru
’l
-
Ra
z
i,
N.
Ch
a
ib
a
k
h
sh
,
L
.
C.
A
b
d
u
ll
a
h
,
S
.
S
.
M
a
d
a
e
n
i,
a
n
d
Z.
Z.
A
b
id
i
n
,
“
M
o
d
e
li
n
g
o
f
m
e
m
b
ra
n
e
b
io
re
a
c
to
r
trea
ti
n
g
h
y
p
e
rs
a
li
n
e
o
il
y
w
a
st
e
w
a
ter
b
y
a
rti
f
icia
l
n
e
u
ra
l
n
e
t
w
o
rk
.
,
”
J
.
Ha
za
rd
.
M
a
ter
.
,
v
o
l.
1
9
2
,
n
o
.
2
,
p
p
.
5
6
8
-
7
5
,
A
u
g
.
2
0
1
1
.
[6
]
W
a
n
g
,
Zh
e
n
-
Hu
a
,
Dia
n
-
Ya
o
G
o
n
g
,
X
u
L
i,
G
u
a
n
g
-
T
a
o
L
i,
a
n
d
Dia
n
-
Hu
a
Zh
a
n
g
,
"
P
re
d
ictio
n
o
f
b
e
n
d
in
g
f
o
rc
e
in
th
e
h
o
t
strip
ro
ll
in
g
p
r
o
c
e
ss
u
sin
g
a
rti
f
icia
l
n
e
u
ra
l
n
e
t
w
o
rk
a
n
d
g
e
n
e
ti
c
a
lg
o
rit
h
m
(
AN
N
-
GA
)
,
"
T
h
e
I
n
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
A
d
v
a
n
c
e
d
M
a
n
u
fa
c
tu
ri
n
g
T
e
c
h
n
o
lo
g
y
,
v
o
l
.
9
3
,
p
p
.
3
3
2
5
-
3
3
3
8
,
2
0
1
7
.
[7
]
Ch
a
tt
e
rjee
,
S
a
n
k
h
a
d
e
e
p
,
S
a
rb
a
rth
a
S
a
rk
a
r,
S
irsh
e
n
d
u
Ho
re
,
Nilan
jan
De
y
,
Am
ira
S
.
A
sh
o
u
r,
a
n
d
V
a
len
ti
n
a
E.
Ba
las
,
"
P
a
rti
c
le
sw
a
r
m
o
p
ti
m
iza
ti
o
n
train
e
d
n
e
u
ra
l
n
e
tw
o
rk
f
o
r
stru
c
tu
ra
l
f
a
il
u
re
p
re
d
icti
o
n
o
f
m
u
lt
isto
ried
RC
b
u
il
d
in
g
s
,
"
Ne
u
ra
l
Co
mp
u
ti
n
g
a
n
d
Ap
p
li
c
a
ti
o
n
s,
v
o
l
.
2
8
,
n
o
.
8
,
p
p
.
2
0
0
5
-
2
0
1
6
,
2
0
1
7
.
[8
]
K.
W
.
Ch
a
u
,
“
A
p
p
li
c
a
ti
o
n
o
f
a
P
S
O
-
b
a
se
d
n
e
u
ra
l
n
e
tw
o
rk
in
a
n
a
l
y
sis
o
f
o
u
tco
m
e
s
o
f
c
o
n
stru
c
ti
o
n
c
lai
m
s,”
Au
to
m.
Co
n
str.
,
v
o
l.
1
6
,
n
o
.
5
,
p
p
.
6
4
2
-
6
4
6
,
A
u
g
.
2
0
0
7
.
[9
]
L
.
Zh
if
e
n
g
,
P
.
Da
n
,
W
.
Jia
n
h
u
a
,
a
n
d
Y.
S
h
u
a
n
g
x
i,
“
M
o
d
e
l
li
n
g
o
f
M
e
m
b
ra
n
e
F
o
u
l
in
g
b
y
P
CA
-
P
S
OB
P
Ne
u
ra
l
Ne
tw
o
rk
,
”
2
0
1
0
I
n
t.
C
o
n
f
.
Co
m
p
u
t.
Co
n
tro
l
In
d
.
E
n
g
.
,
v
o
l.
3
4
,
p
p
.
3
4
-
3
7
,
2
0
1
0
.
[1
0
]
E.
Ra
sh
e
d
i,
H.
Ne
z
a
m
a
b
a
d
i
-
p
o
u
r,
a
n
d
S
.
S
a
ry
a
z
d
i,
“
G
S
A
:
A
Gra
v
it
a
ti
o
n
a
l
S
e
a
rc
h
A
lg
o
rit
h
m
,
”
In
f.
S
c
i.
(
Ny
).
,
v
o
l.
1
7
9
,
n
o
.
1
3
,
p
p
.
2
2
3
2
-
2
2
4
8
,
Ju
n
.
2
0
0
9
.
[1
1
]
E.
Ra
sh
e
d
i,
H.
Ne
z
a
m
a
b
a
d
i
-
p
o
u
r,
a
n
d
S
.
S
a
ry
a
z
d
i,
“
F
il
ter
m
o
d
e
li
n
g
u
si
n
g
g
ra
v
it
a
ti
o
n
a
l
se
a
rc
h
a
l
g
o
rit
h
m
,
”
En
g
.
Ap
p
l
.
Arti
f
.
In
te
l
l.
,
v
o
l
.
2
4
,
n
o
.
1
,
p
p
.
1
1
7
-
1
2
2
,
F
e
b
.
2
0
1
1
.
[1
2
]
L
i,
Ch
a
o
sh
u
n
,
Na
n
Zh
a
n
g
,
X
i
n
ji
e
L
a
i,
Jia
n
z
h
o
n
g
Zh
o
u
,
a
n
d
Ya
n
h
e
X
u
,
"
De
sig
n
o
f
a
f
ra
c
ti
o
n
a
l
-
o
rd
e
r
P
ID
c
o
n
tro
ll
e
r
f
o
r
a
p
u
m
p
e
d
sto
ra
g
e
u
n
it
u
sin
g
a
g
ra
v
it
a
ti
o
n
a
l
se
a
rc
h
a
lg
o
rit
h
m
b
a
se
d
o
n
t
h
e
Ca
u
c
h
y
a
n
d
Ga
u
ss
ian
m
u
tatio
n
,
"
In
fo
rm
a
t
io
n
S
c
ien
c
e
s,
v
o
l
.
3
9
6
,
p
p
.
1
6
2
-
1
8
1
,
2
0
1
7
.
[1
3
]
J.
K.
Ra
jes
h
,
S
.
K.
G
u
p
ta,
G
.
P
.
Ra
n
g
a
iah
,
a
n
d
A
.
K.
Ra
y
,
“
M
u
lt
i
o
b
jec
ti
v
e
Op
ti
m
iza
ti
o
n
o
f
S
tea
m
Re
f
o
r
m
e
r
P
e
rf
o
rm
a
n
c
e
Us
in
g
G
e
n
e
ti
c
A
lg
o
rit
h
m
,
”
In
d
.
E
n
g
.
Ch
e
m.
Res
.
,
v
o
l.
3
9
,
n
o
.
3
,
p
p
.
7
0
6
-
7
1
7
,
M
a
r.
2
0
0
0
.
[1
4
]
E.
Ho
p
p
e
r
a
n
d
B.
T
u
rto
n
,
“
A
g
e
n
e
ti
c
a
lg
o
rit
h
m
f
o
r
a
2
D
in
d
u
stri
a
l
p
a
c
k
in
g
p
ro
b
lem
,
”
Co
mp
u
t.
In
d
.
En
g
.
,
v
o
l.
3
7
,
n
o
.
1
-
2
,
p
p
.
3
7
5
-
3
7
8
,
Oc
t.
1
9
9
9
.
B
I
O
G
RAP
H
I
E
S
O
F
AUTH
O
RS
Za
k
a
r
ia
h
Yusu
f
re
c
e
iv
e
d
h
is
B.
En
g
.
Ho
n
s (E
lec
tri
c
a
l)
a
n
d
M
S
c
.
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
f
ro
m
Un
iv
e
rsiti
T
e
k
n
o
lo
g
i
M
A
RA
,
M
a
la
y
sia
in
2
0
0
7
a
n
d
2
0
1
2
re
sp
e
c
ti
v
e
l
y
.
He
is
c
u
rre
n
tl
y
w
o
rk
in
g
to
w
a
rd
h
is
P
h
D
in
p
ro
c
e
ss
c
o
n
tro
l
a
t
t
h
e
F
a
c
u
lt
y
o
f
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
,
Un
iv
e
rsiti
T
e
k
n
o
lo
g
i
M
a
la
y
sia
.
H
e
h
a
s
a
f
e
w
y
e
a
rs’
e
x
p
e
rien
c
e
s
in
c
o
n
tro
l
a
n
d
in
stru
m
e
n
tatio
n
f
ro
m
v
a
rio
u
s
in
d
u
stries
.
His
c
u
rre
n
t
re
se
a
rc
h
in
t
e
re
sts
in
c
lu
d
e
m
e
m
b
ra
n
e
f
il
tratio
n
a
n
d
a
p
p
li
c
a
ti
o
n
o
f
so
f
t
c
o
m
p
u
ti
n
g
in
th
e
a
re
a
o
f
p
ro
c
e
ss
m
o
d
e
li
n
g
a
n
d
c
o
n
tro
l
Ir
.
Dr
.
N
o
r
h
a
li
z
a
Abd
u
l
W
a
h
a
b
is
c
u
rre
n
tl
y
a
n
A
s
so
c
iat
e
P
ro
f
e
ss
o
r
a
t
Un
iv
e
rsiti
T
e
k
n
o
lo
g
i
M
a
la
y
sia
(UT
M
).
S
h
e
is
c
u
rre
n
tl
y
th
e
He
a
d
De
p
a
rtme
n
t
o
f
Co
n
tro
l
a
n
d
M
e
c
h
a
tro
n
ics
En
g
in
e
e
rin
g
De
p
a
rtm
e
n
t
a
t
th
e
F
a
c
u
lt
y
o
f
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
,
UT
M
.
S
h
e
c
o
m
p
lete
d
h
e
r
P
h
D
in
E
lec
tri
c
a
l
En
g
in
e
e
rin
g
m
a
j
o
rin
g
in
Co
n
tro
l
in
J
u
ly
2
0
0
9
.
S
h
e
is
a
c
ti
v
e
l
y
in
v
o
lv
e
d
in
re
se
a
rc
h
in
g
a
n
d
tea
c
h
in
g
i
n
th
e
f
ield
o
f
in
d
u
strial
p
ro
c
e
ss
c
o
n
tro
l.
He
r
e
x
p
e
rti
se
is
in
m
o
d
e
ll
in
g
a
n
d
c
o
n
tr
o
l
o
f
in
d
u
strial
p
ro
c
e
ss
p
lan
t.
Re
c
e
n
tl
y
sh
e
h
a
s
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
N
eu
r
a
l Net
w
o
r
k
Mo
d
el
Dev
el
o
p
men
t wi
th
S
o
ft C
o
mp
u
tin
g
Tech
n
iq
u
es fo
r
…
(
N
o
r
h
a
liz
a
A
b
d
u
l
)
2623
w
o
rk
e
d
p
rim
a
ril
y
o
n
d
if
f
e
r
e
n
t
ty
p
e
s
o
f
d
o
m
e
stic
a
n
d
in
d
u
strial
w
a
s
tew
a
ter
trea
t
m
e
n
t
tec
h
n
o
l
o
g
y
to
wa
rd
s
o
p
ti
m
iza
ti
o
n
a
n
d
e
n
e
rg
y
sa
v
in
g
s
y
ste
m
.
S
h
a
fis
h
u
h
a
z
a
S
a
h
la
n
g
ra
d
u
a
ted
f
ro
m
Un
iv
e
rsit
y
o
f
S
h
e
ff
ield
,
U
K
in
2
0
0
2
w
it
h
a
n
M
En
g
in
Co
n
tro
l
S
y
st
e
m
s
En
g
in
e
e
rin
g
.
In
2
0
1
0
,
s
h
e
g
ra
d
u
a
ted
f
ro
m
Un
iv
e
rsit
y
o
f
W
e
ste
rn
A
u
stra
li
a
u
p
o
n
c
o
m
p
letio
n
o
f
h
e
r
stu
d
ies
m
a
jo
rin
g
in
C
o
n
tr
o
l
S
y
ste
m
A
lg
o
rit
h
m
.
Cu
rre
n
tl
y
,
sh
e
is
w
o
rk
in
g
in
th
e
Co
n
tro
l
&
M
e
c
h
a
tro
n
ics
En
g
in
e
e
rin
g
De
p
a
rtm
e
n
t
in
F
a
c
u
lt
y
o
f
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
in
Un
iv
e
rsiti
T
e
k
n
o
lo
g
i
M
a
la
y
sia
,
S
k
u
d
a
i,
Jo
h
o
r
a
s
a
s
e
n
io
r
lec
tu
re
r.
He
r
re
se
a
rc
h
in
tere
st
in
c
lu
d
e
s
Co
n
t
ro
l
A
lg
o
rit
h
m
a
n
d
it
s
a
d
v
a
n
c
e
m
e
n
t,
a
s
w
e
ll
a
s an
ima
l
b
io
tec
h
n
o
lo
g
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.