Internati
o
nal
Journal of P
o
wer Elect
roni
cs an
d
Drive
S
y
ste
m
(I
JPE
D
S)
Vol
.
5
,
No
. 2, Oct
o
ber
2
0
1
4
,
pp
. 21
9~
22
9
I
S
SN
: 208
8-8
6
9
4
2
19
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
/
IJPEDS
Comparison Analysis
of Indirect FOC Induction Motor Drive
using PI, Anti-Windup and Pre Filter Schemes
M.H.N Talib*, Z
.
Ibrahim**,
N.
Abd.
Rahim***, A.S.
A. Hasim
*
***
*,** Dep
a
rtement of
Electr
i
cal
E
ngineer
ing,
Univ
ersiti
Tekn
ikal
Mala
y
s
ia
Mel
a
k
a
***
UMPEDAC,
Universiti Malay
a
, K
u
ala
Lum
p
ur, M
a
l
a
y
s
ia
****
Faculty
of Engineering,
Un
iv
ers
iti
P
e
rt
ahan
an Nas
i
ona
l M
a
l
a
y
s
i
a
,
Kual
a
Lu
m
pur, M
a
la
y
Article Info
A
B
STRAC
T
Article histo
r
y:
Received
May 12, 2014
Rev
i
sed
Jun
19,
201
4
Accepte
d
J
u
l 10, 2014
This paper pr
esents the speed p
e
r
f
orman
ce an
al
ys
is
of indire
ct F
i
e
l
d Orient
ed
Control (FOC) induction motor
drive b
y
apply
i
n
g
Proportional I
n
tegral (PI)
controll
er, PI wi
th Anti-Windup (PIAW) and Pre
-
Filter (PF). Th
e objectiv
e
of this experim
e
nt is to have qua
n
tita
tive
com
p
arison between th
e control
l
er
s
t
rateg
i
es
towar
d
s
the perform
ance of the m
o
tor in term
of
s
p
eed track
ing
and load r
e
j
ect
i
on capab
ili
t
y
in
low,
medium and rated sp
eed
operation. In
the first p
a
rt, PI
controll
er is
applied to
the FOC induction motor
drive which
the ga
in is ob
t
a
ined b
a
sed on
determ
in
ed In
duction Motor
(IM) m
o
tor
parameters
. Secondly
an
AWPI strate
g
y
is added to
the ou
ter loop
and
finally
,
PF is ad
ded to th
e s
y
stem. Th
e Space Vector Puls
e Width Modulation
(SVPWM) technique is used to cont
rol the v
o
ltag
e
s
ource inverter and
complete vector
control s
c
heme
of the
IM drive
is tested
b
y
usin
g a DSpace
1103
controller board. The analy
s
is
of
th
e resu
lts shows that,
the PI an
d
AW
P
I
controlle
r s
c
hem
e
s
produce s
i
m
ilar p
e
rform
ance
at
low s
p
eed
operation. However, for
the medium and rated
speed operatio
n
the AWPI
scheme shown significant impro
v
emen
t in redu
cing the oversho
ot problem
and improving the setting time.
The PF
scheme on the oth
e
r han
d
, produces
a
slower speed and torque response for a
ll tested speed operation
.
All schemes
show similar per
f
ormance for
lo
ad
disturban
c
e reject
ion capab
ility.
Keyword:
Fi
el
d O
r
i
e
nt
ed
C
ont
r
o
l
(
F
O
C
)
SVP
W
M
I
ndu
ctio
n Mo
t
o
r Dr
iv
e
PI con
t
ro
ller
Spee
d C
ont
r
o
l
Copyright ©
201
4 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
:
M
.
H.
N Talib
,
Facu
lty of Electri
cal Engineering,
Un
i
v
ersiti Tekn
ik
al Malaysia Melak
a
,
H
a
ng
Tu
ah
Jay
a
, 761
00
Du
r
i
an
Tun
g
g
a
l, Melak
a
, Malaysia.
Em
a
il: h
a
iru
l
n
i
za
m
@
u
t
em
.ed
u
1.
INTRODUCTION
V
ect
or c
o
nt
r
o
l
or
fi
el
d
o
r
i
e
nt
e
d
c
ont
rol
(F
O
C
) d
r
i
v
e
i
s
on
e o
f
t
h
e
m
o
st
po
p
u
l
a
r c
h
oi
ces o
f
vari
a
b
l
e
spee
d
dri
v
e a
p
pl
i
cat
i
on i
n
d
u
s
t
ri
es. Si
nce t
h
e ad
ve
nt
o
f
i
n
di
rect
F
O
C
i
n
7
0
'
s
, t
h
e
p
r
op
ort
i
o
nal
i
n
t
e
gr
al
(P
I)
cont
rol
l
e
r
sc
he
m
e
has
bee
n
wi
del
y
use
d
i
n
va
ri
abl
e
s
p
ee
d
d
r
i
v
e
m
o
t
o
r
.
H
o
weve
r
,
t
h
e
r
e are
se
veral
t
y
pes
of
cont
rol
l
e
r
sche
m
e
such
as P
I
cont
rol
,
f
u
zzy
l
ogi
c c
ont
ro
l,
artificial in
telli
g
e
n
t
con
t
ro
l an
d v
a
riab
le st
ru
cture
co
n
t
ro
lled wh
i
c
h
can
be u
tilized
to
g
e
t the b
e
st
p
e
rforman
ce o
f
th
e
m
o
to
r [1
]-[7
]
.
Th
e m
a
in
reaso
n
PI
cont
roller is
well accepted is due t
o
t
h
e si
m
p
le struct
ure
which can
be
easily underst
o
od a
n
d im
ple
m
ented.
Thi
s
t
ech
ni
q
u
e
i
s
a
b
l
e
t
o
i
n
d
e
pen
d
e
n
t
l
y
co
nt
r
o
l
t
h
e
t
o
r
q
u
e
a
n
d
t
h
e
fl
u
x
-
p
r
o
d
u
ci
n
g
co
m
ponent
o
f
t
h
e st
at
o
r
current i
n
a
wi
de s
p
ee
d ra
nge
.
H
o
w
e
v
e
r
,
in
o
r
d
e
r
to
e
n
s
u
r
e
th
e
PI
c
o
n
t
ro
l
l
e
r
t
o
w
o
rk
ef
ficien
tly
, th
e v
a
l
u
e o
f
propo
rtion
a
l g
a
in
(Kp
)
and i
n
t
e
g
r
al
ga
i
n
(Ki
)
m
u
st
be t
une
d co
rrect
l
y
.
The per
f
o
r
m
ance of t
h
e
m
o
t
o
r real
l
y
depen
d
s
on t
h
e
gai
n
of
the PI c
ontrollers.
Howe
ver
,
in m
o
st cases,
these
gain
s are d
e
term
in
ed
by a trial and
error tun
i
ng
techn
i
qu
e
whic
h re
quires
practical expe
rience a
nd m
a
y lead to ti
m
e
con
s
um
pt
i
on.
Eve
n
t
h
ou
g
h
, t
h
ere a
r
e
num
bers o
f
t
uni
n
g
t
e
c
h
ni
q
u
e s
u
c
h
as
Zi
egl
e
r
-
Ni
c
hol
s
m
e
t
h
o
d
s a
n
d
fi
rs
t
or
der
pl
us t
i
m
e
del
a
y
m
e
t
hod
, ce
rt
ai
n
kn
o
w
l
e
d
g
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-86
94
I
J
PED
S
Vo
l.
5
,
No
.
2
,
O
c
t
o
b
e
r 201
4 :
2
19 –
22
9
22
0
o
f
p
r
o
cess
co
n
t
r
o
l
is r
e
qu
ir
ed an
d
ev
en
t
h
at w
ill
no
t
en
su
r
e
th
e b
e
st
con
t
ro
l p
e
rf
or
m
a
n
ce
[2
], [
3
], [
8
]-[1
0
]
. O
n
t
h
e ot
he
r
han
d
,
t
h
e ge
neral
se
con
d
or
der m
e
tho
d
of
fers
sim
p
ler technique
and m
o
re m
a
the
m
atical form
u
l
ation
ap
pro
ach
ed meth
od
.
Th
u
s
, th
is m
e
th
o
d
h
a
d
b
e
en app
lied in gettin
g all
th
e PI v
a
l
u
es fo
r th
e an
alysis in th
is
pape
r
.
Indirect FOC
method
itself
faced
a
proble
m
with param
e
ter varia
tion cause
d by
the
m
o
tor
heating
phe
n
o
m
e
non
a
n
d
sat
u
rat
i
o
n
[
1
1]
.
T
h
i
s
va
ri
a
t
i
on
cause
s
de
t
uni
n
g
p
r
obl
e
m
i
n
t
h
e
dec
o
upl
i
n
g
o
p
e
r
at
i
o
n
a
n
d
produce e
r
rors
in the m
o
tor out
put
va
lues.
Thus, a robust
controller
des
i
gne
d is neces
sary to ada
p
t with the
p
a
ram
e
ter v
a
ri
atio
n
s
and
d
e
co
up
ling
op
eratio
n. In
add
itio
n, it ab
le to
p
r
od
u
ce
robu
st so
l
u
tio
n
b
y
app
l
yin
g
t
h
e
integral
of tim
e
m
u
ltiplied by the abs
o
lute
of t
h
e error (IT
AE
) criteri
on
m
e
thod [2], [10].
C
o
nve
n
tional or
l
i
n
ear P
I
c
ont
r
o
l
l
e
r
does
n
o
t
have
o
u
t
p
ut
m
a
gni
t
u
de l
i
m
i
t
e
rs,
w
h
i
c
h c
o
ul
d ca
use
dam
a
ge t
o
t
h
e
real
s
y
st
e
m
d
u
e
t
o
relatively lar
g
e ou
tpu
t
v
a
lu
e. In
tro
d
u
c
ing
in
teg
r
ato
r
limiter an
d
sat
u
ration
li
miter p
r
o
v
i
de so
m
e
protection to the system
. Howeve
r
,
t
h
is sat
u
ration lim
iter accum
u
lates error
,
t
hus
producing lar
g
e
ove
rshoot
,
slo
w
settlin
g time an
d so
m
e
ti
mes in
stab
ility to th
e system
[3
], [4
],
[12
]-[15
].
Thu
s
, PI
co
n
t
ro
ller with an
ti
wind up wa
s introduce
d
.
T
h
ere are se
v
e
ral
An
ti-W
i
nd
up
PI con
t
ro
llers
to
so
lv
e t
h
is wind
up
ph
eno
m
en
o
n
suc
h
as
A
W
PI
wi
t
h
dea
d
zo
ne,
A
W
P
I
co
n
d
i
t
i
on,
A
W
P
I
wi
t
h
t
r
ac
ki
n
g
and m
a
ny
m
o
re. M
o
st
o
f
t
h
e
pape
r
s
di
scuss
o
n
t
h
e
ant
i
wi
n
d
u
p
sc
hem
e
i
n
sol
v
i
n
g wi
nd
up
p
h
e
nom
eno
n
i
ssue
and i
t
s
im
pro
v
e
m
e
nt
. B
a
sed on t
h
e
co
m
p
arativ
e st
u
d
y
on th
e an
ti wi
n
dup
strateg
i
es, th
e
A
W
PI cond
itio
n techn
i
qu
e
fo
und
t
o
b
e
th
e m
o
st suitab
l
e
fo
r
us
ual
a
ppl
i
cat
i
on
d
u
e t
o
t
h
e
pe
rf
orm
a
nce res
u
l
t
s
,
si
m
p
l
e
st
ruct
ure
a
n
d l
e
ss
param
e
ter c
o
nt
rol
l
e
d
[
3
]
-[4]
,
[1
2]
-[
1
3
]
,
[
1
6]
. M
o
st
of
t
h
e
pape
rs
di
sc
uss
e
d
onl
y
on
t
h
e
PI a
n
d
A
n
t
i
-
wi
n
d
u
p
per
f
o
r
m
ance at
rat
e
d
spe
e
d
rang
e.
In th
is
p
a
p
e
r
,
th
e
PF
an
alysis is ad
ded
in
t
h
e an
aly
s
is in
v
a
riou
s sp
eed
ran
g
e
d
e
man
d
s
.
Th
e
p
r
e filter
schem
e
i
s
abl
e
t
o
get
ri
d
t
h
e
u
n
wa
nt
ed
zer
o i
n
t
h
e
cl
ose
d
l
o
op
sy
st
em
[3]
,
[1
0]
.
In t
h
i
s
p
r
o
j
ect
,
t
h
e PI c
o
nt
rol
l
er desi
g
n
i
s
a
d
o
p
t
e
d
base
d
on t
h
e sec
o
n
d
or
der
sy
st
em
desi
gn
w
h
i
c
h
has a
sim
p
ler
technique
and direct
m
a
the
m
atical form
ulat
i
on i
n
c
o
m
p
ari
s
o
n
t
o
t
h
e
cl
assi
cal
gai
n
t
uni
n
g
m
e
t
hod
o
r
sy
m
m
e
t
r
i
c
opt
i
m
um
cri
t
e
ri
on[3]
,
[9]
-
[
1
0
]
,
[1
7]
.
The pe
rf
orm
a
nce
res
u
l
t
s
o
f
m
o
t
o
r beha
vi
o
r
s un
de
r
wid
e
sp
eed
ran
g
e
op
erati
o
n
an
d lo
ad
d
i
stu
r
b
a
n
c
e are
an
al
yzed
b
a
sed
on th
e PI,
An
ti-W
i
n
dup
an
d Pre-Filter
tech
n
i
qu
es.
A
s
f
a
r
as th
e au
t
h
or
s' k
now
ledg
e,
no
wor
k
has b
e
en
r
e
po
r
t
ed
o
n
an
alyzin
g th
e sp
eed
co
n
t
r
o
l
m
o
t
o
r perf
o
r
m
a
nce base
d on t
h
i
s
t
h
ree
cont
rol
t
ech
ni
que
s t
oget
h
er
i
n
di
f
f
ere
n
t
spee
d dem
a
nd
ran
g
e
d
q
u
a
n
titativ
ely
.
2.
IND
I
RE
CT F
I
ELD O
R
IE
NTED
CON
T
ROL DR
IV
E
Th
e FOC im
i
t
a
t
es th
e co
n
c
ep
t o
f
sep
a
rately e
x
cited
d
c
m
o
to
r d
r
i
v
e. Thro
ugh
th
is co
n
c
ep
t,
th
e to
rq
ue
an
d th
e
f
l
ux
ar
e con
t
ro
lled b
y
two
ind
e
p
e
nd
en
t or
thog
on
al
v
a
r
i
ab
les kn
own
as t
h
e ar
m
a
tu
r
e
an
d f
i
eld
cur
r
ent
s
. Fi
gu
r
e
1
sh
ow
s t
h
e
bl
oc
k
di
ag
ram
of
i
n
di
rect
F
O
C
schem
e
.
Fi
gu
re 1.
I
ndi
r
ect
FOC
bl
oc
k di
ag
ram
By applying s
p
ace vector transform
a
tion to a three-phase sy
ste
m
, th
e dyna
mic behavior
of induction
m
o
to
r can
b
e
rep
r
esen
ted in
math
e
m
atica
l
eq
u
a
tion
s
as in (1
)-(4
) in
syn
c
hrono
u
s
ro
tatin
g fram
e
[18
]
,
[19
]
.
St
at
or v
o
l
t
a
ge equat
i
o
ns:
̅
V
R
I
̅
ω
φ
(1
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
PED
S
I
S
SN
:
208
8-8
6
9
4
Co
mp
arison
An
a
l
ysis
o
f
I
n
d
i
rect FO
C
I
ndu
ctio
n
Mo
to
r Drive u
s
i
n
g PI, Anti-
Win
d
u
p
and… (M.H
.N
Ta
li
b
)
22
1
Ro
to
r Vo
ltag
e
eq
u
a
tion
s
:
V
0R
I
̅
dφ
dt
ω
ω
φ
V
0R
I
̅
dφ
dt
ω
ω
φ
(2
)
Stato
r
Flux
equ
a
tio
ns:
φ
L
I
̅
L
I
̅
φ
L
I
̅
L
I
̅
(3
)
Ro
to
r Flux
equatio
n
s
:
φ
L
I
̅
L
I
̅
φ
L
I
̅
L
I
̅
(4
)
Whe
r
e
V
,
I
̅
,
φ
,
are t
h
e voltages, current a
n
d
flux
.
Mean
wh
ile sub
s
cri
p
t d,
q
represen
t t
h
e
dq
ax
is
wh
ile s and
r
rep
r
ese
n
t
st
at
o
r
an
d r
o
t
o
r c
o
m
ponent
. T
h
e
st
at
or an
d
rot
o
r resi
st
ance
an
d i
n
duct
a
nce a
r
e de
n
o
t
e
d as
R
s
, R
r
and Ls, L
r
,
whereas Lm
is
the
m
u
tual inductance.
ω
s a
n
d
ω
r
r
e
pr
esen
t th
e
syn
c
hr
ono
us
s
p
eed a
n
d m
echanical
spee
d re
spectively.
In the s
p
ace
vector approac
h
ed, the electrom
a
gnetic
torque
, Te produced by the motor can be
exp
r
esse
d i
n
t
e
rm
s of fl
ux
an
d
cu
rre
nt
as
fol
l
ows;
T
3
2
P
2
φ
I
̅
φ
I
̅
T
T
J
dω
dt
B
ω
(5
)
Whe
r
e P
,
T
L
,
J and
B d
e
no
te
th
e nu
m
b
er of
p
o
l
es, ex
tern
al
lo
ad, in
ertia an
d
friction
o
f
t
h
e IM cou
p
l
ed with
t
h
e pe
rm
anent
m
a
gnet
dc
-m
achi
ne
res
p
ect
i
v
e
l
y
.
In th
is system
, th
e
ro
tating
co
ord
i
n
a
te
referen
ce
fram
e
h
a
v
i
ng
d
i
rect ax
i
s
is align
e
d wi
th
th
e ro
tor
fl
u
x
vect
or t
h
a
t
rot
a
t
e
s at
t
h
e
st
at
or
fre
que
n
c
y
.
If t
h
e
q-c
o
m
ponent
of t
h
e rot
o
r
fl
u
x
i
s
assum
e
zero a
nd t
h
e
el
ect
rom
a
gnet
i
c
t
o
r
q
ue e
x
p
r
es
si
on
bec
o
m
e
s:
T
3
2
P
2
L
L
I
̅
I
̅
(6
)
B
a
sed o
n
t
h
e r
o
t
o
r v
o
l
t
a
ge
q
u
ad
rat
u
re a
x
i
s
equat
i
o
n
of
I
M
, t
h
e r
o
t
o
r fl
ux l
i
nka
ge ca
n
be est
i
m
at
ed
usi
n
g t
h
i
s
f
o
rm
ul
a;
1
(7
)
Whe
r
e,
τ
is th
e
ro
tor tim
e co
n
s
tan
t
.
The slip freque
ncy
ω
is ob
tain
ed
fro
m
th
e ro
t
o
r
v
o
ltag
e
d
i
rect ax
is equ
a
tio
n
b
y
:
(8
)
Th
e ro
tor flux
p
o
s
ition
,
θ
fo
r coo
r
di
nat
e
t
r
an
s
f
o
r
m
i
s
generat
e
d f
r
om
t
h
e i
n
t
e
grat
i
o
n o
f
r
o
t
o
r s
p
ee
d,
ω
and
slip fr
equen
c
y,
ω
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-86
94
I
J
PED
S
Vo
l.
5
,
No
.
2
,
O
c
t
o
b
e
r 201
4 :
2
19 –
22
9
22
2
(9
)
The F
O
C
i
s
co
m
posed o
f
t
w
o
i
nne
r cu
rre
nt
l
o
o
p
s
fo
r fl
ux a
nd t
o
r
q
u
e
co
nt
r
o
l
.
T
h
e o
u
t
e
r s
p
eed l
o
o
p
i
s
cascad
ed
with
th
e to
rq
u
e
cu
rren
t
lo
op
. Th
e ou
tpu
t
o
f
t
h
is
curre
nt loop re
gulate are tr
ansf
o
r
m
e
d i
n
t
o
st
at
iona
ry
refe
rence
fram
e
v
o
ltage by
dq t
o
αβ
tra
n
s
f
orm
a
tion. T
h
en, t
h
ese re
ference
voltage
s
are fe
d to SVPW
M
m
odul
at
i
on
pr
o
cess t
o
ge
nerat
e
p
u
l
s
e
wi
t
h
m
o
d
u
l
a
t
i
o
n
si
g
n
a
l
fo
r i
n
ve
rt
er.
3.
CONTROLLER DE
SIGN
Based
o
n
th
e
math
e
m
atica
l
m
o
d
e
l o
f
t
h
e
th
ree
ph
ase
IM, all th
e cu
rren
t l
o
op
and
sp
eed
o
f
PI
cont
rol
l
e
r
are
c
a
l
c
ul
at
ed
by
us
i
ng a
sec
o
n
d
o
r
de
r sy
st
em
for
a st
ep
i
n
put
.
Al
l
t
h
e
val
u
es
fo
r
pr
o
p
o
r
t
i
o
n
a
l
(K
p)
an
d in
tegral
(Ki)
g
a
in
s of t
h
e thr
ee
PI con
t
ro
llers are determin
ed
b
y
com
p
ari
n
g
t
h
e
ge
neral
sec
o
n
d
or
der
sy
st
em
wi
t
h
t
h
e cl
ose l
o
o
p
bl
ock
di
a
g
ram
t
r
ansfe
r
fu
nct
i
o
n
.
3.1. PI Contr
o
ller
Scheme
B
a
sed
on
t
h
e
m
o
t
o
r Eq
uat
i
o
n
(1
), i
n
sy
nch
r
o
n
ous
re
fere
n
ce fram
e
t
h
e
b
l
ock
di
a
g
ram
of
t
o
r
q
ue a
n
d
fl
u
x
c
o
m
pone
n
t
l
o
o
p
ca
n
be si
m
p
li
fi
ed as i
n
Fi
gu
re
2 a
n
d Fi
gu
re
3.
Fi
gu
re
2
.
Si
m
p
l
i
f
i
e
d t
o
r
q
ue c
o
m
pone
nt
cu
rr
ent
l
o
op
co
nt
r
o
l
Fi
gu
re
3
.
Si
m
p
l
i
f
i
e
d
fl
u
x
c
o
m
ponent
c
u
r
r
e
n
t
l
o
op
co
nt
r
o
l
The cl
o
s
ed l
o
o
p
s e
quat
i
o
ns
f
o
r t
o
r
q
ue an
d
fl
u
x
com
p
onent above a
r
e s
h
own i
n
E
quati
on (10) a
n
d
(1
1)
.
I
∗
I
1
R
s
1
1
1
R
s
1
(1
0)
I
∗
I
1
s
1
1
1
s
1
(1
1)
Whe
r
e
τ
a
n
d
τ
is the stator a
n
d
rot
o
r tim
e constant
res
p
ectively. T
h
e s
p
ee
d loop
bloc
k
diagram
is illu
strated
in Figu
re 4 is b
a
sed
o
n
th
e m
echan
ical m
o
to
r eq
u
a
tion
.
1
R
s
1
I
q
s
I
q
s
1
s
1
I
ds
*
I
ds
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
PED
S
I
S
SN
:
208
8-8
6
9
4
Co
mp
arison
An
a
l
ysis
o
f
I
n
d
i
rect FO
C
I
ndu
ctio
n
Mo
to
r Drive u
s
i
n
g PI, Anti-
Win
d
u
p
and… (M.H
.N
Ta
li
b
)
22
3
Fi
gu
re
4
.
Si
m
p
l
i
f
i
e
d s
p
ee
d l
o
o
p
c
o
nt
r
o
l
Whe
r
e
τ
a i
s
t
h
e m
o
t
o
r m
echani
cal
t
i
m
e
cons
t
a
nt
an
d t
o
r
que
co
nst
a
nt
,
Kt
i
s
gi
ve
n a
s
:
K
3
2
P
2
L
L
I
̅
(1
2)
The s
p
ee
d clos
ed loop tra
n
s
f
e
r
function is
given as
below:
ω
∗
ω
K
s
1
1
K
s
1
(1
3)
The
de
nom
i
n
at
or
o
f
t
h
e
ge
ne
r
a
l
seco
nd
o
r
de
r
sy
st
em
i
s
go
ve
rne
d
by
;
s
2
ς
ω
ω
(1
4)
Whe
r
e
ω
n i
s
the nat
u
ral
fre
q
u
ency
o
f
t
h
e c
l
osed
-l
o
op sy
st
em
and
ς
is th
e d
a
m
p
in
g
ratio
. By co
m
p
arin
g
th
e
den
o
m
i
nat
o
r o
f
t
h
e cl
ose
d
l
o
o
p
t
r
a
n
sfe
r
f
unct
i
o
n
wi
t
h
Eq
uat
i
on
(1
4)
,
t
h
e val
u
e
of
Kp a
nd
Ki
can b
e
d
e
term
in
ed
. The g
a
in
s of th
e PI con
t
ro
ller are sh
own
in
Ta
ble 1. T
h
e values are obtain
based on the equation
above wit
h
ς
i
s
set
at
1 and
ω
n i
s
set
at
10
0Hz
,
1
0
Hz a
n
d 1
H
z f
o
r t
o
r
q
ue l
o
op
, fl
u
x
l
o
o
p
an
d s
p
eed
l
oop
respectively.
Tabl
e 1. PI
C
o
nt
r
o
l
l
e
r
Pa
ram
e
t
e
rs
PI Controller
Kp
Ki
Speed Controller
0.
13
0.
4252
Flux Contr
o
ller
4.
65
8.
94
T
o
r
que Contr
o
ller
13.
4
197.
45
3.2. PI Contr
o
ller
with
Anti-Wind Up Scheme
Th
e m
a
in
o
b
j
ectiv
e o
f
th
e AW
sch
e
m
e
is
t
o
avo
i
d
th
e over v
a
lu
e or satu
ration
v
a
l
u
e in
in
teg
r
at
o
r
whi
c
h ca
uses
hi
g
h
ove
rs
ho
ot
an
d l
o
n
g
set
t
l
i
ng t
i
m
e. Larg
e st
ep c
h
a
nge
or
l
a
rge
e
x
t
e
rn
al
l
o
ad
di
st
u
r
b
a
nce
appl
i
e
d ca
uses
t
h
e PI c
ont
rol
l
er sat
u
rat
e
. Thi
s
wi
n
d
u
p
phe
n
o
m
e
non
resul
t
s
i
n
i
n
co
nsi
s
t
e
n
c
y
bet
w
een t
h
e
real
pl
ant
i
n
put
an
d t
h
e co
nt
r
o
l
l
e
r o
u
t
p
ut
. I
n
or
der t
o
o
v
erc
o
m
e
t
h
e wi
nd
up
pr
obl
em
, t
h
e i
n
t
e
gral
st
at
e i
s
sep
a
rately con
t
ro
lled
[4
],
[15
]
, [18
]
. Th
us, ad
d
ition
a
l in
tegr
al con
t
ro
l in
ad
d
e
d
to
ju
stified
on
th
e
PI co
ntro
ll
er
out
put
i
s
sat
u
ra
t
e
d o
r
n
o
t
base
d
on
t
h
e a
n
t
i
-
w
i
nd
up
st
r
u
ct
u
r
e
i
n
Fi
gu
re
5.
Fig
u
re
5.
AWPI C
o
nd
itio
n
a
l
In
teg
r
ation
Sche
m
e
ε
rr
0
U
o
K
s
1
ω
r
*
ω
r
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-86
94
I
J
PED
S
Vo
l.
5
,
No
.
2
,
O
c
t
o
b
e
r 201
4 :
2
19 –
22
9
22
4
3.3.
Anti Wind Up Sc
heme
with Pre Filte
r
In
o
r
d
e
r to
h
a
v
e
a
p
u
re seco
nd
o
r
d
e
r syste
m
in
th
e sp
eed
clo
s
ed-loo
p, a pre-filter as sho
w
n
i
n
Equ
a
tio
n (1
5) i
s
add
e
d
in seri
es with th
e syst
e
m
[2
],
[3
],
[1
0].
G
K
k
S
K
(1
5)
By in
sertin
g
the p
r
e-filter b
l
ock
,
th
e b
e
h
a
v
i
o
r
of th
e clo
s
ed
lo
op
sp
eed
lo
op
system is
eq
u
a
l to
th
e
desi
re
d
pu
re se
con
d
o
r
de
r sy
st
em
. It
i
s
abl
e
to
can
cel th
e unwan
ted
zero fro
m
th
e lo
op
g
a
in
.
4.
RESULTS
A
N
D
DI
SC
US
S
I
ON
Th
e
p
e
rfo
r
m
a
n
ce co
m
p
arison b
e
tween PI co
n
t
ro
ller,
PI co
n
t
ro
ller with
an
ti-wi
n
dup
an
d pre-filter
schem
e
s i
s
cond
uct
e
d
usi
n
g
Dspace
11
0
3
cont
rol
l
e
r.
A
th
ree parallel i
n
su
lated
b
i
po
l
a
r tran
sistor (IGBT)
intelligent power m
odule (SEMiX252GB
126HDs
) are
use
d
for the inve
rter. T
h
e param
e
ters of a 1.5kW
in
du
ctio
n m
o
to
r
ar
e
show
n
i
n
Tab
l
e
2
.
Th
e
vo
ltag
e
su
pp
ly is set at rated vo
ltag
e
3
8
0
Vrm
s
an
d
th
e switch
i
ng
fre
que
ncy
i
s
se
t
at
8k
Hz.
The
sam
p
l
i
ng t
i
m
e
i
s
50µ
s. T
h
e t
e
st
s are c
o
n
d
u
ct
ed t
o
eval
uat
e
t
h
e pe
rf
orm
a
nc
e of
t
h
e m
o
t
o
r u
n
d
e
r va
ri
o
u
s s
p
e
e
d o
p
e
r
at
i
on
dem
a
nds an
d
l
o
ad
di
st
ur
ba
n
ce reject
i
o
n.
Fi
gu
re 6
sh
o
w
s t
h
e
har
d
ware e
x
pe
ri
m
e
nt
al
set
up
fo
r TL
I a
n
d FL
I
dri
v
e
sy
st
em
.
Fi
gu
re
6
.
T
h
e
har
d
ware e
x
pe
ri
m
e
nt
al
set
up
Tabl
e 2. In
d
u
ct
i
on
M
o
t
o
r
Pa
ra
m
e
t
e
rs
Motor Specif
i
cations
Value
Rated Voltage
380 V
Rated Fr
equency
50 Hz
Poles 4
Rated Speed
1430 r
p
m
Stator
Resistance
3.
45
Ω
Rotor
Resistance
3.
6141
Ω
Stator
I
nductance
0.
3246 H
Rotor
I
nductance
0.
3252 H
M
a
gnetizing I
nductance
0.
3117 H
I
n
er
tia 0.
02kgm
2
Viscous Fr
iction
0.
001 Nm
/(
r
a
d/s)
4.
1.
Oper
ati
o
n
under
Wide
S
p
eed O
p
eration
Thi
s
t
e
st
i
s
c
o
nd
uct
e
d
du
ri
n
g
no
l
o
a
d
c
o
n
d
i
t
i
on.
F
o
r
t
h
i
s
ex
peri
m
e
nt
set
up,
1
.
5
k
W
B
a
l
dor
t
h
ree
p
h
a
se IM m
o
to
r is coup
led
with
2
.
2
k
W B
a
ld
or
p
e
rm
an
en
t m
a
g
n
e
t DC
mach
in
e.
In
cremen
tal o
p
tical
en
cod
e
r
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
PED
S
I
S
SN
:
208
8-8
6
9
4
Co
mp
arison
An
a
l
ysis
o
f
I
n
d
i
rect FO
C
I
ndu
ctio
n
Mo
to
r Drive u
s
i
n
g PI, Anti-
Win
d
u
p
and… (M.H
.N
Ta
li
b
)
22
5
is u
s
ed
to
m
easu
r
e th
e sh
aft sp
eed wh
ich h
a
s 50
0 pu
lses
per revo
lu
tion
.
Fo
r th
is test, t
h
e m
o
to
r is req
u
ired
to
o
p
e
rate at th
ree d
i
fferen
t
cond
itio
n
s
wh
ich
are stand
s
till,
fo
rward
d
i
rectio
n
and
rev
e
rse d
i
rection
at 50
0rp
m
,
10
0
0
r
p
m
and
14
0
0
r
p
m
oper
a
t
i
ng s
p
eed
. E
v
ery
t
e
st
i
s
re
peat
ed
fo
r t
h
r
ee t
i
m
e
s wi
t
h
di
ffe
re
nt
co
nt
rol
l
e
rs'
schem
e
s. The
fi
rst
t
e
st
i
s
co
n
duct
e
d
usi
n
g c
o
n
v
e
n
t
i
ona
l
p
r
o
portio
n
a
l co
ntro
ller (PI) contro
ller
with
limiters.
Propo
rtion
a
l co
n
t
ro
ller
with
an
ti wind
up
(AWPI) sch
e
m
e
fo
r sp
eed
con
t
ro
ller is ap
p
l
ied
for th
e seco
nd
test
co
nd
itio
n. Th
en
, t
h
e
p
r
e filter (PF) is
add
e
d
i
n
cascad
e
with
th
e sp
eed loo
p
for th
e fi
n
a
l exp
e
rim
e
n
t
al test
.
Fig
u
r
e
7
shows th
e sp
eed
r
e
sp
on
ses at 50
0r
p
m
, 1
0
00r
p
m
an
d
14
00r
p
m
. Th
e
m
o
to
r
is r
e
qu
ir
ed
to
o
p
e
rate fro
m
st
an
dstill to
fo
rward
d
i
rection
at 0
.
97
5s an
d
rev
e
rse it d
i
rect
io
n
at 4
.
2
2
6
s
. Based
on
th
e resu
lts,
th
e m
o
to
r track
s
th
e co
mm
a
n
d sp
eed
with al
m
o
st zer
o sp
eed error
d
u
rin
g
stead
y
state con
d
ition
for all
cont
rol
l
e
rs
. H
o
we
ve
r, di
f
f
er
ent
t
r
ansci
e
nt
beha
vi
o
u
r
s
are
notified s
u
ch
as rise
ti
m
e
, p
e
r
cen
t ov
er
shoo
t and
settlin
g
ti
m
e
. Th
e
d
e
tails p
e
rfo
r
m
a
n
ce resu
lts fro
m
zer
o sp
eed
to fo
rward
d
i
rection
are sh
own
in
Tab
l
e 3
bel
o
w:
Fig
u
re
7
.
Sp
eed
respon
se experim
e
n
t
resu
lts du
ri
n
g
stan
dstill, fo
rward
an
d rev
e
rse
d
i
rectio
n
at
5
00rpm
,
10
0
0
r
p
m
and 14
0
0
r
p
m
speed
o
p
erat
i
o
n (a)
Ove
r
al
l
pe
rf
or
mances (b) Closed
up s
p
eed
re
spo
n
se
at 5
0
0
r
p
m
(c)
Clo
s
ed
up
sp
eed
r
e
spon
se at
10
00r
p
m
(
d
)
Clo
s
ed
u
p
sp
eed
r
e
spon
se at
140
0rp
m
Table
3.
Performance analysis
of PF, PI
a
n
d A
W
co
ntr
o
ller fo
r fo
rwa
r
d
Di
rection
Test Condition
Controller
%OS
Tr(s
)
Ts(s)
500r
p
m
PI 11.
6%
1.
073
1.
743
AW 12.
2%
1.
077
1.
718
PF 0%
1.
625
2.
006
1000r
pm
PI 14.
3%
1.
107
1.
811
AW 7.
4%
1.
091
1.
727
PF 0%
1.
585
1.
977
1400r
pm
PI
18.
3%
1.
101
1.
861
AW 5.
0%
1.
130
1.
640
PF 0%
1.
610
1.
990
M
eanw
h
i
l
e
Ta
bl
e 4
sh
o
w
s t
h
e per
f
o
rm
ance res
u
l
t
s
fr
om
for
w
ar
d t
o
re
ve
rse
ope
rat
i
o
n a
t
4.
22
6s
o
f
t
h
e s
p
eed
dem
a
nd
cha
n
ged:
Table
4.
Performance analysis
of PF, PI
a
n
d A
W
co
ntr
o
ller fo
r reve
rse Dir
ection
Test Condition
Controller
%OS
Tr(s
)
Ts(s)
-
500r
p
m
PI 24.
2%
4.
334
5.
154
AW 11.
6%
4.
362
5.
022
PF 0%
5.
027
5.
427
-
1000r
p
m
PI 43.
4%
4.
383
5.
333
AW 5.
3%
4.
423
4.
923
PF 0%
5.
041
5.
411
-
1400r
p
m
PI
35.
64%
4.
445
5.
375
AW 3.
71%
4.
447
4.
897
PF 0%
5.
045
5.
415
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
5.5
6
-2,000
-1,500
-1,000
-
500
0
500
1,000
1,500
2000
Ti
m
e
(
s
)
Speed (
r
p
m
)
Wr
*
Wr
P
F
Wr
P
I
Wr
AW
1400r
pm
1000r
pm
500r
pm
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-86
94
I
J
PED
S
Vo
l.
5
,
No
.
2
,
O
c
t
o
b
e
r 201
4 :
2
19 –
22
9
22
6
Based
on
th
e resu
lts, th
e
AWPI and
PI co
n
t
ro
ller
produced alm
o
st sim
ilar tim
e
rise response, T
r
.
Mean
wh
ile, PF co
n
t
ro
ller
p
r
od
u
c
es slower
rise ti
m
e
re
sp
o
n
se as well as the settlin
g
ti
m
e
.
Fo
r t
h
e fo
rward
and
r
e
v
e
r
s
e op
eratio
n
s
at 140
0r
pm d
e
m
a
n
d
,
PI co
n
t
ro
ller
sche
m
e
r
eco
rd
ed
th
e h
i
gh
est p
e
r
cen
t ov
er
shoot w
ith
1
8
.3% an
d
3
5
.6
4% resp
ectively. No
ov
ersho
o
t
resu
lts from th
e p
r
e filter co
n
t
ro
ller fo
r
th
e d
e
m
a
n
d
s
ch
ang
e
d
.
Mean
wh
ile, the AW
PI produce lo
wer
o
v
e
rsh
o
o
t
fo
r tho
s
e
co
nd
itio
ns at 5
.
0
%
an
d
3
.
7
1
% resp
ectiv
ely
.
Th
e
b
e
st ch
aracteri
s
tic o
f
AWPI co
n
t
ro
ller is it cap
ab
ilities
to
produ
ce lo
wer
percen
t
o
v
e
rsh
o
o
t
wh
ile
m
a
in
t
a
in
in
g
th
e rise tim
e a
n
d
im
p
r
ov
ing
t
h
e settin
g ti
m
e
. Fro
m
th
e AWPI con
t
ro
ller p
e
rfo
rm
an
ce resu
lts,
b
y
in
creasin
g
t
h
e spee
d
ran
g
e
dem
a
nd, t
h
e perce
n
t
ove
rsh
o
o
t
pa
ram
e
t
e
r i
s
re
duce
d
com
p
ared t
o
t
h
e co
n
v
e
n
t
i
onal
P
I
cont
roller.
I
n
PI c
ont
roller,
a large ste
p
s
p
eed
dem
a
nd
refe
rence
caus
e
the output
of the s
p
ee
d c
o
ntroller
reaches
the
sa
turate limit of curre
nt,
Iq. T
h
is a
n
ti-wi
nd
up phe
nom
enon ca
n
be c
ont
rolled by t
h
e
AWPI
schem
e
. The
A
W
PI sc
hem
e
co
nt
rol
t
h
e i
n
t
e
gral
pa
rt
s f
r
o
m
keep up
i
n
t
e
grat
i
n
g t
h
e er
ro
r a
nd t
h
e c
o
nt
r
o
l
l
e
r
out
put
from
increased. T
h
e
details analysis of the i
n
te
g
r
al
st
age
be
havi
o
r
s i
n
P
I
a
n
d
A
W
P
I
are
e
xpl
a
i
ne
d
base
d
on
t
h
e si
m
u
l
a
t
i
on re
sul
t
s bel
o
w
.
Fi
gu
re
8.
Si
m
u
l
a
t
i
on res
u
l
t
c
o
m
p
ari
ng t
o
rq
ue
cu
rre
nt
res
p
on
se usi
n
g
P
I
a
n
d
A
W
PI
Fi
gu
re 8 s
h
o
w
s t
h
e si
m
u
l
a
t
i
on res
u
l
t
s
of
curr
ent
t
o
rq
u
e
com
pone
nt
s resp
on
se w
h
i
c
h com
p
are
s
bet
w
ee
n co
n
v
e
n
t
i
onal
P
I
an
d
A
W
PI
whe
n
a
st
ep f
unct
i
o
n d
e
m
a
nd i
s
ap
pl
i
e
d f
r
om
zero s
p
eed t
o
1
4
0
0
r
p
m
i
n
fo
rwa
r
d an
d re
verse
di
rect
i
o
n
.
Sim
i
l
a
r param
e
t
e
rs, cont
r
o
l
l
ers and s
p
ee
d dem
a
nd are us
ed i
n
t
h
i
s
sim
u
l
a
t
i
on
an
d ex
p
e
r
i
m
e
n
t
. In th
e conven
tio
n
a
l
PI sche
m
e
, th
e in
te
g
r
al state b
e
co
mes larg
e at t
h
e
star
t of
lin
ear
reg
i
on
because it accum
u
lates the speed e
r
ror,
e
v
e
n
in sat
u
ration regi
on. Thus
,
it produces excessive inte
gra
l
state
resu
lts i
n
a larg
e
o
v
e
rsh
o
o
t
.
Mean
wh
ile in
th
e AWPI the
in
teg
r
al state
work on
ly wh
en
th
e inp
u
t
and ou
tpu
t
satu
ration
d
i
fferen
t
is
v
a
rn
ish
e
d
.
Th
erefo
r
e, it is ab
le
to
redu
ce ov
ershoo
t
sig
n
i
fican
tly an
d
also
m
a
in
tai
n
s the
ri
se t
i
m
e
respo
n
se, T
r
r
e
sul
t
s
as PI c
o
nt
rol
l
er. T
h
e s
p
eed
cont
rol
per
f
o
r
m
a
nce i
s
m
u
ch i
m
prove
d
by
A
W
P
I
sch
e
m
e
with
reg
a
rd
s t
o
th
e larg
e sp
eed
ch
ange.
(a)
(b
)
Fig
u
r
e
9. (
a
)
Si
m
u
latio
n
r
e
su
l
t
co
m
p
ar
in
g tor
q
u
e
cu
rr
en
t r
e
sp
on
se u
s
i
n
g
PI
at
1
000
rp
m
a
n
d 140
0rp
m
d
e
m
a
n
d
;
(b
) R
e
feren
ce step
i
n
pu
t
d
e
m
a
n
d
an
d pre
filter outp
u
t
d
e
m
a
n
d
Fi
gu
re 9
(
a) sh
ow t
h
e si
m
u
l
a
ti
on res
u
l
t
s
co
m
p
ari
ng t
o
r
q
u
e
curre
nt
com
p
o
n
e
n
t
s
fo
r PI
cont
r
o
l
l
e
r at
10
0
0
r
p
m
and
14
0
0
r
p
m
dem
a
nd
. D
u
ri
n
g
f
o
r
w
ar
d
op
erat
i
o
n
,
t
h
e i
n
t
e
g
r
a
l
st
at
e out
p
u
t
i
s
2.
85
A a
n
d
5A
f
o
r
1
000
rp
m
an
d
14
00r
p
m
r
e
sp
ectiv
ely. Th
ese resu
lts 14
.3
% an
d
18
.3
%
o
f
speed
p
e
r
c
en
t over
s
hoo
t r
e
sp
ect
iv
ely.
It
m
eans t
h
at
,
t
h
e
hi
g
h
e
r
c
h
ange
d
o
f
s
p
ee
d
dem
a
nd
pr
o
duce
hi
g
h
er
o
v
ers
h
oot
.
Ho
w
e
ver
,
du
ri
n
g
r
e
verse
operation, the
percent of
overshoo
t becam
e 43.4% and
35.64% for
20
00rpm
and 2800rpm
speed changed
0
0.
5
1
1.5
2
2.5
3
3.
5
4
4.5
5
5.5
6
-1
0
-8
-6
-4
-2
0
2
4
6
8
10
Ti
m
e
(
s
)
Iq
re
f (A
)
I
n
t
e
g
r
al
s
t
a
t
e A
W
Iq
r
e
f A
W
I
n
t
e
g
r
al
s
t
a
t
e PI
Iq
r
e
f P
I
0
0.
5
1
1.
5
2
2.
5
3
3.
5
4
4.
5
5
5.
5
6
-1
0
-5
0
5
10
Ti
m
e
(
s
)
Iq
re
f
fo
r P
I
c
o
n
t
ro
ll
e
r
(A
)
Iq
r
e
f
1
4
0
0
r
p
m
I
n
t
e
g
r
al
s
t
ate 1
4
0
0
r
p
m
Iq
r
e
f
1
0
0
0
r
p
m
I
n
t
e
g
r
al
s
t
ate 1
0
0
0
r
p
m
0
1
2
3
4
5
6
-1
500
-1
000
-
500
0
500
1
000
1
500
Ti
m
e
(
s
)
S
p
e
e
d
(rp
m
)
Wr s
t
e
p
r
e
f
Wr r
e
f
PF
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
PED
S
I
S
SN
:
208
8-8
6
9
4
Co
mp
arison
An
a
l
ysis
o
f
I
n
d
i
rect FO
C
I
ndu
ctio
n
Mo
to
r Drive u
s
i
n
g PI, Anti-
Win
d
u
p
and… (M.H
.N
Ta
li
b
)
22
7
resp
ectiv
ely. Th
is resu
lt was
affected
b
y
in
t
e
g
r
al limiter in
teg
r
al li
m
iter
wh
ich
is set at 1
0
A. Th
e i
n
tegral state
o
u
t
p
u
t
is clam
p
at
1
0
A
for
n
e
g
a
tiv
e
1
400
rpm
sp
eed
fro
m
4
.
3
s
u
n
til 4.55
s. Th
is actio
n, co
n
t
ro
l t
h
e Iq
referen
ce
dem
a
nd f
r
om
pr
o
duci
n
g
hi
gh
er o
v
e
r
sh
o
o
t
c
o
m
p
ared t
o
ne
gat
i
v
e
10
0
0
r
p
m
dem
a
nd cha
nge
.
As a
res
u
l
t
,
l
o
we
r
p
e
rcen
t o
v
e
rsho
o
t
with
h
i
gher
sp
eed
d
e
m
a
n
d
ch
ang
e
s h
a
pp
en
ed.
The
in
teresting
p
a
rt o
f
th
e p
r
e
filter
co
n
t
ro
ller is it
cap
ab
ilities to
main
tain
zero
o
v
e
rsh
o
o
t
fo
r
all th
e sp
eed
de
m
a
n
d
ran
g
e
. Fig
u
re 9(b) sho
w
s the
si
m
u
latio
n
com
p
ariso
n
b
e
t
w
een
step
i
n
pu
t referen
ce and
sp
eed
referen
c
e after th
e
p
r
e-filter p
r
o
cess
used
as
t
h
e refe
re
nce si
gnal
i
n
PF sc
h
e
m
e
. Due t
o
t
h
e negat
i
v
e e
x
p
one
nt
i
a
l
spee
d
refe
rence
dem
a
nd
by
ad
di
n
g
t
h
e PF
,
slo
w
er speed
respo
n
s
e
resu
lts for PF sch
e
m
e
. Th
is situ
ation
h
a
pp
en
ed
to
all th
e step
referen
ce
rang
e
de
m
a
nd
and
res
u
l
t
s
no
ove
rs
ho
ot
resul
t
s b
u
t
sl
o
w
er
s
p
eed
res
p
on
se.
Figure 10 show the torque curren
t
com
p
o
n
e
nt
, I
q
an
d p
h
a
s
e A cur
r
e
n
t
,
I
a
expe
ri
m
e
nt
r
e
sul
t
s
of t
h
e
step res
p
onse
dem
a
nds. F
r
om
the stand still condition, th
e
m
o
tor operat
ed at 1400rp
m
in forward dire
ction t
o
14
0
0
r
p
m
rever
s
e op
erat
i
o
n. B
a
sed
on t
h
e res
u
l
t
,
alm
o
st
sim
i
l
a
r t
o
rq
ue c
u
r
r
e
nt
, I
q
p
e
r
f
o
r
m
a
nce can
be
no
t
i
f
y
for PI a
n
d
AWPI sc
hem
e
. The torque c
u
rre
n
t reach lim
ited 10A set at t
h
e
spee
d c
o
ntroller trem
endousl
y
. PF
schem
e
pro
d
u
ce sl
ower t
o
r
que c
u
r
r
ent
r
e
sp
onse a
nd
onl
y
reac
h 4
A
am
pl
i
t
ude du
ri
n
g
f
o
r
w
ar
d spe
e
d
com
m
a
nd.
Thi
s
also
results i
n
slowe
r
s
p
eed res
p
onse
of
PF schem
e
. The
sim
ilar results effect ca
n
be s
een for
the phase
a stat
or curre
nt,
Ia.
(a)
(b
)
Fig
u
re
10
. Experim
e
n
t
resu
lts du
ri
n
g
stand
s
t
ill, fo
rwar
d and
rev
e
rse
d
i
rectio
n
at
1
400
rpm
sp
eed
op
erat
io
n
,
(a)
To
rqu
e
cu
rr
en
t r
e
sp
on
se (b
)
P
has
e A c
u
rre
nt
resp
onse
4.
2.
Oper
ati
o
n
under
Load
Conditi
on
Th
e l
o
ad rej
e
ctio
n
cap
a
b
ilitie
s of th
e design
were
inv
e
stig
ated
with th
e no
m
i
n
a
l lo
ad d
i
sturb
a
n
c
e
applied
duri
ng rated spee
d
operation as s
h
own i
n
Figu
re
11. T
h
e loa
d
dis
t
urbance
opera
tion is accom
p
lished
by using a
DC
m
achine attached wit
h
the
l
o
ad bank
. T
h
e
arm
a
ture terminals of t
h
e
pe
rm
anent m
a
gnet DC
machine are c
o
nnected t
o
the resistor
bank. The exte
rnal
resistor
of the
DC
m
ach
in
es i
s
set to
p
r
odu
ce rated
cur
r
ent
l
o
a
d
o
f
IM
. The m
o
t
o
r i
s
ope
rat
e
d a
t
140
0
r
pm
and
sud
d
e
n
rat
e
d l
o
ad
di
st
ur
ba
nc
e i
s
appl
i
e
d at
2.
5s.
From
t
h
e res
u
l
t
s, t
h
e
spee
d
dr
op
pe
d a
b
o
u
t
1
8
0
r
pm
and rec
ove
r
fr
om
t
h
e un
de
rsh
o
o
t
si
t
u
at
i
on
wi
t
h
i
n
1s
fo
r t
h
e
en
tire sch
e
m
e
s
.
It
p
r
ov
en th
at
, all th
e sch
e
mes h
a
s sam
e
ca
p
a
b
ility o
f
lo
ad
d
i
stu
r
b
a
n
ce
rej
ection
during rated
lo
ad
.
Ho
w
e
v
e
r, th
e sp
eed
ch
an
g
e
is
n
o
t
large en
oug
h
t
o
turn-on
th
e sat
u
ratio
n
cond
itio
n o
f
th
e an
ti-w
i
n
dup
syste
m
.
Fig
u
re
11
. Experim
e
n
t
resu
lts with
rated sp
eed
with
n
o
m
in
al lo
ad
d
i
rst
ubu
n
c
e
0
0.5
1
1.5
2
2.
5
3
3.5
4
4.5
5
5.
5
6
6
-12
-10
-8
-6
-4
-2
0
2
4
6
8
10
12
12
Ti
m
e
(
s
)
Iq
(A
)
Iq PF
Iq PI
Iq A
W
0
0.5
1
1.5
2
2.5
3
3.5
4
4.
5
5
5.5
6
6
-12
-10
-8
-6
-4
-2
0
2
4
6
8
10
12
12
Ti
m
e
(
s
)
Iq
(
A
)
Ia PF
Ia PI
Ia A
W
0
1
2
3
4
5
6
0
200
400
600
800
1,000
1,200
1,400
Ti
m
e
(
s
)
Spe
e
d(r
p
m)
WrPF
WrAW
WrPI
Wr*
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-86
94
I
J
PED
S
Vo
l.
5
,
No
.
2
,
O
c
t
o
b
e
r 201
4 :
2
19 –
22
9
22
8
5.
CO
NCL
USI
O
N
Thi
s
pape
r
pre
s
ent
s
t
h
e
spee
d
pe
rform
ance analysis of IFOC perfor
m
a
n
ce resu
lts b
e
tween
PI,
AWPI
and
PF sc
hem
e
cont
rol
at
l
o
w, m
e
di
um
and rat
e
d sp
eed
dem
a
nd. Al
l
t
h
e t
h
ree P
I
c
ont
rol
l
e
r
desi
g
n
a
r
e usi
n
g
seco
nd
o
r
de
r s
y
st
em
desi
gn a
p
p
r
oach
. F
r
om
t
h
e a
n
al
y
s
i
s
, th
e
AWPI is able to
redu
ce
o
v
ershoo
t prob
lem
b
y
co
n
t
ro
lling
th
e in
tegral p
a
rts
fro
m
k
eep
u
p
in
teg
r
ating
t
h
e erro
r at a set
p
r
ed
eterm
i
n
e
li
m
i
ter. Th
e resu
lt is
m
o
re si
gni
fi
ca
nt
especi
al
l
y
in t
e
rm
of per
cent
o
v
ers
h
oot
red
u
ct
i
on at
t
h
e hi
ghe
r spee
d ra
nge
dem
a
nd
. I
n
ad
d
ition
,
th
is
great sch
e
m
e
is
ab
le to
m
a
in
tain
th
e rise tim
e
an
d
im
p
r
ov
ing
th
e settin
g
ti
me co
m
p
ared
to
th
e PI
schem
e
. M
eanwhi
l
e
,
PF sc
he
m
e
resul
t
s
sl
o
w
er
spee
d a
n
d
t
o
r
que
res
p
ons
e pe
rf
orm
a
nces. H
o
weve
r,
t
h
e PF
i
s
abl
e
t
o
pr
od
uc
e pure sec
o
nd
orde
r sy
st
em
i
n
al
l speed
ran
g
e dem
a
nd
s. It
i
s
abl
e
to get
zero
ov
ersh
o
o
t
respon
se with reason
ab
le settlin
g
ti
m
e
respo
n
s
e. Fin
a
l
l
y, fo
r th
e load
d
i
sturb
a
n
c
e rej
ection
ab
ility, al
l
sch
e
m
e
s sh
ow
si
m
ilar p
e
rfo
r
man
ce cap
ab
ility to
with
stan
d th
e
rated
l
o
ad
d
i
stu
r
b
a
n
ce.
ACKNOWLE
DGE
M
ENT
The aut
h
ors
would like to acknowle
d
ge
their gr
atitude to Faculty of Electrical Engineeri
n
g
Un
i
v
ersiti Tekn
ik
al Malaysia Melak
a
for
p
r
o
v
i
d
i
ng
th
e reso
urces and
supp
ort in th
is st
ud
y.
REFERE
NC
ES
[1]
GW Chang, G Espinosa-Perez,
E Mendes, R Ortega, "Tun
ing rules for the PI
gain
s of field-or
iented controllers of
induction
motors,"
IEEE Transactions on I
ndustrial Electronics,V
ol. 47
, No
. 3
,
pp
. 592-602, Jun. 2
000.
[2]
RC Dorf, RH B
i
shop,
Modern C
ontrol Syst
ems,
Tenth Ed
ition
,
P
r
entice H
a
ll, 200
5.
[3]
J Espina, A Arias, J Balcells,
C Or
tega. Speed anti-windup P
I
strateg
i
es re
view for field or
iented
control of
permanent magn
et s
y
nchronous
machines.
Compatibility and
Po
wer Electronics,
2009; 279-285.
[4]
C Jong-Woo, L
Sang-Cheol. Antiwindup St
rateg
y
for PI-Ty
p
e S
p
eed Controller
,"
IEEE Transactions on Industrial
Electronics
. 200
9; 56(6): 2039-2
046.
[5]
Z Ibrahim
,
E
L
e
vi. A
com
p
arat
ive an
al
ys
is
of f
u
zz
y
log
i
c
and
PI speed contro
l in high-p
e
rfor
m
ance AC driv
es
using exper
i
mental approach.
IEEE Transactions
on Industry App
lications
. 2002
;
38(5): 1210-121
8.
[6]
S Kuo-Kai, S
Hsin-Jang. Vari
able
struc
t
ure
c
u
rrent cont
rol f
o
r induction m
o
tor drives b
y
sp
ace vol
tag
e
vec
t
or
PWM.
IEEE T
r
a
n
sactions on Ind
u
strial El
ec
troni
cs.
1995; 42(6):
572-578.
[7]
K Saty
anar
ay
ana, P Surekha, P Vija
y
a
Prasuna. A
new FOC approach of
i
nduction
motor drive usin
g DTC strateg
y
for the minimization of
CMV.
International Journ
a
l of Powe
r
Electronics and
Drive Systems.
2013; 3(2): 241-2
50.
[8] SN
Nise.
Contro
l System Eng
i
neering
, John
Wiley
& Sons, Inc., 2
011.
[9]
M Zele
chowski,
MP Kazm
ierkowski, F Blaab
je
rg. Contro
ller d
e
sign for dir
ect torque
controlled space vecto
r
modulated (DTC-SVM) induction motor drives
.
IEEE International Symposiu
m
on Industrial Electronics,
2005
:
951-956.
[10]
G Foo, CS Goo
n
, MF Rahman. Analy
s
is and d
e
sign of
the S
V
M
direct torqu
e
and
flux contro
l scheme for IPM
s
y
nchronous mo
tors.
Internation
a
l Conferen
ce o
n
Electrical Machines and S
y
stems.
2009: 1-6
.
[11]
KB Nordin, DW Novotn
y
, D
S
Zinger
.
Th
e
Influence
of M
o
tor Parameter
Deviations in F
eedforward Field
Orienta
tion Driv
e S
y
st
em
s.
IEEE Transactions on
Industry App
l
ications.
1985
; IA
-21(4): 1009-10
15.
[12]
C Bohn, DP Atherton. An an
aly
s
is
packag
e co
mparing PID an
ti-windup str
a
tegies.
I
EEE Control Systems
. 199
5;
15(2): 34-40
.
[13]
S Hwi-Beon. New antiwindup PI contro
ller fo
r variable-speed
motor drives.
IEEE T
r
ansactio
ns on Industrial
Electronics.
199
8; 45(3): 445-45
0.
[14]
A Scottedward
Hodel, CE Ha
ll.
Variable-structu
re PID contro
l to
prevent in
tegr
ator windup.
IEEE Transactions on
Industrial Electronics
. 2001
; 48(
2): 442-451.
[15]
HB Shin. Comp
arison and evalu
a
tion of an
ti-win
dup PI controller
s
.
Journal of Po
wer Electronics
. 2011; 11(1): 45-
50.
[16]
Z Ibrahim. Fuzzy
Log
i
c Con
t
rol
of PWM Inverter-Fed Sinusoid
a
l Perman
ent Magnet S
y
nchrono
us Motor Drives.
Ph. D
.
Dissertation, Liverpool
Jo
hn Moores.
1999
.
[17]
S Eun-Chul, P
Tae-Sik
,
O Won-H
y
un,
Y Ji-Yo
on. A design method of PI c
ontr
o
ller for
an
indu
ction motor with
parameter
var
i
ation.
Annua
l Con
f
erence of the IEEE
Industria
l Electr
onics Society.
2003: 408-41.
[18]
M Talib
, Z Ibrah
i
m, N Abdul Rahi
m, A Hasim. Implementati
on o
f
Anti-Windup PI Speed Controller for Induction
Motor Drive Using dSPACE and
Matlab/Simulin
k Environment.
Australian Jour
nal of
Ba
sic
&
Applied
Sc
ienc
e
s
.
2013; 7(1).
[19]
ZM Salem Elb
a
rbar
y
,
M Mohamed Elkholy
.
Performance
an
aly
s
is of field or
ientation of
indu
ction motor dr
iv
e
under open
gate of IGBT fault.
International Jo
urnal of
Power
Electronics and
Drive Systems
.
2013; 3(3): 304-
310.
Evaluation Warning : The document was created with Spire.PDF for Python.