Internati
o
nal
Journal of Ele
c
trical
and Computer
Engineering
(IJE
CE)
V
o
l.
6, N
o
. 2
,
A
p
r
il
201
6, p
p
.
63
0
~
63
8
I
S
SN
: 208
8-8
7
0
8
,
D
O
I
:
10.115
91
/ij
ece.v6
i
2.8
884
6
30
Jo
urn
a
l
h
o
me
pa
ge
: h
ttp
://iaesjo
u
r
na
l.com/
o
n
lin
e/ind
e
x.ph
p
/
IJECE
Modified Predictive Con
t
rol fo
r a Class of Electro-Hydraulic
Actu
at
or
Abd
u
lrahm
a
n
A.
A.
Emhem
e
d*,
R
o
sbi Bi
n Ma
ma
t*
, Ahma
d ‘A
thif
M
o
hd
Faudzi
**
,
Mohd
Ridz
uan Joh
a
r
y
**, K
h
airud
d
in Os
man
*
**
* Department of
Control
and Mechatronic Eng
i
neering,
Univ
ersiti
Teknologi Malay
s
ia, 81300
Skudai, Malay
s
ia
** Centr
e
for
Ar
tificial Int
e
ll
igen
ce
and Robo
tics
(CAIRO
),
Universiti Teknolog
i Malay
s
ia, 81300
Skudai, Mal
a
y
s
i
a
*** Departm
e
n
t
of Industrial
Electroni
cs, Univ
er
siti T
e
kni
k
a
l
Malay
s
ia Melak
a
, 7
6100
Durian
Tu
nggal, Mal
a
y
s
ia
Article Info
A
B
STRAC
T
Article histo
r
y:
Received Aug 23, 2015
Rev
i
sed
D
ec 28
, 20
15
Accepted Dec 12, 2016
Man
y
model pr
edictiv
e con
t
rol (
M
PC) algorithms have been
pro
posed in the
liter
a
tur
e
depend
ing on the condi
t
i
onali
t
y
of the s
y
stem
m
a
trix and the tuning
control par
a
meters. A modified
p
r
edictiv
e con
t
rol method is p
r
op
osed in
this
paper. Th
e modified pr
edictiv
e method is b
a
sed on the con
t
rol matrix
formulation co
mbined with op
timized
move suppression coefficient. Poor
d
y
namics and high nonlinear
ities are parts of
the
difficu
lti
es in th
e control of
the Electro-H
y
d
r
aulic Actuator
(EHA) functions, which make the proposed
m
a
trix an attr
act
ive solution. Th
e de
velop
e
d con
t
roller is designed based on
simulation model of a position control
EHA to r
e
duce
the ov
ershoot of th
e
s
y
s
t
em
and to ac
hieve bet
t
er and
s
m
oother trackin
g. The perform
a
n
ce of the
designed contro
ller a
c
hiev
ed quick re
sponse and accura
te beh
a
vior of the
track
ing com
p
ar
ed to
th
e prev
iou
s
s
t
ud
y.
Keyword:
M
odi
fi
e
d
pre
d
i
c
t
i
v
e
co
nt
r
o
l
Electro-hydra
u
lic actuator
Po
sition
co
n
t
rol
Copyright ©
201
6 Institut
e
o
f
Ad
vanced
Engin
eer
ing and S
c
i
e
nce.
All rights re
se
rve
d
.
Co
rresp
ond
i
ng
Autho
r
:
Ab
d
u
l
r
a
h
m
a
n A.
A. Em
hem
e
d,
Depa
rt
m
e
nt
of
C
ont
r
o
l
a
n
d
M
echat
r
oni
c
En
g
i
neeri
n
g,
Fac
u
l
t
y
of El
ect
ri
cal
En
gi
nee
r
i
n
g,
U
n
i
v
ersiti Tekn
o
l
o
g
i
Malaysia, 813
00
Skudai, Malaysia.
Em
a
il: ab
d
o
_
8
3
f
@yaho
o
.com
1.
INTRODUCTION
Gen
e
ralized
Pred
ictiv
e Con
t
ro
l (GPC
) is co
nsid
er
ed
o
n
e o
f
th
e m
o
st p
opu
lar Model Pred
ictiv
e
C
ont
r
o
l
(M
PC
) al
g
o
ri
t
h
m
s
i
n
i
n
d
u
st
ry
.
GP
C
wi
t
h
i
n
t
e
g
r
a
l
act
i
on i
s
de
ri
ved
ba
sed
o
n
t
h
e m
i
nim
i
zat
i
on
of
a
m
odi
fi
ed p
r
edi
c
t
i
v
e per
f
o
rm
ance cri
t
e
ri
on
.
An i
m
port
a
nt
a
d
v
a
n
t
ag
e of th
is typ
e
of
p
r
ed
ictiv
e con
t
ro
l is its
ability to cope with hard constraint
s on controls and states. Recently, se
veral pre
d
ictive
control approaches
wh
ere
p
r
op
o
s
ed
to so
l
v
e EHA
Actu
ators
p
r
o
b
l
em
s. EHA
p
o
s
ition
con
t
rol syste
m
s are very im
p
o
r
tan
t
fo
r t
h
e
industrial application of c
o
ntrol system
s, e.
g., airc
raft
flight control,
re
m
o
te r
obot position control, due
t
o
their cha
r
acteri
s
tics of fast response
, accurate positioni
ng,
and so on. Ele
c
tro-Hydr
a
u
licsyste
m
s are com
p
lex,
n
o
n
lin
ear and
d
i
fficu
lt to
id
en
tify and
co
n
t
ro
l.
A clos
e inve
stigation
has sugge
sted t
h
at the
problem
s
are
m
o
stl
y
related
to
th
e
n
a
t
u
re
o
f
h
ydrau
lic fu
n
c
tion
s
. Fi
rst
o
f
all, in
EHA,
flex
ib
le con
n
ecting
ho
ses, large
vol
um
e of
fl
ui
d
un
de
r c
o
m
p
ressi
o
n
a
n
d t
r
appe
d
ai
r i
n
t
h
e
hy
d
r
aul
i
c
f
l
ui
d l
e
a
d
t
o
hi
gh
c
o
m
p
l
i
a
nce. T
h
i
s
p
a
p
e
r’s m
a
in
c
o
n
t
ribu
tio
n
is
m
o
d
i
fied
p
r
ed
i
c
tiv
e co
n
t
ro
l
desig
n
e
d
to
enhan
ce po
sition
co
n
t
ro
l issu
e for EHA
pl
ant
.
T
h
e
pr
o
pos
ed m
e
t
hod
i
s
usi
n
g a
new
m
a
t
r
i
x
com
b
i
n
ed
wi
t
h
o
p
t
i
m
al
t
uned m
ove s
u
p
p
re
ssi
o
n
fact
o
r
.
Th
e GPC m
e
t
h
od
design
ed
an
d
im
p
l
e
m
en
t
e
d
to
ov
er
co
m
e
v
a
r
i
ou
s co
n
t
r
o
l pro
b
l
em
s i
n
on
e algo
r
ithm [
1
].
They a
r
e ca
pa
ble to stabilize proces
ses
with
varia
b
le
para
m
e
t
e
rs. It
ca
n
al
so a
d
apt
wi
t
h
m
odel
or
der
whi
c
h
ch
ang
e
s immed
i
ately p
r
ov
id
ed
th
at th
e in
p
u
t/ou
t
pu
t
d
a
ta are v
a
lu
ab
le to
allow well-fou
nded
p
l
an
t
id
en
tificatio
n.
Big
d
e
li and
Haeri i
m
p
l
e
m
en
ted
GPC algo
ri
th
m
fo
r Po
sitio
n
C
o
n
t
ro
l
o
f
an
Ultrason
ic
Mo
to
r
(USM
). T
h
e m
o
tor posses
s
heavy nonli
n
ear, and loa
d
depe
ndent c
h
a
r
acteristics suc
h
as dea
d
-z
one. These
pr
o
p
ert
i
e
s ha
v
e
m
a
de bi
g ch
al
l
e
nge f
o
r P
o
si
t
i
on an
d vel
o
ci
t
y
cont
rol
o
f
t
h
e pl
ant
.
T
h
e GPC
im
pro
v
e
d t
h
e
p
e
rform
a
n
ce of th
e m
o
to
r
fo
r
bo
th
po
sitio
n track
i
n
g
an
d d
i
sturb
a
n
c
e rej
ecti
o
n
[2
]. Th
e
p
r
esen
ted
GPC
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
IJEC
E V
o
l
.
6, No
. 2, A
p
ri
l
20
16
:
63
0 – 6
3
8
63
1
p
o
s
ition
con
t
ro
l sch
e
m
e
p
r
ov
id
es
d
a
m
p
in
g n
o
i
se,
fast
resp
on
se, and
go
od
rej
ection
o
f
u
n
c
ertain
ties for
in
du
ctio
n
m
o
tor. Th
e po
sition
lo
op
is reg
u
l
at
ed
with
a lin
ear GPC con
t
ro
ller and
th
e cu
rren
t
lo
op
s i
n
corp
orate
the Space Vec
t
or Pulse
W
i
dt
h Modulation
(SVPW
M
)
with
PI control [3]. T
h
e In
vestigation has
achieve
d
p
e
rfect resu
lts b
a
sed
on
time d
e
lay en
han
cem
en
t fo
r
h
ydrau
lic po
si
tio
n
con
t
ro
l syste
m
(HPCS) u
s
ing
m
odi
fi
ed pre
d
i
c
t
i
v
e cont
r
o
l
t
echni
q
u
e. B
o
t
h
se
nso
r-t
o-
cont
rol
l
e
r (
S
-
C
) an
d co
nt
r
o
l
l
e
r-t
o-act
uat
o
r
(C
-A
)
net
w
or
k-i
n
d
u
c
e
d
del
a
y
s
are
m
odel
e
d as
M
a
rk
ov
chai
n
s
[
4
]
.
A
resea
r
ch
g
r
o
u
p
f
r
o
m
The Uni
v
e
r
si
t
y
of
Man
ito
b
a
, Can
a
d
a
ap
p
lied
GPC con
t
ro
l to
a class o
f
p
o
s
ition
con
t
ro
l of h
ydrau
lic
m
a
n
i
p
u
l
ato
r
. The
i
nvest
i
g
at
e
d
ex
peri
m
e
nt
al
has been d
o
n
e t
h
r
o
u
g
h
dat
a
connection
betwee
n the com
pute
r
and the real plant
t
h
r
o
u
g
h
acq
ui
si
t
i
on card
.
An
on
-l
i
n
e est
i
m
a
ti
on o
f
t
i
m
e
vary
i
ng pl
ant
pa
ra
m
e
t
e
rs and i
d
e
n
t
i
f
i
cat
i
on achi
e
ved
.
The i
n
fl
uence
of G
P
C
desi
g
n
param
e
t
e
rs was al
so expe
ri
m
e
nt
al
l
y
st
udi
ed
and o
b
se
rve
d
[5]
.
M
P
C
ad
va
nt
ages
are d
e
tailed
in [6
], [7
]. Desig
n
i
n
g
a pred
ictiv
e co
n
t
ro
lle
r for in
tegratin
g p
r
o
c
esses is a b
i
g
ch
alleng
ing
step
because of the num
ber
of
adjusta
b
le para
m
e
ters (p
re
di
ction horizon, cont
rol
horizon, m
ove suppressi
on
coefficient, a
n
d the
sam
p
ling tim
e
) that affe
ct the cl
ose
d
-l
oo
p
per
f
o
rm
ance. Th
e n
o
v
el analytical
m
e
thod has
been
achi
e
ved
go
o
d
per
f
o
r
m
a
nce i
n
cl
ude
(se
t
poi
nt
t
r
ac
ki
n
g
,
o
v
ers
h
oot
l
i
m
i
t
a
t
i
on, a
n
d
d
i
st
urba
nce
re
je
ct
i
on)
base
d o
n
si
m
u
l
a
t
i
on
res
u
l
t
s
fo
r va
ri
es
pr
ocesses
[8]
.
T
h
e s
h
i
p
dy
na
m
i
c posi
t
i
oni
n
g
sy
st
em
has som
e
characte
r
istics suc
h
as nonli
n
ear, larg
e d
e
lay an
d
stro
ng
co
up
ling
,
wh
ich is d
i
fficu
lt to
stru
cture th
e accu
rate
math
e
m
atica
l
m
o
d
e
l an
d
v
e
ry co
m
p
lica
t
ed
to
m
o
d
e
l it. S
o
th
e fu
zzy pred
ictiv
e con
t
rol u
s
ed
to
con
t
ro
l th
e
sh
ip
d
y
n
a
m
i
c p
o
s
ition
i
ng
syste
m
(SDPS) in th
ree
d
e
g
r
ees
o
f
freedo
m
(DOF) h
a
s b
e
en
i
n
v
e
stig
ated
and
the
si
m
u
latio
n
resu
lts sh
ow th
at
th
e fu
zzy pred
ictiv
e co
n
t
ro
l
can
orien
t
th
e vessels effectively [9
]. An
imp
r
ov
ed
robu
st m
o
d
e
l pred
ictiv
e con
t
ro
ller (RMPC
)
is p
r
esen
te
d
ba
sed
o
n
m
odel
r
e
fere
nce a
d
a
p
t
i
v
e sy
st
em
(M
R
A
S)
for three
d
e
g
r
ee freedo
m
sate
llite. Th
e con
t
ro
l d
e
sign
ed b
e
cau
se
o
f
ex
tern
al d
i
st
u
r
b
a
n
c
e is co
m
p
en
sat
e
d
on
th
e stab
ility an
d
p
e
rfo
r
m
a
n
ce o
f
cl
o
s
ed
loo
p
syste
m
. Then
th
e
resu
lts o
f
propo
sed
RMPC co
m
p
ared
t
o
gene
ral
i
zed i
n
crem
ent
a
l
pred
i
c
t
i
v
e cont
r
o
l
(
G
IPC
)
a
nds
h
o
w
t
h
at
t
h
e R
M
PC
m
e
t
hod i
s
m
o
re rob
u
st
t
h
an t
h
e
GIPC
m
e
t
hod
[1
0]
.
A
n
El
e
c
t
r
o-
Pne
u
m
a
t
i
c
cl
ut
ch act
uat
o
r c
o
nt
r
o
l
l
e
d
on/
of
f
val
v
es
usi
n
g P
u
l
s
e
W
i
dt
h
M
o
d
u
l
a
t
i
on
(P
WM
) an
d
No
nl
i
n
ear M
odel
Predi
c
t
i
v
e C
o
nt
r
o
l
(NM
P
C
)
app
r
oac
h
i
s
appl
i
e
d t
o
des
i
gn a
n
ex
p
licit reference track
ing
.
The p
e
rform
a
n
ce o
f
t
h
e d
e
si
gn
ed
con
t
ro
ller
h
a
s b
e
tter
q
u
a
lity in
co
m
p
arison to
an
ex
p
licit qu
an
tized
NMPC con
t
ro
ller wit
h
out PW
M
[11
]
. Th
e
p
r
esen
ted article h
a
s th
ese
o
u
tlin
es:
Sect
i
on
I:
I
n
t
r
o
duct
i
o
n a
n
d
ba
ckg
r
ou
n
d
st
u
d
i
e
s o
f
t
h
e
E
H
A
cont
rol
l
e
d
by
p
r
edi
c
t
i
v
e c
o
nt
r
o
l
.
Sect
i
on II:
T
h
e
e
xpe
ri
m
e
ntal
set
u
p
an
d i
d
ent
i
f
i
cat
i
o
n
m
e
t
hods
f
o
r
vari
es o
r
de
rs of
t
h
e EH
A ar
e
prese
n
t
e
d
.
Sect
i
on
II
I:
T
h
e m
odi
fi
ed
pre
d
i
c
t
i
v
e c
ont
r
o
l
,
w
h
ere
t
h
e
pr
o
pos
ed
m
a
t
r
i
x
are e
xpl
ai
ne
d.
Sect
i
on
I
V
:
Th
e t
uni
ng
st
rat
e
gy
an
d c
o
m
put
at
i
ons
of
t
h
e
co
nt
r
o
l
pa
ram
e
t
e
r
s
are
p
r
esent
e
d
.
Sectio
n
V
:
Si
m
u
la
tio
n
d
e
sign
o
f
t
h
e pro
posed
co
n
t
r
o
ller
an
d co
m
p
ar
ison
w
i
t
h
pr
ev
ious stud
y b
a
sed
on
m
o
d
e
l o
f
po
sitio
n con
t
ro
l EHA.
Sect
i
on
VI:
Di
scussi
o
n
a
n
d c
oncl
u
si
o
n
pres
ent
s
i
n
resul
t
s
are sum
m
ari
z
ed an
d
di
scu
sse
d i
n
t
h
e l
i
ght
o
f
o
u
r ob
j
ectiv
es.
2.
MODEL IDENTIFICATION OF THE E
H
A
In
EH
A sy
st
e
m
operat
i
on,
hy
d
r
aul
i
c
act
u
a
t
o
r ca
uses l
i
n
ear act
i
o
n f
r
o
m
di
fferent
pr
essur
e
i
n
t
h
e
cylin
d
e
r
b
y
p
u
sh
ing
ag
ain
s
t t
h
e
p
i
ston
.
Generally, th
e actuato
r con
s
ists of two o
il ch
am
b
e
rs, sep
a
rated
b
y
the
pi
st
o
n
. T
h
e
re
sul
t
o
f
oi
l
fl
o
w
s m
ovi
n
g
i
n
t
o
a
n
d
o
u
t
of t
h
e c
h
am
bers d
r
i
v
i
n
g t
h
e pi
st
on
an
d
ge
nera
t
e
t
h
e
requ
ired
p
r
essu
res to
m
o
v
e
th
e lo
ad
of th
e actu
a
to
r [12
]
. A lin
ear typ
e
actu
a
tio
n
of th
e EHA system
u
s
ing
a
si
ngl
e-e
n
ded
c
y
l
i
nder a
nd c
o
nt
r
o
l
l
e
d wi
t
h
ON/
OFF
val
v
e i
s
con
s
i
d
ere
d
i
n
t
h
e ex
pe
r
i
m
e
nt
al
desi
gn
. Th
e
expe
ri
m
e
nt
al
set
up a
n
d t
h
e
i
d
ent
i
f
i
cat
i
on
res
p
o
n
ses
o
f
t
h
e E
HAs
h
o
w
n
i
n
fi
gu
re
1.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJECE
ISS
N
:
2088-8708
Mod
ified
Pred
i
ctive Co
n
t
ro
l fo
r
a
Class
o
f
Electro
-H
yd
rau
lic Actu
a
t
o
r
(Ab
d
u
l
ra
hma
n
A.A. Emh
e
med
)
63
2
Figu
re
1
.
(a
) R
eal expe
rim
e
ntal setup
o
f
th
e
EH
A
,
(b
) Ide
n
t
i
fication
res
p
o
n
ses of
o
f
t
h
e EHA
In
th
e system id
en
tificatio
n p
r
o
c
ess, prev
iou
s
ly, th
is EHA
p
r
o
cess is id
en
tified
u
s
in
g
ARX
id
en
tificatio
n
t
ech
n
i
q
u
e
. In
t
h
is p
a
p
e
r t
h
e
id
en
tifica
t
i
on
descri
bed
base
d o
n
c
u
r
v
e r
e
act
i
on a
p
p
r
o
x
i
m
at
i
on
t
echni
q
u
es
.T
he
fi
rst
or
der
st
ep
res
p
o
n
se
sy
st
em
i
d
en
t
i
f
i
cat
i
on m
e
t
hods
ap
p
r
o
x
i
m
at
ed t
h
e param
e
t
e
rs bas
e
d
o
n
th
e inpu
t/o
u
t
pu
t th
e m
o
d
e
l is determ
in
e
d
b
y
ap
p
l
ying
n
on-p
a
ram
e
t
r
ic system
id
en
tificatio
n b
a
sed
on
expe
ri
m
e
nt
al
resp
onse
f
o
r
t
h
e EH
A.
T
h
e
pr
ocess
gai
n
de
s
c
ribes
base
d
on the
steady
state effect
of the input
change t
o
the c
h
ange
of the
output. T
h
e
dea
d
tim
e can be
dir
ectly r
ead
f
r
om
th
e ou
tpu
t
r
e
sp
on
se [1
3
]
, [14
]
.
1.
587
.
(1
)
The
fi
rst
or
d
e
r a
p
pr
oxi
m
a
ti
on
w
o
ul
d
b
e
de
ri
ve
d
as
a
p
r
ocess
gai
n
, d
e
ad tim
e
. T
h
e
app
r
oxi
m
a
t
i
on fr
om
an e
xpe
ri
m
e
nt
al
t
e
st
of t
h
e
dy
nam
i
c sy
st
em
and
c
o
m
p
ared
wi
t
h
t
h
e i
d
e
n
t
i
f
i
e
d
fo
urt
h
o
r
d
e
r
system
i
d
en
tification
i
n
un
it step
resp
on
se. Tim
e
delay adjusted t
o
be
m
o
re tha
n
zero to achi
e
ve a
satisfacto
r
y m
a
tch
and
sim
p
lif
y th
e con
t
ro
l
param
e
ters an
alysis.
0.1
5
51
0
71
0
5
1
0
1055
2.5
10
3
.
4
1
0
0
.
5
(2
)
3.
MO
DIFIE
D
P
R
EDI
C
TI
VE CO
NTROL
Th
e fo
llowing
lin
ear, d
i
screte ti
m
e
,
sin
g
l
e
inp
u
t/sing
l
e o
u
t
pu
t
AR
X
m
o
d
e
l represen
tatio
n
b
y
(3
):
1
(3
)
whe
r
e
A
and
B
are polynom
i
als
in
t
h
e b
ackward shi
f
t operat
or
as below:
1
⋯
(4
)
⋯
(5
)
whe
r
e
nb
na
,
d
is ti
m
e
d
e
lay,
na
is
nu
m
b
er
o
f
po
les,
nb
is nu
m
b
er
of
zer
o
s.
Th
e
p
r
ed
ictiv
e
co
n
t
ro
l al
g
o
rith
m
co
n
s
ists
o
f
ap
p
l
ying
a
contro
l seq
u
e
n
c
e th
at m
i
n
i
mizes a m
u
lt
istag
e
co
st fun
c
tio
n of th
e fo
rm
,
,
δ
j
|
λ
Δ
1
(6
)
whe
r
e
|
is an
op
ti
m
u
m
j
-
step
ah
ead
p
r
ed
ictio
n
o
f
t
h
e system o
u
t
pu
t o
n
data u
p
to
tim
e
t.
is
the f
u
tu
re
refe
rence
tra
j
ecto
r
y
,
i
s
t
h
e
mi
n
i
mu
m c
o
s
t
i
n
g
h
o
r
i
zo
n
1
,
t
h
e
ma
x
i
mu
m c
o
s
t
i
n
g
ho
ri
zo
n i
s
, and
kn
ow
n as
t
h
e co
nt
r
o
l
hor
izon
, r
e
sp
ectiv
ely.
δ
j
, and
λ
are
weigh
tin
g seq
u
en
ces.
Since the c
ont
rol signal that i
s
actually sent to th
e proce
ss
is the first element of vect
or u (recedi
n
g
st
rat
e
gy
),
i
t
i
s
gi
ve
n
by
:
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
IJEC
E V
o
l
.
6, No
. 2, A
p
ri
l
20
16
:
63
0 – 6
3
8
63
3
Δ
(7
)
whe
r
e
is th
e dyn
amic
m
a
trix
,
whe
r
e
is d
i
ag
on
al m
a
trix
,
is a vector
of the refe
rence t
r
ajectory,
and
i
s
a m
a
t
r
ix cal
cul
a
t
e
d
us
i
ng t
h
e
Di
o
pha
nt
i
n
e e
quat
i
on
.
In
ou
r p
r
o
p
o
se
d m
a
t
r
i
x
anot
h
e
r st
rat
e
gy
ef
f
ect
s di
ffe
rent
t
y
pe of m
odel
s
such as
fi
rst
,
seco
nd
, an
d
h
i
gh
o
r
d
e
r
syste
m
s, an
d g
i
v
e
s m
o
re q
u
a
lity respo
n
s
es co
m
p
ared
t
o
th
e
prev
iou
s
st
u
d
y
esp
ecially in
term
s o
f
ove
rs
ho
ot
. T
h
e
pr
o
pose
d
m
a
t
r
i
x
de
pen
d
s
o
n
t
h
e su
p
p
ressi
o
n
coe
ffi
ci
e
n
t
, th
e
d
i
ago
n
a
l
m
a
trix
, an
d oth
e
r
val
u
es c
r
eat
ed
fr
om
t
h
e open
-
l
o
o
p
res
p
on
se
of t
h
e o
r
i
g
i
n
al
sy
st
em
.The pr
op
ose
d
st
rat
e
g
y
i
s
im
pl
em
ent
e
d t
o
diffe
re
nt proc
esses to e
v
aluate the
performance of a
new st
rategy a
p
proach of t
h
e m
odified
predictive
m
e
t
hod.
Let
’
s
assum
e
i
s
t
h
e ope
n
l
o
o
p
st
ep resp
o
n
se
sam
p
l
i
ng dat
a
of
t
h
e sy
st
em
m
odel
as:
…
…
.
(8
)
whe
r
e
- step response
coe
ffi
cients.
Th
e m
o
d
i
fied
matrix
Φ
with
in M=6
is:
Φ
(9
)
Φ
1/2
1/2
0
0
1/2
0
0
1/2
0
.
0
1/2
0
0
1/2
(1
0)
4.
TUNI
NG
ST
RATEG
Y
Mo
d
e
l pred
ictiv
e con
t
ro
l fam
i
ly
m
o
stl
y
u
s
ed
for ind
u
s
t
r
ial p
r
o
cesses. Th
e
GPC p
e
rfo
r
m
a
n
ce obj
ectiv
e
is v
e
ry si
m
ilar
to
th
e DMC b
u
t
is min
i
m
i
z
e
d
v
i
a recu
rsi
o
n
on
th
e Dioph
an
tin
e iden
tity b
y
Clark
e
[1], [1
5
]
.
N
e
v
e
r
t
h
e
less,
G
P
C
r
e
du
ces t
o
th
e D
M
C algo
r
ith
m
p
o
l
ynomial th
at
m
o
d
i
f
i
ed
the pr
ed
icted
ou
tpu
t
tr
ajecto
r
y is
assu
m
e
d
to
b
e
un
ity; th
erefore, t
h
e
DMC tu
n
i
n
g
strateg
y
can
b
e
d
i
rect
ly ap
p
lied to
GPC con
t
ro
ller [16
]
.
An
ot
he
r
st
udy
base
d on
C
o
op
er’s
st
rat
e
gy
w
a
s
i
m
pl
em
ent
e
d fo
r Dual
A
d
a
p
t
i
o
n
Ge
ne
ral
i
zed
P
r
e
d
i
c
t
i
v
e C
ont
r
o
l
(DA-GPC
)
by [17]. The m
odi
fied
predict
i
ve m
e
thod
will have t
h
e s
a
m
e
pre
d
iction
horizon equation a
s
pr
o
pose
d
by
C
o
o
p
e
r
[1
8]
i
n
t
e
rm
s of
prediction horizon and will
be
c
o
m
p
ared
with C
o
ope
r’s
m
e
thod. T
h
e
tu
n
i
ng
pr
o
c
edur
e for
co
n
t
ro
l
par
a
m
e
ter
s
,
, a
n
d
as fo
llow:
Predi
c
t
i
on h
o
r
i
zon
:
no
rm
all
y
chose
n
base
d
on a
ri
se t
i
m
e
of
95
% o
f
t
h
e
pr
ocess st
ea
dy
st
at
e. A m
o
re
ex
p
licit tun
i
ng
co
rrelatio
n was propo
sed b
y
[1
8
]
for t
h
e selectio
n
o
f
P
as follo
ws:
5
√
10
1
(1
1)
C
ont
r
o
l
h
o
ri
z
o
n
:
I
n
t
h
e
G
P
C
fo
rm
ul
at
i
on, s
e
l
ect
i
ng a
val
u
e o
f
t
h
e
co
nt
r
o
l
ho
ri
zo
n M
>
1
pr
od
uces
a
m
o
re rob
u
st
cont
rol
l
e
r at
t
h
e cost
of i
n
crea
sed com
put
at
i
on l
o
a
d
. Se
ver
a
l
researche
r
s
have
pr
o
pose
d
t
o
set th
e
v
a
lu
e
o
f
(
M
=1
) [1
],
[1
5
]
,[
19
].
Ro
ssiter
[2
0
]
stated
(
M
3),
A
b
u-
Ay
y
a
d [2
1]
, [
22]
st
at
ed
5
10
.
In
th
is p
a
p
e
r co
m
p
ariso
n
will
b
e
p
r
esen
ted
with
p
r
ev
iou
s
stud
y
b
y
Coop
er pro
v
i
d
e
d an
alytical
fo
rm
ula for
M
as f
o
llow
s
.
√
10
1
(1
2)
In th
e m
o
d
i
fied
p
r
ed
ictiv
e meth
od
th
e con
t
ro
l horizon
ch
osen
as (
5
1
0
).
Th
e mo
ve
supp
ression co
efficien
t
: seve
ral researc
h
e
r
s
enha
nced
th
e pred
ictiv
e con
t
ro
l b
a
sed
on
pr
o
pose
d
t
u
ni
ng m
e
t
hods
f
o
r t
une
. Fi
rst
t
uni
n
g
f
o
rm
ula sug
g
est
e
d b
y
a t
e
am
from
Depa
rt
m
e
nt
of
C
h
em
i
c
al
En
gi
neeri
n
g
,
U
n
i
v
e
r
si
t
y
o
f
Al
be
rt
a
E
d
m
ont
on
, C
a
nada
[1
7]
as
.
1
.
In
add
itio
n, t
h
e
val
u
e
o
f
can
be
cal
c
u
l
a
t
e
d b
a
sed o
n
t
une
d
du
ri
n
g
o
p
erat
i
o
n
t
o
i
m
prove
t
h
e ove
ral
l
pe
r
f
o
r
m
a
nce
an
d
1
is t
h
e
no
m
i
n
a
t
o
r
p
o
l
yno
m
i
al.
Ban
e
r
j
ee and
Sh
ah
[2
3
]
suggested
sim
p
le rob
u
s
t
tun
i
ng gu
i
d
elin
es fo
r
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
Mod
ified
Pred
ictive Co
n
t
ro
l fo
r
a
Class
o
f
Electro
-Hyd
rau
lic Actu
a
t
o
r
(Ab
d
u
l
ra
hma
n
A.A. Emh
e
med
)
63
4
a proce
ss u
nde
r Gene
ral
i
zed
Pre
d
i
c
t
i
v
e C
o
n
t
rol
usi
n
g
fre
quency res
p
ons
e technique where the sm
a
l
l gain
criterio
n
pro
v
i
d
e
s
u
s
efu
l
i
n
fo
rm
atio
n
about th
e st
ab
ility and
p
e
rform
a
n
ce
o
f
GPC an
d th
e system
stability unde
r GPC
only if t
h
e val
u
e
of
approxim
a
tely
exceeds 1. Anot
he
r resea
r
c
h
ers propose
d
to
t
une
t
h
e
m
ove
s
u
p
p
re
ssi
o
n
wei
g
ht
base
d on
t
h
e
fol
l
o
wi
ng
rel
a
t
i
ons
hi
p:
. Where
,
are
con
s
t
a
nt
s
[2
4]
.
A
ne
w t
e
c
hni
que
ba
sed
o
n
r
e
gres
si
o
n
m
o
d
e
l
t
echni
ques
f
o
r
o
p
t
i
m
al
su
gge
st
ed by
[2
5
]
,
Analytical expression
for
bas
e
d
on
C
o
o
p
er
’s
de
ri
vat
i
o
n i
s
p
r
esent
e
d as
f
o
l
l
o
ws:
.
(1
3)
0
.
08
1
(1
4)
1
(1
5)
whe
r
e
i
s
p
r
oce
ss gai
n
,
is ti
m
e
d
e
lay,
is samp
lin
g ti
m
e
.
5.
C
O
N
T
ROL DESIGN AND
D
I
SC
USSION
Seve
ral
i
ssues
i
n
p
r
edi
c
t
i
v
e
c
ont
rol
t
uni
ng
b
ecom
e
appare
n
t
and
fi
rst
l
y
ca
n
be e
xpl
ai
ne
d
w
h
en t
h
e
strateg
y
is
d
e
si
g
n
e
d and
tested
on
sim
u
latio
n
b
e
fore tested on
real tim
e a
p
p
lication
.
These issu
es
related
t
o
t
h
e set
-
p
o
i
n
t
t
r
acki
n
g,
ove
rs
h
oot
, a
n
d di
st
u
r
bance
s
. P
r
evi
o
us i
n
vest
i
g
at
i
o
ns ha
ve est
a
bl
i
s
he
d t
h
at
a si
g
n
i
f
i
cant
ad
v
a
n
t
ag
e
o
f
MPC o
v
e
r
PID co
n
t
ro
ller
s
is th
at PID
con
t
r
o
l
m
u
st b
e
tu
n
e
d
to
r
e
j
ect d
i
stu
r
b
a
n
ces to
h
a
v
e
good
cont
rol, which
is not
neces
sary in
m
a
ny instances
when usi
n
g pre
d
ictive
c
ont
rol st
rategie
s
[16].
The
objective
of si
m
u
l
a
t
i
on desi
g
n
i
s
t
o
m
odi
fi
ed t
h
e p
r
e
d
i
c
t
i
v
e co
nt
rol
and t
e
st
i
ng t
h
e cont
r
o
l
l
e
r be
fo
re ve
ri
fy
i
t
in real
-
t
i
m
e
experi
m
e
nt
. Thi
s
i
s
im
port
a
nt
whe
n
de
al
i
ng wi
t
h
c
o
m
p
li
cat
ed i
n
st
rum
e
nt
whi
c
h i
s
not
easy
t
o
u
s
e an
d
in
stall. Th
e si
m
u
la
tio
n
an
aly
s
is and
design
h
e
lp
s t
o
redu
ce th
e tim
e tak
e
n
th
an
exp
e
ri
men
t
ally in
d
e
sig
n
i
n
g
,
t
uni
n
g
a
n
d
up
g
r
adi
n
g
t
h
e c
o
nt
rol
l
e
r.
B
a
sed o
n
t
h
e p
o
si
t
i
on m
odel
st
ep res
p
onse
t
h
e pl
a
n
t
can
b
e
app
r
oxi
m
a
t
e
d by
a
fi
rst
-
or
d
e
r pl
us
dea
d
ti
m
e
(FOPTD) in
teg
r
ating
as
th
e pro
c
ess g
a
i
n
1
.
5
8
and t
h
e time delay estim
a
ted as
d
=0.
05 t
o
get
t
h
e
v
a
lu
e fo
r m
o
ve su
pp
ression
.
Th
e pred
ictiv
e h
o
r
izon
was
P
=2
0, t
h
e co
nt
rol
h
o
ri
z
on
wa
s
M
=5 fo
r Co
ope
r’
s
m
e
thod. T
h
e m
ove
s
u
ppre
ssion
for the
Cooper m
e
thod is
=0
.1
5, an
d fo
r the op
ti
m
a
l tu
n
i
ng
is
=0.03,
The t
uni
ng
va
lue for t
h
e m
ove
s
u
ppressi
on c
o
efficient
is esti
m
a
ted
em
p
i
rically a
n
d the
param
e
t
e
r x e
s
t
i
m
a
t
e
d as x=
20
. B
e
st
pe
rf
o
r
m
a
nce achi
e
v
e
d
wi
t
h
0
.
0
3
whe
r
e
faster
reac
hing to t
h
e
setp
o
i
n
t
th
at is clear fro
m
th
e yello
w lin
es to si
m
p
lify
t
h
e est
i
m
a
ti
on o
f
t
h
e t
r
acki
n
g.
The
s
e t
uni
ng
val
u
e
s
wi
t
h
th
eir p
e
rform
a
n
ces b
a
sed
on
rise ti
m
e
an
d
settlin
g
ti
me
are p
r
esen
ted
in
th
e figu
res (2-4) an
d
th
e co
m
p
arison
anal
y
s
i
s
o
f
t
h
e
t
uni
n
g
pe
rf
orm
a
nce descri
pes i
n
Fi
g
u
r
e 5.
Fi
gu
re
2.
Di
f
f
e
r
ent
val
u
es
o
f
t
h
e
param
e
t
e
r t
uni
ng
base
d
o
n
0
.
3
7
0
0.
2
0.
4
0.
6
0.
8
1
1.
2
1.
4
1.
6
1.
8
2
0
0.
2
0.
4
0.
6
0.
8
1
1.
2
ti
m
e
(
s
e
c
)
y
=1
=2
=4
=
0
.
37
x
=
2
0
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
IJEC
E V
o
l
.
6, No
. 2, A
p
ri
l
20
16
:
63
0 – 6
3
8
63
5
Fi
gu
re
3.
Di
f
f
e
r
ent
val
u
es
o
f
t
h
e
param
e
t
e
r t
uni
ng
base
d
o
n
0
.
1
5
Fi
gu
re
4.
Di
f
f
e
r
ent
val
u
es
o
f
t
h
e
param
e
t
e
r t
uni
ng
base
d
o
n
0
.
0
3
Fi
gu
re
5.
A
n
al
y
s
i
s
of
di
f
f
ere
n
t
val
u
es
o
f
t
h
e
param
e
t
e
r t
uni
ng
base
d
on
di
ffe
rent
The a
n
al
y
s
i
s
t
e
st
t
o
eval
uat
e
t
h
e m
odi
fi
ed
p
r
edi
c
t
i
v
e m
e
t
hod
prese
n
t
e
d
wi
t
h
di
ffe
rent
val
u
es
of
with
v
a
ries v
a
lu
es o
f
th
e
param
e
ter
tu
n
i
ng
as show
n in Figu
r
e
5
.
Best m
o
d
i
f
i
ed
p
r
ed
ictiv
e
r
e
spon
se
ach
iev
e
d
with
4
and
0
.
0
3
beca
use
o
f
i
t
s
fast
t
o
reac
h t
h
e
set
poi
nt
.
Th
e po
sition
m
o
d
e
l resu
lts with
0
.15,
0
.
0
3
and
=4 p
r
esent
e
d i
n
Fi
gu
re 6 a
n
d t
h
e
perform
a
nceanalysis in Table 1.
All
the m
o
dified
pre
d
ictive responses a
r
e dem
onstrati
ng t
h
e shortest settling
t
i
m
e
, ri
se t
i
m
e
an
d n
o
o
v
ers
h
oot
.
O
n
t
h
e
o
t
h
er ha
n
d
C
oop
er’s m
e
th
o
d
resp
on
se is
un
attractiv
e b
ecau
s
e o
f
its
ove
rs
ho
ot
a
n
d
sl
ow t
r
acki
n
g
c
o
m
p
are t
o
t
h
e
m
odi
fi
ed p
r
e
d
i
c
t
i
v
e m
e
t
hod.
0
0.
2
0.
4
0.
6
0.
8
1
1.
2
1.
4
1.
6
1.
8
2
0
0.
2
0.
4
0.
6
0.
8
1
Ti
m
e
(
s
e
c
)
y
=1
=2
=4
=
0
.
03
x
=
2
0
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
Mod
ified
Pred
ictive Co
n
t
ro
l fo
r
a
Class
o
f
Electro
-Hyd
rau
lic Actu
a
t
o
r
(Ab
d
u
l
ra
hma
n
A.A. Emh
e
med
)
63
6
Fi
gu
re
6.
C
l
ose
l
o
o
p
c
o
nt
r
o
l
re
spo
n
ses
ba
sed
on
di
ffe
rent
c
o
nt
r
o
l
schem
e
s wi
t
h
vari
es
of
the E
H
A system
Tabl
e
1. C
o
m
p
ari
s
o
n
R
e
s
u
l
t
s
bet
w
ee
n t
h
e
M
odi
fi
ed
Method
an
d Coop
er’
s
Meth
od
f
o
r
The EHA
System.
Si
m
u
lation
per
f
orm
a
nce
=0.
15
=0.
03
Cooper
’
s
m
e
thod
m
odified
m
e
thod
Cooper
’
s
m
e
thod
m
odified
m
e
thod
Over
shoot (
%
)
4.
4
0
4.
5
0
Rise tim
e
(
s
ec)
0.
30
0.
29
0.
22
0.
19
Settling ti
m
e
(sec)
0.59
0.43
0.40
0.29
6.
CO
NCL
USI
O
N
Th
e m
o
d
i
fied
p
r
ed
ictiv
e con
t
ro
l is cho
s
en
as represen
tativ
e
m
o
d
e
l p
r
ed
ictiv
e co
n
t
ro
l tech
n
i
q
u
e
and
is
ap
p
lied to
simu
latio
n
st
ud
ies o
f
po
sition
co
n
t
ro
l m
o
d
e
l o
f
EHA. Th
e meth
o
d
o
l
og
y o
f
an
alysis
con
s
id
ers
a
di
m
e
nsi
onl
ess param
e
t
e
r i
n
order t
o
ge
t
ge
ne
ral
resul
t
s
w
h
i
c
h co
ul
d be exe
r
cised to any s
p
ecific real case. The
cont
rol perform
a
nce of the
m
odified
controller is com
p
ared with C
o
ope
r’s
m
e
thod
usi
ng
fast reacting proce
ss.
The close
d
-l
oop res
p
onses a
r
e com
p
ared and the di
ffe
renc
es and sim
ilarities are explained on the
basis
of the
stru
cture of the co
n
t
ro
l sch
e
mes. Th
e si
m
u
latio
n
tests fo
r th
e co
n
t
ro
l p
a
ra
m
e
ters in
so
me tu
n
i
n
g
cases fo
r th
e
m
odel
EHA p
o
si
t
i
on c
ont
rol
are gi
ve
n t
o
veri
fy
t
h
e e
ffe
ct
i
v
eness
of t
h
e
m
odi
fi
ed
pr
edi
c
t
i
v
e co
nt
r
o
l
.
The
m
odi
fi
ed m
e
t
h
od
d
o
es
n
o
t
ha
ve
ove
rs
ho
ot
,
f
a
st
t
o
reac
h t
h
e
set
p
o
i
n
t
,
an
d
sm
oot
her t
h
a
n
C
o
o
p
er
’s m
e
t
hod
.
REFERE
NC
ES
[1]
Clarke DW
, M
o
htadi C, Tuffs
P
S
.
Generali
zed p
r
edic
tive con
t
rol
—
P
a
rt I. The ba
s
i
c algorithm
.
A
u
tomatica
. 1987;
23(2): 137-148
.
[2]
Bigdeli N, Haer
i M.
Position control of an ultrasonic motor
u
s
ing generalized
predictive control
. In
Contro
l,
Automation, Ro
botics
and Visio
n
Conferen
ce (I
CARCV). 2004;
1957-1962.
[3]
Egiguren PA, Caramazana OB, Etxeb
e
rria JAC.
Linear generalized predictive p
o
sition control o
f
induction moto
r
dr
ives
. In
IECO
N 2011-37th An
nual Conf
eren
ce on IEEE
Industr
ial Electronics
S
o
ciety
.
2011; 19
22-1927.
[4]
Yu B,
Shi Y.
GPC-based remote control for hy
draulic position
control systems in a networked
environment
. I
n
American Contr
o
l Confer
ence. 2
009; 1: 5558-55
63.
[5]
Sepehri N, Wu G. Exp
e
rimental
evalu
a
tion
of
gener
a
li
zed
pr
edic
tive
contro
l
appli
e
d
to a
h
y
drau
li
c a
c
tu
ato
r
.
Robotica
. 1998
;
16(4): 463-474
.
[6]
Camacho EF
, B
o
rdons C. Model Predic
tive Control. Spring
er
Lo
ndon. 2004
.
[7]
Maciejowski JM. Predictiv
e Con
t
rol: With
Constr
aint
s. Prentice Hall-
Techno
lo
g
y
&
Eng
i
neer
ing. 2002.
[8]
Danielle D, Cooper DJ. Tuning guidelin
es of a d
y
n
a
mic ma
trix controller for
integr
ating (no
n
-self-regulating
)
proc
e
sse
s.
Indus
trial &
engineeri
ng chemistry res
e
arch
.2003; 42(
8): 1739-1752.
[9]
Pirouzmand F. Robust model predictive con
t
rol
based
on MRAS for satellite a
ttitude control s
y
s
t
em.
Internationa
l
Journal of Electrical and
Co
mputer
Eng
i
neer
ing
(
I
JECE)
. 2014;
4(1):81-92.
[10]
Liy
un W, Jian
mei X, Xihua
i W. Ship dy
n
a
mic positionin
g
sy
stem
s based on fuzzy
pr
edictive con
t
rol.
TELKOMNIKA Indonesian Journ
a
l
of Electrical
Engineering
. 20
13; 11(11): 6769
-6779.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
IJEC
E V
o
l
.
6, No
. 2, A
p
ri
l
20
16
:
63
0 – 6
3
8
63
7
[11]
Grancharova A.
Explicit MPC of Constraine
d Nonlinear S
y
stems with Quantized Inputs.
L
ectur
e N
o
tes
in Contr
o
l
and Information
Scien
ces
. 2009;
384: 371-380.
[12]
Ghazal
i R, S
a
m
Y, Rahm
at M
,
H
a
s
Z.
Adaptive D
i
screte S
liding
Mode Cont
rol fo
r a Non-minimum Phase Electro
-
Hydraulic Actua
t
or
System
. Inter
n
ation
a
l Conf
erence on
Robotic,
Visi
on, Signal P
r
ocessing & Power Application
s
.
Springer. 2014
;
1:3-14.
[13]
Seborg DE, Mellichamp DA,
Edg
a
r TF
, Do
y
l
e III
FJ. Process d
y
n
a
mics a
nd
control. John Wiley
&
Sons.2010
[14]
Emhemed AAA
,
Mamat R,
Hanafi D.
Comparison of boiler plant model and
real plant responses based on
proportional Integral controller
.
In
Sustainabl
e Utili
zat
ion and
Devel
opm
ent i
n
Engineering and
Technolog
y
(STUDENT), 20
12 IEEE Conf
er
ence on. 2012; 1
:
117-122.
[15]
Clarke DW
, Mo
thadi C
,
Tu PS.
General
i
z
e
d pre
d
ictiv
e con
t
rol P
a
rt II
. Ext
e
nsion
s
and interpr
e
t
a
ti
ons.
Automati
ca
.
1987; 23(2): 149
-160.
[16]
Shridhar R, Co
oper DJ. A
tu
ning strateg
y
f
o
r
unconstrained SISO model predictiv
e
con
t
rol.
Industria
l
&
Engineering Chemistry Research
.
1997; 36(3)
: 72
9-746.
[17]
Ho YK, Yeoh H
K
, Mjal
li FS.
Genera
liz
ed Pred
i
c
tiv
e Contro
l Al
gorithm
with R
e
al-T
im
e Sim
u
lta
neous Modeling
and Tun
i
ng.
Ind
u
strial
&
Engineering Chemistry
Research
. 2014;
53(22): 9411-94
26.
[18]
Dougherty
D
,
C
ooper DJ. Tunin
g
guidelin
es of a d
y
n
a
mic ma
trix controller for
integr
ating (non
-self-regulating
)
proc
e
sse
s.
Indus
trial &
engineeri
ng chemistry res
e
arch
. 2003
; 42(
8): 1739-1752.
[19]
Mc
Intosh AR,
Sha
h
SL
,
Fishe
r
DG.
Selection o
f
tuning parameters for adap
tive generalized predictive con
t
rol
.
In American
Co
ntrol C
onf
eren
ce, 1989; 1:1846-1
851.
[20]
Rossiter JA. Mo
del-based
pred
ictive contro
l:
a p
r
actical
approach. CRC press. 201
3.
[21]
Abu-Ay
y
a
d M,
Dubay
R
,
Kemb
er GC. SISO extended pr
edictive control—fo
rmulation
and th
e bas
i
c algorithm.
IS
A
Trans
. 2006; 45(
1): 9-20.
[22]
Abu-Ay
y
a
d M,
Dubay
R.
Rea
l
-time
c
o
mpa
r
i
s
on of a
numbe
r of pre
d
i
c
t
i
v
e
c
ont
roll
e
r
s.
IS
A Trans
. 2007;
46(3): 411
-
418.
[23]
Ba
ne
rje
e
P, Sha
h
SL
.
Tuning
guidelin
es for r
obust ge
neraliz
ed predictive
control
. In
Decis
i
on and Con
t
ro
l,
Proceedings of
t
h
e 31st I
E
E
E
Co
nferenc
e
on
. 199
2; 1:
3233-3234.
[24]
Rani KY, Unbeh
a
uen H
.
Stud
y
o
f
pr
edictiv
e
contr
o
ller
tuning
methods.
Automat
i
c
a
. 1997; 33(12):
2243-2248.
[25]
Emhemed AAA, Mamat R, Faudzi AAM.
A New Pr
edicti
ve Co
ntr
o
l T
echnique
for
For
ce Contr
o
l of Pneumati
c
Actuator
Plan
t.
The 10th
Asian C
ontrol Conf
eren
ce (ASCC 2015).
2015; 1:1-6
.
BIOGRAP
HI
ES OF
AUTH
ORS
Abdulrahman A.A. Emhemed
is a
lecturer
at C
o
lleg
e
of
Electr
onic
Techno
log
y
-Ban
i Walid,
Lib
y
a. Currently he is a PhD
Candidate at Contr
o
l and Mechat
ro
nic Engin
eering
Department,
Universiti
Te
ch
nologi Mal
a
y
s
ia
. He h
a
s presen
ted m
a
n
y
pap
e
r
s
at conf
eren
ces
and journ
a
ls.
Mr.Em
h
em
ed is
a m
e
m
b
er for
man
y
in
tern
ation
a
l
institutes and e
ngineer
ing associations such as
Institute
of El
ec
tric
al and Ele
c
tr
onics
Eng
i
neer
i
ng
(IEE
E
),
Scie
nce and Engine
ering
Inst
itut
e
(SCIEI), and
th
e Intern
ation
a
l
Association of
ENGineers (IA
ENG).
His area of interesting
is
artif
icial intellig
ent, microcon
tro
llers
a
pplications, robotics, and
pr
ocess contro
l.
Ros
b
i M
a
m
a
t is
an as
s
o
ciat
e prof
es
s
o
r at Depart
ment of Control and Mechatronic
Engineering at
the Faculty
o
f
Electrical
Engin
e
ering, Univer
siti Teknolog
i Malay
s
ia. He
obtain
e
d his PhD at
1996 in con
t
rol
engineering fro
m University
of
Sh
effield
,
UK.
He has au
thored
and
co-author
ed
m
o
re m
a
n
y
p
a
p
e
rs in in
tern
atio
nal
and lo
cal
jo
urnals and
conf
erenc
e
s. His r
e
search
int
e
rests
includ
e Em
bedd
ed S
y
st
em
s, Arti
fici
al In
tellig
en
t
Control, Roboti
c
s and Mechatron
i
c S
y
s
t
em
s.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
Mod
ified
Pred
ictive Co
n
t
ro
l fo
r
a
Class
o
f
Electro
-Hyd
rau
lic Actu
a
t
o
r
(Ab
d
u
l
ra
hma
n
A.A. Emh
e
med
)
63
8
Ahmad `Athif Mohd Faudzi was born in 1982.
He rece
ived the B.
Eng. and
th
e M.
Eng.
d
e
grees
from Universiti Teknolog
i Malay
s
ia, Malay
s
ia a
nd th
e Dr.
Eng. in S
y
stem I
n
tegration from
Okay
ama University
, Japan
in 2
004, 2006, and
2010
respectively
.
H
e
is now
attached with the
Centre for Artif
ici
a
l Intellig
ence and Robotics
(CAIRO), Universiti T
e
knologi
Malay
s
i
a
as a
Senior Lecturer
. Dr. Athif is th
e leader for
of
the Actuator and
Automation Research Group
(A2RG). He is
m
a
inl
y
engag
e
d
in th
e res
e
arch
fields
of pn
eu
m
a
tic a
c
tu
ators
,
s
o
ft ac
tuators
,
robotics autom
a
t
i
on
and
th
eir app
lic
ations.
Mohd Ridzuan
Johar
y
is a m
a
s
t
er student
at
U
n
iversiti Teknol
ogi
Malay
s
ia.
Mr. Johar
y
is a
member of the
Actuator and Automation Resear
ch Group (A2R
G) and supervised b
y
Dr. Ahmad
`Athif Mohd Faudzi. He
is main resear
ch
in th
e in
ellegent con
t
rol integr
ated
with Electro-
H
y
draulic Actu
ators and
their
ap
plications.
Khairuddin Osman
is a Lectur
er
at Dep
a
rtment
of Industr
ial Electronics, Facu
lty of Electrical
Engine
ering,
Universiti
Tekn
ikal
Mala
ysia Me
la
ka. He r
ece
ived
his MS.cEng an
d PhD degrees
from
Universiti Teknologi Mal
a
y
s
i
a
, Mal
a
y
s
ia. He has presented m
a
n
y
pap
e
r
s
at local and
intern
ation
a
l co
nferenc
e
s and j
ournals. His are
a
s of interst in
c
l
ude S
y
st
em
Identifi
c
a
tion,
as
Instrum
e
ntation
and
Autom
a
tion
S
y
stem
s.
Evaluation Warning : The document was created with Spire.PDF for Python.