Int
ern
at
i
onal
Journ
al of
P
ow
er El
ectron
i
cs a
n
d
Drive
S
ys
te
m
(I
J
PE
D
S
)
Vo
l.
9
, No
.
4
,
Decem
ber
20
18
, p
p.
1967
~
1975
IS
S
N: 20
88
-
8
694,
DOI: 10
.11
591/
ij
peds
.
v9
.i
4
.
pp
1967
-
19
75
1967
Journ
al h
om
e
page
:
http:
//
ia
escore.c
om/j
ourn
als/i
ndex.
ph
p/IJPE
D
S
Self
-
tu
nin
g
Fu
zz
y
Lo
gi
c
Con
troll
er
Based o
n Ta
kagi
-
Sugeno Appli
ed to
In
du
ction Mot
or Driv
es
Nabil
Fa
r
ah,
M.
H.
N. Tali
b,
Z
. I
br
ah
im
,
J
.
M
. L
az
i, M
aa
s
pa
li
z
a Az
ri
Facul
t
y
of Electr
ic
a
l
Eng
ineeri
ng
,
Cen
te
r
for
Rob
oti
cs
and
Industr
ia
l
Autom
at
ion
(
CeRIA),
Univ
er
siti
T
ekni
k
al Ma
lay
s
ia
Mela
ka
,
Ma
lay
si
a
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
A
ug
8
, 2
01
8
Re
vised
Sep 2
1
, 2
01
8
Accepte
d
Oct
10
, 201
8
Fuzz
y
logic
con
trol
ler
has
be
en
the
m
ai
n
foc
us
for
m
an
y
rese
a
rch
ers
and
industri
es
in
m
ot
or
drive
s.
Th
e
p
opula
rity
o
f
Fuzz
y
Logic
Contro
ll
er
(FLC)
is
due
to
it
s
re
li
ab
i
li
t
y
and
ab
il
i
t
y
t
o
handl
e
par
amete
rs
ch
ange
s
dur
ing
loa
d
or
disturba
nc
e.
Fuzz
y
logi
c
de
si
gn
ca
n
be
visual
i
ze
d
in
two
ca
te
gor
ie
s
,
m
amdani
design
or
Ta
k
agi
-
Sug
eno
(TS).
Mam
dani
t
y
p
e
c
an
f
ac
i
li
t
at
e
the
design
proc
ess,
howeve
r
it
req
u
i
re
high
computational
burde
n
espe
cially
wit
h
big
num
ber
of
rule
s
and
exper
imenta
l
te
st
ing.
Thi
s
p
ape
r
,
de
vel
op
S
elf
-
Tuni
ng
(ST)
m
ec
han
ism
base
d
on
Ta
k
agi
-
Sug
eno
(TS)
fuz
z
y
t
y
pe.
Th
e
m
ec
hani
sm
tunes
the
input
sc
al
i
ng
fac
tor
of
spe
e
d
fuz
z
y
con
trol
of
Induc
ti
on
Motor
(IM)
drive
s
Based
on
t
he
spee
d
err
o
r
and
cha
ng
es
of
err
or.
A
compari
son
study
is
done
b
et
we
en
the
stand
ard
TS
and
the
ST
-
TS
base
d
on
sim
ula
ti
ons
appr
oac
hes
consi
der
i
ng
diffe
r
ent
spee
d
oper
at
i
ons.
Speed
response
cha
ra
cteri
sti
cs
such
as
rise
ti
m
e,
over
s
hoot,
and
settlin
g
ti
m
e
are
compare
d
for
ST
-
TS
and
TS.
It
was
sho
wn
tha
t
ST
-
TS
has
opti
m
um
result
s
compare
d
to
the
standa
rd
TS.
T
he
signifi
c
anc
e
of
the
proposed
m
et
hod
is
tha
t
,
opt
imum
computat
iona
l
bur
den
red
u
ct
ion
is
ac
hi
eve
d
.
Ke
yw
or
d:
Com
pu
ta
ti
on
al
burde
n
IM drive
ST
-
TS
TS
Copyright
©
201
8
Ins
t
it
ut
e
o
f
Ad
vanc
ed
Engi
n
e
er
ing
and
S
cienc
e
.
Al
l
rights
reserv
ed
.
Corres
pond
in
g
Aut
h
or
:
Nab
il
Fa
ra
h
,
Faculty
of Elec
tric
al
Engineer
ing
,
Ce
nter fo
r
R
obotics an
d Ind
ust
rial
A
ut
om
ation
(CeR
IA),
Un
i
ver
sit
i Te
knikal M
al
ay
sia
Mel
aka,
Me
la
ka,
Mal
ay
sia
Em
a
il
:
Nab
il
-
f
arah1
1@ho
tm
ai
l.com
1.
INTROD
U
CTION
Fu
zzy
log
ic
co
ntr
ol
has
at
tract
ed
an
intensiv
e
at
te
ntion
in
the
fiel
d
of
m
oto
r
dri
ves
[
1],
wh
et
her
i
n
sp
ee
d
con
t
ro
l
[2
]
or
c
urren
t
con
tr
ol
[3
]
.
I
t
has
been
a
reli
able
al
te
rn
at
ive
of
the
pro
portio
nal
integral
con
t
ro
ll
ers
,
du
e
to
it
s
capab
il
it
y
of
handlin
g
non
-
li
near
it
y
and
un
ce
rtai
nty
[4
]
.
Var
i
ou
s
res
earches
ha
ve
pro
ve
d
the opti
m
u
m
p
erfor
m
ance of
fu
zzy
l
og
ic
c
ontr
ollers in
com
par
ison
to
conv
e
ntio
nal c
ontrolle
rs [5
]
,
[
6].
The
w
orka
bili
ty
of
f
uzzy
lo
gi
c
is
achieved
in
identic
al
m
ann
e
r
to
hum
a
n
be
ha
vior.
F
uz
zy
syst
e
m
ta
kes
inputs
as
cl
assic
al
data
and
c
onve
rts
them
into
fu
zz
y
data.
Finall
y
the
data
is
process
to
ob
ta
i
n
the
desire
d
outp
ut
base
d
on
a
set
of
desig
ne
d
r
ul
es
[7
]
.
T
he
re
a
re
tw
o
po
pu
la
r
ty
pe
of
fu
zzy
interface
syst
e
m
1)
m
a
m
dan
i
[8
]
and
2)
Ta
k
a
g
i
-
S
ug
e
no
[9
]
.
Ma
m
dan
i
fu
zzy
typ
e
is
widely
use
d
by
m
any
re
searche
rs.
[10].
[11]
,
Howe
ver,
inc
r
easi
ng
the
c
om
pu
ta
ti
on
al
bu
rd
e
n
ca
n
be
a
big
iss
ue
es
pe
ci
al
ly
wh
en
de
sign
i
ng
bi
g
nu
m
ber
of
ru
le
s
f
or
ex
perim
ental
te
sti
ng
.
T
his
prob
le
m
can
be
m
ini
mize
d
by
us
i
ng
TS
ty
pe
with
t
he
sam
e
nu
m
ber
of
ru
le
s a
nd
funct
ion
al
it
y.
Ma
ny
stu
dies
ha
ve
util
iz
ed
TS
fu
z
zy
ty
pe
to co
nt
ro
l
t
he
IM d
ri
ve
syst
em
[1
3]
-
[
15
]
.T
he
m
ai
n
featu
r
e
of
the
TS
is
that
s
in
gleto
n
ou
t
pu
t
m
e
m
be
rsh
i
p
MFs
ca
n
be
us
e
d
.
Mos
t
of
the
stu
dies
relat
ed
to
i
nducti
on
m
oto
r
dr
i
ves
ut
i
li
ze
TS
fu
zz
y
ty
pe
as
s
pee
d
c
on
t
ro
l
[
16
]
or
as
tu
ning
t
oo
l
f
or
ot
her
f
uzzy
ty
pe
[12]
.
T
he
eff
ect
ive
ness
of
sel
f
-
tu
ning
f
uzzy
ty
pe
has
been
discuss
e
d
by
m
asi
al
a
[12].
T
he
pro
posed
syst
em
util
iz
ed
com
bin
at
ion
of
m
a
m
dan
i
an
d
T
S
F
uzzy
ty
pe
syst
em
.
Ma
m
dan
i
fu
zz
y
is
ap
plied
t
o
t
he
m
ai
n
sp
eed
c
on
t
ro
l
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8
694
I
nt J
P
ow
Ele
c
&
Dri
Syst
, Vol.
9
, N
o.
4
,
D
ece
m
ber
2018
:
1967
–
1975
1968
wh
il
e
the
TS
f
uzzy
app
li
e
d
to
the
tu
ning
m
echan
ism
.
The
syst
e
m
uti
li
ze
s
TS
fu
zzy
wit
h
25
r
ules
to
achieve
bette
r per
form
ance a
nd r
e
duc
e the c
om
pu
ta
ti
on
al
bur
den on the
h
a
r
dw
a
re
.
Ther
e
is
a
la
ck
of
stud
ie
s
of
ut
il
iz
ing
the
TS
ty
pe
as
sp
eed
con
t
ro
l
of
induc
ti
on
m
oto
r
dr
iv
e
and
tu
ne
it
with
the
sam
e
TS
ty
pe.
Ho
we
ve
r,
a
stu
dy
[1
7]
has
pro
po
s
ed
a
sel
f
-
tu
ning
ST
ba
sed
on
TS
fu
zzy
t
ype
for
process
perf
orm
ance
i
m
pr
ov
e
m
ent.
The
st
ud
y
has
discusse
d
in
ge
ner
al
the
i
m
ple
m
entat
ion
of
Self
-
Tu
ning
Taka
gi
-
S
ug
e
no
(S
T
-
TS
)
to
tu
ne
a
Ta
kag
i
-
S
ugen
o
fu
zzy
ty
pe
consi
der
i
ng
a
gen
e
ral
pr
oces
s
syst
e
m
.
How
ever,
the
pro
po
se
d
s
yst
e
m
app
li
ed
to
the
First
Order
a
nd
Sec
ond
ord
er
wit
h
delay
tim
e
t
ran
s
fer
f
unct
io
n.
Th
e
eff
ect
ive
ness
of
the
pro
posed
con
t
r
oller
ap
pl
ie
d
to
the
higher
order
syst
e
m
su
ch
as
IM
dr
i
ve
syst
e
m
i
s
sti
ll
undisc
ov
e
red.
This
pa
per
ai
m
s
to
util
iz
e
the
featu
res
of
T
S
f
uzzy
ty
pe
i
n
reducin
g
t
he
c
om
pu
ta
ti
on
al
bur
de
n.
The
TS
f
uzzy
ty
pe
are
a
ppli
ed
to
the
m
ai
n
sp
ee
d
c
on
tr
oller
a
nd
Self
-
T
un
i
ng
m
echan
is
m
.
The
Sel
f
-
tu
ning
m
echan
ism
is
us
e
d
to
tun
e
f
uzzy
the
input
scal
ing
facto
r
s.
An
y
va
riat
ion
of
on
the
sp
ee
d
error
will
be
com
pen
sat
ed
by
the
ST
m
e
chan
ism
to
update
the
scal
in
g
facto
rs
t
o
a
dap
t
to
t
he
sy
stem
var
ia
ti
ons
.
T
hi
s
appr
oach
e
d
is
aim
ed
to
ac
hieve
op
ti
m
u
m
com
pu
ta
ti
on
al
bu
rd
e
n
re
du
ct
io
n
achie
ved
an
d
ben
e
fici
al
espe
ci
al
ly
for real
experi
m
ental
test
.
The
pa
per
di
vid
ed
int
o
fi
ve
s
ect
ion
s,
sect
io
n
I
rev
ie
w
the
fu
zzy
l
og
ic
co
ntr
ol
in
IM
dr
i
ve
syst
em
an
d
discusse
s
the
e
ff
ect
ive
ness
of
TS
f
uzzy
ty
pe
.
Sect
io
n
I
I,
pr
esents
the
m
at
hem
atical
m
od
el
li
ng
of
the
I
M
dr
i
ve
syst
e
m
a
lon
g
with
associat
e
d
eq
uations
.Se
ct
ion
III
f
ocus
on
the
desig
n
pro
ducers
of
Ta
kag
i
–
Suge
no
(
TS)
as
well
as
the
Se
lf
-
T
un
i
ng
Ta
ka
gi
-
S
ugen
o
(ST
-
TS
)
al
ong
w
it
h
co
rr
es
pond
ing
m
e
m
ber
s
hi
p
functi
ons
a
nd
ru
le
bases.
Sect
io
n
IV
pr
e
sents
t
he
si
m
ulati
on
s
re
su
lt
s
of
com
par
iso
n
bet
wee
n
the
Stan
dard
T
S
f
uzzy
an
d
th
e
sel
f
-
tun
e
d
S
T
-
T
S fuzzy
. Lastly
, se
ct
ion
V
c
on
cl
ud
e
the
of the
s
tud
y a
nd h
i
gh
li
gh
te
d
the
m
ai
n
fin
ding
obta
in
ed.
2.
INDU
CTIO
N
MOTO
R M
O
DEL
The
in
duct
ion
m
oto
r
is
m
at
hem
at
ic
ally
m
od
el
le
d
an
d
can
be
represe
nted
i
n
va
rio
us
ref
e
r
ence
f
ram
e.
Diff
e
re
nt
ref
e
r
ences
f
ram
e
of
in
du
ct
io
n
m
otor
is
disc
us
s
ed
in
[18]
,
[
22]
.
Fig
ur
e
1
s
hows
t
he
sim
plifie
d
equ
i
valent
q
-
a
xis a
nd d
-
axis
ci
rcu
it
of s
quir
rel
cage
in
du
ct
ion
m
oto
r
in ro
ta
ry r
efe
rence
fram
e [2
0].
(a)
(b)
Figure
1. Eq
ui
valent circ
uit o
f
in
duct
ion m
oto
r
i
n
r
otary
refe
ren
ce
fram
e (a)
q
-
a
xis fram
e
, (b) d
-
a
xis fra
m
e
Re
fer
ri
ng
t
o
th
e
equ
iv
al
ent
ci
rcu
it
of
i
nduction
m
oto
r
re
pr
e
sented
in
Fi
gur
e
1,
volt
age
in
q
-
a
xis
an
d
d
-
a
xis
of roto
r a
nd stat
or can
be
e
xpresse
d
a
s foll
ow
:
()
d
r
l
r
d
r
m
d
s
d
r
L
i
L
i
i
(1)
qs
q
s
s
q
s
e
d
s
d
V
R
i
dt
(2)
ds
ds
s
ds
e
ds
d
V
R
i
dt
(3)
\
\
RS
V
qs
V
qr
e
Y
ds
(
e
-
r
)
Y
dr
L
I
s
=
Ls
-
Lm
L
I
r
=
Lr
-
Lm
Rr
Lm
\
\
RS
V
ds
V
dr
e
Y
qs
(
e
-
r
)
Y
qs
L
I
s
=
Ls
-
Lm
L
I
r
=
Lr
-
Lm
Rr
Lm
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
P
ow
Ele
c
&
D
ri
Syst
IS
S
N: 20
88
-
8
694
Self
-
tu
ning F
uzzy
Logic C
on
tr
oller Ba
sed
on Tak
ag
i
-
Su
geno
A
ppli
ed
t
o Ind
uction M
oto
r…
(
Nab
il
F
ar
ah
)
1969
()
qr
q
r
r
q
r
e
r
d
r
d
V
R
i
dt
(4)
()
dr
d
r
r
d
r
e
r
q
r
d
V
R
i
dt
(5)
And
qr
V
,
dr
V
=0
an
d
t
he
f
lu
x
e
quat
ion as f
ollo
w:
()
q
s
ls
q
s
m
q
s
q
r
L
i
L
i
i
(6)
()
q
r
lr
r
m
q
s
q
r
L
i
L
i
i
(7)
()
d
s
l
s
d
s
m
d
s
d
r
L
i
L
i
i
(8)
The
el
ect
r
om
a
gn
et
ic
to
r
qu
e
c
an be e
xpresse
d
as
foll
ow
:
3
()
22
m
e
d
r
q
s
q
r
d
s
r
L
P
T
i
i
L
(9)
The
num
ber
of
inducti
on
m
otor
po
le
s
is
r
epr
ese
nted
by
P,
once
the
ve
ct
or
co
ntr
ol
is
achieved
d
fr
am
e
of
r
otor
side
is
zer
o.
Hen
ce
,
the
m
oto
r
to
r
qu
e
is
c
on
t
ro
ll
ed
by
q
fr
am
e
of
sta
to
r
side
as
m
od
el
le
d
i
n
equ
at
io
n 1
0:
3
()
22
m
e
d
r
q
s
r
L
P
Ti
L
(
10)
The
fu
ll
dri
ve
syst
e
m
of
in
duct
ion
m
oto
r
is
presente
d
in
Fig
ur
e
2.
The
syst
e
m
con
sis
ts
of
sp
e
e
d
con
t
ro
ll
er,
pha
se co
nversi
on, hyst
eresis c
urr
ent contr
oller,
inv
e
rter,
m
oto
r a
nd en
c
oder
.
Figure
2. I
nduc
ti
on
m
oto
r
dr
i
ve
b
ase
d Hyst
er
esi
s curre
nt contr
oller
The
FO
C
dr
iv
e
syst
e
m
pr
ese
nted
in
Fig
ur
e
2
is
based
on
hyste
resis
cu
rrent
con
t
ro
ll
er
[
21
]
.
T
his
is
du
e
t
o
it
s
sim
ple
struct
ur
e
,
f
ast
respo
ns
e
a
nd
good
acc
uracy
.
The
in
ve
r
te
r
pulse
s
are
gen
e
rated
by
ut
il
iz
ing
hys
te
resis b
an
d.
T
he
sp
ee
d
c
on
t
ro
ll
er
is
des
ign
e
d
us
in
g
fuzzy
log
ic
c
on
t
r
oller.
T
he
sp
ee
d
er
r
or
b
et
w
ee
n
act
ua
l
m
oto
r
sp
ee
d
a
nd
re
fer
e
nce
s
peed
are
proc
essed
t
hroug
h
sp
ee
d
c
on
tr
ol
le
r
to
produce
the
to
rque
c
urren
t
ref
e
ren
ce
c
urr
ent
iq*
.
T
he
iq*
c
urren
t
is
gathe
red
with
con
sta
nt
id*
and
tra
nsfo
rm
ed
into
th
ree
ph
a
se
qu
a
ntit
ie
s.
T
he
se
thre
e
ph
ase
quantit
ie
s
are
the
ref
e
rence
currents
w
hich
are
t
hen
c
ompare
d
with
t
he
thre
e
S
pe
e
d
c
on
t
r
ol
l
e
r
E
nc
ode
r
r
r
*
DQ
-
A
B
C
I
qs
*
I
ds
*
H
ys
t
e
r
e
s
i
s
Ia
b
c
V
S
I
P
u
l
s
es
A
B
C
-
DQ
T
he
t
a
c
a
l
a
c
ul
a
t
i
on
I
a
b
c
q
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8
694
I
nt J
P
ow
Ele
c
&
Dri
Syst
, Vol.
9
, N
o.
4
,
D
ece
m
ber
2018
:
1967
–
1975
1970
ph
a
se
act
ual
m
oto
r
cur
re
nts
.
The
res
ultant
of
the
com
par
iso
n
betwee
n
the
act
ual
and
ref
ere
nce
cu
rrents
is
then
fe
d
into
hyste
resis
cu
rrent
con
t
ro
ll
er
to
gen
e
rate
th
e
require
d
swi
tc
hin
g
pulse
s
for
the
three
ph
a
se
Vo
lt
age
S
ourc
e I
nv
e
rter
(VSI
)[19
]
,
[
23
]
.
3.
SPEE
D CO
N
TROLL
ER
D
ESIGN
The
f
uzzy
lo
gi
c
con
t
ro
ll
er
util
iz
ed
as
sp
ee
d
con
t
ro
l
wh
ic
h
ta
ke
tw
o
i
np
ut
(
error
a
nd
cha
nge
of
e
rror)
and
pr
ocess
t
hem
to
pro
duce
the
ou
t
pu
t
sig
nal
.T
her
e
are
three
ste
ps
i
n
f
uzzy
lo
gic
c
ontrolle
r,
f
uzzific
at
ion
wh
ic
h
c
onve
rt
the
cris
p
data
into
fuzzy
dat
a
or
MF.
I
nter
face
e
ng
i
ne
will
com
bin
e
the
MFs
wit
h
des
ign
e
d
fu
zzy
ru
l
es
t
o
ob
ta
in
the
fuzz
y
ou
t
pu
t.
Last
ly
the
de
fu
zzi
ficat
ion
will
conver
t
bac
k
the
f
uzzy
data
i
nto
cris
p
ou
t
pu
t
data.
Fi
gure
3
s
hows
t
he
blo
c
k
diag
r
a
m
of
fu
zzy
l
ogic
co
ntr
oller
ste
ps
.
The
re
a
r
e
two
ty
pe
of
fu
zzy
interface
syst
e
m
(F
IS
)
,
m
a
md
ani
or
Ta
kag
i
-
Suge
n
o
(TS).
TS
fu
zzy
ap
pli
ed
a
sing
le
to
n
MFs
for
the
outp
ut
fu
zzy
.
T
he
ad
va
ntages
of
this
TS
fu
zzy
ty
pe
sel
f
-
tun
i
ng
co
ntr
oller
are
to
reduce
the
co
m
pu
ta
ti
on
al
burd
e
n
of
the contr
oller.
In the
fo
ll
owin
g
sect
io
n, the
de
sign p
ro
ce
ss
f
or ST m
echan
i
sm
w
il
l be f
urt
h
er
d
isc
us
se
d.
Figure
3. F
uzz
y l
og
ic
blo
c
k d
ia
gr
am
3.1. St
an
d
ard
Takagi
-
Su
geno
(TS)
Desi
gn
The
sta
ndar
d
TS
is
sam
e
as
m
a
m
dan
i
ty
pe
for
th
e
in
put
va
riables.
F
or
i
nductio
n
m
otor
tw
o
i
nputs
fu
zzy
with
li
ne
ar
MFs
and
on
e
ou
tp
ut
fu
zzy
with
co
ns
ta
nt
MFs.
For
each
inp
ut
an
d
outpu
t
there
is
a
scal
ing
factor
t
o
ad
j
ust
the
factor
.
Fi
gure
4
pr
ese
nt
s
the
bl
oc
k
dia
gr
am
of
fu
zzy
log
ic
co
ntro
ll
er
util
iz
ed
for
sp
ee
d
con
t
ro
l of
in
duct
ion
m
oto
r.
T
wo
in
pu
t var
ia
bles
s
peed
er
ror
(e
)
a
nd
cha
ng
e
of
s
peed
er
ror
(
e) which
ne
ed
t
o
be
c
onve
rted
i
nto
f
uzzy
va
ria
bles
(
f
uzzifica
ti
on),
t
hen
the
i
nterf
ace
e
ng
i
ne
com
bin
es
the
f
uzzy
r
ules
with
th
e
MFs
of
the
va
riables
to
pr
oduce
the
f
uz
zy
ou
tp
ut.
Th
e
defuzzifi
cat
ion
process
c
onve
rt
bac
k
the
fu
zzy
var
ia
bles to c
ri
sp
var
ia
bles
util
iz
ing
fuzzy
ru
l
es an
d
si
ng
le
to
n ou
t
pu
t M
Fs.
Figure
4.St
an
da
rd TS f
or
IM
dr
i
ve
In
the
pr
e
proce
ssing
par
t,
the
crisp
in
pu
ts
of
the
sp
ee
d
error,
e
and
it
s
change
of
sp
ee
d
error
e
ar
e
conve
rted
i
nto
to their
corres
pondin
g f
uzzy v
ariable a
nd the
y are
def
ine
d a
s:
F
u
z
z
i
f
i
e
r
I
n
t
e
r
f
a
c
e
e
n
g
i
n
e
K
n
o
w
l
e
d
g
e
B
a
s
e
P
r
o
c
e
s
s
D
e
f
u
z
z
i
f
i
e
r
*
Ge
Z
^
-
1
G
c
e
G
c
u
IM
r
Iq
*
F
u
z
z
y
Z
^
-
1
P
r
e
-
p
r
o
c
c
e
s
s
i
n
g
P
r
o
c
e
s
s
i
n
g
P
o
s
t
-
p
r
o
c
c
e
s
s
i
n
g
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
P
ow
Ele
c
&
D
ri
Syst
IS
S
N: 20
88
-
8
694
Self
-
tu
ning F
uzzy
Logic C
on
tr
oller Ba
sed
on Tak
ag
i
-
Su
geno
A
ppli
ed
t
o Ind
uction M
oto
r…
(
Nab
il
F
ar
ah
)
1971
(
)
=
(
∗
(
)
−
(
)
)
=
(
)
(11)
∆
(
)
=
(
(
)
−
(
−
1
)
(12)
Fr
om
the
a
bove
eq
uatio
n,
ωr
*
a
nd
ω
r
sta
nd
for
ref
e
ren
ce
a
nd
act
ual
s
pee
d
resp
e
ct
ively
.
Me
anwhil
e,
(k)
an
d (k
-
1) re
pr
ese
nt the c
urre
nt and pre
vio
us
sta
te
o
f
t
he
error. Ts
re
pre
sents for
t
he
sa
m
pl
ing
ti
m
e. T
he
Ge
and
Gce
de
note
th
e
e
rror
an
d
the
c
hange
of
error
gai
n
scal
i
ng
fact
or
.
T
he
m
axi
m
u
m
Ge
gain
is
determ
ined
to
cov
e
r
t
he
rate
d spee
d usin
g
t
he
foll
ow
i
ng equati
on.
=
1
|
|
(13)
Wh
e
re,
is
the
m
axi
m
u
m
erro
r
for
the
rate
d
s
peed
op
e
rati
on
to
e
nsure
high
e
noug
h
ga
in
app
li
ed
t
o
c
over
the
rate
d
s
pe
ed
ope
rati
on
and
norm
al
iz
e
d
the
in
put
val
ue.
F
or
the
c
ha
ng
e
of
er
ror
ga
in,
Gc
e
and
outp
ut
gai
n,
Gc
u
the
m
em
ber
sh
ip
f
unct
ion
range
opte
d
to
fit
the
rated
s
peed
operati
on.
T
he
m
e
m
ber
sh
i
p
functi
on fo
r
er
r
or
(e)
an
d
c
ha
nge
of
e
rro
r
e
and incr
em
enta
l ou
t
pu
t
gain
, c
u
are
prese
nted
in
Fi
gure
5.
S
e
ven
m
e
m
ber
sh
ip
f
unct
ions are
u
ti
li
zed for the i
nputs e a
nd
e
w
hile 7 s
i
ng
le
t
on m
e
m
ber
sh
ip
functi
ons a
re
ut
il
iz
ed
for
the
outp
ut.
Un
ive
rse
of
di
scourse
f
or
in
pu
t
is
dec
om
po
se
d
into
se
ve
n
f
uzzy
MFs
f
or
in
put
rangin
g
from
Neg
at
ive
Bi
g
(
NB)
i
nto
Posi
ti
ve
Bi
g
(P
B
).
Wh
il
e
for
t
he
ou
t
pu
t
seve
n
sing
le
to
ns
(c
onsta
nt)
MF
s
are
s
el
ect
ed
wh
ic
h
im
plies
the
feat
ur
e
of
Taka
gi
-
S
ug
e
no
(TS),
fu
zzy
ty
pe.
T
his
ca
n
m
ini
m
iz
e
the
exe
cutio
n
ti
m
e
of
t
he
fu
zzy
syst
em
.
Figure
5.
I
nput
s and
outp
ut m
e
m
ber
sh
i
p fun
ct
ion
s
var
ia
ble
desig
n 4
9
r
ules
The
7x7
m
a
m
ber
sh
i
p
f
unct
io
ns
m
a
trix
res
ults
in
49
f
uzzy
r
ule
set
.
T
he
fuzzy
ru
le
s
set
s
desig
ne
d
ar
e
pr
ese
nted
i
n
Table
1
f
or
49
ru
le
s.
T
he
r
u
le
s
are
devel
ope
d
us
i
ng
TS
-
ty
pe
f
uzzy
inference.
T
he
r
ules
are
interp
reted
b
as
ed on t
he
ta
ble
s for
t
he rules
desig
n usin
g F
uzzy L
og
ic
To
ols Mat
la
b.
Table1
. Fuzzy
Rule (
49)
e
NB
NM
NS
ZE
PS
PM
PB
NB
NB
NB
NB
NB
NM
NS
ZE
NM
NB
NB
NB
NM
NS
ZE
PS
NS
NB
NB
NM
NS
ZE
PS
PM
ZE
NB
NM
NS
ZE
PS
PM
PB
PS
NM
NS
ZE
PS
PM
PL
PB
PM
NS
ZE
PS
PM
PB
PL
PB
PB
ZE
PS
PM
NB
PB
PB
PB
Thro
ugh de
fu
z
zi
ficat
ion
, t
he ou
t
pu
t c
urre
nt
Iq*
is cal
c
ulate
d usin
g
t
he
f
ollow
i
ng equati
on
∗
(
)
=
∗
(
−
1
)
+
(
∆
(
)
)
(14)
The
scal
in
g
fac
tor
f
or
t
he
f
uzz
y
designed
a
re
m
anu
al
ly
tun
e
d
to
ob
ta
in
t
he
op
ti
m
u
m
per
form
ance
fo
r
the
G
ce
value.
Wh
il
e
Ge
val
ue
is
c
onsta
nt
base
d
on
the
m
axi
m
u
m
sp
eed
e
rror
cal
cul
at
ion
.
Last
ly
,
the
G
c
u
value
is
set to
1.
1
0
-
0
.
5
0
.
5
-
0
.
75
-
0
.
25
0
.
25
0
.
75
1
0
-
1
NB
NM
NS
ZE
PS
PM
PB
1
0
-
0
.
5
0
.
5
-
0
.
75
-
0
.
25
0
.
25
0
.
75
-
1
1
NB
NM
NS
ZE
PS
PM
PB
E
,
CE
Cu
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8
694
I
nt J
P
ow
Ele
c
&
Dri
Syst
, Vol.
9
, N
o.
4
,
D
ece
m
ber
2018
:
1967
–
1975
1972
3.2. Sel
f
-
T
une
d S
T
Desig
n (ST
-
TS)
The
pr
e
vious
s
ect
ion
discuss
e
d
the
desig
n
process
of
sta
nda
rd
T
S
f
uzzy
or
fixe
d
gai
ns
fuzzy
.
It
wa
s
ob
s
er
ved
that
the
value
of
outpu
t
gain
(
Gcu)
is
fixe
d
wh
ic
h
m
ake
the
syst
e
m
un
a
ble
to
a
dap
t
onli
ne
wit
h
a
ny
changes
occ
urs.
T
he
propose
d
Sel
f
-
T
unin
g
m
echan
ism
(S
T
-
TS
)
fo
c
us
e
d
on
tu
ning
t
he
input
scal
in
g
f
act
ors
wh
ic
h
ena
ble
the
in
pu
t
scal
in
g
facto
rs
Ge
a
nd
Gce
to
a
da
pt
onli
ne
in
ac
corda
nce
to
th
e
process
c
hanges
f
or
perform
ance
i
m
pr
ov
em
ent.
Tw
o
fu
zzy
syst
e
m
hav
e
been
design
e
d
to
tu
ne
the
inputs
scal
ing
facto
r.
Figure
6
pr
ese
nts t
he block dia
gr
am
o
f
the
pro
po
se
d s
el
f
-
tu
ning m
echan
ism
.
Figure
6. Pro
pose
d
Self
-
T
un
i
ng (
S
T_
TS) m
echan
ism
As
discu
ssed
e
arli
er,
the
pro
pose
d
ST
-
T
S
to
tun
e
the
inputs
scal
ing
facto
rs
by
desig
ning
the
ru
le
s
def
i
ned
in
te
r
m
s
of
e
an
d
C
e
for
updatin
g
the
scal
in
g
fa
ct
or
s.
Self
-
tu
nin
g
m
et
ho
d
bas
ic
al
ly
m
eans
t
hat
th
e
sel
f
-
tu
ning
of
input
gains
ba
sed
on
er
ror
a
nd
c
ha
ng
e
in
error.
Acc
ordi
ng
to
t
his
ST
m
echn
asi
m
,
th
e
input
scal
ing
facto
rs ca
n be c
om
pu
te
d by util
iz
ing
the foll
owin
g
e
qu
at
io
ns
:
E (k)=
(α.G
e
).
e
(k)
Ce
(
k)=(
β.
Gce)
ek)
α
and
β
a
re
the
updatin
g
facto
rs
wh
ic
h
us
e
d
t
o
c
on
ti
nu
ously
ad
j
ust
the
i
nputs
scal
in
g
factor
s
Ge
a
nd
Gce
bas
ed o
n
the erro
rs
an
d chan
ge of
e
rrors. H
e
nce,
t
he
inputs scal
in
g
f
act
or
s a
re v
a
rie
d
an
d
a
dju
ste
d on
li
ne
with a
ny ch
a
ng
es to t
he
syst
e
m
accor
dingly
. Th
e
m
e
m
ber
sh
ip
f
un
ct
io
n for α a
nd β a
re
presented
in Fi
gu
re
7.
Figure
7. Me
m
ber
s
hip f
unct
io
n for
updatin
g fact
ors α, β
4.
SIMULATI
O
N RESULTS
MATLAB/S
I
MULI
NK
e
nv
i
ronm
ent
has
be
en
util
iz
ed
to
de
velo
p
t
he
f
uzzy
syst
em
a
s
well
as
to
m
od
el
the
IM
dr
i
ve
syst
em
.
Tw
o
dif
fer
e
nt algorit
hm
s
have
been
d
esi
gn
e
d
w
hich
t
he
sta
nd
a
r
d
Taka
gi
-
Sugeno
(S
T
)
a
nd
Self
-
Tu
ning
Ta
ka
gi
-
Suge
no
(
ST
-
TS).
Bot
h
m
eth
ods
te
ste
d
un
der
dif
fe
ren
t
s
peed
ope
rati
ons
a
nd
su
bject
e
d
to
lo
ad.
49
r
ules
f
uz
zy
wer
e
util
iz
ed
thr
ough
al
l
the
te
sti
ng
.T
he
adv
a
ntage
of
ST
-
TS
is
pro
duci
ng
*
Ge
Z
^
-
1
G
c
e
G
c
u
IM
r
Iq
*
F
u
z
z
y
Z
^
-
1
R
u
l
e
b
a
s
e
(
b
)
R
u
l
e
b
a
s
e
(
a
)
1
.
5
0
.
5
0
1
-
0
.
25
0
.
25
0
.
75
1
.
25
1
-
0
.
5
S
B
.
5
2
3
4
-
1
1
S
M
B
VB
E
,
CE
E
,
CE
a
b
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
P
ow
Ele
c
&
D
ri
Syst
IS
S
N: 20
88
-
8
694
Self
-
tu
ning F
uzzy
Logic C
on
tr
oller Ba
sed
on Tak
ag
i
-
Su
geno
A
ppli
ed
t
o Ind
uction M
oto
r…
(
Nab
il
F
ar
ah
)
1973
le
sser
com
pu
ta
ti
on
al
bur
den
util
iz
ed
du
e
to
the
sing
le
ton
ou
t
pu
t
m
e
m
be
rsh
i
p
functi
ons
.
In
ad
diti
on,
ST
-
T
S
sys
tem
is able to in
c
rease t
he a
ccur
acy
of the
syst
e
m
p
erfo
r
m
ance.
4.2. N
o
-
l
oad
Opera
tio
ns
The
sp
ee
d
perf
or
m
ance
of
TS
and
ST
at
400,9
00
a
nd
1400
rp
m
with
no
lo
ad
is
pr
ese
nted
in
Figure
8,
9, 10
res
pecti
vely
. Th
e
sele
ct
ed
s
peed ra
nges c
over
from
low to
rated
s
peed o
pe
rati
ons.
Figure
8.Spee
d res
pons
e
at 4
00r
pm
Figure
9. S
pee
d respo
ns
e at
900r
pm
Figure
10.S
pee
d respo
ns
e at
1400
rp
m
w
it
h
cl
os
e
up
view
Figure
11
s
hows
t
he
ph
ase
A
ou
t
pu
t
c
urre
nt
f
or
the
both
co
ntr
ollers
at
1400
rp
m
with
m
axi
m
u
m
current
at
sta
rting
10A.
In
a
ddit
ion
,
the
tor
qu
e
ou
tp
ut
of
TS
and
S
T
-
T
S
are
com
par
ed
an
d
pr
ese
nted
i
n
Figure
12,
in
wh
ic
h
t
he
ST
-
TS
is
be
tt
er
than
T
S
with
le
sser
t
orqu
e
rip
ple.
B
ot
h
tor
que
an
d
current
perf
orm
ance
sh
ow
n
ty
pical
correla
ti
on
be
tween
t
orq
ue
,
curre
nt
and
sp
eed
ba
hav
i
our
of
m
oto
r
dr
ive
syst
em
.
Th
us,
validat
ed
the
s
i
m
ulati
on
syst
e
m
dev
el
ope
d.The
perf
or
m
ance
analy
sis
f
or
the
tw
o
al
gori
thm
s
in
te
rm
s
of
ris
e
tim
e,
set
tling
t
i
m
e
and
ov
e
rs
hoot
f
or
the
s
peed
res
pons
e
in
re
ve
rse
a
nd
f
orwa
rd
ope
ra
ti
on
is
prese
nted
i
n
Table
2.
T
he
s
peed
pe
rfo
rm
a
nce
a
naly
sis
shows
that,
ST
-
T
S
is
s
up
e
rio
r
i
n
te
rm
of
rise
t
i
m
e,
set
tl
ing
ti
m
e
as
well
as
pe
rcen
t
ov
e
rs
hoot.
T
he
ST
-
T
S
over
s
hoot
with
only
15
r
pm
beyond
the
ref
e
ren
ce
wh
il
e
TS
over
sh
oot
with
49 rpm
b
ey
ond
the
re
fer
e
nce.
Figure
11. O
utp
ut
of
ph
ase
a
current
Figure
12. T
or
qu
e
outp
ut
0
2
4
6
8
10
-
5
0
0
0
500
T
i
m
e
(
s
e
co
n
d
s
)
RP
M
S
p
e
e
d
T
S
-
T
u
n
n
e
d
TS
Re
f
0
.
9
1
1
.
1
1
.
2
250
300
350
400
450
T
i
m
e
4
.
9
5
5
.
1
5
.
2
-
4
0
0
-
3
0
0
-
2
0
0
T
i
m
e
0
2
4
6
8
10
-
1
0
0
0
-
5
0
0
0
500
1000
T
i
m
e
(
s
e
co
n
d
s
)
S
p
e
e
d
RP
M
T
S
-
t
u
n
e
d
TS
Re
f
1
1
.
2
700
800
900
T
i
m
e
4
.
9
5
5
.
1
5
.
2
-
9
0
0
-
8
0
0
-
7
0
0
T
i
m
e
0
2
4
6
8
10
-
1
5
0
0
-
1
0
0
0
-
5
0
0
0
500
1000
1500
T
i
m
e
(
s
e
co
n
d
s
)
S
p
e
e
d
RP
M
T
S
-
T
u
n
e
d
TS
Re
f
0
.
9
1
1
.
1
1
.
2
1
.
3
1100
1200
1300
1400
T
i
m
e
5
5
.
5
-
1
4
0
0
-
1
2
0
0
-
1
0
0
0
T
i
m
e
0
.
9
5
1
1
.
0
5
1
.
1
1
.
1
5
1
.
2
-
1
0
-5
0
5
10
T
i
m
e
(
s
e
co
n
d
s
)
A
m
p
li
t
u
d
e
Cu
r
r
e
n
t
T
S
-
T
u
n
e
d
TS
0
2
4
6
8
10
-
4
0
-
2
0
0
20
40
T
i
m
e
(
s
e
co
n
d
s
)
A
m
p
li
t
u
d
e
T
o
r
q
u
e
TS
T
S
-
T
u
n
e
d
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8
694
I
nt J
P
ow
Ele
c
&
Dri
Syst
, Vol.
9
, N
o.
4
,
D
ece
m
ber
2018
:
1967
–
1975
1974
4.2.
L
oaded
O
pera
tion
A
loa
d
wa
s
ap
plied
to
t
he
syst
e
m
to
ob
se
rve
the
reli
abili
ty
of
the
c
ontr
ollers
an
d
t
heir
sta
bili
ty
to
changes
in
te
r
m
s
of
sp
ee
d,
c
urren
ts
as
well
as
torque
pro
du
ce
d.
T
he
s
pe
ed
res
pons
e
a
t
1400
rp
m
with
load
app
li
ed
is
pr
es
ented
i
n
Fi
gure
13.
Figure
13. Spe
ed resp
onse at
1400r
pm
w
it
h l
oad
The
S
T
-
T
S
spe
ed
has
s
peed
dro
p
of
24
r
pm
i
n
0.5
1s
be
fore
return
t
o
it
s
ste
ady
sta
te
co
nd
i
ti
on
,
w
hile
the
T
S
sp
ee
d
dro
pped
is
35
rpm
with
reco
ve
r
y
tim
e
of
0.
56s
.Th
e
ph
ase
A
ou
t
pu
t
cu
rr
e
nts
and
tor
ques
duri
ng
load
a
re
pr
e
sen
te
d
in
Fig
ur
es
14 and
15
respec
ti
vely
.
Figure
14.Cu
rrents re
spo
ns
e
with loa
d
Figure
15.T
orq
ues res
pons
e
w
it
h
load
5.
CONCL
US
I
O
N
In
c
on
cl
us
io
n,
this
pap
e
r
pr
esented
a
c
omparati
ve
st
ud
y
betwee
n
sta
ndar
d
f
uzzy
lo
gi
c
and
Self
-
tun
in
g
fu
zzy
l
og
ic
base
d
on
Taka
gi
-
S
ug
e
no
fu
zzy
inter
fac
e
ty
pe
app
li
ed
to
the
sp
ee
d
con
t
ro
l
of
in
duct
ion
m
oto
r
dr
i
ve.
Taka
gi
-
S
ug
e
no
(TS)
is
fu
zzy
ty
pe
that
util
i
zes
sing
le
to
n
ou
t
pu
t
m
e
m
ber
sh
i
p
f
unct
io
n
an
d
reduces
the
fuzzy
com
pu
ta
tio
nal
burd
e
n.
S
peed,
tor
que
a
nd
c
urre
nts
pe
rfor
m
ance
ha
ve
been
c
om
par
ed
f
or
two
c
ontr
ollers
in
te
rm
s
rise,
set
tl
ing
ti
m
e
an
d
per
ce
nt
ov
e
rs
hoot.
In
al
l
cases,
ST
-
TS
co
ntr
oller
sh
owe
d
su
p
erio
r
pe
rform
ance
du
e
to
i
ts
abili
ty
to
adjust
the
input
sc
al
ing
facto
rs
onli
ne
in
acco
rdance
to
va
riat
ion
s
i
n
sp
ee
d
er
r
or
a
nd c
hange
of
s
pe
ed
e
rror o
f
the
syst
e
m
.
APPE
ND
I
X
A
:
I
NDU
CTIO
N MOTO
R
P
ARA
METE
R
S
Vs
(
rated
)=
380V,
fs(rate
d)
=5
0H
z
,P(p
o
le
s)=4
,
ωr
(r
a
te
d)
=
1400,Rs=
3.45
,
Rr=
3.6
141
,
Ls=
0.3
252H,
Lr=0.3
252H,
L
r=0.3
117H,
J=
0.325
2H
=
0.0
2kgm
^2
ACKN
OWLE
DGME
NTS
The
a
uthor
s
w
ou
l
d
li
ke
t
o
grat
efu
ll
y
ack
nowled
ge
the
f
unding
s
upport
pro
vid
e
d
by
U
T
eM
and
th
e
Mi
nistry of E
ducat
ion
Ma
la
ys
ia
u
nde
r
the
r
es
earch
grant
N
o:
FRGS
/
1/2015
/TK0
4/F
KE/0
2/
F0
02
58
.
0
2
4
6
8
10
-
2
0
0
0
-
1
0
0
0
0
1000
2000
T
i
m
e
(
s
e
co
n
d
s
)
RP
M
S
p
e
e
d
wi
t
h
l
o
a
d
T
S
-
T
u
n
e
d
TS
Re
f
2
.
8
3
3
.
2
1300
1400
1500
T
i
m
e
6
.
9
7
7
.
1
-
1
4
0
0
-
1
2
0
0
T
i
m
e
0
.
9
5
1
1
.
0
5
1
.
1
1
.
1
5
1
.
2
-
1
0
-5
0
5
10
T
i
m
e
(
s
e
co
n
d
s
)
A
m
p
li
t
u
d
e
Cu
r
r
e
n
t
T
S
-
T
u
n
e
d
TS
0
2
4
6
8
10
-
4
0
-
2
0
0
20
40
T
i
m
e
(
s
e
co
n
d
s
)
A
m
p
li
t
u
d
e
T
o
r
q
u
e
r
e
s
p
o
n
s
e
T
S
-
T
u
n
e
d
TS
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
P
ow
Ele
c
&
D
ri
Syst
IS
S
N: 20
88
-
8
694
Self
-
tu
ning F
uzzy
Logic C
on
tr
oller Ba
sed
on Tak
ag
i
-
Su
geno
A
ppli
ed
t
o Ind
uction M
oto
r…
(
Nab
il
F
ar
ah
)
1975
REFERE
NCE
S
[1]
Uddin,
M.
Nasir,
Ta
wfik
S.
Rad
wan,
and
M.
Az
iz
ur
Rahman.
"P
erf
orm
anc
es
of
f
uzzy
-
log
ic
-
bas
e
d
indi
rect
vector
cont
rol
for induc
ti
on
m
otor
dr
ive.
" IE
EE
Tr
ansa
c
tions
on
Industr
y
Applic
a
ti
ons 38.5 (2002):
1219
-
1
225.
[2]
Barr
ero
,
F.,
et
al
.
"S
pe
ed
con
t
rol
of
indu
ct
io
n
m
otors
using
a
nove
l
fuz
z
y
slidi
ng
-
m
ode
s
truc
tur
e.
"
I
EE
E
Tra
nsac
ti
ons on
Fuzz
y
S
y
st
ems
10.
3
(2002):
375
-
383.
[3]
Ab
Ghani,
Mohd
Ruddin,
et
al.
"Invest
iga
ti
on
Stud
y
of
Three
-
Le
ve
l
Casca
d
ed
H
-
bridge
Multi
le
ve
l
Inve
rter."
Te
lkomnika
15.
1
(2017):
125
.
[4]
Ibra
him,
Zul
k
ifilie,
and
Emil
Le
vi
.
"A
compara
t
ive
an
aly
sis
of
fuz
z
y
logic
and
PI
spee
d
cont
rol
in
h
ig
h
-
per
form
anc
e
AC
drive
s
using
ex
per
imental
app
r
oac
h.
"
IEEE
T
r
ansa
ctions
on
In
dustr
y
Appl
ications
38.
5
(2002)
:
1210
-
1218.
[5]
Sahu,
Benudha
r
,
K.
B.
Mohant
y
,
and
Sw
aga
t
Pati
.
"A
compara
tive
stud
y
on
fuz
z
y
and
PI
spee
d
cont
rollers
for
fie
ld
-
or
ie
nt
ed
i
nduct
ion
m
otor
drive."
Moder
n
El
e
ct
r
ic
Pow
er
S
y
st
ems
(MEPS
),
2010
Proce
ed
ings
of
t
he
Inte
rna
ti
ona
l
S
ym
posium
.
IEE
E, 2010.
[6]
Yongcha
ng,
Zh
ang,
and
Zha
o
Z
hengming.
"Com
par
at
ive
stud
y
of
PI,
slidi
ng
m
ode
and
fuz
z
y
l
ogic
cont
ro
ller
f
or
rotor
fi
el
d
or
ie
n
te
d
cont
rol
le
d
i
nduct
ion
m
otor
drive
s."
E
lectr
i
c
al
Ma
chi
nes
an
d
S
y
stems
,
2
00
8.
ICEMS
2008
.
Inte
rna
ti
ona
l
Co
nfe
ren
c
e
on
.
I
EEE,
2008
.
[7]
Bai
,
Ying,
and
Dali
W
ang.
"F
undamenta
ls
of
fuz
z
y
logic
cont
r
ol
—
fuz
z
y
sets,
f
uzzy
ru
le
s
and
def
uzzifi
c
ations.
"
Advanc
ed
Fuz
z
y
Logic
Te
chno
lo
gie
s in
Industri
al Appli
c
at
ions.
S
pringe
r London,
2006.
17
-
36
.
[8]
Mam
dani
,
Ebr
ah
im
H.,
and
Sed
r
ak
As
sili
an
.
"A
n
expe
rimen
t
in
linguis
ti
c
s
y
nth
esi
s
with
a
fu
zzy
l
ogic
con
trol
l
er.
"
Inte
rna
ti
ona
l jou
rna
l
of
m
an
-
m
achine
stud
ie
s 7
.
1
(1975):
1
-
13
.
[9]
Ta
kag
i,
Tomohi
ro,
and
Michi
o
Sugeno.
"F
uzzy
ide
nti
fi
cation
o
f
sy
st
ems
and
i
ts
appl
icati
ons
t
o
m
odel
ing
and
cont
rol
.
" IEEE
tr
ansa
ctions on
s
ystems
,
m
an,
and c
y
b
ern
e
ti
cs
1
(1
985):
116
-
132.
[10]
Fara
h,
N.
S.
Y.,
et
al.
"Expe
r
imenta
l
inv
esti
ga
ti
o
n
of
diffe
ren
t
ru
l
es
size
of
fuz
z
y
logi
c
con
trol
l
er
f
or
vec
tor
con
trol
of
induction
m
ot
or
driv
es.
"
Journal
of
Fundam
ental and
Appl
ie
d
S
ci
en
ce
s 10
.
6S (2
018):
1696
-
1717
.
[11]
Ta
li
b,
M.
H.
N.
,
et
al.
"S
implif
ied
self
-
tun
ing
Fu
zzy
Logi
c
Speed
cont
ro
ll
er
for
in
duct
ion
m
otor
dr
ive
.
"
Pow
er
and
Ene
rg
y
(PECon)
,
2016
IEEE
In
ternat
ion
al
Conf
er
enc
e
on.
IE
EE
,
2
016.
[12]
Masial
a,
Mavu
ngu,
et
al.
"F
uzzy
self
-
tuni
ng
spee
d
cont
ro
l
of
an
indi
re
ct
field
-
orie
nt
ed
cont
ro
l
induc
ti
on
m
ot
or
drive
.
"
IEEE Tra
nsac
ti
ons on
Ind
ustr
y
Applicati
o
ns 44.
6
(2008)
: 1732
-
1740.
[13]
Kus
agur
,
As
hok,
S.
F.
Kodad,
and
BV
Sankar
Ram.
"Regul
ar
pape
r
Nov
e
l
d
esign
of
a
T
akagi
-
Sugeno
fuz
z
y
strat
eg
y
for indu
ct
ion
m
otor
spee
d
cont
ro
l.
" J
.
E
lectr
i
ca
l
S
y
st
ems
6
(2010).
[14]
Lope
z
-
Tor
ibi
o,
C.
J.,
R.
J.
Pat
to
n,
and
S.
Daley
.
"A
m
ult
ipl
e
m
od
el
appr
oa
ch
to
fa
ult
-
tolera
n
t
cont
r
ol
using
ta
kagi
-
sugeno
fuz
z
y
m
odel
li
ng
:
re
al
ap
pli
c
at
ion
to
an
i
nduct
ion
m
otor
drive
s
y
st
em."
Control
Confer
e
nce
(ECC)
,
199
9
Europe
an
.
I
EE
E
,
1999.
[15]
Moez
,
Allouc
h
e,
et
a
l.
"Ta
k
agi
-
Sugeno
Fuzz
y
Control
of
Induc
ti
on
Motor.
"
Proc.
Int.
Journa
l
of
El
ectrical
an
d
El
e
ct
roni
cs
Eng
g
2.
1
(2009).
[16]
Shaukat
,
Ne
e
lo
far
,
et
al.
"
Ta
k
agi
-
Sugeno
Fuz
z
y
log
ic
b
ase
d
spee
d
con
trol
o
f
Induc
t
ion
Mot
or.
"
Frontiers
o
f
Inform
at
ion
T
echnolog
y
(FIT), 2
016
Inte
rn
at
ion
a
l
Confer
ence
on
.
IEEE, 2016.
[17]
Pati
l,
Neha
K.
,
and
Bhagsen
J.
Parva
t.
"S
elf
Tu
ning
Mec
han
ism
using
Input
Scal
ing
Fa
c
tors
of
PI
li
ke
Fuz
z
y
Control
le
r
for
I
m
prove
d
Proce
ss
Perform
anc
e.
"
Inte
rna
ti
ona
l
Jo
urna
l
of
Adv
anced
Rese
arc
h
in
Com
pute
r
Scie
n
ce
and
E
le
c
tronics
Engi
ne
eri
ng
(
IJA
RCS
EE
)
2.
11
(
2013):
pp
-
729.
[18]
Bric
e
,
C.
W
.
,
et
al
.
"Refe
ren
ce
f
rames
for
si
m
ula
ti
on
of
el
e
ct
r
ic
m
otors
and
driv
es.
"
Proce
edi
ngs
of
Inte
rna
ti
on
al
Confer
ence
on
E
le
c
tri
c
Mac
h
ine
s
.
2000
.
[19]
Ab
Ghani,
Mohd
Ruddin,
Nab
il
Fara
h,
and
M.
R.
Ta
m
ji
s
.
"F
ie
l
d
orie
nt
ed
con
tr
ol
of
6/4
SR
M
f
or
torque
r
ippl
e
m
ini
m
ia
zt
ion
.
"
El
e
ct
ri
ca
l
,
E
le
c
t
ronic
s,
and
Opt
imiza
ti
on
Te
chn
ique
s
(ICEE
OT
),
Inte
rn
at
ion
al
Confer
ence
on.
IEE
E
,
2016
.
[20]
Ta
li
b,
Md
Hair
ul
Niza
m
,
et
al.
"Com
par
ison
Anal
y
sis
of
Indi
rec
t
FO
C
Induc
t
ion
Motor
Driv
e
using
PI,
Ant
i
-
W
indup
and
Pre
Filt
er
Schemes."
Inte
rn
at
ion
al Journal
of
Pow
er
E
le
c
troni
cs
and
D
rive
S
y
ste
m
s 5.
2
(2014):
219
.
[21]
Ta
li
b
,
M.
H.
N.,
et
al.
"H
y
ster
esis
cur
ren
t
con
trol
of
induc
ti
on
m
ot
or
drive
s
using
dSP
ACE
D
SP
co
ntrol
ler."
Pow
er
and
En
erg
y
(PE
Con),
2016
IEEE
Int
ern
ationa
l Confere
nc
e
on
.
I
EE
E
,
2016
.
[22]
Gae
id,
Khal
af
Sall
oum
,
Hew
W
ooi
Ping,
and
H
ai
der
A
F
Moham
ed.
"S
imulink
rep
rese
nt
at
ion
o
f
induc
ti
on
m
oto
r
ref
ere
n
ce fra
m
es
.
" Tec
hni
cal
Pos
t
gra
duates (TEC
HP
OS
),
2009
Inte
rna
t
iona
l
Confe
ren
ce for. I
E
EE, 2009
[23]
Fara
h,
Nabil
,
e
t
al
.
"M
ult
i
le
ve
l
Inve
rte
r
Fed
Sw
it
che
d
Rel
u
c
ta
nc
e
Motors
(SRM
s):
6/4,
8/6
and
10/8
S
R
M
Geom
et
ric
T
y
pe
s."
Inte
rn
at
ion
al
Journal
of
Pow
e
r
Elec
tron
ic
s
and
Drive
S
y
stems
(
IJP
EDS) 8.
2
(20
17).
Evaluation Warning : The document was created with Spire.PDF for Python.