Internati
o
nal
Journal of P
o
wer Elect
roni
cs an
d
Drive
S
y
ste
m
(I
JPE
D
S)
Vol.
4, No. 4, Decem
ber
2014, pp. 517~
527
I
S
SN
: 208
8-8
6
9
4
5
17
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
A Novel Direct Torque Contro
l for Induction Machine Drive
System
with Low Torqu
e
a
nd Flux Ripples using XSG
S
o
uh
a
Bo
uk
ad
id
a
,
So
u
f
ien G
d
a
i
m,
A
b
d
e
lla
t
if
Mt
ib
aa
Laborator
y
EµE
of the FSM, University
of
Monas
tir, Tunisia
Article Info
A
B
STRAC
T
Article histo
r
y:
Received
May 6, 2014
Rev
i
sed
Sep
27
, 20
14
Accepted Oct 11, 2014
The conv
entional Direct To
rque
Cont
rol (DTC)
is known to produce
a quick
and robust r
e
sponse in AC driv
es. Howeve
r, dur
ing stead
y
s
t
ate, stator f
l
ux
and el
ec
trom
agneti
c torqu
e
whic
h res
u
lts
in
inco
rrect
s
p
eed
es
ti
m
a
tions
an
d
acous
ti
cal no
is
e.
A m
odified Direct
Torque Control (DTC) b
y
us
ing Space
Vector Modulation (DTC-SVM)
for induc
tion
machine is prop
osed in this
paper. Using th
is control
strat
e
g
y
,
th
e ripp
les
int
r
oduced
in torq
ue and
flux
are redu
ced
. This paper pr
esents a novel approach to
design and
implementation
of a high p
e
rfromane
torq
ue contro
l (DTC-SVM) of
induction mach
ine using Field
Progr
ammable gate
array
(F
PGA). The
performance of
the proposed control sc
heme is evaluated thro
ugh digital
simulation usin
g Matlab
\
Simulink and
Xilinx S
y
stem Generator.
The
sim
u
lation result
s are used to verif
y
th
e effectiv
eness of the prop
osed control
stra
te
gy
.
Keyword:
DTC-S
V
M
FPGA
I
ndu
ctio
n m
a
c
h
in
e
Matlab
/
Si
m
u
li
n
k
XS
G
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
:
Souha
Boukadi
d
a,
Laboratory Eµ
E of the
FSM
Un
i
v
ersity of
Mo
n
a
stir, Tu
n
i
sia
Em
a
il: b
o
u
k
a
d
i
d
a
so
uh
a@yaho
o.fr
1.
INTRODUCTION
Sin
ce its in
cep
tio
n
,
th
e Direct To
rqu
e
Con
t
ro
l h
a
s g
a
in
ed
p
opu
larity fo
r
in
du
ctio
n
m
a
c
h
in
e driv
es.
Ind
e
ed
, th
e con
t
ro
l v
a
riab
les th
at are th
e st
ato
r
fl
u
x
and
t
o
rqu
e
are calcu
lated
fro
m
th
e q
u
a
n
tities related
to
th
e stato
r
witho
u
t
t
h
e in
terv
en
tio
n of
m
echanical sens
or.
The
res
p
onse
of t
h
e
DTC is
fast,
howe
ve
r it ha
s
som
e
draw
bac
k
s s
u
c
h
as
not
a
b
l
e
t
o
r
q
ue a
n
d
fl
u
x
ri
ppl
es a
n
d t
h
e
va
ri
abl
e
com
m
ut
ati
on f
r
eq
ue
ncy
be
ha
vi
o
r
o
f
t
h
e i
nve
rt
er. M
a
ny
pa
pers
pre
s
ent
e
d
di
ffe
re
n
t
appr
oac
h
es t
o
m
i
nim
i
ze
t
h
e fl
ux a
nd t
o
r
q
u
e
ri
ppl
es
[1]
-
[
4
]
. I
n
[1]
an
d [
3
]
,
el
ect
rom
a
gnet
i
c
t
o
r
que a
n
d fl
u
x
are c
ont
rol
l
e
d di
r
ect
l
y
by
t
h
e sel
ect
i
on
of
a swi
t
c
hi
n
g
v
ect
or
fr
om
a t
a
bl
e sel
ect
i
on. Ne
vert
hel
e
ss, t
h
e selected vector is not always the
best one beca
use only the sector is
considere
d
, where the
flux
sp
ace v
ector lies
with
ou
t co
nsidering
its lo
cati
o
n.
To o
v
erc
o
m
e
t
h
e several
di
sad
v
ant
a
ges o
f
DTC
a new
cont
rol
t
ech
ni
que cal
l
e
d Di
r
ect
Torq
ue
Co
n
t
ro
l –
Sp
ace V
ecto
r
Modulated
(
D
TC-SV
M
)
[5
]-
[6
] is
d
e
v
e
l
o
pp
ed
. In th
is n
e
w
m
e
th
o
d
, th
e d
i
sadv
an
tag
e
s
of t
h
e DTC a
r
e elim
inated. The DTC-SVM
strategies ar
e
based
on t
h
e sa
me fundam
entals as classical DTC;
i
t
pr
ovi
des
dy
n
a
m
i
c beha
vi
or
com
p
arabl
e
wi
t
h
cl
assi
cal
DT
C
.
In
p
r
actice, t
h
e v
ector con
t
ro
l alg
o
rith
m
fo
r an
indu
ction
m
ach
in
e is im
p
l
e
m
en
ted
u
t
ilizin
g
d
i
g
ital
si
gnal
p
r
oces
s
o
r (
D
SP
). T
h
e
DSP co
nt
r
o
l
pr
oce
d
u
r
e i
s
p
e
rfo
r
m
e
d
seq
u
en
tially;
th
is
may resu
lt in
a slo
w
er
cy
cl
i
ng pe
ri
o
d
i
f
com
p
l
e
x al
go
ri
t
h
m
s
are inv
o
l
v
e
d
. Em
pl
oy
i
ng
fi
el
d p
r
og
ram
m
abl
e
gat
e
array
(FP
G
A
)
i
n
im
pl
em
ent
i
ng vect
o
r
co
nt
r
o
l
st
rat
e
gi
es p
r
o
v
i
des ad
vant
a
g
e
s
suc
h
as si
m
p
l
e
r har
d
ware a
nd s
o
ft
ware
de
si
gn
,
rapi
d pr
ot
ot
y
p
i
n
g
,
hence
fast
swi
t
c
hi
n
g
fre
q
u
ency
a
n
d hi
g
h
spee
d
c
o
m
put
at
i
on [7]
-
[
8]
.
The
pape
r
devotes to a c
o
m
p
arativ
e study between t
h
e pe
rform
a
nces of t
w
o a
p
proac
h
es
: (i) Classical
DTC (ii) DTC-SVM.
Th
ese
strateg
i
es are d
e
si
gn
ed u
s
i
n
g Xilinx
System
Gen
e
rat
o
r (XSG) and
M
a
t
l
a
b/
Sim
u
l
i
nk
so
ft
wa
re
pa
ckage
s
a
n
d
i
m
pl
em
ent
e
d o
n
FPG
A c
o
nt
rol
l
er.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-86
94
I
J
PED
S
Vo
l.
4
,
No
.
4
,
D
ecem
b
er
2
014
:
51
7 – 527
51
8
2.
BASI
C P
R
I
N
CIPLE O
F
D
T
C
The m
a
i
n
i
d
ea
of
DTC
i
s
t
o
reco
ve
r t
h
e
r
e
duct
i
o
n
o
f
t
h
e ri
p
p
l
e
s
of t
o
rq
ue a
n
d
fl
ux
,
an
d t
o
ha
ve
sup
e
ri
o
r
dy
na
m
i
c perfo
rm
ances. Fi
g
u
re 1
prese
n
t
a po
ssi
bl
e schem
a
t
i
c
of Di
rect
Tor
q
ue C
o
nt
rol
.
T
h
ere are
t
w
o
di
f
f
ere
n
t
l
o
o
p
s c
o
rres
p
o
ndi
ng
t
o
t
h
e
m
a
gni
t
ude
s
of
t
h
e st
at
o
r
fl
u
x
a
n
d
t
o
r
q
ue.
The e
r
r
o
r
bet
w
een
t
h
e
est
i
m
a
t
e
d st
at
or fl
ux
m
a
gni
t
u
de
φ
s
a
n
d t
h
e
refe
rence
st
at
o
r
fl
ux
m
a
gni
t
u
de
φ
s
*i
s
t
h
e
i
n
put
o
f
a t
w
o
l
e
vel
hysteresis comparat
or whe
r
ea
s the e
rror bet
w
een the estimated torque T
e
and the re
ference torque T
e
* i
s
t
h
e
in
pu
t of a three lev
e
l h
y
steresis co
m
p
arator. The
outputs
of t
h
e stator fl
ux e
r
r
o
r an
d t
o
r
q
ue er
ro
r
hy
st
eresi
s
bl
oc
ks,
t
o
get
h
e
r
wi
t
h
t
h
e
p
o
si
t
i
on
o
f
t
h
e
st
at
o
r
fl
ux
are
use
d
as i
n
put
s
o
f
t
h
e
swi
t
c
hi
ng
t
a
bl
e.
Fi
gu
re
1.
B
l
oc
k
di
ag
ram
of D
T
C
The sel
ect
i
on
vect
o
r
i
s
based
on t
h
e
hy
st
ere
s
i
s
cont
r
o
l
of t
h
e t
o
r
q
ue an
d t
h
e st
at
or fl
ux
. In t
h
e
basi
c
fo
rm
the stator
flu
x
φ
s
is estimated
with
:
0
()
t
ss
s
VR
i
d
t
(1
)
The st
at
or
vol
t
a
ge and st
at
o
r
cu
rre
nt are calculated from the stat
e of three phase (Sa ,Sb ,Sc
)
a
nd
measured curre
nts
(ia, ib
, ic).
24
33
0
24
33
2
(,
,
)
(
)
3
2
(,
,
)
(
)
3
jj
sa
b
c
a
b
c
jj
sa
b
c
a
b
c
VS
SS
E
S
S
e
S
e
ii
i
i
i
i
e
i
e
(2
)
Phase
an
gl
e a
n
d st
at
o
r
fl
u
x
a
m
pli
t
ude
are
c
a
lculated in expressi
on (3).
22
()
s
s
s
ss
s
ar
c
t
g
(
3
)
The devel
o
ped
el
ect
ro
m
a
gnetic t
o
rque T
e
of t
h
e m
a
chi
n
e can be e
v
al
uat
e
d by
Equat
i
on
(4
):
3
()
2
es
s
s
s
Tp
i
i
(4
)
The st
at
or
fl
u
x
vect
or i
s
m
ovi
ng al
o
n
g
a st
ra
i
ght
axi
s
col
i
n
ear t
o
t
h
at
of t
h
e v
o
l
t
a
ge vec
t
or re
qui
red
by
t
h
e i
nve
rt
er:
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
PED
S
I
S
SN
:
208
8-8
6
9
4
A N
o
vel
Di
rect
Tor
q
ue C
ont
r
o
l
f
o
r
I
n
d
u
ct
i
o
n M
a
c
h
i
n
e
Dri
ve Syst
e
m
w
i
t
h
Low
T
o
r
q
ue…
(
S
o
u
h
a
B
o
uka
d
i
da)
51
9
Fi
gu
re
2.
St
at
o
r
fl
ux
vect
or
ev
ol
ut
i
o
n i
n
t
h
e
f
i
rst
sect
or
3.
DTC SP
ACE VECTO
R
M
O
D
U
LATIO
N
Th
e DTC algorith
m
is b
a
sed
o
n
th
e instan
tan
e
ou
s
v
a
lu
es an
d
d
i
rectly cal
cu
lated
th
e g
a
t
e
sig
n
a
ls fo
r
th
e in
verter. Th
e con
t
ro
l algo
rith
m
in
DTC
-
SVM is
b
a
se
d on a
v
e
r
age
va
lues where
a
s the s
w
itching s
i
gnals
(Sa,
Sb a
n
d Sc
) for the
invert
er are
calcu
late
d by
space vector
m
odulator [9]-[11].
3.
1.
Pri
n
ci
pl
e
o
f
V
ect
or
ML
I
For eac
h pe
ri
o
d
o
f
m
odul
at
i
on
of t
h
e i
n
ve
rt
er, t
h
e t
h
ree
pha
se v
o
l
t
a
ges
pr
ovi
ded
by
t
h
e co
nt
r
o
l
al
go
ri
t
h
m
can be e
x
p
r
esse
d i
n
a
fi
xe
d
refe
re
nce
link
e
d to th
e stator
, t
h
roug
h th
eir pr
oj
ect
io
n
s
V
s
α
and
V
s
β
.
The i
n
vert
er
h
a
s si
x swi
t
c
hi
ng cel
l
s
, gi
vi
n
g
ei
ght
p
o
ssi
b
l
e swi
t
c
hi
ng c
o
n
f
i
g
urat
i
o
ns
. These ei
g
h
t
swi
t
c
hi
n
g
c
o
nf
i
g
u
r
at
i
ons
ca
n
be e
x
p
r
esse
d i
n
t
h
e
pl
a
n
e
(
α
,
β
)
by
8
vect
ors
t
e
nsi
o
ns.
Knowing
th
at i
n
th
e
gradu
a
tion
ph
as
e
voltages (Va
,
Vb, Vc
) are
re
prese
n
t
e
d in t
h
e
plane
by a ve
ctor
V
s
. T
h
e
p
r
i
n
ci
pl
e
of
vect
or
M
L
I i
s
t
o
pr
o
j
ect the
desire
d stator voltage
vector
V
s
on
t
h
e t
w
o a
d
jace
nt
vect
ors
co
rresp
ond
ing
to
two
switch
i
n
g
states
o
f
the in
verter
. T
h
e
values
of the
s
e projections
provide t
h
e
de
sired
comm
utation times.
3.2.
Ge
neral
Struc
t
u
re
of the Control
DTC-SVM
M
o
st
exi
s
t
i
ng
bl
oc
ks i
n
t
h
e c
ont
rol
D
T
C
-
S
V
M
are i
d
en
tical to
th
o
s
e of co
n
t
ro
l DTC as sh
own
in
th
e
fo
llowing
figu
re
(3
).
Th
e n
e
w b
l
o
c
k
s
will
b
e
d
i
scu
s
sed
b
e
l
o
w.
Fi
gu
re
3.
B
l
oc
k
di
ag
ram
of D
T
C
-
SVM
3.
3.
Cal
c
ul
ati
o
n
of
ti
me
o
f
a
ppl
i
c
ati
o
n
of
t
h
e st
at
us
of
t
h
e i
n
ver
t
er
Each m
odul
at
i
on pe
ri
o
d
T
mo
d
of the invert
er, the project
ed vector V
s
on the two adja
cent vectors
assu
res th
e switch
i
n
g
tim
e o
f
calcu
latio
n
.
The
key
st
ep
of t
h
e S
V
M
t
echni
que
i
s
t
h
e
det
e
rm
i
n
at
i
on of
T
i
and T
i+1
du
ri
n
g
eve
r
y
m
odul
at
i
o
n
peri
od T
mo
d
. To illustrate the
m
e
thodol
ogy
we conside
r
the case where
V
s
can be c
o
m
p
o
u
nde
d by
t
h
e act
i
v
e
vol
t
a
ge vect
o
r
s
V
1
and V
2
.
T
h
e
p
r
o
j
ect
i
o
n of
t
h
e refe
renc
e
vol
t
a
ge vect
or on V
1
and
V
2
is illu
strated
in
th
e
fo
llowing
figu
re:
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-86
94
I
J
PED
S
Vo
l.
4
,
No
.
4
,
D
ecem
b
er
2
014
:
51
7 – 527
52
0
Fi
gu
re 4.
Pr
o
j
e
c
t
i
on of
t
h
e ref
e
rence
v
o
l
t
a
ge vect
o
r
on
V
1
an
d V
2
The active
voltage
vectors
V1 an
d
V2
are
gi
ven
as
fol
l
o
w:
0
1
3
2
2
..
3
2
..
3
j
j
VE
e
VE
e
(5
)
Ex
pressi
ng
t
h
e
v
o
l
t
a
ge
vect
o
r
Vs i
n
t
h
e
gra
d
uat
i
o
n
(
α
,
β
)
w
e
ha
ve:
12
12
mo
d
m
o
d
ss
s
TT
VV
j
V
V
V
TT
(6
)
Exp
a
nd
ing
th
is equ
a
tio
n
it is
p
o
s
sib
l
e to
exp
r
ess th
e tim
e
T
1
and T
2
i
n
t
e
r
m
s
o
f
V
s
α
and
V
s
β
. T
h
e
co
ndu
ctio
n time will b
e
exp
r
essed
as fo
llows:
mo
d
1
mo
d
2
31
(.
)
.
22
2.
.
ss
s
T
TV
V
E
T
TV
E
(
7
)
To
facilitate
th
e
calcu
latio
n
s
, we n
o
rm
alize
t
h
e v
o
ltag
e
s Vs
α
and Vs
β
b
y
po
sing
:
^
s
s
^
s
s
V
V2
E
V
V2
E
(
8
)
C
onse
q
uent
l
y
,
t
h
e d
u
t
i
e
s e
x
p
r
essi
ons
are
gi
v
e
n as
f
o
l
l
o
w
s
:
^^
ss
1
^
s
2
01
2
31
.V
.
V
22
V
1
D
D
DD
D
(9
)
The space
vec
t
or in sector 1 is shown in figure
(5).T
h
e tim
e
duration
of zero
vectors is divide
d
equal
l
y
i
n
t
o
(
V
0
,
V
1
, V
2
,
V
7
, V
2
,
V1
, V
0
), w
h
erea
s t
h
e
t
i
m
e
durat
i
on
of eac
h n
onz
er
o vect
or i
s
di
st
ri
b
u
t
e
d
in
to
two
p
a
rts.
Th
is sequ
en
ce
can
en
sure th
at
is o
n
e
ph
ase switch
e
s wh
en
th
e switc
hi
ng
p
a
t
t
e
rn swi
t
c
hes
,
t
hus
can
reduce t
h
e
harm
onic c
o
mpone
n
t of t
h
e
output
cu
rre
nt
a
n
d
t
h
e l
o
ss
o
f
s
w
i
t
c
hi
n
g
devi
c
e
s.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
PED
S
I
S
SN
:
208
8-8
6
9
4
A N
o
vel
Di
rect
Tor
q
ue C
ont
r
o
l
f
o
r
I
n
d
u
ct
i
o
n M
a
c
h
i
n
e
Dri
ve Syst
e
m
w
i
t
h
Low
T
o
r
q
ue…
(
S
o
u
h
a
B
o
uka
d
i
da)
52
1
Fi
gu
re 5.
Se
qu
ences of
t
h
e sw
itches states in
sector
N1
The
duties of e
ach
pha
se
of t
h
e i
nve
rter are presente
d
as
follows:
12
12
12
0.
5
(
1
)
0.
5
(
1
)
0.
5
(
1
)
a
b
c
SD
D
SD
D
SD
D
(
1
0
)
4.
SIMULATION AND RESULT
The
DTC
a
n
d
DTC
-
S
V
M
sc
hem
e
for i
n
d
u
c
t
i
on m
achi
n
e
are si
m
u
l
a
t
e
d usi
n
g M
a
t
l
a
b/
Sim
u
l
i
nk a
n
d
Xi
l
i
nx Sy
st
em
Gene
rat
o
r an
d
t
h
ei
r res
u
l
t
s
ha
ve bee
n
c
o
m
p
ared
. The m
achi
n
e pa
ram
e
t
e
rs use
d
f
o
r si
m
u
l
a
t
i
on
are
gi
ve
n i
n
t
h
i
s
t
a
bl
e.
Tabl
e 1.
I
n
d
u
c
t
i
on
M
a
c
h
i
n
e p
a
ram
e
t
e
rs
Vo
ltag
e
220
/3
80
v
S
t
ato
r
r
e
si
st
an
ce R
s
5.
71
7
Ω
R
o
t
o
r re
si
st
an
ce
R
r
4.
28
2
Ω
Stator
in
duc
tanc
e
L
s
0.46
4
H
Ro
tor
in
duc
t
a
n
c
e
L
r
0.46
4
H
Mutua
l
indu
ctan
ce
M
0.44
1 H
Mome
nt o
f
ine
r
tia
J
0.00
49
Kg
.m
2
4.
1.
Sim
u
link
Model
of Dir
ect Tor
que
Contr
o
l
Th
e sim
u
latio
n of
D
T
C
w
a
s co
ndu
cted u
s
i
n
g Sim
u
l
i
nk\
M
A
TLAB
.
T
h
e i
n
vert
er
s
w
i
t
c
hi
n
g
pul
ses
a
r
e
obt
ai
ne
d
fr
om
t
h
e swi
t
c
hi
n
g
t
a
bl
e w
h
i
c
h
de
ci
des t
h
e
p
u
l
s
e
s
fr
om
t
h
e err
o
r
si
g
n
al
s o
f
t
o
r
q
ue an
d
fl
u
x
.
The
ove
ral
l
DTC
m
odel
i
s
sh
o
w
n
i
n
Fi
gu
re
6.
Fi
gu
re 6.
Si
m
u
l
i
nk
M
odel
of
DTC
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-86
94
I
J
PED
S
Vo
l.
4
,
No
.
4
,
D
ecem
b
er
2
014
:
51
7 – 527
52
2
4.2. Sim
u
link Model of Space
Vec
t
or
Modul
at
ed
Direct Torque Cont
rol obtaine
d
with
Matlab\Simulink
Fig
u
re 7
illu
st
rate th
e sim
u
la
tio
n
b
l
o
c
k
o
f
t
h
e DTC
-
SVM
co
n
t
ro
l. Th
e syste
m
is co
m
p
o
s
ed
of th
e
machine, PI
cont
rollers
, three phase voltage sour
ce i
nve
rt
er,
refe
re
nce fram
e
t
r
ansf
orm
a
t
i
on b
l
ocks
Conc
ordia and Park. The
Ins
u
lated-ga
te bi
polar tra
n
sistor
IGBT s
w
itche
s
are controlle
d using space
vector
m
odul
at
i
on t
e
c
hni
que
.
Fi
gu
re 7.
Si
m
u
l
i
nk
M
odel
of
DTC
-
S
V
M
Th
e sim
u
latio
n of th
is tech
n
i
qu
e is m
a
d
e
throu
g
h
t
h
e
fo
llowi
n
g
m
o
d
e
l:
Fi
gu
re
8.
Si
m
u
l
i
nk M
odel
of
bl
oc
SVM
4
.
3
.
Simulink Model
of Space
Vecto
r
Modula
t
ed Di
rect
To
rque Co
ntro
l o
b
ta
ined with X
ilinx
Sy
stem
Generator
In
itially, an
algo
rith
m
is d
e
sig
n
e
d
and
sim
u
la
ted
at th
e system lev
e
l with
th
e flo
a
ting
-
p
o
i
nt Si
m
u
lin
k
bl
oc
kset
s. A har
d
ware re
pr
esent
a
t
i
o
n
o
f
FPG
A
i
m
pl
em
ent
a
t
i
o
n
i
s
t
h
en deri
ved
us
i
ng XS
G.
T
h
e
X
S
G
provides a
bit-accurate m
odel
of FP
GA circ
uits and a
u
tom
a
tically gene
ra
tes a synthesiz
a
ble VHDL c
o
de for
i
m
p
l
e
m
en
tatio
n
in
Xilin
x
FPGA. For DTC
-
SVM m
o
d
e
lin
g
,
th
e b
l
o
c
k
s
used
are m
o
stly
m
u
ltip
liers, a
d
d
e
rs,
Cordic sin cos, etc. The detailed st
eps are s
h
own in the following diag
ra
m
i
n
Fi
gure 9.
The XS
G des
i
gn
o
f
pr
o
pose
d
DTC
-
SVM
i
s
sh
ow
n i
n
Fi
g
u
re 1
0
.
The bl
oc
k C
a
l
c
ul
_
V
sal
p
ha
_
V
sbet
a i
s
use
d
t
o
pro
j
ect
t
h
e t
h
ree-
p
h
a
se vo
ltag
e
s in
th
e repo
sito
ry (
α
,
β
)
by
per
f
o
r
m
i
ng t
h
e
pr
ocessi
ng C
l
arke as
sh
ow
n
i
n
Fi
g
u
re
1
0
(
a
). T
h
e
bl
oc
k S
V
M
ge
nerat
e
s a se
ri
es of
pul
ses t
o
be use
d
s
u
b
s
e
que
nt
l
y
t
o
carr
y
out
t
h
e co
nt
r
o
l
si
gnal
s
use
d
i
n
t
h
e
m
odel
of t
h
e
i
nve
rt
er as
sh
o
w
n i
n
Fi
gu
re
1
0
(
b
)
an
d
10
(c)
.
The
X
S
G
de
si
gn
o
f
t
o
rq
ue
and
fl
u
x
est
i
m
at
or i
s
sho
w
n i
n
Fi
gu
r
e
1
1
-
1
2.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
PED
S
I
S
SN
:
208
8-8
6
9
4
A N
o
vel
Di
rect
Tor
q
ue C
ont
r
o
l
f
o
r
I
n
d
u
ct
i
o
n M
a
c
h
i
n
e
Dri
ve Syst
e
m
w
i
t
h
Low
T
o
r
q
ue…
(
S
o
u
h
a
B
o
uka
d
i
da)
52
3
Fi
gu
re
9.
I
n
d
u
c
t
i
on m
achi
n
e
d
r
i
v
e c
o
nt
rol
l
e
r
desi
g
n
a
n
d i
m
pl
em
ent
a
t
i
on p
r
ocess
Fig
u
re 10
. Xilin
x
Mod
e
l o
f
SVM
Fi
gu
re 1
0
(a
).
C
a
l
c
ul
Vsal
pha
Vsbet
a
Fi
gu
re 1
0
(
b
)
.
VM
bl
oc
i
n
XS
G
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-86
94
I
J
PED
S
Vo
l.
4
,
No
.
4
,
D
ecem
b
er
2
014
:
51
7 – 527
52
4
Figu
re
1
0
(c
). C
a
lcul Sc
Fi
gu
re
1
1
. M
o
del
o
f
el
ect
r
o
m
a
gnet
i
c
t
o
r
que
Fi
gu
re
1
2
. M
o
del
o
f
fl
u
x
e
s
t
i
m
at
or
4
.
4
.
SIMUL
A
TION
RESULTS
Th
e p
e
rfo
r
m
a
n
ce o
f
th
e i
n
du
ction
m
a
c
h
in
e und
er differen
t
op
erat
in
g
cond
itio
ns was also
i
nvest
i
g
at
e
d
i
n
or
der t
o
veri
f
y
t
h
e rob
u
st
ne
ss of t
h
e
pr
o
p
o
se
d co
nt
rol
s
c
hem
e
. The steady
st
at
e beh
a
vi
o
r
of
in
du
ctio
n m
a
c
h
in
e
with
t
h
e co
nv
en
tio
n
a
l
DTC an
d
DTC-SVM are illu
strated
in
Figu
re 13
-15
.
It is
p
o
ssi
b
l
e t
o
see in
Figu
re 13
(a), (b
), (c) an
ap
preciab
le redu
ctio
n
o
f
electro
m
a
g
n
e
tic to
rqu
e
ri
p
p
l
e
ha
s
been
o
b
t
a
i
n
e
d
usi
n
g t
h
e
DTC
-
SV
M
.
F
o
r
t
h
e
DT
C
,
t
o
rq
ue
va
ri
at
i
on
of
t
h
e
hy
s
t
eresi
s
ba
n
d
e
q
ual
t
o
1.
1.
The
hi
gh
ri
p
p
l
e
o
b
se
rve
d
i
n
t
h
e
DTC
i
s
re
duce
d
w
h
e
n
we
use t
h
e
DTC
-
S
V
M
,
be
cause i
n
S
V
M
,
m
a
ny
vectors (IGBT
states) are selected to
adjust
the flux a
nd t
o
rque ripple
in each sam
p
le tim
e, whereas i
n
DTC
ju
st
one ve
ct
or
i
s
sel
ect
ed t
o
adj
u
st
ri
p
p
l
e
i
n
si
de hy
st
e
r
esi
s
ban
d
s o
f
fl
u
x
. Usi
ng S
V
M
cont
r
o
l
pr
o
v
i
d
es t
h
e
sy
st
em
wi
t
h
m
i
ni
m
u
m
ri
ppl
e f
o
r
fl
u
x
a
s
s
h
o
w
n i
n
Fi
g
u
r
e
14
,
whe
r
e t
h
e fl
ux
ri
p
p
l
e
pe
rce
n
t
a
ge i
s
a
b
out
0.
9
2
%.
The
DTC-SVM of induction m
achine prese
n
ts the
ad
va
nc
ed
perform
a
nce to ac
hieve t
r
acking
of the
desi
re
d sm
oot
h
ci
rcul
ar
t
r
a
j
ect
ory
of
st
at
or
fl
ux
l
o
c
u
s s
h
ow
n i
n
Fi
g
u
r
e
15
.
Fi
gu
re
1
3
. El
ec
t
r
om
agnet
i
c
t
o
r
que
,
(a):
DTC
usi
n
g M
A
TL
A
B
, (
b
)
DTC
-
S
V
M
u
s
i
n
g M
A
TLAB
,
(c
):
DT
C
-
SVM
usi
n
g XS
G
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
PED
S
I
S
SN
:
208
8-8
6
9
4
A N
o
vel
Di
rect
Tor
q
ue C
ont
r
o
l
f
o
r
I
n
d
u
ct
i
o
n M
a
c
h
i
n
e
Dri
ve Syst
e
m
w
i
t
h
Low
T
o
r
q
ue…
(
S
o
u
h
a
B
o
uka
d
i
da)
52
5
Fig
u
r
e
13
4.
Stato
r
f
l
ux
D
T
C usin
g
MA
TLA
B
,
(b
)
:
D
T
C
-
SVM
u
s
ing
M
A
TLA
B
,
(c):
DTC-
SVM
usi
n
g
XS
G
Fi
gu
re 1
5
.
T
r
a
j
ect
ory
of st
at
or
fl
u
x
:
DTC
usi
n
g
M
A
T
L
AB
,
(b
):
D
T
C
-
S
V
M
usi
n
g
M
A
T
L
AB
, (c):
DTC
-
SVM
usi
n
g XS
G
Table
2. T
h
e
p
e
rcenta
ge
flu
x
and
to
rq
ue e
r
r
o
r f
o
r D
T
C a
n
d
DTC-S
V
M
Control strategies
Flux ripple (
%
)
Torque ripple (%
)
DTC using Matla
b
5.
52
11
DTC-SVM
using
Matlab
0.
92
1
DTC-SVM
using
X
S
G
1.
84
2
The best
res
u
l
t
s
are gi
ve
n by
DTC
-
S
V
M
us
i
ng M
A
T
L
AB
\
S
IM
UL
I
N
K
,
t
h
i
s
i
s
due t
o
t
h
e ar
bi
t
r
ary
choice
of the
num
b
er of
bits a
t
XSG.
5.
FPGA SIMULATION RE
SULTS OF
DTC-SVM
The ab
o
v
e des
i
gne
d m
odel
i
s
im
pl
em
ent
e
d
usi
n
g FP
GA E
d
i
t
o
r
.
FP
GA E
d
i
t
o
r rea
d
s t
h
e
NC
D fi
l
e
gene
rat
e
d
by
t
h
e M
a
p o
r
Pl
a
ce & R
out
e p
r
ocess,
whi
c
h c
ont
ai
n
s
t
h
e l
o
g
i
c and r
out
i
ng
of t
h
e
desi
g
n
m
a
pped
to
co
m
p
on
en
ts, su
ch
as CLBs
an
d IO
Bs.Th
e
i
n
tern
al stru
cture of
FPGA
is sh
own
in Figur
e 16
.
Figu
re
1
4
6
.
I
n
t
e
rnal st
ruct
ure
of
FP
GA
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-86
94
I
J
PED
S
Vo
l.
4
,
No
.
4
,
D
ecem
b
er
2
014
:
51
7 – 527
52
6
Th
e
resu
lt o
f
the reso
urces
u
s
ed
is sho
w
n
i
n
t
h
e
fo
llowing
tab
l
e:
Tabl
e
3. T
h
e
re
sul
t
o
f
t
h
e
res
o
urces
Slice logic utilizati
o
n
Used
Available
Utilization
Nu
m
b
er
of slices
L
U
T
s
10,
511
44,
800
23%
Nu
m
b
er
used as
Logic
9,
869
44,
800
22%
Nu
m
b
er
of DSP48
E
s
109
128
85%
Nu
m
b
er
of slice
r
e
gister
s
655
44,
800
1%
6.
CO
N
C
LUS
I
ON
Thi
s
pape
r
has
bee
n
dev
o
t
e
d
t
o
t
h
e
com
p
ar
i
s
on
bet
w
een
t
h
e
per
f
o
r
m
a
nces o
f
t
h
e
DTC
an
d
DTC
-
SVM
st
rat
e
gy
.
The st
eady
st
at
e feat
ures
of t
h
e i
nduct
i
o
n m
a
chi
n
e as wel
l
as t
h
e t
r
ansi
ent
beha
vi
o
r
u
n
d
er
bot
h
approaches
ha
ve bee
n
commented and
com
p
ared.
T
h
e sim
u
lation result clearly indicates the high
per
f
o
r
m
a
nce of
DTC
-
SVM
.
The
pr
o
p
o
s
ed hi
gh
pe
rf
orm
a
nce sche
m
e
i
s
desi
gn
ed u
s
i
n
g X
S
G an
d
M
a
t
l
a
b/
Sim
u
l
i
nk
bl
oc
kset
s a
nd i
m
pl
em
ent
e
d o
n
Xi
l
i
nx
V
i
rt
ex 5
FP
GA
.
Num
e
ri
cal
si
m
u
l
a
t
i
ons ha
v
e
been
carried
out showi
ng t
h
e a
d
vantages
of
th
e
DTC-SVM
with
resp
ect to
t
h
e DTC. Th
is
work is th
e fi
rst step
to
ward
s im
p
l
emetatio
n
o
n
FPGA o
f
DTC-SVM. Fu
ture wo
rk
will ex
ten
d
th
is ex
p
e
ri
men
t
al v
a
lid
ati
o
n
to
t
h
e
st
udy
.
REFERE
NC
ES
[1]
Z Li, L W
a
ng,
S
Zhang, C Zha
ng, J A
hn. Torque Ripple Redu
ction in Dire
ct
Torque Controlled Brushless DC
Motor.
IE
EE
Trans. El
ec
trical
Machines.
2011; 1-4.
[2]
Y Cho, D
Kim,
K Lee, Y Lee, J
Song.
Torque Ripple Reduction and Fast To
rque Response Strateg
y
of Direct
Torque Con
t
rol f
o
r Permanent-
Magnet S
y
nchro
nous Motor.
I
E
EE Trans.Ind.
El
ec
tr
onics
.
2013
; 1-
6.
[3]
J
Beerten
,
J
Ver
v
eckken
,
J
Dries
e
n. P
r
edi
c
t
i
ve Di
rect
Torque Con
t
rol for Flux and
Torque Ripp
le
Reduction
.
IE
EE
Trans.Ind.Electronics
. 2010
; 57(
1).
[4]
T Sutikno, N Ru
mzi, N Idris
,
A J
i
din, N Cirstea.
An Im
proved FPGA Implementation of
Dir
ect To
rque Control for
Induction
Machines.
IEEE Trans.Ind.Electronics.
2013; 1280-129
0.
[5]
KN Achari,
B Gururaj,
DV Asho
k Kumar,
M
Vijay
a
Ku
ma
r.
A N
ove
l
MAT
L
A
B/
Si
mul
i
nk Model of PMSM
Drive
using Direct
Tor
que Contro
l with
SVM.
IEE
E
Mu
ltimedia
Comput
ing and S
y
stems.
2012; 1069-107
5
.
[6]
B Metidj
i
, F
Ta
zrart
,
A Az
ib,
N Taib
,
T Rekioua.
A New Fuzzy
Dir
ect Torq
ue Control Strateg
y
for Indu
ctio
n
Machine Based
on Indir
e
ct M
a
trix
Converter.
International Journal
of Rese
arch and Reviews in Computin
g
Engineering.
20
11;
1(1).
[7]
MW Naouar, E Monmasson,
AA Naas
s
a
ni. F
P
GA-bas
e
d current con
t
roll
ers
for AC m
achine
drives
-A revi
e
w
.
IEEE Transactio
ns on
Industrial Electronics
. 200
7; 54(4): 1907- 1
925.
[8]
JJ Rodriguez-Andina, MJ Moure, MD
Valdes
.
F
eatures
, d
e
s
i
gn
tools
,
and app
l
ication domains of FPGAs.
IEEE
Transactions on
Indus
trial Electroncis
. 2007
; 54(
4): 1810-1823.
[9]
Habetler TG, Profumo F, Pastorelli
M. Direct
torque contro
l of
inducti
on mach
ines over
a wid
e
speed r
a
nge.
Pr
oceed
ings
of
I
EEE-I
A
S Conf
er
ence
.
1992
; 600-
606.
[10]
Casadei D, Serr
a G, Tani A. Implem
entation of a direct control
algorithm
for induction motors
based on discrete
space v
e
c
t
or m
odulation
.
I
E
EE T
r
ansactions on
P
o
wer
El
ectr
oni
cs
.
2000; 15: 769-
777.
[11]
Tsung-Po Chen, Yen-Shin Lai,
Chang-Hu
an Liu. A new space vector modulati
on techn
i
que for
inverter contro
l.
Power Electronics Specialist Co
nference
. 1999;
2: 777-782.
BIOGRAP
HI
ES
OF AUTH
ORS
Souha BOUKADIDA received
the degr
ee in
Elect
r
i
cal Engin
e
ering from Natio
nal School of
Engine
ering of
Monastir, Tunisi
a in 2012. In 201
3 she receiv
e
d hi
s M.S degree in Autom
a
tic and
Diagnostic from Moanstir Univ
ersity
.
Her curr
ent res
ear
ch int
e
res
t
s
includ
e rapid protot
yping
and re
configur
ab
le
archi
t
e
c
ture
fo
r rea
l
-tim
e
cont
r
o
l app
lic
ations
o
f
el
ect
ric
a
l
s
y
s
t
e
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.