TELKOM
NIKA Indonesia
n
Journal of
Electrical En
gineering
Vol. 12, No. 9, September
2014, pp. 65
6
0
~ 656
5
DOI: 10.115
9
1
/telkomni
ka.
v
12i9.463
5
6560
Re
cei
v
ed O
c
t
ober 5, 20
13;
Revi
se
d Apr
27, 2014; Accepted June 4,
2014
Fuzzy PID Speed Control of Two Phase Ultrasonic
Motor
Ma Qiujie, S
h
i Jingz
huo
*
Hen
an Un
ivers
i
t
y
of Scie
nce a
nd T
e
chnol
og
y
No.26
3
Kai
y
u
a
n
Dad
ao, Lu
o
y
ang 4
7
1
023, C
h
in
a. F
a
x: +
86-
379-
642
31
91
0
*Corres
p
o
ndi
n
g
author, e-ma
i
l
: sjzne
w
@
1
6
3
.com
A
b
st
r
a
ct
Becaus
e of ultrason
ic mot
o
r
’
s time-var
iab
l
e
nonl
i
n
e
a
rity, an on
lin
e ad
a
p
tive correcti
o
n of PID
control
para
m
eters bas
ed
o
n
fu
zz
y
lo
gic
i
s
prese
n
ted. T
he sp
ecia
l n
a
ture of u
l
traso
n
i
c motor
’
s s
p
e
e
d
control is fully t
a
ken i
n
to acco
unt in des
ign
i
n
g
fu
zz
y
rul
e
s. T
he amplit
ude
of drivin
g volta
ge is use
d
as the
control p
a
ra
me
ter to reali
z
e t
he co
ntrol
met
hod. T
he ex
pe
riments in
dicat
e
that t
he cont
rol char
acterist
ics
are much b
e
tte
r than that of PID controll
er.
Ke
y
w
ords
: ultr
ason
ic motor, spee
d c
ontro
l,
fu
zz
y
ada
ptive, PID
Copy
right
©
2014 In
stitu
t
e o
f
Ad
van
ced
En
g
i
n
eerin
g and
Scien
ce. All
rig
h
t
s reser
ve
d
.
1. Introduc
tion
Becau
s
e
of ultraso
n
ic m
o
tor’s time
-vari
abl
e, nonli
n
e
a
rity, the stro
ng co
uplin
g
betwe
en
control vari
ab
les a
nd othe
r
rea
s
on
s, the
motion’s
co
ntrol pe
rforman
c
e i
s
not e
a
sy
to improve.
As
a kin
d
of in
telligent cont
rol
strategy,
a fuzz
y co
ntrolle
r of p
r
operly d
e
si
g
ned h
a
s
strong
robu
stne
ss, and can wea
k
en the im
p
a
ct of the o
b
je
ct chan
ge
to the control effect, so
it is
suiTa
b
le fo
r
nonlin
ear, ti
me-varyin
g
o
b
ject
s, such
as
ultra
s
o
n
i
c
moto
r. But
the tra
deoff
o
f
para
m
eters
b
e
twee
n robu
stne
ss an
d
control
preci
s
i
on al
so
exist
in fu
zzy
con
t
rol de
sig
n
, it is
difficult to indepen
dent ap
plied to the servo co
ntro
l
occa
sion
s of high prec
i
s
io
n, thus we often
combi
ne fuzzy control with
the traditional
control m
e
thod.
The fuzzy co
ntrol is t
r
ied t
o
apply to ultraso
n
ic m
o
tor
at home a
nd
abro
ad. Th
e pape
r [1,
2], resp
ectiv
e
ly combi
ne
the fuzzy co
ntrol with
ne
ural n
e
two
r
k control an
d the slidin
g-m
ode
control, an
d i
s
u
s
e
for ultra
s
oni
c m
o
tor control.
In
the
pape
r [3], de
sign
ed
of the
ultrasoni
c m
o
tor
spe
ed
control
l
er ba
se
d on
fuzzy lo
gic
was d
e
si
gne
d,
and it is
co
nverted to PI co
ntrol to imp
r
o
v
e
the problem
of poo
r
stead
y-state
p
e
rfo
r
mance of fu
zzy control
wh
en the
erro
r i
s
small. In th
e
pape
r [4], the position
co
ntrol metho
d
th
at using
a fuzzy cont
rolle
r to adju
s
t the p
hase differe
n
c
e
of two-pha
se
voltage of th
e motor i
s
p
r
ese
n
ted,
for t
he p
r
oble
m
o
f
the ultra
s
on
ic moto
r cont
rol
dead
-ba
nd. T
he pa
per [5] use fu
zzy co
ntrolle
r to adj
ust the frequ
ency to
reali
z
e sp
eed
co
ntrol,
and o
n
line
correct fu
zzy
control rule
s based o
n
g
enetic algo
rithms, so con
t
rol
algo
rithm
is
compl
e
x
.
PID co
ntrolle
r is d
e
si
gne
d t
o
control
the
rota
ting
sp
ee
d of ultrasoni
c moto
r by
a
d
justin
g
voltage ampl
itude. Beca
u
s
e of the ti
me-vari
ng a
n
d
nonli
nea
rity cha
r
acte
ri
stics of ult
r
a
s
onic
motor, onlin
e
adaptive co
rre
ction of P
I
D cont
rol p
a
ram
e
ters b
a
se
d on fuzzy logic i
s
a
l
so
desi
gne
d. The spe
c
ial nat
ure of ultra
s
o
n
ic moto
r’s speed
control are fully take
n into accou
n
t,
appli
c
ative
fu
zzy rule
s
i
s
d
e
sig
ned and amend
ed by
the actu
al te
st. The
experiments i
ndi
ca
te
that
the spe
ed control
p
e
rform
a
n
c
e h
a
s obviou
s
i
m
provem
ent comp
ared wit
h
traditio
nal PID
control alon
e.
2. Design of
Fuzz
y
PID Speed Contr
o
ller
2.1. Speed Control Ch
ara
c
teris
t
ics Ba
sed on Volta
g
e Amplitud
e Adjus
t
men
t
In this p
ape
r, the travelin
g wave
ultra
s
oni
c m
o
t
o
r
spe
ed f
u
zzy
P
I
D co
nt
rol
sy
st
e
m
stru
cture is
shown in Figu
re 1.
Speed cl
ose
d
loop
co
ntrol a
c
hieve
d
by the PID controlle
r, fuzzy
regul
ator i
s
use
d
for
onli
ne corre
c
ting
paramet
e
r
s
of PID co
ntroller to i
m
prove the
cont
rol
perfo
rman
ce.
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Fuzzy PID Speed Control o
f
Two Phase Ultra
s
oni
c Mo
tor (Ma Qi
ujie
)
6561
For a
c
hievin
g effective control of ultraso
n
ic m
o
tor spee
d, the controlled va
riable
ca
n
cho
o
se plus drive voltage amplitude,
frequen
cy or
pha
se dif
f
eren
ce of motor; The
r
e
i
nto,
adju
s
ting the
amplitude of
the voltage can a
c
hi
eve
a good line
a
rity, high precision of sp
ee
d
control. In thi
s
pa
per,
we
u
s
e the
driving
voltage
ampl
itude a
s
the
controlle
d vari
able, that is t
he
controlled va
riable of the
o
u
tput of the PID co
ntrolle
r i
s
the voltag
e
amplitude
of set value
Ur
e
f
,
and thro
ugh
clo
s
ed
-loo
p control of the
two -
phase voltage amplitude of the inne
r loo
p
to
achi
eve the adju
s
tment o
f
the driving voltage am
pli
t
ude. The in
put of fuzzy regul
ator is t
h
e
ultrasoni
c mo
tor spe
ed e
r
ror
e
and e
r
ror variation
ec
,
and the outpu
t is the incre
m
ental
∆
K
p
,
∆
K
i
,
∆
K
d
of PID control pa
ramet
e
rs
K
p
, K
i
, K
d
.
Usi
ng fu
zzy rules to
adju
s
t
the PID
cont
rol
paramete
r
s i
s
a
comm
only control
method.
Ho
wever, t
r
a
v
eling wave
ultrasoni
c mo
tor spee
d
co
ntrol that
ba
se on volta
ge
amplitude
adj
ust
has some
ru
nning
characteristic
s
different from
the
co
nventional
PID control, thus ma
ke t
h
e
desi
gn
of fu
zzy regul
ator e
s
pe
cially th
e
desi
gn
of
the
fuzzy
rul
e
s a
r
e different fro
m
the
co
mmon
method obvio
usly.
The process
of the spee
d step re
sp
on
se is startin
g
from the moto
r static (sp
e
e
d
is 0),
for exam
ple, the out
put con
t
rolled va
riabl
e of conv
enti
onal PID cont
rol i
s
g
ene
rall
y starting
fro
m
0, and g
r
ad
u
a
lly increa
se
s un
der th
e
controlle
r a
c
tion, and eve
n
tually
rea
c
h
e
s steady state
value, and
th
e ste
ady-state value
great
er tha
n
th
e
ini
t
ial value 0. B
u
t be
cau
s
e
of sp
eci
a
l runni
ng
mech
ani
sm
of traveling
wave ult
r
a
s
o
n
ic m
o
tor,
so it must
be
exerted
hig
h
freq
uen
cy
high
voltage to
start. To e
n
sure the
USM n
o
rmal
sta
r
t, the initial valu
e of the
set
value
Ur
ef
of
the
motor d
r
ive v
o
ltage a
m
plitude u
s
u
a
lly large
r
, such
a
s
30
0V, the i
n
itial value i
s
usu
a
lly grea
ter
than the valu
e
Ur
ef
that th
e motor
stea
dy-state
ope
ration ne
ede
d
after sta
r
ting.
Thu
s
, the ou
tput
controlled va
riable
stea
dy-state valu
e
of the PI
D controlle
r is l
e
ss th
an initial
value, and t
h
e
variation tren
d is diffe
rent f
r
om the
co
nventional PI
D
control. On th
e other han
d, also i
n
o
r
de
r to
ensure th
e USM normal
start, the initial value
of motor drive fre
quen
cy mu
st be set a hi
gh
enou
gh valu
e to maintain the motor normal
star
ting. However, if runnin
g
in this hi
gher
freque
ncy, th
e sp
eed
ra
ng
e is limite
d
b
y
adjustin
g
th
e voltage a
m
plitude a
c
hi
e
v
ed. If the sp
eed
set value is hi
gher, we nee
d to redu
ce m
o
tor drive fre
quen
cy to an approp
riate value. In orde
r to
redu
ce
the resp
on
se
reg
u
lation time
of the spee
d, the sy
stem i
s
de
sig
ned
a
s
follo
ws, i
n
the
spe
ed a
s
cent
stage
after
motor
startin
g
the freq
uen
cy is re
du
ced
open
-loo
p to
the de
sire
d v
a
lue
as a fixed
ste
p
. Thu
s
, the
spe
ed
clo
s
ed
loop PI
D
co
ntrol ba
se
d o
n
the voltage
regul
ation al
ong
with the ope
n-loo
p
frequ
e
n
cy red
u
ctio
n
at the
same time, have to increa
se the spee
d co
ntrol
compl
e
x
i
t
y
at
t
he sam
e
t
i
me,
also m
a
ke
s t
h
i
s
PID control
proce
s
s is
different from t
he
conve
n
tional PID.
2.2. PID Con
t
rol Algorit
h
m
Figure 1
sho
w
s th
e PID
controlle
r u
s
es in
cr
eme
n
tal PID algo
ri
thm, the formula a
s
follows
.
pi
d
u
k
K
N
e
K
Ns
p
K
Ne
Ne
la
s
t
(1)
P
I
D c
ontr
o
ll
er
A
pha
se
volt
a
ge
a
m
pli
t
ude
c
ontr
o
ll
er
B
pha
se
volt
a
ge
a
m
pli
t
ude
c
ontr
o
ll
er
Tw
o-pha
se
H
-
br
idge
dri
v
e
r
ci
rc
ui
t
F
u
zz
y c
ontr
o
ll
e
r
US
M
s
Nr
e
f
n
+
-
+
-
Ur
e
f
Figure 1. Structure of Fuzzy PID Speed
Control Syst
em of Ultra
s
o
n
ic Moto
r
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 9, September 20
14: 65
60 – 656
5
6562
Her
e
,
Ns
p=Nr
ef-
n
, i
s
th
e
current moto
r
spe
ed
error,
Nref
i
s
a give
n spee
d,
n
i
s
the a
c
tual
sp
eed.
Ne
=N
sp
-N
sp
0
is th
e
current speed
error variatio
n.
Nsp0
i
s
th
e
sp
eed
error on the
previou
s
moment.
Nel
a
st
is the spe
ed error va
ria
t
ion of the pre
v
ious mom
e
n
t
.
2.3. Fuz
z
y
Online Adjustment of PID
Control Parameters
As mention
e
d
above, the input of fu
zzy regulato
r
is th
e spe
ed erro
r
e
and error v
a
riation
ec
, an
d th
e
ou
tp
u
t
is th
e in
c
r
e
m
en
ta
l
∆
K
p
,
∆
K
i
,
∆
K
d
of PID
cont
ro
l pa
ramete
rs
K
p
, K
i
, K
d
. All
of
the fuzzy lan
guag
e value
of five input a
nd outp
u
t
vari
able
s
is take
n as 5: NB
(n
egative big
)
,
NS
(neg
ative sm
all), ZO
(zero
)
, PS (po
s
itive sm
all)
a
nd
PB (po
s
itive big). Mem
b
e
r
ship fu
nctio
n
is
sele
cted t
r
ia
ngula
r
fun
c
tion, and
ea
ch of varia
b
le
s mem
b
e
r
shi
p
functio
n
d
e
fined in th
e
unit
domain
of di
scourse
di
stribution
are
shown in
Fig
u
re
2, Fig
u
re 3, respe
c
tively. The fu
zzy
inferen
c
e
met
hod of the fu
zzy regul
ator
is M
AX-MIN
method, an
d
the defu
zzifi
cation metho
d
is
Height method.
In orde
r to make the a
c
tual values
of input variable
s
mapp
e
d
to unit domain of
discou
rse [-1,1], we need to set and qu
antify the
quantizatio
n factors. In orde
r to make these
variable
s
on
different ope
rating point p
o
siti
on a
r
e n
o
t sen
s
itive unde
r the sp
eed ra
nge,
we
con
s
id
er the rang
e of variation and ch
ara
c
teri
stic
of
chang
e of input variable
values in a
c
tual
control, a
nd
set the
q
uant
itat
ive factor
of sp
eed
e
r
ror
e
a
s
th
e
speed
set val
ue
Nr
ef
, and
the
quantitative factor of
spee
d error vari
ation is 10.
Figure 2. Membershi
p
Fun
c
tion Di
strib
u
tion of
e
and
ec
Figure 3. Membershi
p
Fun
c
tion Di
strib
u
tion
of
∆
K
p
,
∆
K
i
,
∆
K
d
The de
sign
of fuzzy cont
rol rule
s sho
u
ld take full accou
n
t of th
e control particularity
whi
c
h is differ from convent
ional co
ntrol
aforem
entio
n
ed, in orde
r to get expecta
nt control effe
ct.
The ultra
s
oni
c
m
o
tor drive
voltage ampl
itude control
variable
s
hav
e different eff
e
cts in th
e
ri
se
and
stabili
zati
on sta
ge of t
he moto
r ste
p
re
spo
n
se, i
n
the ri
se
sta
ge, the big
g
e
r
of the
cont
rolled
variable the f
a
ster in
crea
sed, but
also e
a
sily ca
use o
v
ersh
oot.
Table 1. Fuzzy Rule
s of
∆
K
p
e
NB NS
ZO
PS
PB
NB
NB NB
NS NS ZO
NS
NB NS
NS ZO PS
ZO NS
NS
ZO
PS
PS
PS
NS
ZO
PS PS PB
PB
ZO
PS
PS PB PB
Table 2. Fuzzy Rule
s of
∆
K
i
、
∆
K
d
e
NB NS ZO
PS
PB
NB
PB PB PS
PS
ZO
NS
PB PS PS
ZO
NS
ZO PS
PS
ZO
NS
NS
PS
PS
ZO NS
NS NB
PB
ZO NS NS
NB NB
If the co
ntroll
ed vari
able
is sm
aller, th
e
re
spon
se
of t
he
system
wi
ll be
slo
w
e
r
,
and i
n
steady sta
g
e
,
the controll
ed variabl
e’s
oversi
ze
ca
n cau
s
e o
s
ci
llation. Whe
n
e
<0,
ec
<
0
, the
motor a
c
tual
spee
d ha
s excee
ded th
e spe
ed se
t value, the error an
d overshoot g
r
ad
ually
increa
se, the
control fun
c
ti
on of th
e volt
age
amp
litud
e shoul
d try t
o
redu
ce
the
overshoot.
When
e
>0,
ec
>0,the motor a
c
tu
al spe
ed ha
s not yet reache
d the sp
eed set value, and the e
rro
r
grad
ually in
crea
sed,
the
function
of t
he vo
ltag
e
amplitude
sh
ould try
to
redu
ce
reverse
overshoot. When
e
>0,
e
c<0, the motor
actual
spe
e
d
tends to the
spe
ed set value a
s
soon
a
s
possibl
e, that is the erro
r should be
eli
m
inated a
s
soo
n
as po
ssible
. When
e
<0,
ec
>
0
, the motor
ec
ec
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Fuzzy PID Speed Control o
f
Two Phase Ultra
s
oni
c Mo
tor (Ma Qi
ujie
)
6563
actual
spe
ed
has ex
cee
d
e
d
the spe
ed
set value,
an
d the error i
s
redu
ced.
Wh
en the error
and
error vari
atio
n decrea
s
e to
zero, the system ent
ers the steady stat
e. At
the same time, beca
u
s
e
of the
K
p
of PID co
ntroll
er
can
sp
eed
up
the sy
stem re
sp
on
se, inte
gral
actio
n
is
mainly u
s
ed t
o
eliminate the
static erro
r of t
he syste
m
, and the d
e
rivative acti
on ca
n fore
see the tren
d of the
deviation vari
ation, and produ
ce
s adva
n
cin
g
co
nt
rol
action. The f
u
zzy rule
s for regulatin
g
∆
K
p
,
∆
K
i
,
∆
K
d
are
sho
w
n
in
Tab
l
e 1
and
Ta
bl
e 2
re
sp
ec
tiv
e
ly. Thereinto
,
the fu
zzy
rul
e
s
of
∆
K
i
is
th
e
same a
s
∆
K
d
, their variatio
n trend a
r
e same, but the rang
eability o
f
∆
K
d
is large.
3. Experimental Adjus
t
m
e
nt of Fu
zz
y
Rules
In this p
ape
r, the de
sign
of DSP program
re
alized
ultrasoni
c mo
tor fuzzy PID spe
e
d
control, the e
x
perime
n
tal motor is
Shi
n
sei US
R60
t
w
o-pha
se tra
v
eling wave
ultrasoni
c mo
tor
,
the co
ntrol
circuit is
H-bri
d
g
e
pha
se
-shift PWM
d
r
ive control
circuit [
6
] based o
n
DSP and
CP
LD,
and the inp
u
t power is
DC1
2
V.
As mention
e
d
above, the d
e
sig
n
of the fuzzy
reg
u
lato
r use the unit
domain
of discou
rse,
its output ne
eds to divide
the corre
s
p
ondin
g
out
pu
t quantizatio
n
factor to ma
p to the actu
a
l
value of
∆
K
p
,
∆
K
i
,
∆
K
d
.
In Figure 4, the motor
spee
d
set value i
s
1
00r/min, the
cha
nge p
r
o
c
e
ss
of spee
d step
respon
se in
different outp
u
t quantiz
atio
n factor
contrast with the
chang
e pro
c
e
s
s
of controlled
variable
Ur
ef
, visibly, different outpu
t quantizatio
n factor m
a
kes different
Ure
f
cha
nge p
r
o
c
ess, thereby,
can produ
ce impact to
spe
ed re
sp
o
n
se p
r
o
c
e
ss.
Determi
ne the
quantitative factor of
∆
K
p
,
∆
K
i
,
∆
K
d
are
0.5, 1, 3.4, re
spe
c
tively, throught experim
e
n
ts
adju
s
tment.
Figure 4. Step Re
spo
n
se of Speed
(Experim
enta
l
result
s)
Figure 5. Step Re
spo
n
se of Speed
(Experim
enta
l
result
s)
Figure 6.
K
p
and
K
i
Adjus
t
ment Proc
ess
(Experim
enta
l
result
s)
Figure 7.
K
p
and
K
i
Adjus
t
ment Proc
ess
(Experim
enta
l
result
s)
The fuzzy ru
les sho
w
n in
Table 1 a
n
d
Table 2
reflect the ex
pectatio
n
s of
desig
n
pro
c
e
ss. Th
e rule
s nee
d to trim as the actual
applicatio
n effect. Figure 5 sh
ows th
e rule
s sh
own
in Tabl
e 1
an
d Tabl
e 2, th
e speed
ste
p
re
spo
n
se
wh
en the
motor
set valu
e spe
ed i
s
10
0r/mi
n
,
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 9, September 20
14: 65
60 – 656
5
6564
and the
chan
ge p
r
o
c
e
ss
curve of
cont
rolled va
riable
Ur
ef
, Figure
6 sho
w
s th
e
co
rre
sp
ondi
ng
adju
s
tment proce
s
s cu
rve
of
K
p
, K
i
, the
adju
s
tment proce
s
s is ba
si
cally in line with expectatio
n
s
(adju
s
tme
n
t p
r
ocess i
s
simi
lar, in o
r
de
r t
o
ma
ke
the
g
r
aphi
cs
clea
rer, not
sho
w
n
in the Fig
u
re
)
.
But Figure 5
sho
w
s that ,
after the
spe
e
d
re
ached
ste
ady state , th
e motor spee
d re
du
ce
s sl
o
w
ly
and d
e
viates
from
set valu
e; Refe
rrin
g
t
o
Figu
re
6, it sh
ows the
a
d
justme
nt proce
s
s of
cont
rol
para
m
eter,
a
nd thi
s
i
s
m
a
inly be
cau
s
e
of
K
i
gradu
all
y
red
u
ce in
steady state,
and th
e
cont
ro
l
effect wea
k
e
ned. Analy
s
is of fuzzy a
d
j
u
stment
process
d
a
ta sho
w
s
the domi
nant
rule whi
c
h
lead to
K
i
gra
dually red
u
ce
in steady sta
t
e is the following fu
zzy rul
e
in Table 2.
if
e=ZO
and
ec=PS
then
k
i
=N
S
Adjus
t
the rule as
follows
:
if
e=ZO
and
ec=PS
then
k
i
=Z
O
The n
e
w adj
ustment
processe
s of
K
p
, K
i
are
sho
w
n in Fi
gure
7, the corre
s
pondi
ng
spe
ed respo
n
se
cu
rve a
r
e sh
own in F
i
gure
8,
and
above p
r
obl
e
m
s di
sap
pea
r. Figure
8 at
the
same
time
sh
ows the
effect of the
spe
e
d
contro
l
whe
n
u
s
ing
fixed
para
m
eters P
I
D control, PID
para
m
eters a
r
e the sa
me
as the initial value of PI
D para
m
eter of
fuzzy PID co
ntrol. Visibly , in
the premi
s
e that the fuzzy PID k
eep the
spe
ed to fast-track varyi
ng
set value, wh
en oversho
o
t is
lowe
red to 0, the cont
rol pe
rforma
nce is
better than it of the PID co
ntrolle
r.
4. Conclusio
n
The
spe
c
ial
n
a
ture
of ultra
s
onic
motor’
s
spe
ed
cont
rol
are
fully take
n into a
c
cou
n
t
in this
pape
r, a o
n
lin
e ada
ptive co
rre
ction
of PID control pa
rameters b
a
se
d on fu
zzy lo
gic i
s
de
sig
n
e
d
,
to deal
with the time-va
r
ying no
nlinea
r
of system
. Th
e experi
m
ent
s indi
cate th
a
t
the fuzzy
rul
e
s
of spe
c
ial
de
sign
achieved
the expe
ctan
t control
pa
ra
meter a
d
ju
stment process. Comp
are
d
with
the PID controller, the co
ntrol pe
rform
a
n
c
e ha
s obvio
us imp
r
oving.
Ackn
o
w
l
e
dg
ements
This
study wa
s su
ppo
rted b
y
the National
Natural S
c
ie
nce Fo
und
ation of Chin
a throu
gh
grant nu
mbe
r
U130
450
1.
Referen
ces
[1]
KT
Chau, SW
Chu
ng, CC
Ch
an. Ne
uro-fuzz
y S
pee
d T
r
acking
Contr
o
l of
T
r
aveling-W
a
v
e
Ultras
o
n
i
c
Motor Driv
es
U
s
ing
Dir
ect Pul
s
e
w
idt
h
Mo
dul
ation.
IEEE
Tr
ansacti
ons on
Industry
Ap
plic
ations
, 20
03
;
39(4): 10
61-
10
69.
[2]
FL Wen, CY Yen, MS Ouy
a
ng.
Fu
zz
y
Slidi
ng-Mo
de
Control for
Thin-Disc
Ultrason
i
c Motor
.
Procee
din
g
s of
the 2
0
0
4
IEE
E
Internati
ona
l
Confer
ence
o
n
Contro
l Ap
plic
ations, U
n
ite
d
States. 200
4:
81-8
6
.
Figure 8. Step respon
se of
spee
d
(Experim
enta
l
result
s)
0.0
0
.5
1.0
1
.5
2.
0
2
.
5
3.
0
3
.
5
0
20
40
60
80
100
120
PID
fuzzy
PID
s
p
eed/(r
/min)
time/s
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Fuzzy PID Speed Control o
f
Two Phase Ultra
s
oni
c Mo
tor (Ma Qi
ujie
)
6565
[3]
Shi T
i
ngn
a,
Xu Sh
ao
hui. F
u
zz
y
-
PI
Du
al-
M
ode
Ada
p
tiv
e
Sp
ee
d C
o
n
t
rol for
Ultras
onic
Motors.
T
r
ansactio
n
s o
f
China El
ectro
t
echnic
a
l Soci
e
t
y
. 2003; 18(3):
1-4.
[4]
Li H
uafe
ng, Z
h
ao C
h
u
n
she
n
g
.
Precise P
o
siti
on C
ontro
l of
Ultraso
nic Mot
o
r B
y
F
u
zz
y C
ontrol.
Jou
r
nal
o
f
H
u
a
z
hong
Un
i
v
e
r
si
ty o
f
Scie
n
c
e
an
d
Te
chn
o
l
o
g
y
. 20
04; 32(5): 22-
24.
[5]
Xi
a C
h
a
ngl
ia
n
g
,
Xu S
h
a
ohu
i. F
u
zz
y Ad
aptiv
e Sp
ee
d C
ontr
o
l for
Ultras
oni
c Motor B
a
sed
on
Gen
e
ti
c
Algorit
hm.
Proceed
ings
of the Csee
. 200
3; 2
3
(3): 99-1
03.
[6]
W
ang H
a
i
y
a
n
,
Shi Ji
ngzh
uo.
Ultraso
nic Mot
o
r’s Ph
ase S
h
i
ft PW
M Control Base
d o
n
CP
LD.
Electric
al
Autom
a
tion
. 20
09; 31(2): 4
8
-5
0.
Evaluation Warning : The document was created with Spire.PDF for Python.