Int
ern
at
i
onal
Journ
al of
P
ower E
le
ctr
on
i
cs a
n
d
Drive
S
ystem
s
(
IJ
PEDS
)
Vo
l.
1
2
,
No.
1
,
M
a
r
202
1
, p
p.
20
~
28
IS
S
N:
20
88
-
8694
,
DOI: 10
.11
591/
ij
peds
.
v
1
2
.i
1
.
pp
20
-
28
20
Journ
al h
om
e
page
:
http:
//
ij
pe
ds
.i
aescore.c
om
Smart in
tegrati
on of
drive s
ystem
for ind
uction m
oto
r
appli
ca
ti
ons in
elec
t
ric
vehicles
Moham
ed
K. Met
w
aly
1
, Mo
ha
mm
ad Alsh
aref
2
,
N
e
hmd
oh
A. S
ab
ih
a
3
,
Eha
b
E. El
att
ar
4
,
Ibra
him
B.
M
. Ta
h
a
5
,
A
m
r
M.
Abd
-
El
h
ady
6
, Nag
y
I.
E
lkalashy
7
1,2,3,4,5,7
El
e
ct
ri
ca
l
Engi
n
ee
ring
De
par
tment, Colleg
e
of Engin
ee
ring
,
T
ai
f
Univer
sity
,
T
ai
f
,
Saud
i
Ara
bia
6
El
e
ct
ri
ca
l
Eng
in
ee
ring
Depa
r
tm
e
nt,
Fa
cul
ty
of En
gine
er
ing, Meno
ufia
Univ
ersit
y
,
Shebin
E
lkom
,
E
gypt
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
hist
or
y:
Re
cei
ved
Sep
22, 202
0
Re
vised
Jan
15
, 202
1
Accepte
d
Fe
b
5,
202
1
In
thi
s
pap
er,
a
smart
driv
e
sys
t
em
of
the
induc
t
ion
mot
or
(IM)
is
proposed
and
ad
apted
for
appl
i
ca
t
ions
in
e
le
c
tri
c
veh
ic
l
es
(
EVs).
Obje
ct
iv
ely,
th
e
EV
drive
sys
tems
ar
e
robust
ov
er
w
ide
spe
ed
and
t
orque
ran
g
es.
T
he
proposed
drive
sys
tem
is
i
ndepe
nden
t
of
e
ncode
r
(en
cod
erless
)
and
concern
ed
with
th
e
torque
cont
ro
l
dr
ive
(TCD)
and
i
ndire
c
t
ro
tor
fiel
d
-
orie
nt
ed
cont
r
ol
(IRFO
C)
using
the
sl
idi
n
g
mode
observe
r
(SM
O).
Thi
s
arr
a
nge
me
nt
of
moni
tori
ng
sys
te
m
and
con
t
rol
t
ec
hniqu
es
ar
e
smar
tl
y
integr
at
ed
for
the
IM
appl
i
ca
t
ions
in
EVs.
Th
e
e
ncode
rl
ess
tech
nique
u
ti
l
izes
SM
O
to
est
im
a
t
e
th
e
st
at
or
cur
ren
t
,
rotor
fl
ux
ang
le
,
and
ro
tor
spe
ed.
Th
e
S
MO
is
ver
ifi
ed
i
n
mot
or
in
g
mode
at
ver
y
lo
w
and
ze
ro
spee
d
conditions.
Th
e
a
ccel
er
at
or
ped
al
is
utilized
for
TCD
to
g
enerat
e
the
ref
ere
n
c
e
torque
r
equi
re
d
to
acce
l
erate
t
he
EV
by
the
drive
r.
The
rot
or
flux
angle
is
est
im
a
te
d
b
ase
d
on
IRFO
C
m
et
hod.
The
la
bora
tory
wav
e
forms
illus
tra
t
e
t
he
robust
ness
of
the
en
code
rl
ess
cont
ro
l
of
the
IM
-
base
d
tor
que
cont
ro
l
driv
e
sys
te
m
in
el
e
ctrical
vehicle
app
li
c
at
ions
a
t
ver
y
low
spe
ed
using
SM
O.
Th
e
l
abor
at
ory
wav
efo
rms
prov
e
th
e
v
al
id
it
y
of
SM
O
with
en
co
der
le
ss
cont
rol
of
a
smar
t
dive
sys
te
m
of
the
IM
in
EV
a
pplications und
er
lo
ad torque
/sp
ee
d
v
ariati
ons.
Ke
yw
or
d
s
:
Ele
ct
ric v
e
hicle
Enc
od
e
rless c
ontr
ol
Ind
uction m
otor
Torq
ue
c
on
t
ro
l
drive
This
is an
open
acc
ess arti
cl
e
un
der
the
CC
BY
-
SA
l
ic
ense
.
Corres
pond
in
g
Aut
h
or
:
M
oha
med
K.
M
et
wal
y
Ele
ct
rical
En
gi
neer
i
ng D
e
par
t
ment, C
ollege
of Enginee
rin
g
Tai
f
U
niv
e
rsity
, 21974
Tai
f,
S
aud
i
A
rab
ia
Emai
l:
m.met
wall
y@
t
u.
e
du.
sa
1.
INTROD
U
CTION
Du
e
to
th
e
i
ncrea
sing dema
nd
f
or
s
mart
ci
ti
es
a
nd
s
us
ta
ina
bl
e
el
ect
rificat
ion
s
us
i
ng r
ene
w
able p
owe
r
gen
e
rati
ons,
th
e
recent
re
sear
ch
tre
nd
s
a
re
pro
po
se
d
to
i
nc
rease
the
penet
rati
on
of
el
e
ct
ric
veh
ic
le
s
(EV
s
).
Howe
ver,
t
he
dr
i
ve
s
ys
te
m
of
E
V
nee
ds
se
ns
ors
f
or
t
he
r
otor
s
pee
d
me
asur
e
ments
.
T
hese
sen
sors
a
re
not
adap
te
d
with
ha
rsh
en
vir
onm
ents
ch
an
ges
s
uch
as
mec
ha
ni
cal
vib
rati
ons,
te
mp
e
ratur
e
c
hanges
,
an
d
du
st.
To
so
lve
these
pro
blems,
the
devel
op
me
nts
of
e
ncode
rless
of
I
nductio
n
Moto
r
(IM)
dr
i
ves
ha
ve
bee
n
i
ncr
e
ased,
recently
.
E
ncoder
le
ss
s
pee
d
e
sti
mati
on
te
ch
niques
a
re
util
iz
ed
in
hosti
le
env
i
ronme
nts
and
al
s
o
for
a
bnormal
conditi
ons in
s
afety
-
c
riti
cal
ap
plica
ti
ons in
c
ase o
f dama
ge of
sp
ee
d sens
or [1
],
[2].
The
I
M
s
ha
ve
been
good
so
l
ut
ion
s
in
the
fiel
d
of
E
V
dr
ive
sy
ste
ms
,
beca
use
the
y
ha
ve
m
any
me
rits
su
c
h
as
r
obus
t
ness,
ri
gid
it
y,
s
imple
co
ns
tr
uc
ti
on
,
le
ss
co
st,
and
le
ss
mai
nt
enan
ce
.
F
urt
he
rm
or
e,
the
I
M
dr
i
ves
hav
e
be
en
util
iz
ed
in
ma
ny
po
sit
io
n
t
rack
i
ng
ap
plica
ti
ons,
wh
ic
h
need
preci
se
c
on
t
r
ol
act
io
n
f
or
e
xam
ple
act
uation,
r
obot
ic
s,
automatio
n
proce
ss,
a
nd
gu
i
ded
process
.
These
merit
s
make
the
I
M
s
a
very
go
od
s
ol
utio
n
for
i
ndus
tria
l
app
li
cat
io
ns
.
T
hu
s
,
preci
se
posit
ion
an
d
spe
ed
c
ontrols
a
re
ma
ndat
ory
in
E
V
dr
i
ve
s
ys
te
ms
durin
g
l
oad torqu
e/
s
pee
d dist
urba
nc
es
[1].
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
P
ow Elec
& Dri S
ys
t
IS
S
N:
20
88
-
8
694
Smart i
ntegr
ation of
dr
iv
e sy
ste
m
fo
r in
duct
io
n mo
t
or
applic
ations in
elec
tri
c
…
(
M
ohame
d
K.
Metw
aly
)
21
Sli
din
g
mode
ob
se
r
ver
(
S
M
O
)
is
one
of
the
models
c
on
ce
rn
i
ng
e
ncode
rless
sp
e
ed
est
imat
ion
te
chn
iq
ues
f
or
the
m
otor
dri
ve
s
ys
te
ms.
It
c
an
r
obus
tl
y
an
d
e
ff
ic
ie
ntl
y
es
ti
mate
the
mo
t
or
s
peed,
r
oto
r
fl
ux
,
ro
t
or
posit
io
n,
an
d
m
oto
r
to
rque,
w
he
re
only
one
obse
r
ver
gai
n
is
ne
eded
[
3].
Als
o,
a
go
od,
sta
ble,
a
nd
rob
us
tness
performa
nce
agai
ns
t
the
pa
ram
et
ers’
se
ns
it
ivit
y
can
be
obta
ined
ov
e
r
wi
de
sp
ee
d
ra
ng
e
of
op
e
rati
on,
as
well
as
ease
of
im
plementa
ti
on
[
4
]
-
[
5].
T
o
kee
p
the
dy
namic
sta
te
s
of
operati
ons
f
or
th
e
est
imat
ed
va
ri
ables,
ti
me
-
va
r
ying
gains
f
or
the
s
witc
hing
f
un
ct
io
ns
of
t
he
obser
ver
ar
e
us
e
d
[6].
H
oweve
r,
chatt
erin
g
is
th
e
main
disad
va
ntage
of
the
S
M
O
va
riable
st
ru
ct
ur
e
s
ys
te
m
.
This
sy
ste
m
chatt
erin
g
is
r
e
du
ce
d
us
in
g
a
sigm
oi
d
f
unct
io
n
a
nd
the
ob
s
er
v
er
is
co
ns
tr
ucted
ba
sed
on
a
bac
k
el
ect
ro
-
mo
ti
ve
force
(E
M
F
)
model
for
im
pro
ving t
he
e
ff
ect
of est
imat
ion
[7
]
-
[
9].
In
[10],
a
no
nl
inear
S
M
O
sc
heme
is
pr
opose
d
f
or
i
nterio
r
pe
rma
nen
t
m
agn
et
s
ynch
ronous
m
oto
r
(I
P
M
S
M)
c
on
siderin
g
ma
ximu
m
to
r
qu
e
per
am
per
e
(
M
TP
A),
t
he
SMO
is
util
iz
ed
as
s
pee
d
e
sti
mato
r
te
chn
iq
ue.
Lya
punov
sta
bili
ty
analysis
is
ach
ie
ved
for
both
the
co
ntr
oller
and
obser
ver.
A
second
-
ord
er
S
M
O
is
us
e
d
to
est
imat
e
the
ro
t
or
sp
ee
d
based
on
a
super
twist
ing
al
gorithm
and
m
od
el
re
fe
ren
ce
ada
ptive
sy
ste
m
cal
culat
ing
pr
i
nciples
[
11]
.
A
co
mp
a
rati
ve
study
betwee
n
f
ull
orde
r
S
M
O
an
d
im
pro
ved
i
nteg
rati
on
for
the
vo
lt
age
m
odel
(VM)
is
pr
ese
nted
i
n
[12].
I
n
that
stu
dy,
t
he
dc
offset
is
est
imat
ed
a
nd
ke
pt
by
pro
po
rtion
al
plu
s
inte
gr
al
(
PI
)
c
ompe
ns
at
or,
i
n
w
hi
ch
al
l
un
wan
te
d
off
set
s
a
nd
dr
ifts
are
c
ompe
ns
at
ed
in
the
ac
qu
i
sit
ion
te
rmin
al
s.
H
oweve
r,
due
to
t
he
inc
orrect
observ
e
r
var
ia
bles,
a
ph
ase
sh
i
ft
is
obser
ve
d
in
qu
a
si
-
S
M
O
(
Q
SMO)
or
discrete
-
ti
m
e
S
M
O
durin
g
the
loa
d
cha
nges
[
13].
T
oward
s
olv
in
g
thi
s
pr
ob
le
m
,
a
n
adap
ti
ve
QSM
O
is
pr
ese
nted
to
ca
lc
ulate
the
e
xt
end
e
d
E
M
F
c
omp
on
e
nts
f
or
t
he
IPMS
M
,
in
wh
ic
h
a
re
util
iz
ed
t
o
determi
ne
t
he
ro
t
or
an
gle
f
or
the
I
PMSM
.
Fo
r
rob
us
t
e
nc
od
e
rless
co
ntr
ol,
a
n
Iterati
ve
S
M
O
(I
S
MO)
is
pro
po
s
ed
i
n
[14
].
The IS
M
O
e
nhances t
he
est
i
m
at
ed
s
peed pe
rformance
by mi
nimizi
ng the
bac
k
E
M
F
er
ror.
In
[15],
a
qua
drat
ur
e
phase
-
l
oc
ked
l
oop
(
QPLL)
-
base
d
on
high
orde
r
S
MO
f
or
IPMS
M
with
MTP
A
is
pr
ese
nted
to
e
nh
a
nce
t
he
enc
od
e
rless
posit
ion
c
ontr
ol
of
IPM
SM.
The
QP
LL
cal
culat
ion
meth
od
has
immu
nity
a
gai
ns
t
noise
a
nd
distor
ti
on
e
ff
e
ct
s.
Im
pro
ving
the
est
imat
ed
ro
t
or
flu
x
a
ng
le
at
very
l
ow
sp
ee
d
op
e
rati
on
base
d
on
S
M
O
is
pr
ese
nted
in
[
16]
.
T
he
pro
po
s
ed
te
ch
ni
qu
e
does
not
nee
d
a
ny
s
peed
ada
pt
at
ion
al
gorithm
an
d
has
i
mm
un
it
y
to
the
dev
ia
ti
on
of
s
pee
d
es
ti
mati
on
.
A
pre
dicti
ve
S
MO
of
th
e
sta
to
r
flu
x
at
a
wide
ra
ng
e
of
op
e
rati
ng
s
pee
d,
t
he
sal
ie
nc
y
of
the
machi
ne
due
t
o
it
is
sat
ur
at
io
n
e
ff
ect
i
s
util
iz
ed
in
[
17]
.
A
change
ov
e
r
al
gorithm
is
program
med
to
com
bin
e
al
l
modes
of
op
e
rati
on
s
.
Sli
ding
mode
reac
hi
ng
la
w
(SMRL)
is
pro
po
s
ed
i
n
[18]
to
a
dap
t
t
he
e
nc
od
e
rless
s
pee
d
c
on
t
ro
l
oper
at
ion
of
t
he
IPMS
M
unde
r
diff
e
rent
disturba
nces
a
nd
un
ce
rtai
nties.
T
he
S
M
RL
ha
s
a
feat
ure
of
ad
aptat
ion
acc
ordin
g
to
the
c
on
tr
ol
syst
em
var
ia
ti
ons,
as
resu
lt
s
i
n
a
re
du
ct
io
n
of
t
he
chatt
eri
ng
pr
ob
le
m
,
a
re
do
ne
on
the
c
ontrol
in
put,
w
hile
the
trackin
g o
per
at
ion
of the
contr
oller is mai
ntained
unc
hange
d.
Nonsing
ular
te
rmin
al
sli
ding
mode
(
NTSM
)
is
integrated
with
hi
gh
orde
r
sli
ding
mode
(HOS
M
)
for
est
imat
ing
the
IPMS
M
s
pee
d
an
d
it
s
po
si
ti
on
[
19].
To
el
imi
nate
the
chatt
erin
g
an
d
to
ens
ur
e
ob
serv
e
r
sta
bili
ty,
t
he
H
OSM
co
ntr
ol
l
aw
is
desig
ne
d.
In
[20],
a
sch
eme
of
t
he
e
nc
od
e
rless
s
peed
est
imat
io
n
at
a
very
low
ope
rati
ng
sp
ee
d
f
or
the
I
M
PS
M
util
iz
ing
S
M
O
is
in
ve
sti
gated.
T
he
de
la
y
ti
me
du
e
t
o
t
he
lo
w
pas
s
filt
er
or
t
he
chatt
eri
ng
pro
blem
is
so
lve
d
us
in
g
a
sigm
oid
f
unct
ion
for
the
si
gn
or
sat
urat
io
n
functi
on.
T
wo
SMOs
with
co
mpo
und
ma
nifo
l
ds
f
or
the
I
M
s
peed/
flux
obser
vation
is
pr
ese
nted
in
[21].
Th
e
encode
rless
co
nt
ro
l
o
f
the IM
-
base
d mo
del
re
fer
e
nc
e ada
ptive s
ys
t
em (MR
AS
)
fo
r
E
V
a
pp
li
cat
io
n
is
pr
e
sente
d
i
n [22].
In
[
23
]
-
[25
],
a
n
a
dap
ti
ve
an
d
rob
us
t
posit
io
n
co
ntr
ol
f
or
t
he
I
M
is
propose
d.
T
he
switc
hi
ng
gai
ns
a
re
sel
ect
ed
co
ns
id
erin
g
the
s
ys
te
m
uncertai
ntie
s.
Als
o,
the
sli
p
sp
ee
d
is
use
d
to
est
imat
e
the
ro
to
r
fl
ux
a
ngle
.
As
early
mentio
ne
d,
t
he
mai
n
disadv
a
ntage
of
S
M
O
is
the
c
hat
te
ring
pe
rform
ance
because
t
he
os
ci
ll
at
ion
s
occur
with
high
f
re
quencies
wh
e
n
sli
din
g
m
ode
t
akes
place.
T
he
s
olu
ti
on
of
SMO
pro
blem
s
is
t
o
el
imi
na
te
the
chatt
erin
g
a
nd
enh
a
nce
the
s
pe
ed
est
imat
ion
accurac
y,
es
pe
ci
al
ly
in
low
-
sp
ee
d
re
gion.
Howe
ver,
the
EV
can
be
dr
ive
n
from
zero
sp
ee
d
up
to
it
s
li
mit
sp
eed.
T
o
fin
d
a
s
peed
co
ntr
ol
unive
rsall
y
ov
e
r
a
ve
ry
wide
ra
ng
e
o
f
sp
ee
d
a
nd in
de
pende
nt of spe
ed
e
ncode
r
is a
ch
al
le
n
ge fo
r
t
he
E
V
a
ppli
cat
ion
s
.
In this
pap
e
r,
main c
on
t
rib
ution
s
are
s
um
ma
rized
by:
•
In
te
gr
at
in
g
a
s
mart
dr
ive
s
ys
te
m of
I
M
.
•
Util
iz
ing
t
he
SMO
of
s
pee
d
est
imat
ion
for
t
he
e
nc
od
e
rl
ess
s
pee
d
-
c
on
t
ro
ll
ed
I
M
with
t
orq
ue
c
ontr
ol
dr
i
ves
s
ys
te
ms
in E
V
a
pp
li
cat
ion
s
.
•
Im
plementin
g
the
in
direct
r
oto
r
fiel
d
-
ori
ente
d
co
ntr
ol
(
IRF
OC)
for
the
e
n
c
o
d
e
r
l
e
s
s
s
p
e
e
d
c
o
n
t
r
o
l
o
f
t
h
e
I
M
d
r
i
v
e
s
y
s
t
e
m
s
i
n
E
V
s
i
n
t
h
e
l
a
b
o
r
a
t
o
r
y
u
s
i
n
g
a
d
i
g
i
t
a
l
s
i
g
n
a
l
p
r
o
c
e
s
s
i
n
g
D
S
p
a
c
e
-
D
S
1
1
0
3
c
o
n
t
r
o
l
boar
d.
•
Evaluate t
he
la
borato
ry w
a
ve
f
orms
unde
r
e
xtreme
distu
rb
a
nc
es of the
loa
d t
orqu
e/
s
peed.
•
Evaluate
the
e
ff
ect
ive
ness
of
the
pro
po
se
d
smart
enc
oderl
ess
IM
dr
i
ve
s
ys
te
m
at
very
low
a
nd
ze
ro
mecha
nical
fre
qu
e
nc
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8
694
In
t J
P
ow
Ele
c
&
Dr
i
S
ys
t
,
V
ol
.
12
, N
o.
1
,
Ma
rch
202
1
:
20
–
28
22
2.
MO
DEL DE
S
CR
I
PTIO
N
O
F IM
The
m
odel
of
IM
i
n
the
sta
ti
on
a
r
y
re
fer
e
nc
e
fr
a
me
ta
ki
ng
the
sta
tor
cu
rrent
an
d
ro
t
or
f
lux
as
sta
te
var
ia
ble is re
pr
esented
by
(1) [2
6
]:
s
s
s
r
s
s
s
r
s
s
u
B
i
A
A
A
A
i
dt
d
+
=
22
21
12
11
(1)
wh
e
re,
s
s
i
is
the
sta
tor
c
urre
nt,
s
r
is
the
r
oto
r
f
lux
,
s
s
u
is
the
sta
tor
vo
lt
age
,
aI
A
=
11
,
J
cT
cI
A
r
r
−
=
12
,
dI
A
=
21
,
J
I
T
A
r
r
+
−
=
1
22
,
T
b
B
0
=
;
+
−
=
r
r
s
m
S
S
T
L
L
L
L
R
a
2
,
s
L
b
1
=
,
r
T
c
1
=
,
r
m
T
L
d
=
,
r
s
m
L
L
L
2
1
−
=
,
r
r
r
R
L
T
=
,
m
r
s
L
L
L
=
;
=
1
0
0
1
I
,
−
=
0
1
1
0
J
R
s
an
d
R
r
a
re
the
sta
to
r
a
nd
r
otor
resist
ances,
res
pecti
ve
ly.
L
s
a
nd
L
r
are
the
sta
tor
an
d
r
otor
inducta
nces
,
r
especti
vely
.
L
m
is
the
mag
ne
ti
zi
ng
in
du
ct
a
nce,
a
nd
r
is
the
r
otor
sp
e
ed.
is
the
le
ak
age
coeffic
ie
nt,
a
nd
T
r
is
the
r
otor ti
me c
on
sta
nt
.
The
e
quat
ion o
f
el
ect
romec
ha
nical
torq
ue
is
il
lustrate
d
by
(
2) an
d (3)
:
r
e
r
L
d
T
J
B
T
dt
=
+
+
(2)
(
)
s
qr
s
ds
s
dr
s
qs
r
m
e
I
I
L
L
P
T
−
=
2
2
3
(3)
wh
e
re
P
is
t
he
IM
pole
pair
s,
s
qs
I
,
s
ds
I
,
s
qr
,
an
d
s
dr
are
qua
dr
at
ur
e
-
axis,
di
rect
-
axis
sta
tor
currents
a
nd
r
otor
flu
xes
in stat
ion
a
r
y
re
fer
e
nce
fr
a
me re
sp
ect
ively
.
T
he
rotor an
gle is
co
m
pute
d
a
s
(
4)
:
(4)
3.
SMO FO
R
P
OSITIO
N AND SPEE
D
ES
TIMATI
ONS
The mo
del
pr
e
sent
e
d
i
n
[
2
6
]
i
s
u
t
i
l
i
z
e
d
t
o
d
e
t
e
r
m
i
n
e
t
h
e
r
o
t
o
r
s
p
e
e
d
a
n
d
p
o
s
i
t
i
o
n
a
s
d
e
s
c
r
i
b
e
d
a
s
(5)
.
(5)
wh
e
re
s
ds
i
,
s
ds
i
ˆ
,
s
qs
i
,
s
qs
i
ˆ
are
t
he
act
ual
a
nd
e
sti
mate
d
sta
to
r
currents
in
dire
ct
an
d
qua
dr
at
ur
e
a
xes,
res
pe
ct
ively.
s
dr
ˆ
,
s
qr
ˆ
are
the
est
imat
ed
r
otor
fl
ux
in
di
rect
an
d
quad
ratu
re
a
xes
resp
e
ct
ively.
F
is
the
ta
n
-
si
gmoid
s
witc
hi
ng
functi
on
descr
i
bed by
(
6)
:
(6)
the
est
imat
ed r
otor
po
sit
io
n
is
d
esc
ribe
d by
(
7)
:
(7)
4.
EV
-
I
NCOR
P
ORATE
D
E
N
CODERLE
SS
SPEED
ESTI
MA
TI
ON B
A
SED ON
SMO WIT
H
TC
D
The
sc
hemati
c
diag
ram
of
th
e
encode
rless
sp
ee
d
est
imat
ion
util
iz
ing
S
M
O
f
or
IRFOC
of
the
I
M
base
d
TC
D
i
n
EV
a
ppli
cat
ions
is
s
how
n
i
n
Figure
1.
T
he
t
orq
ue
dema
nd
sign
al
is
inse
rt
ed
i
nto
the
I
M
dri
ve
rr
dt
=
(
)
(
)
=
−
−
−
ˆˆ
ˆˆ
ˆ
s
s
s
s
s
s
r
d
s
d
s
q
r
q
s
q
s
d
r
F
i
i
F
i
i
d
t
2
2
F
(
s
)
=
ta
n
s
ig
(
S)
1
1
S
e
−
=−
+
=
ˆ
ˆ
rr
dt
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
P
ow Elec
& Dri S
ys
t
IS
S
N:
20
88
-
8
694
Smart i
ntegr
ation of
dr
iv
e sy
ste
m
fo
r in
duct
io
n mo
t
or
applic
ations in
elec
tri
c
…
(
M
ohame
d
K.
Metw
aly
)
23
sy
ste
m
us
i
ng
a
ccel
erator
pe
da
l
-
base
d
to
r
que
co
ntr
ol
dri
ve
te
chn
i
que
in
EV
a
ppli
cat
ions
by
t
he
dr
i
v
er
.
T
his
dr
i
ve
s
ys
te
m
f
or
t
he
E
V
is
s
mart
as
it
is
in
dep
e
ndent
of
the
e
ncoders.
A
s
de
picte
d
in
t
he
fi
gure,
t
he
r
efere
nc
e
q
-
a
xis
c
urren
t
com
pone
nt
i
n
s
ynch
ron
ous
ref
e
re
nce
fr
a
me
is
e
valuat
ed
from
t
he
e
le
ct
ro
ma
gn
et
ic
tor
que
form
ula as
(8)
[22]
:
*
2
*
*
5
.
1
d
m
r
q
i
PL
T
L
i
=
(8)
Also
,
the
ref
e
r
ence
d
-
a
xis
cu
rr
e
nt
co
mpo
ne
nt
in
s
yn
c
hro
nous
re
fer
e
nce
fr
ame
is
de
pe
ndant
of
t
he
ref
e
ren
ce
roto
r
f
lu
x
a
mp
li
tu
de
*
r
an
d mag
netiz
ing i
nducta
nce
as
(
9)
:
m
r
d
L
i
*
*
=
(9)
The
d
-
a
xis
an
d
q
-
a
xis
re
fer
e
nc
e
curre
nts
co
mpon
e
nts
in
c
ombinati
on
with
the
est
imat
e
d
r
otor
fl
ux
ang
le
us
i
ng
S
M
O
are
util
iz
e
d
to
gen
e
rate
t
he
ref
e
ren
ce
th
ree
-
ph
a
se
sta
to
r
c
urren
ts
*
a
b
c
i
.
Tw
o
vo
lt
a
ge
a
nd
t
wo
current
tra
ns
du
cers
a
re
util
iz
ed
t
o
s
ense
the
two
-
phase
sta
tor
volt
ages
a
nd
t
wo
-
ph
ase
st
at
or
cu
rr
e
nts
sign
al
s
.
The
thir
d
ph
a
s
e
vo
lt
age
a
nd
current
is
e
valuated
base
d
on
the
three
-
phas
e
balance
c
on
c
ept.
T
he
act
ual
three
-
ph
a
se
m
otor
c
urren
ts
-
ba
sed
trans
ducers
sig
nals
are
co
mpa
red
with
t
he
r
efere
nce
th
ree
-
ph
a
se
m
otor
c
urren
ts
and
the
dev
ia
t
ion
s
betwee
n
t
hese
si
gn
al
s
a
r
e
fed
into
t
he
hy
ste
resis
band
c
urren
t
c
on
t
ro
ll
er
(H
BC
C)
.
The
ou
t
pu
ts
of
the
HBCC
are
t
he
input
gate
pule
s
of
t
he
th
r
ee
-
phase
i
nv
e
r
te
r
us
i
ng
I
GB
T
switc
he
s
.
T
h
e
d
r
i
v
e
s
y
s
t
e
m
i
s
d
r
i
v
e
n
w
i
t
h
s
p
e
e
d
c
o
m
m
a
n
d
s
i
g
n
a
l
o
b
t
a
i
n
e
d
f
r
o
m
p
e
r
m
a
n
e
n
t
m
a
g
n
e
t
s
y
n
c
h
r
o
n
o
u
s
m
o
t
o
r
(
P
M
S
M
)
.
T
h
e
r
e
f
e
r
e
n
c
e
s
p
e
e
d
i
s
t
h
e
o
u
t
p
u
t
o
f
V
/
F
s
p
e
e
d
-
c
o
n
t
r
o
l
l
e
d
i
n
v
e
r
t
e
r
a
n
d
t
h
e
I
M
r
o
t
o
r
s
p
e
e
d
i
s
s
c
a
l
e
d
u
p
/
d
o
w
n
b
a
s
e
d
o
n
t
h
e
g
e
a
r
b
o
x
t
u
r
n
s
r
a
t
i
o
.
T
h
e
w
h
e
e
l
s
p
e
e
d
f
o
r
E
V
d
r
i
v
e
a
p
p
l
i
c
a
t
i
o
n
s
i
s
t
h
e
o
u
t
p
u
t
s
p
e
e
d
f
r
o
m
t
h
e
gear b
ox
dev
ic
e.
Figure
1. Sc
he
mati
c d
ia
gram
of the
SMO
f
or I
RF
OC of t
he e
ncode
rless IM
b
ase
d
t
orqu
e
c
on
t
ro
l
dr
i
ve
5.
LABOR
ATO
RY S
YS
TE
M
SET
UP
The
sc
hem
at
ic
diag
ram
of
t
he
pro
posed
en
c
od
e
rless
I
M
dri
ve
usi
ng
S
MO
s
pee
d
est
im
at
or
util
iz
ing
TDC
ba
sed
on
IRFO
C
st
rategy
of
EV
ap
pl
ic
at
ion
s
ha
s
be
en
im
plemente
d
in
t
he
la
bora
tor
y
us
i
ng
DSpace
-
DS110
3
c
on
tr
ol
bo
a
r
d
f
or
5.5
kW
I
M
as
sho
wn
in
Fi
gure
2.
The
m
ac
hin
e
unde
r
te
st
was
op
e
rated
co
nce
rn
i
ng
encode
rless
to
rque
-
c
ontr
olled
dr
ive
co
ndit
ion
s
.
T
he
dr
iv
e
sp
ee
d
is
im
po
s
ed
by
the
sp
ee
d
-
c
on
t
ro
ll
ed
loa
d
dynam
om
et
e
r
PM
S
M.
The
s
pecifica
ti
ons
of
the
I
M
t
hat
are
us
e
d
in
al
l
l
abor
at
ory
te
sts
are
gi
ven
i
n
T
able
1.
The
I
M
is
fed
from
a
3
-
phase
V
S
-
i
nv
e
rter,
wh
ic
h
has
six
isolat
ed
gate
bi
po
la
r
tra
ns
ist
ors
(IGBT
’s)
s
witc
he
s
and
it
s
pulse
s
are
ob
ta
ine
d
by
util
iz
ing
a
gate
dr
i
ver
board.
T
he
D
S1103
co
ntr
ol
boar
d
is
m
ou
nted
i
n
a
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8
694
In
t J
P
ow
Ele
c
&
Dr
i
S
ys
t
,
V
ol
.
12
, N
o.
1
,
Ma
rch
202
1
:
20
–
28
24
per
s
onal
c
omp
uter
(P
C
).
This
Di
gital
Sig
nal
Proc
essi
ng
(
D
SP)
boa
rd
ha
s
a
set
of
on
-
boa
rd
pe
rip
heral
s
su
c
h
as
dig
it
al
-
to
-
a
nalo
g
(
D/A
),
analo
g
-
to
-
dig
it
al
(A
/D
)
c
onver
te
r
s
an
d
posit
ion
e
nc
od
e
r
inter
faces.
I
t
al
so
pro
vid
es
the
ne
cessar
y
dig
it
a
l
in
pu
t/
outp
ut
(I
/O
)
po
rts,
ti
mer
f
unct
ion
I
/O
ca
pture,
a
nd
PWM
g
at
e
pu
lse
s
gen
e
rati
on
port
s.
O
ver
PC,
al
l
cal
culat
ion
s
a
re
do
ne
an
d
pe
rformed
us
in
g
the
M
at
la
b/Si
mu
li
nk
en
vir
onme
nt.
The
Sim
ulin
k
con
t
ro
l
m
odel
s
are
c
ompil
ed
us
in
g
a
Te
xas
In
st
ruments
Ma
tl
ab/Simuli
nk
real
-
ti
me
inte
rf
ace
env
i
ronme
nt a
nd do
wn
l
oad
e
d t
o
the
DS
pace
-
DS1
103 b
oard
u
ti
li
zi
ng
DSpa
ce so
ftwa
re too
ls.
Table
1.
T
he
r
a
te
d
val
u
es a
nd
sp
eci
ficat
io
ns
of the
I
M
Sp
eciticatio
n
Valu
e
Rated
ou
tp
u
t po
we
r
5
.5 k
W
Rated
vo
ltag
e
1
8
6
V
F
5
0
Hz
P
2
R
s
0
.29
4
Ω
R
r
0
.14
3
2
5
Ω
L
s
5
7
.3 m
H
L
r
5
7
.3 m
H
L
m
5
6
.43
m
H
J
0
.12
kg
.m
2
B
0
.01
8
N.
m
.s/rad
Fo
r
the
e
xp
e
ri
mental
e
xecu
ti
on,
a
real
-
ti
me Simulin
k
c
ontr
ol
m
odel
is d
one
a
nd
th
en
co
mp
il
ed
to
the
DS
P
bo
a
r
d
util
iz
ing
softwa
re
too
ls.
T
he
acce
le
rator
ped
al
is
util
iz
ed
for
in
serti
ng
t
he
refe
ren
ce
to
r
qu
e
i
n
EV
app
li
cat
io
n
by
the
dr
i
ver,
t
he
ref
e
ren
ce
c
urre
nts
∗
an
d
∗
are
uti
li
zed
to
ge
ner
a
te
the
re
fer
e
nc
e
ph
ase
c
urre
nts
∗
,
∗
∗
us
i
ng
the
in
ve
rse
P
ar
k’
s
tra
nsfo
rmati
on
f
ormula.
The
m
ot
or
volt
ages
an
d
c
urre
nts
are
sense
d
us
in
g
LA
25
-
N
P
Hall
-
e
ff
ect
c
urren
t
sen
sors
and
L
V25
-
P
H
al
l
-
eff
ect
volt
age
se
nsors
,
res
pecti
vely
.
T
he
mo
to
r
measu
reme
nts
are
se
nt
bac
k
to
t
he
DSP
c
on
trol
boar
d
via
the
A/D
ports.
The
dif
fere
nce
s
betwee
n
t
he
sense
d
ph
a
se
c
urren
ts
an
d
t
heir
c
orr
esp
onding
ref
e
ren
ce
phase
c
ur
ren
t
are
fe
d
as
an
in
put
to
hyste
rese
s
ba
nd
s
to
gen
e
rate
pu
lse
s
f
or
t
he
th
ree
-
phase
i
nv
e
rter
.
Th
e
ge
ne
rated
pu
lse
s
trig
ge
r
the
I
GBT
ga
te
s
of
the
i
nv
e
rter
to
feed
the
I
M
wi
th
s
uitable
volt
ages
an
d
fr
e
qu
ency.
The
op
ti
cal
enc
od
e
r
wi
th
1024
-
pu
lse
reso
l
ution
is
ut
il
iz
ed
to
se
ns
e
the
r
ot
or
s
pee
d/posit
ion
,
this
se
ns
e
d
sig
nal
is
util
iz
ed
as
a
re
fer
e
nce
si
gn
al
for
com
par
is
on
pu
rpose
s
with the
esti
ma
te
d
s
peed sig
na
l. Th
e
IM is c
ouple
d
to
a s
pee
d
-
c
ontrolle
d l
oa
d dynam
ome
te
r
(
PMS
M
).
To
ve
rif
y
a
nd
pro
ve
t
he
e
ff
ic
ie
ncy
of
the
pr
opos
e
d
TCD
a
nd
e
ncode
rless
sp
e
ed
est
imat
ion
base
d
on
the
S
MO
obser
ver
wit
h
E
V
a
pp
li
cat
io
ns
,
the
la
borat
ory
wa
veforms
are
ta
ken
at
va
rio
us
op
e
rati
ng
c
ondi
ti
on
s
.
These
c
onditi
ons
incl
ude
the
IM
dri
ve
at
ve
ry
l
ow
an
d
ze
r
o
s
peeds
in
m
ot
or
in
g
a
nd
re
ge
n
erati
ng
mod
es
an
d
with
sud
den
l
oa
d
va
riat
ion
s
.
Also
,
t
he
r
unni
ng
at
ze
ro
el
ec
tric
al
fr
eq
ue
nc
y
is
al
so
e
valu
at
ed.
T
he
res
ul
ts
are
ta
ken
un
der
IR
FO
C
of
the
en
cod
e
rless
I
M
dri
ve.
T
he
IM
w
as
r
unni
ng
un
de
r
TC
D
c
onditi
on
wit
h
sta
nd
a
rd
dq
current
lo
op
s
,
and
the
est
ima
te
d
fl
ux
a
ng
le
is
us
e
d
f
or
vec
tor
t
ran
s
f
or
m
at
ion
of
t
he
ro
t
or
-
or
ie
nted
dq
-
fr
ame
.
The dri
ve
s
ys
te
m w
a
s c
onnect
ed
to
a s
pee
d
c
on
t
ro
ll
ed
P
M
S
M
whose
sp
e
e
d
is
monit
or
e
d.
Figure
2. Sc
he
mati
c d
ia
gram
of the la
borato
ry setu
p fo
r
S
M
O
-
based s
pe
ed
est
imat
io
n o
f
IRF
OC e
nc
oderless
IM drive
u
s
i
ng D
S
pace
-
D
S11
03 contr
ol
bo
a
r
d
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
P
ow Elec
& Dri S
ys
t
IS
S
N:
20
88
-
8
694
Smart i
ntegr
ation of
dr
iv
e sy
ste
m
fo
r in
duct
io
n mo
t
or
applic
ations in
elec
tri
c
…
(
M
ohame
d
K.
Metw
aly
)
25
6.
RESU
LT
S
AND DI
SCUS
S
ION
S
6.1.
Very l
ow
an
d
zero s
peed
oper
at
i
on
The
sta
rtin
g
pe
rformance
of
the
enc
oderles
s
IM
with
TC
D
us
i
ng
the
S
M
O
f
or
s
pee
d
est
imat
ion
at
ver
y
lo
w
sp
ee
d
of
20
r
pm
unde
r
rate
d
loa
d
co
ndit
ion
is
sh
ow
n
in
Fi
gure
3.
In
Fig
ur
e
3.
a
the
uppe
r
gr
a
ph
pr
ese
nts
t
he
rot
or
s
pee
d
ref
e
r
ence
obta
ined
from
t
he
posit
ion
enc
oder
(
bl
ack)
an
d
t
he
e
sti
mate
d
s
peed
us
i
ng
the
S
MO
(gar
y
-
dott
ed),
w
hile
the
er
ror
betw
een
t
hese
t
wo
s
ign
al
s
is
dep
ic
t
ed
i
n
t
he
mid
dle
grap
h
in
r
pm.
It
is
no
te
d
that
t
he
diff
e
re
nce
bet
ween
t
hese
tw
o
sp
ee
ds
is
ap
pro
ximate
ly
13
r
pm
du
rin
g
s
ta
rting
w
hich
dies
out
after
0.5
sec
.
T
his
pr
ov
es
the
accurac
y
of
est
imat
ed
s
peed
s
ign
al
a
nd
the
e
ff
ect
ive
nes
s
of
the
S
MO.
Thi
s
can
be
sma
rtly
ac
hi
eved
with
ou
t
any
e
nc
oder.
The
loa
d
c
urre
nt
(
i
q
)
is
dep
ic
te
d
in
the
l
ower
grap
h.
The
sta
tor
currents
i
n
αβ
fr
ame
s
are
sho
wn
i
n
the
uppe
r
gra
ph
of
Fig
ur
e
3(
b
)
.
In
or
der,
to
pr
es
ent
the
ad
va
ntages
of
t
he
SMO,
t
he
r
ot
or
fl
ux
an
gles
a
nd
the
dif
fer
e
nc
e
betwee
n
t
he
ref
e
rence
(
bla
ck)
a
nd
est
ima
te
d
fl
ux
(
gr
a
y
-
do
tt
ed
)
ang
le
s
a
re
depi
ct
ed
in
the
mi
dd
le
a
nd
lo
wer
gr
a
phs
in
Fig
ur
e
3(
b
)
.
T
he
f
igure
prese
nts
the
r
otor
flu
x
ang
le
smartly
obta
in
ed
from
the
e
nc
od
e
rless
c
ontr
ol
met
hod
us
i
ng
the
S
MO
te
c
hn
i
qu
e
(
gr
a
y
-
dott
ed)
an
d
t
he
flu
x
ang
le
f
r
om
the
sens
or
-
base
d
(
ref
e
ren
ce
)
cu
rrent
model
(
black).
It
is
no
te
d
that
the
est
imat
ed
flu
x
an
gle
track
s
the r
e
fer
e
nce
one sm
artl
y
a
nd
smoothl
y
wit
h smal
l flu
x
a
ng
l
e d
e
viati
on
a
s
s
how
n
in
the l
ower
gra
ph
of F
igure
3(
b
)
.
T
he
a
ppli
cat
ion
of
S
M
O
pro
vid
es
bette
r
matc
hing
to
the
real
val
ues
of
t
he
ro
t
or
flu
x
a
ng
le
with
out
an
y
encode
r.
Fig
ure
.
4
sho
ws
the
la
borat
ory
wa
vefo
rms
du
rin
g
t
he
dy
namic
sp
ee
d
r
esp
on
ses
of
t
he
enc
oderless
IM
dri
ve
usi
ng
th
e
S
MO
te
chn
i
q
ue
at
rated
lo
ad
an
d
sp
e
ed
c
ha
ng
e
.
T
he
I
M
dr
i
ve
s
yst
em
was
ope
ra
te
d
at
sp
ee
d
ref
e
re
nc
e
20
r
pm
unde
r
rated
loa
d
c
onditi
on,
the
n
a
su
dde
n
sp
e
ed
c
hange
t
o
ze
ro
sp
ee
d
was
a
pp
l
ie
d
at
t=
18
s
ec,
th
en
bac
k
to
20
r
pm
at
t=
48
se
c.
T
he
e
ncode
rless
dr
i
ve
op
erates
i
n
t
he
mo
to
rin
g
m
ode.
It
is
ob
s
er
ved
t
hat
t
he
enc
oderless
d
ri
ve
sho
ws
a sign
ific
a
nt p
e
rformance
w
it
h excel
le
nt sp
ee
d
est
imat
io
n
acc
ur
ac
y
with
a
ppr
oxim
at
el
y
8
rpm
de
viati
on
wh
ic
h
dies
ou
t
s
oon
after
tra
ns
ie
nts.
Also,
t
he
de
viati
on
betwee
n
t
he
ref
e
ren
ce
flu
x
ang
le
a
nd
the
est
imat
ed
flu
x
ang
le
(b
a
sed
on
t
he
S
MO
te
chn
i
qu
e
)
is
il
lust
rated
i
n
th
e
lowe
r
gr
a
ph
of
Fi
gur
e
4(
b
)
.
As
it
i
s
obse
rv
e
d,
th
e
an
gle
de
viati
on
sta
ys
withi
n
a
fe
w
de
gre
es
as
de
pict
ed
in
the
lowe
r gr
a
ph of
Figure
4(
b
)
.
(a)
(b)
Figure
3.
Lab
orat
ory
e
valu
at
ion o
f fixe
d
lo
w
sp
ee
d
at
20 rp
m du
rin
g
sta
rti
ng p
e
rio
d
at
t
he
r
at
ed
loa
d; (
a
)
the
su
bfi
gu
res
a
re
a compa
rison
of act
ual
(b
la
c
k), esti
mate
d (
grey
-
dott
ed)
sp
ee
ds
, spee
d
e
sti
mati
on
e
rror (
rpm),
and loa
d
to
r
que cu
rr
e
nt (p.
u)
,
(
b)
t
he
s
ubfig
u
res
a
re
-
c
urren
t
,
compa
ris
on of
ref
e
ren
ce
(blac
k) an
d
est
imat
ed
(
gr
e
y
-
dott
ed)
flu
x ang
le
s
, a
nd f
lu
x
a
ng
le
e
rro
r
(a)
(b)
Figure
4.
Lab
orat
ory
e
valu
at
ion at
step
sp
ee
d
c
hanges
(20 r
pm
→
0 rpm
→
20 rpm
)
at
t
he rat
ed
l
oad
;
(a)
sp
ee
d
c
omp
ariso
n,
e
rr
or, a
nd loa
d
to
r
qu
e
current,
(
b)
-
c
urren
t,
f
lu
x
ang
le
c
ompa
ris
on, a
nd er
ror
0
2
4
6
8
0
10
20
S
p
e
e
d
s
[
r
p
m
]
0
2
4
6
8
0
.
9
1
1
.
1
t
i
m
e
(
s
e
c
)
I
q
[
p
.
u
]
0
2
4
6
8
0
5
10
15
E
r
r
o
r
[
r
p
m
]
0
2
4
6
8
-4
-2
0
2
4
t
i
m
e
(
s
e
c
)
F
l
u
x
e
r
r
o
r
[
d
e
g
.
]
0
2
4
6
8
-
2
0
0
0
200
F
l
u
x
a
n
g
l
e
s
[
d
e
g
.
]
0
2
4
6
8
-1
0
1
i
[
p
.
u
]
15
20
25
30
35
40
45
50
0
10
20
S
p
e
e
d
s
[
r
p
m
]
15
20
25
30
35
40
45
50
-
1
0
0
10
E
r
r
o
r
[
r
p
m
]
15
20
25
30
35
40
45
50
0
.
9
1
1
.
1
t
i
m
e
(
s
e
c
)
i
q
[
p
.
u
]
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8
694
In
t J
P
ow
Ele
c
&
Dr
i
S
ys
t
,
V
ol
.
12
, N
o.
1
,
Ma
rch
202
1
:
20
–
28
26
6.2.
Sudd
e
n lo
ad d
isturb
an
ces
at
v
ery
lo
w an
d
zero s
peeds
The
la
borato
ry
wa
veforms
s
howi
ng
the
pe
rformance
of
t
he
enc
od
e
rless
IM
ba
sed
TC
D
us
in
g
t
he
SMO
with
s
ud
den
ap
ply
i
ng
a
nd
re
movin
g
t
he
rated
l
oad
t
orq
ue
f
or
20
r
pm
sp
ee
d
ref
e
r
ence
at
t=
8
se
c
an
d
t=
41
sec
,
respec
ti
vely
in
the
mo
to
rin
g
m
ode
are
s
how
n
in
Fig
ur
e
5.
T
he
La
borato
ry
wav
e
f
or
m
s
showi
ng
app
l
ying
a
nd
rem
ov
i
ng
the
rate
d
-
l
oad
to
r
que
for
refe
re
nc
e
sp
ee
d
zer
o
are
al
so
pres
ented
as
de
picte
d
in
Fig
ure
6.
A
ra
te
d
loa
d
to
rqu
e
ste
p
is
a
ppli
ed
a
t
t=
12
sec
an
d
rem
oved
at
t=
32
sec.
A
s
it
is
no
te
d,
a
go
od
matc
hing
bet
w
een
the
meas
ured
a
nd
est
im
at
ed
s
pee
ds
is
ac
hieve
d
sinc
e
the
est
imat
e
d
sp
ee
d
f
ollo
ws
the
measu
red
one
an
d
rec
overs
qu
ic
kly
unde
r
loa
d
im
pact.
The
est
imat
e
d
sig
nals
ca
n
su
ccess
fu
ll
y
tr
ack
the
measu
red
on
e
s
duri
ng
the
tra
ns
ie
nt
an
d
ste
ady
sta
te
s.
A
c
onside
rab
le
re
duct
ion
of
the
s
peed
est
imat
io
n
e
rror
of
a
ppr
oximat
el
y
20
rpm
duri
ng
tra
ns
ie
nt
operati
on
to
ze
ro
rpm
i
n
ste
ad
y
sta
te
is
obser
ve
d.
It
is
e
vid
e
nt
that
the
SMO
wor
ks
pro
per
ly
,
es
pe
ci
al
ly
in
the
mo
to
rin
g
m
od
e
at
very
low
and
ze
r
o
sp
ee
ds
unde
r
sud
de
n
loa
d
disturba
nces.
It
can
be
see
n
th
at
the
s
pee
d,
th
e
cu
rr
e
nt,
the
r
otor
flu
x
a
ngle
are
co
rr
e
ct
ly
obser
ve
d.
T
he
a
ct
ual
ro
t
or
flu
x
a
ngle
tracks
the
e
sti
mate
d
r
otor
flu
x
a
ng
le
of
the
S
M
O
te
chn
i
qu
e
,
with
only
li
mit
ed
e
rror
as
dep
ic
te
d
in
t
he
mid
dle
an
d
l
ower
grap
hs
of
Figure
5(
b
)
an
d
Fi
gure
6(
b
)
r
especti
vely
.
T
he
dev
ia
ti
on
be
tween
the
act
ual r
ot
or
flu
x
a
ng
le
a
nd
the
est
imat
ed
f
lu
x
a
ng
le
sta
ys
within
a few d
eg
rees.
T
his
a
lso
in
dicat
es
th
at
the
sp
ee
d
c
on
tr
ol
of e
ncoder
le
ss
IM
base
d
TC
D usin
g
t
he
S
MO
works
well
.
(a)
(b)
Figure
5. Lab
orat
ory
e
valu
at
ion d
uri
ng s
udde
n
a
pp
li
cat
io
n and re
moval
of the
rated
l
oad
tor
qu
e
at spee
d o
f
20 rpm;
(a) spe
ed
c
omparis
on,
erro
r,
a
nd loa
d t
orq
ue
c
u
r
r
e
n
t
,
(
b
)
-
c
u
r
r
e
n
t
,
f
l
u
x
a
n
g
l
e
c
o
m
p
a
r
i
s
o
n
,
a
n
d
e
rror
(a)
(b)
Figure
6. Lab
orat
ory
e
valu
at
ion d
uri
ng s
udde
n
a
pp
li
cat
io
n and re
moval
of the
rated l
oad
tor
qu
e
at zer
o
s
peed
;
(a)
sp
ee
d
c
omp
ariso
n,
e
rro
r,
a
nd loa
d
to
r
qu
e
current,
(
b)
-
c
urren
t,
f
lu
x
ang
le
c
ompa
ris
on, a
nd er
ror
6.3.
Speed re
vers
al
a
t
v
ery
l
ow
s
peed
The
la
borat
ory
wa
ve
forms
il
lustrate
the
pe
r
forma
nce
of
th
e
e
ncoder
le
ss
IM
base
d
TCD
us
i
ng
the
SMO
duri
ng
ve
ry
lo
w
s
pee
d
rev
e
rsals
a
re
s
how
n
i
n
Fig
ure
7.
T
he
I
M
is
op
e
rated
at
rated
l
oad
c
onditi
on.
T
he
ref
e
ren
ce
s
pee
d
is
sud
den
l
y
c
hange
d
f
rom
-
20
rpm
to
20
r
pm
at
t
=
7
s
e
c,
then
bac
k
to
-
20
r
pm
at
t=
21
sec
,
and
c
hange
d
t
o
20
rpm
at
t
=
34
sec.
T
he
encode
rless
dr
i
ve
operates
fir
st
in
t
he
re
generati
ng
m
ode
(
t=
0
→
7
sec),
t
hen
in
t
he
mo
t
or
i
ng
m
ode
(t
=
7
→
21
sec),
agai
n
in
the
re
ge
ner
at
in
g
m
ode
(t=
21
→
34
s
ec
),
a
nd
finall
y
in
the
m
otorin
g
mode
(t
=
34
→
40
sec)
as
il
l
us
trat
ed
in
F
igure
7.
It
has
been
ob
se
r
ved
that
the
enc
od
erless
5
10
15
20
25
30
35
40
45
-
2
0
0
20
40
60
S
p
e
e
d
s
[
r
p
m
]
5
10
15
20
25
30
35
40
45
-
2
0
0
20
E
r
r
o
r
[
r
p
m
]
5
10
15
20
25
30
35
40
45
0
0
.
5
1
t
i
m
e
(
s
e
c
)
i
q
[
p
.
u
]
10
15
20
25
30
35
-
2
0
0
20
E
r
r
o
r
[
r
p
m
]
10
15
20
25
30
35
0
0
.
5
1
t
i
m
e
(
s
e
c
)
i
q
[
p
.
u
]
10
15
20
25
30
35
-
4
0
-
2
0
0
20
40
S
p
e
e
d
s
[
r
p
m
]
10
15
20
25
30
35
-1
0
1
i
[
p
.
u
]
10
15
20
25
30
35
-
2
0
0
0
200
F
l
u
x
a
n
g
l
e
[
d
e
g
.
]
10
15
20
25
30
35
-5
0
5
t
i
m
e
(
s
e
c
)
F
l
u
x
e
r
r
o
r
[
d
e
g
.
]
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
P
ow Elec
& Dri S
ys
t
IS
S
N:
20
88
-
8
694
Smart i
ntegr
ation of
dr
iv
e sy
ste
m
fo
r in
duct
io
n mo
t
or
applic
ations in
elec
tri
c
…
(
M
ohame
d
K.
Metw
aly
)
27
IM
ba
sed
TC
D
us
in
g
t
he
S
M
O
s
how
a
si
gn
i
ficant
performa
nce
with
e
xcell
ent
est
im
at
ion
acc
ur
a
cy
duri
ng
very
lo
w
s
pee
d
rev
e
rsals
in
t
h
e
mo
to
rin
g
a
nd
reg
e
nerat
ing
modes
of
oper
at
ion
.
Also,
th
e
est
imat
ed
r
oto
r
flu
x
ang
le
t
rack
s
th
e ref
e
re
nce
on
e
w
it
h smal
l er
r
or as s
how
n
i
n t
he
mid
dle a
nd
low
e
r gr
a
phs
of Fig
ur
e
7(
b
)
.
(a)
(b)
Figure
7. Lab
orat
ory
e
valu
at
ion at
v
e
r
y
lo
w spee
d
re
ve
rsal
±20 rp
m at
rated loa
d
;
(a)
sp
ee
d
c
omparis
on,
error, a
nd loa
d t
orqu
e
curre
nt,
(b)
-
c
urre
nt,
fl
ux angle c
ompa
rison, a
nd
error
7.
CONCL
US
I
O
N
In
this
pap
e
r,
a
sma
rt
i
nteg
rati
on
of
dri
ve
s
yst
em
base
d
on
encode
rless
I
M
,
TC
D,
an
d
i
nd
i
rect
ro
t
or
fiel
d
-
or
ie
nted
con
t
ro
l
IRFO
C
us
in
g
th
e
S
MO
is
util
iz
ed
for
sp
e
ed
est
imat
ion
i
n
E
V
ap
pl
ic
at
ion
s.
T
he
S
M
O
is
util
iz
ed
to
est
i
mate
the
ro
t
or
sp
ee
d
from
sta
tor
cu
rr
e
nts/v
ol
ta
ges
me
asu
re
ments.
T
he
ma
themat
ic
al
m
odel
of
SMO
is
presen
te
d,
a
nd
la
bora
tor
y
im
plemen
te
d.
T
he
TCD
is
us
e
d
as
acce
le
rati
ng
t
orq
ue
w
hich
ge
ner
at
es
th
e
ref
e
ren
ce
tor
que
c
urren
t
i
n
EV
by
the
dri
ver.
I
n
t
he
a
bse
nce
of
enc
oder
sens
or,
T
CD
ba
sed
on
IRFOC
requires
enc
oderless
s
peed
est
imat
or
f
or
flu
x
an
gle
de
te
rmin
at
io
n.
Lab
or
at
or
y
w
avefor
ms
c
onf
irm
the
eff
ect
ive
ness
of
t
he
pro
posed
enc
oderless
I
M
base
d
TCD
us
i
ng
t
he
S
MO
at
ve
r
y
l
ow
and
ze
ro
s
peeds.
T
he
ro
t
or
sp
ee
d/flux a
ng
le
e
sti
mati
on
acc
uracy
is
no
te
d u
nd
e
r
l
oa
d
to
r
qu
e/
s
pee
d variat
io
ns
.
ACKN
OWLE
DGE
MENTS
This
w
ork
was
sup
ported
in
par
t
by
Scie
nti
fic
Re
searc
h
D
eans
hip
,
Tai
f
Un
i
ver
sit
y
unde
r
Re
sea
rch
Group 6
147
-
440
-
1.
REFERE
NCE
S
[1]
A.
Borisevich,
a
nd
G
Schull
eru
s
,
“
En
erg
y
Eff
ic
i
e
nt
Control
of
an
Induc
ti
on
M
ac
hi
ne
Under
Torqu
e
Step
Ch
ange
s
,”
IEE
E
Tr
ansacti
o
ns on
Ene
rg
y
Co
nve
rs
ion
,
vo
l. 31
,
no
.
4
,
pp
.
1295
-
1303,
2016
.
[2]
M.
Pac
as,
“
Sens
orle
ss
driv
es
in
industri
al
appl
i
c
at
ions,
”
IEEE
In
dustrial
El
e
ct
ro
nic
s
Magazin
e
,
vol.
5,
no
.
2
,
pp
.
16
-
23,
2011
.
[3]
S.
Xe
papa
s,
A.
Kale
tsanos,
F
.
Xepa
pas
and
S.
Mania
s,
“
Slidi
ng
-
mode
observ
er
for
spee
d
-
sen
sorless
induc
ti
o
n
mot
or
dr
ive
s,
”
I
EE
Proceedi
ngs
-
Control
Theory an
d
Applications
,
vol
.
150
,
no
.
6
,
pp
.
611
-
617
,
20
03.
[4]
M.
S.
Z
aky,
M
.
M.
Kha
te
r
,
S.
S.
Shokralla
,
and
H.
A.
Yasi
n,
“
Wi
de
-
Spe
ed
-
Range
Esti
m
at
i
on
with
Onlin
e
Para
meter
Id
entificat
ion
Schemes
of
Sensorles
s
Induc
ti
on
Mo
tor
Drive
s,
”
IE
EE
Tr
ansacti
on
s
on
Industrial
El
e
ct
ronics
,
vol
.
56,
no.
5,
pp
.
16
99
-
1707,
2009
.
[5]
Y.
Fan,
L
.
Zha
n
g,
M.
Cheng
,
an
d
K.
T.
Chau
,
”
S
ensorle
ss
SV
PWM
-
FA
DTC
of
a
New
Flux
-
Modulated
Perm
ane
n
t
-
Magne
t
Wh
ee
l
Motor
Based
on
a
Wi
de
-
Spe
ed
Sl
idi
ng
Mode
Obs
erv
er,
”
I
EE
E
Tr
ansacti
o
ns
on
Industrial
El
e
ct
ronics
,
vol
.
62,
no.
5,
pp
.
31
43
-
3151,
2015
.
[6]
F.
Chen
and
M.
W
.
Dunnigan
,
“
Compa
r
at
iv
e
st
udy
of
a
sl
idi
ng
-
mode
observe
r
and
Ka
lman
filt
ers
for
ful
l
state
esti
mation
in an induct
ion
m
a
c
h
i
n
e
,
”
I
E
E
P
r
o
c
e
e
d
i
n
g
s
-
E
l
e
c
t
r
i
c
P
o
w
e
r
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
1
4
9
,
n
o
.
1
,
p
p
.
5
3
-
64
,
2002.
[7]
Z.
Qiao
,
T
.
Shi,
Y.
Wa
ng
,
Y.
Yan,
C.
Xi
a,
an
d
X.
He,
“
New
Slidi
ng
-
Mode
Obs
erv
er
fo
r
Pos
it
ion
Sensorles
s
Control
of
Per
m
ane
nt
-
Magn
et
S
ynchr
onous
Motor,
”
I
EE
E
Tr
ansacti
ons
on
Indus
trial
E
le
c
tronic
s
,
vol
.
60,
no.
2
,
pp
.
710
-
719
,
20
13.
[8]
Y.
Zh
ao,
W
.
Qi
ao,
and
L
.
Wu
,
“
Improve
d
Roto
r
Pos
it
ion
and
S
pee
d
Estimat
ors
for
Sensorless
C
ontrol
of
In
te
r
io
r
Perma
nen
t
-
Mag
net
Synchronous
Mac
h
ine
s,
”
I
EEE
Journal
o
f
Eme
rging
and
Se
lec
te
d
Topics
in
Po
wer
Elec
tronic
s
,
vol.
2
,
no
.
3
,
pp
.
627
-
639,
2014
.
[9]
H.
Kim
,
J.
Son,
and
J
.
Lee,
“
A
High
-
Speed
Sl
id
ing
-
Mode
Obs
er
ver
fo
r
the
Sens
orle
ss
Speed
Co
ntrol
of
a
PM
SM
,
”
IEE
E
Tr
ansacti
o
ns on
Industrial Elect
ronics
,
vol
.
58,
no.
9,
pp
.
40
69
-
4077,
2011
5
10
15
20
25
30
35
40
-
2
0
0
20
S
p
e
e
d
s
[
r
p
m
]
5
10
15
20
25
30
35
40
-
2
0
0
20
E
r
r
o
r
[
r
p
m
]
5
10
15
20
25
30
35
40
1
1
.
0
2
t
i
m
e
(
s
e
c
)
i
q
[
p
.
u
]
5
10
15
20
25
30
35
40
-1
0
1
i
[
p
.
u
]
5
10
15
20
25
30
35
40
-
2
0
0
0
200
F
l
u
x
a
n
g
l
e
[
d
e
g
.
]
5
10
15
20
25
30
35
40
-5
0
5
t
i
m
e
(
s
e
c
)
F
l
u
x
e
r
r
o
r
[
d
e
g
.
]
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8
694
In
t J
P
ow
Ele
c
&
Dr
i
S
ys
t
,
V
ol
.
12
, N
o.
1
,
Ma
rch
202
1
:
20
–
28
28
[10]
G.
Foo,
and
M. F
.
Rahman,
“
Se
nsorless Slidi
ng
-
Mode
MTPA
Control
of
an
IPM
Synchronous Motor
Drive
Us
ing
a
Slidi
ng
-
Mode
O
bserve
r
and
HF
Signal
In
je
c
ti
on
,
”
IE
EE
Tr
ansacti
ons
on
Industri
al
E
le
c
tronic
s
,
v
ol.
57
,
no
.
4,
pp
.
1270
-
1278,
201
0.
[11]
L.
Zha
o
,
J.
Hua
n
g,
H.
Li
u,
B
.
Li
and
W
.
Kon
g,
“
Second
-
Ord
er
Slid
ing
Mode
Obs
erv
er
with
Online
Para
m
eter
Ide
nti
f
ic
a
ti
on
for
Sensorless
Indu
ct
ion
Motor
Dri
ves,
”
IE
EE
Tr
ansacti
ons
on
Indu
strial
E
lec
tronics
,
vol
.
61
,
no.
10
,
pp
.
5280
-
5289
,
2014.
[12]
C.
L
asc
u
and
G.
D.
Andree
scu
,
s
pp
Slidi
ng
-
Mode
Obs
erv
er
and
I
m
prove
d
Int
egr
at
o
r
Wi
th
DC
-
Offs
et
Compe
nsa
ti
on
for
Flux
Esti
m
ation
in
Sensorles
s
-
Control
le
d
Ind
uct
ion
Motors,
”
IEE
E
Tr
ansacti
o
ns
on
Industrial
El
e
ct
ronics
,
vol.
53,
no
.
3
,
pp
.
78
5
-
794,
2006
.
[13]
Y.
Zha
o,
W.
Q
ia
o,
and
L.
Wu
,
“
An
Adapti
v
e
Quasi
Slid
ing
-
Mode
Rotor
Po
siti
on
Obs
erv
er
-
Based
Sensorl
ess
Control
for
Int
er
ior
Perma
n
ent
Magne
t
Synchro
nous
Mac
hine
s,
”
IEE
E
Tr
ansactions
on
Industrial
Elec
troni
cs
,
vo
l.
28,
no
.
12
,
pp
.
5
618
-
5629,
2013
.
[14]
H.
L
ee
and
J.
Lee,
“
Design
of
Iterat
iv
e
Slid
ing
M
ode
Obs
erv
er
fo
r
Sensorless
PM
SM
Control
,
”
I
E
EE
Tr
ansacti
ons
on
Control
S
ystem
s Tec
hnology
,
vol.
21
,
no
.
4
,
pp
.
1394
-
1399
,
20
13.
[15]
G.
Wa
ng
,
Z.
L
i
,
G.
Zha
ng,
Y
.
Yu,
and
D.
Xu,
“
Qu
adr
at
ur
e
PLL
-
Based
High
-
Order
Slid
ing
-
Mode
Obs
erv
er
for
IPM
S
M
Sensorless
Control
with
Online
MTPA
Control
Str
at
egy
,
”
IEEE
Tr
ansac
ti
ons
on
Ene
rgy
Conve
rs
ion
,
vol.
28,
no
.
1
,
pp
.
21
4
-
224,
2013
.
[16]
G.
Foo,
and
M.
F.
Rah
ma
n
,
“
Dire
ct
Torque
Control
of
an
I
PM
-
Sy
nchr
onou
s
Motor
Drive
a
t
Very
Low
Spe
ed
Us
ing
a
Slid
ing
-
Mode
Stat
or
F
l
u
x
O
b
s
e
r
v
e
r
,
”
I
E
E
E
T
r
a
n
s
a
c
t
i
o
n
s
o
n
P
o
w
e
r
E
l
e
c
t
r
o
n
i
c
s
,
v
o
l
.
2
5
,
n
o
.
4
,
p
p
.
9
3
3
-
942
,
2010
.
[17]
R.
Moral
es
-
Cap
ora
l,
E
.
B
.
Huer
t
a,
C
.
Hern
andez,
M.
A.
Arjona
,
a
nd
M.
Pa
ca
s,
“
T
ran
sduce
rl
ess
Acquisit
ion
of
the
Rotor
Pos
it
ion
for
Predic
t
ive
Torque
Contro
lled
PM
Synchronous
Mac
hine
s
Based
on
a
DS
P
-
FPGA
Digit
al
Sys
te
m,
”
IE
EE
Tr
ansacti
ons on Indus
trial
Info
r
matic
s
,
vo
l. 9, n
o.
2
,
pp
.
799
-
80
7,
2013
.
[18]
X.
Zha
ng
,
L.
Sun,
K.
Zha
o
,
and
L
.
S
u
n
,
“
N
o
n
l
i
n
e
a
r
S
p
e
e
d
C
o
n
t
r
o
l
f
o
r
P
M
S
M
S
y
s
t
e
m
U
s
i
n
g
S
l
i
d
i
n
g
-
M
o
d
e
C
o
n
t
r
o
l
a
n
d
D
i
s
t
u
r
b
a
n
c
e
C
o
m
p
e
n
s
a
t
i
o
n
T
e
c
h
n
i
q
u
e
s
,
”
I
E
E
E
T
r
a
n
s
a
c
t
i
o
n
s
o
n
P
o
w
e
r
E
l
e
c
t
r
o
n
i
c
s
,
v
o
l
.
2
8
,
n
o
.
3
,
p
p
.
1
3
5
8
-
1365,
2013.
[19]
Y.
Feng
,
J.
Zheng,
X
.
Yu
,
and
N.
V.
Truong
,
“
Hybrid
Te
rm
in
al
Slidi
ng
-
Mode
Obs
erv
er
Desi
gn
Method
for
a
Perma
nen
t
-
Mag
net
Synchronous
Motor
Control
Sys
te
m,
”
I
EE
E
Tr
ansacti
ons
on
Industrial
El
e
ct
r
onic
s
,
vol
.
56
,
n
o.
9,
pp
.
3424
-
343
1,
2009
.
[20]
L.
Yu
an,
F
Xiao
,
J
Shen
,
M
Che
n,
Q
.
Shi,
L
.
Q.
f
eng,
“
Sensorl
ess
con
trol
of
high
-
power
in
te
r
ior
p
erm
an
ent
m
agnet
synchronous m
o
tor
driv
es
a
t
v
ery
low
spee
d
,
”
IET
Elec
tri
c Powe
r
Appl
ic
a
ti
ons
,
vo
l.
7
,
no
.
3
,
pp
.
19
9
-
206,
2013
.
[21]
M.
Com
ane
scu
,
“
Single
and
dou
ble
com
pound
ma
nifol
d
slid
ing
mode
observ
ers
for
flux
and
spee
d
estimation
o
f
the
induc
t
ion mo
tor
driv
e,
”
I
ET
E
le
c
tric
Powe
r
Ap
pli
cations
,
vo
l. 8
,
no
1,
pp.
29
-
38
,
2014
.
[22]
E.
Aza
d
,
S.
Ga
doue,
D.
Atkins
on,
H
.
Slater,
P.
Barr
ass,
and
F.
Bl
aa
bj
erg
,
“
Se
nsorless
Control
of
IM
Based
o
n
Stat
or
-
Volt
age
MRA
S
for
Li
mp
-
Home
EV
Appl
ic
a
ti
ons
,”
IEEE
Tr
ansacti
ons
on
Powe
r
Elec
troni
cs
,
vol
.
33,
no
.
3
,
pp
1911
-
1921,
2
018.
[23]
M.
K.
Metwa
l
ly
,
“
Contro
l
of
F
our
Sw
it
ch
Thr
ee
Phase
Inv
erter
Fed
Induc
t
io
n
Motor
Drive
s
Based
Speed
a
nd
Stat
or
Resista
n
c
e
Esti
m
ation
,”
I
nte
rnational
Jou
rnal
of
Powe
r
Elec
troni
cs
and
Dr
iv
e
S
yste
ms
(IJ
P
EDS)
,
vol.
4
,
no
.
2,
pp
.
192
-
203
,
2014.
[24]
M.
K.
Metwa
ll
y
,
“
Dire
ct
Torqu
e
Con
trol
of
Four
Sw
it
ch
Th
ree
Phase
Inve
r
te
r
fed
Indu
ct
ion
Motor
Sensorles
s
Spe
ed
Drive
,”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
P
o
w
e
r
E
l
e
c
t
r
o
n
i
c
s
a
n
d
D
r
i
v
e
S
y
s
t
e
m
s
(
I
J
P
E
D
S
)
,
v
o
l
.
5
,
n
o
.
2
,
p
p
.
1
5
3
-
165
,
2014
.
[25]
O.
Bar
am
bon
es
and
P.
Alkort
a,
“
Pos
it
ion
Co
ntrol
o
f
the
Ind
uct
ion
Motor
Us
ing
an
Adapt
i
ve
Slid
ing
-
Mode
Control
le
r
and
O
bserve
rs,
”
IEEE T
rans
a
ct
ions o
n
Industrial
E
le
c
tronic
s
,
vo
l. 6
1,
n
o.
12
,
pp
.
6556
-
6565,
2014
.
[26]
M.
S.
Za
ky
,
M
.
K.
M
et
waly
,
S
ai
d
A
.
Der
az,
a
nd
H.
Z.
Az
az
i
,
“
A
New
Adapt
ive
SM
O
for
Speed
Esti
m
ation
of
Sensorless
Induc
ti
on
Motor
Dr
ive
s
a
t
Ze
ro
a
nd
Very
Low
Freque
ncies,
”
I
EE
E
Tr
ansacti
o
ns
on
Industria
l
El
e
ct
ronics
,
vol
.
65,
no.
9,
pp
.
69
01
-
6911,
2018
.
Evaluation Warning : The document was created with Spire.PDF for Python.