Internati
o
nal
Journal of P
o
wer Elect
roni
cs an
d
Drive
S
y
ste
m
(I
JPE
D
S)
Vol
.
7
,
No
. 2,
J
une
2
0
1
6
,
pp
. 42
2~
43
0
I
S
SN
: 208
8-8
6
9
4
4
22
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 Optimal PI Parameter
Tuning Strategy to Improve
Constant Switching Performanc
e of Direct Torque Control
R. Sun
dram
1
, Auz
a
ni Jidin
2
, To
le Su
t
i
k
n
o
3
1,2
Department of
Power Elec
troni
cs and Dr
ives,
Universiti
T
e
knika
l Mal
a
y
s
ia
Mela
ka (UTeM)
, Ma
l
a
y
s
i
a
3
Department of Electrical
Eng
i
n
eering
,
Univ
ersitas Ahmad Dahlan (UAD), Indon
esia
Article Info
A
B
STRAC
T
Article histo
r
y:
Received Feb 2, 2016
Rev
i
sed
May 11
, 20
16
Accepted
May 20, 2016
This paper pr
esents a novel m
e
thod of
optimal Propotional-In
tergral (PI)
controller’s par
a
meter t
uning
strateg
y
in-ord
er to improve the constant
switching perfor
m
ance of 3-phase direct
torque
control (DTC) shceme. Th
e
DTC sheme is
acknowledg
ed to provide
fast
decoupled contr
o
l over th
e
torque outpu
t an
d stator flux
via a si
m
p
le con
t
ro
l s
t
ructur
e. How
e
ver,
DTC
sheme has two major downsides, whic
h ar
e the inconsistent inverter
switching frequ
ency
and high torque out
put ripple. Th
e main reason that
contribut
es
to t
h
es
e tribu
l
at
ions
is
the us
age o
f
h
y
s
t
eres
is
co
m
p
arators
i
n
order to con
t
rol the outpu
t torq
ue. Th
e
realization of PI based
controller
method as replacement of
h
y
ter
i
sis contro
ller in
DTC s
y
stem ab
le to prov
ide
signific
a
nt solut
i
ons to over com
e
the fall bac
k
while reta
inin
g the sim
p
le
control structur
e of conv
entional DTC.
Th
e
co
mbination usag
e of high
er
sam
p
ling contro
ller DS1004 an
d also 3-level cascaded H-bridg
e
m
u
ltilev
e
l
inverters (CHMI) in th
is s
y
stem can fu
rth
e
r minimize the output tor
que ripple
b
y
prov
iding h
i
g
h
er resolu
tion with lower d
i
gital
error and
greater
number of
vectors. This p
a
per presen
ts detail
e
xplanation and calcula
tion
o
f
optim
al PI
param
e
ter
tunin
g
s
t
rat
e
g
y
cons
e
c
utiv
el
y
to enh
a
nce
the p
e
rform
ance
of 3-
level DTC s
y
stem. In order to v
e
rif
y
th
e feasibility
of th
e propo
sed method
experimentation
,
the proposed
method
is compared with
convention DTC
s
y
stem via simulation
and
exper
iment results.
Keyword:
Cascaded H-bridge
C
onst
a
nt
s
w
i
t
c
hi
n
g
Di
rect
t
o
rq
ue
c
ont
rol
I
ndu
ctio
n m
a
c
h
in
e
Mu
ltilev
e
l in
v
e
rter
Op
tim
al PI p
a
ra
m
e
ter
Copyright ©
201
6 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
s:
Auza
ni
Ji
di
n,
Depa
rt
m
e
nt
of
Power Electron
ics and Dr
ives
,
Un
iv
ersiti Tekn
ik
al Malaysia Melak
a
7
610
0 Du
rian
Tun
g
g
a
l, Melak
a
, Malaysia
Em
a
il: au
zan
i@u
t
em
.ed
u
.
m
y
To
le Su
tikn
o
Depa
rt
m
e
nt
of
El
ect
ri
cal
Engi
neeri
n
g
,
Fac
u
l
t
y
of
I
n
d
u
st
ri
al
Tech
nol
ogy
,
U
n
i
v
e
r
si
t
a
s A
h
m
a
d
Dahl
a
n
UA
D 3r
d
C
a
m
pus
,
Ja
nt
u
r
a
n
, Um
bul
har
j
o, Y
ogy
a
k
art
a
5
5
1
6
4
,
In
d
onesi
a
Em
a
il: to
le@ee.u
a
d
.
ac.i
d
1.
INTRODUCTION
There a
r
e t
w
o m
o
st
general
ac dri
v
e
s
co
nt
r
o
l
sche
m
e
s
that are being c
o
mm
only researched. One
of it
i
s
fi
el
d ori
e
nt
e
d
co
nt
r
o
l
(F
O
C
) w
h
i
c
h
was
pr
o
pose
d
by
F.
B
l
aschke a
nd
t
h
e f
o
l
l
o
wi
ng i
s
di
rect
t
o
rq
ue
cont
rol
(DTC
) w
h
i
c
h
was p
r
op
ose
d
by
I. Ta
ka
has
h
i
and T
.
N
o
gu
chi
[
1
,
2]
. The
r
e are t
w
o m
a
jor
d
o
w
n
si
des
of F
O
C
com
p
ared to t
h
e DTC. T
hos
e dra
w
back
s are th
e torq
u
e
ou
tpu
t
is con
t
rolled
in
d
i
rectly an
d
n
ecessity
o
f
t
h
e
pulse e
n
coder
[3]. FOC sche
me contro
ls the torque indire
ctly because c
ont
roller gi
ves
priority to the flux
vect
o
r
. FOC
sc
hem
e
requi
res
t
h
e pul
se enc
o
der i
n
o
r
der t
o
acqui
re t
h
e spe
e
d an
d p
o
si
t
i
on of t
h
e
rot
o
r.
These
make the DTC
as an alternative schem
e
in
whic
h gaine
d
t
h
e interest of num
e
rous re
sea
r
chers
recently. This is
because of its s
i
m
p
le structure
by
excl
usion
of pulse e
n
coder and sim
p
le
algorithm
with lesser de
pe
ndenc
e
on
m
o
t
o
r param
e
ters (
onl
y
re
qu
i
r
es val
u
e
of
st
at
or resi
st
an
ce R
s
and phase current
)
[1, 4]. Rece
ntly, the
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-86
94
I
J
PED
S
Vo
l. 7,
No
.
2,
Ju
ne 20
16
:
422
–
4
30
42
3
u
tilizatio
n
o
f
m
u
l
tilev
e
l in
v
e
rter top
o
l
o
g
y
i
n
DTC sch
e
m
e
h
a
s
g
a
in
ed
po
pu
larity m
a
in
ly in
th
e m
e
d
i
u
m
an
d
h
i
gh
vo
ltag
e
ap
p
lication
s
. In
g
e
n
e
ral, th
ere
are th
ree
typ
e
s o
f
m
u
ltilev
e
l i
n
v
e
rter topo
log
i
es. Th
ose typ
e
s are
th
e n
e
u
t
ral
p
o
i
n
t
clam
p
e
d
(NPC),
flyin
g
cap
acito
r
(FC
)
an
d
cascad
ed
H-bridg
e
m
u
ltile
v
e
l inv
e
rters (CHMI)
[6].
The
be
ne
fit of usi
n
g m
u
lt
ilevel inve
rter
is its acce
ssibility of larger num
b
er of
volta
ge vectors
i
n
whi
c
h
al
l
o
ws t
h
e
sel
ect
i
on
of m
o
st
opt
i
m
al
and
sui
t
a
bl
e v
o
l
t
a
g
e
vect
o
r
s i
n
o
r
de
r t
o
c
o
nt
r
o
l
fl
u
x
a
nd t
o
r
q
ue
by
r
e
du
cing
t
h
e slo
p
e
m
a
g
n
itude o
f
th
e
f
l
ux
an
d to
rqu
e
. M
o
re
over, this
features als
o
c
ont
ributes towards
o
p
tim
u
m
switc
h
i
ng
strateg
i
es
in
wh
ich
in
creases th
e effi
ciency by reducing the sw
itchi
ng fre
quency and also
i
m
p
r
ov
e th
e outp
u
t
v
o
ltage
q
u
ality (b
y redu
ci
n
g
th
e
rate o
f
ch
ang
e
of p
h
a
se
vo
ltag
e
, dV/d
t
)
[6
].
In c
o
n
v
e
n
t
i
o
na
l
DTC
,
t
h
e va
r
i
abl
e
swi
t
c
hi
n
g
f
r
eq
ue
ncy
of
t
h
e t
o
r
que
o
u
t
put
a
nd
p
o
we
r
swi
t
c
hes i
s
in
ev
itab
l
e
d
u
e
to
th
e
no
n
lin
ear effects
o
f
app
lied
v
o
ltag
e
vecto
r
s
on
torque and
fl
u
x
v
a
ri
atio
n
s
, in
restri
ctin
g
the va
riations
within
hystere
s
is bands.
As t
h
e stator
flux s
p
ace vect
or
form
s
a circ
ular locus, one
of t
h
e two
app
r
op
ri
at
e vo
l
t
a
ge vect
ors t
h
at
i
s
t
h
e
m
o
st t
a
ngent
i
a
l
t
o
t
h
e st
at
or fl
u
x
gi
ves hi
ghe
r r
a
t
e
of t
o
r
que c
h
an
ge
co
m
p
are to
less tan
g
e
n
tial v
o
ltag
e
v
ector [3]. It can
b
e
as
serted
th
at th
e ap
p
lication
of th
is v
ect
o
r
is do
m
i
n
a
n
t
(i.e. lo
ng
er ti
me ap
p
licatio
n) wh
en
th
e flux
m
o
v
e
s clo
s
er
to the sector bo
rd
er in
th
e stato
r
flux
p
l
an
e, wh
ile
t
h
e t
w
o s
u
i
t
a
bl
e vect
ors
bec
o
m
e
l
e
ss t
a
nge
n
t
i
a
l
and s
w
i
t
c
h
m
o
re oft
e
n
wh
en t
h
e fl
ux
vec
t
or t
r
avel
s
ar
ou
nd
t
h
e
middle of sectors
(
i.e. the rat
e
of
to
rqu
e
in
crem
en
t is
lesse
r). So
th
e torqu
e
switch
i
n
g
(o
r fl
u
x
switch
i
n
g
) is
m
o
re freq
u
e
n
t in the b
o
r
de
r (
o
r m
i
ddle) o
f
t
h
e secto
r
com
p
are t
o
the m
i
ddle
(or bor
de
r) of the sect
or. These
occu
rre
nces m
a
jo
rl
y
cont
ri
b
u
t
e t
o
t
h
e i
n
con
s
i
s
t
e
nt
swi
t
c
hi
ng
fre
que
ncy
i
n
DTC
sc
hem
e
. The real
i
zat
i
on
o
f
con
s
t
a
nt
swi
t
c
hi
n
g
m
e
t
hod c
a
n o
v
erc
o
m
e
the i
n
co
nsi
s
t
e
n
t
swi
t
c
hi
ng p
r
obl
em
s i
n
DTC
schem
e
[3, 4, 5]
.
Earl
i
e
r t
h
e co
n
s
t
a
nt
fre
que
nc
y
t
o
rq
ue co
nt
r
o
l
l
e
r (C
FTC
)
m
e
t
hod wa
s pr
op
ose
d
by
[
3
]
i
n
or
der t
o
red
u
ce t
h
e
out
put
t
o
r
q
ue
r
i
ppl
e.
I
n
t
h
i
s
a
p
p
r
oach
, t
w
o t
r
i
a
n
gul
ar
car
ri
er
waves
we
re
i
n
ject
e
d
a
f
t
e
r
t
h
e t
o
rq
ue e
r
r
o
r
no
de
an
d
two
co
m
p
arato
r
s were
used
to
g
e
n
e
rate to
rqu
e
st
atu
s
. Th
e low frequ
en
cy torqu
e
erro
r
o
s
cillatio
ns stil
l
ex
ist ev
en
thou
gh
th
is m
e
th
o
d
redu
ced
the to
rqu
e
ri
p
p
l
e.
Th
is is d
u
e
to
th
e stato
r
flux
h
y
steresis b
a
sed
co
n
t
ro
ller in
wh
ich
u
s
ed
to regu
late flux arou
nd
its
re
fere
nce val
u
e.
This err
o
r is less significa
nt an
d
n
e
g
lig
i
b
le if t
h
e PI p
a
ram
e
ter is calcu
lated
co
rrectly.
Thi
s
pa
pe
r re
vi
ews t
h
e i
m
p
l
em
ent
a
t
i
on o
f
C
F
TC
i
n
m
i
ni
m
i
zi
ng t
h
e
out
put
t
o
r
que
ri
p
p
l
e
wi
t
h
a
co
nstan
t
switch
i
ng
frequ
en
cy
as
p
r
op
o
s
ed in [3
]. So
m
e
ex
ten
s
ion
s
of t
h
e
work in u
tilizin
g CFTC
are do
n
e
b
y
ad
d
i
n
g
3-lev
e
l
cascad
ed
CHMI and
also
i
n
crease th
e sa
m
p
lin
g
frequ
en
cy o
f
th
e
DSp
a
ce DS1
004
b
y
utilizin
g
C p
r
og
ramm
in
g
langu
ag
e. B
y
u
tilizin
g
th
is m
e
th
o
d
, it is
p
o
s
sib
l
e to fu
rth
e
r increase t
h
e efficien
cy
of DTC
sch
e
m
e
b
y
pro
v
i
d
i
ng
con
s
tan
t
switch
i
ng
, eli
m
in
ati
n
g
t
h
e l
o
w frequen
c
y torqu
e
erro
r o
s
cillation
,
and
opt
i
m
i
z
i
ng t
h
e vol
t
a
ge vect
or
sel
ect
i
on base
d o
n
t
h
e
t
o
r
q
u
e
dem
a
nd. The
det
a
i
l
e
d expl
a
n
at
i
on
rega
rdi
ng t
h
e
im
pact
of
v
o
l
t
a
ge
vect
o
r
sel
e
ct
i
on i
n
i
n
fl
ue
nci
n
g t
h
e
m
i
nim
i
zat
i
on i
n
t
o
r
que
ri
ppl
e a
n
d
swi
t
c
hi
ng
f
r
e
que
ncy
are al
so
gi
ven
i
n
t
h
i
s
pa
per
.
2.
PRINCIPLE
OF BASIC T
H
REE PHASE DTC SHE
M
E
The basi
c pri
n
ci
pl
es of DTC
sy
st
em
are based on t
h
e est
i
m
a
t
i
ons of el
ect
rom
a
gnet
i
c
t
o
rqu
e
and st
at
or
fl
ux i
n
d
-
q a
x
i
s
. B
o
t
h
of t
h
e
val
u
es are est
i
m
a
ted usi
ng t
h
e i
n
form
ati
on of ap
pl
i
e
d vol
t
a
ge vect
or an
d
phas
e
current
. The
b
e
havi
or o
f
t
h
re
e phase i
nduct
i
on m
achi
n
es
in DTC
dri
v
es
can be descri
b
e
d i
n
t
e
r
m
s of space
v
ecto
r
s b
y
th
e
eq
u
a
tio
n
s
th
at
are written
in
stat
o
r
stat
io
n
a
ry referen
ce frame as sh
own
b
e
lo
w:
Vs
r
i
dΨ
dt
(1)
Ψ
L
i
L
i
(2)
Ψ
L
i
L
i
(3)
T
3
2
PΨ
i
Ψ
i
(4)
Ψ
v
i
r
dt
(5)
Ψ
v
i
r
dt
(6)
Where P
i
s
t
h
e num
ber of
pol
e pai
r
s;
Ls (st
a
t
o
r i
n
ductance), Lr (rotor indu
ctance), and
L
m
(
m
utual
i
nduct
a
nce) ar
e t
h
e i
nduct
a
n
ces of t
h
e m
o
t
o
r,
Ψ
and
Ψ
are the stator and rotor
flux a
n
d
δ
is the angular
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
Opt
i
m
al
PI
P
a
ra
met
e
r Tu
ni
n
g
St
r
a
t
e
gy t
o
I
m
prove
C
onst
ant
Sw
i
t
c
hi
n
g
…
(R. Su
nd
ram
)
42
4
di
fference bet
w
een st
at
or fl
u
x
l
i
nkage and i
s
stator current space vector . As in
equation
5 and
6,
the stator flux
vect
or i
s
wri
t
t
eni
n
d-
q axi
s
com
ponent
s. Vol
t
a
ge vect
or
s for t
h
e swi
t
c
hi
ng pat
t
e
rns
of t
h
e v
o
l
t
a
ge sourc
e
i
nvert
er S
a
, S
b
, and
S
c
(ca
n
be
ei
t
h
er 0 or
1),
o
b
t
a
i
n
ed by
usi
n
g d
-
q t
r
ans
f
o
r
m
a
ti
on equat
i
o
n are gi
ve
n bel
o
w:
v
1
3
V
2S
S
S
(7)
v
1
√
3
V
S
S
(8)
Fig
u
re
1
(
a) illustrates th
e
3-ph
ase i
n
v
e
rter co
nn
ected
to the star
wind
ing
o
f
3-ph
ase m
a
ch
in
e. Th
ere
are 6
switch
e
s in
to
tal wh
ere
3
up
p
e
r switches an
d
3
lo
wer switch
e
s. Th
e u
p
p
e
r switch
e
s are co
m
p
le
m
e
n
t
ary
with
lo
wer swi
t
ch
es.
Wh
en
the u
p
p
e
r switch is in
ON (1
) po
sitio
n, lo
wer switch
will b
e
in
OFF (0
) po
sitio
n.
The switc
hing state is corresponds to [S
a
, S
b
, S
c
]
.
The com
b
i
n
at
i
ons
of t
h
i
s
swi
t
c
hi
n
g
st
at
es pr
od
uces
6 no
n
-
zero active vol
t
age vectors and two zer
o
vo
ltag
e
v
ector
s. Fig
u
r
e
2
(
a
)
sh
ow
s th
e vo
ltag
e
v
ector
pr
odu
ce b
y
th
e 2-
lev
e
l
vo
ltag
e
sou
r
ce i
n
ver
t
er
(
V
SI)
sh
ow
n in
Figu
r
e
1.
(a)
(b
)
Figu
re
1.
(a
) s
h
ows
T
h
ree
-
p
h
a
s
e VS
I c
o
nnect
ed to
3-phase
machine a
n
d (b)
Basic structure
of
DTC-
hy
st
eresi
s
base
d
m
o
t
o
r dri
v
e
s
y
st
em
Fig
u
re
1
(
b
)
sho
w
s th
e Basic stru
ct
u
r
e
o
f
DTC-h
y
steresi
s
b
a
sed
m
o
to
r d
r
i
v
e system as in
itially
pr
o
pose
d
by
[
1
]
.
The vol
t
a
ge
vect
o
r
an
d o
u
t
put
st
at
or
vol
t
a
ge are ap
pl
i
e
d
base
d on t
h
e l
ook
-
up t
a
bl
e sel
ect
i
o
n
of t
h
e s
w
itching states (Sa
,
Sb, Sc
).
The
voltage vect
or
s a
r
e sel
ect
ed base
d o
n
t
h
e t
o
rq
ue
, st
at
or
fl
u
x
de
m
a
nd
an
d th
e stator
flux
po
sition
i
n
d
-
q
p
l
an
e.
Th
e
d
ecisio
n
s as to
wh
eth
e
r i
n
crease
o
r
d
e
crease th
e torq
u
e
an
d
/
o
r
the fl
ux are m
a
de
by their res
p
ective hystere
s
is
com
p
arators.
(a)
(
b
)
Fi
gu
re
2.
The
s
e
l
ect
i
on
of a
p
p
r
o
p
ri
at
e
vol
t
a
g
e
vect
o
r
(a)
v
o
l
t
age vect
ors
o
f
2-l
e
vel
V
S
I
wi
t
h
co
rre
sp
o
ndi
ng
switch
i
ng
co
mb
in
ation
s
, and
(b)
3
-
ph
ase secto
r
d
e
fi
n
itio
n
V
1
[1 1 0]
V
2
[1 1 0]
V
7
[1 1
1]
V
0
[0 0
0]
V
6
[1 1 0]
60
O
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-86
94
I
J
PED
S
Vo
l. 7,
No
.
2,
Ju
ne 20
16
:
422
–
4
30
42
5
3.
IMPLEME
N
TATION OF
CFTC CONT
ROLLER IN
3
-
LEVEL
DT
C SCHE
ME
The C
F
TC
wa
s im
pl
em
ent
e
d i
n
D
T
C
sy
st
em
i
n
or
der
t
o
red
u
ce t
h
e o
u
t
p
ut
t
o
r
q
ue ri
ppl
e
w
h
i
l
e
m
a
i
n
t
a
i
n
i
ng a
con
s
t
a
nt
s
w
i
t
c
hi
n
g
fre
qu
enc
y
as p
r
op
ose
d
in
[3
]. Th
is is don
e
b
y
inj
e
ctin
g
t
w
o trian
g
u
l
ar
carri
er
wave
s at
t
h
e t
o
rq
ue e
r
r
o
r
no
de a
nd
t
h
e res
u
l
t
a
nt
si
gnal
are
passe
d t
h
r
o
ug
h t
w
o
com
p
arat
ors.
Thes
e
com
p
arat
ors a
r
e use
d
t
o
ge
ne
rat
e
t
h
e t
o
rq
ue
err
o
r
st
at
us si
gnal
(Tst
at
).
T
h
e C
F
TC
(sh
o
w
n i
n
Fi
gu
re
3
)
was
use
d
t
o
re
pl
ace
t
h
e t
o
r
que
hy
s
t
eresi
s
con
t
ro
ller as
sho
w
n in
Fig
u
re
1
(
t
h
e area m
a
rk
ed
with
red do
tted
line).
Fi
gu
re
3(
b
)
s
h
ows
t
h
e
C
F
T
C
whi
c
h
was
m
odi
fi
ed f
o
r
t
h
e
pu
r
pose
o
f
im
pl
em
ent
i
n
g
i
t
i
n
3
-
l
e
vel
C
H
M
I
DTC
s
y
st
em
. The
m
odi
fi
ed C
F
TC
consi
s
t
s
o
f
si
x t
r
i
a
ng
ul
ar
gene
rat
o
rs, si
x
com
p
arat
ors
and a
pr
o
p
o
r
t
i
onal
-
i
n
t
e
gral
(PI
)
c
o
nt
r
o
l
l
e
r.
In
p
r
i
n
ci
pl
e,
t
h
e t
o
r
que
er
ro
r
st
at
us si
gnal
(T
stat
)
gene
rat
e
d
f
r
o
m
t
h
e
m
o
d
i
fied
CFTC is si
milar to
a 7
-
lev
e
l h
y
steresis
com
p
arator, which ca
n
be in
one
of t
h
ree states;
−
3
,
-2
,-
1, 0,
1,
2
o
r
3.
T
h
er
e i
s
n
o
m
odi
fi
cat
i
on
of
t
h
e
o
r
i
g
i
n
al
l
o
o
k
-
u
p
t
a
bl
e i
s
re
qui
r
e
d.
As
a
res
u
l
t
,
t
h
e
si
m
p
l
e
cont
r
o
l
structure of hysteresis-based
DTC ca
n
be
retained. The T
stat
si
gnal
ge
ne
rat
e
d
by
t
h
e c
o
m
p
arat
ors
i
n
Fi
gu
re
3(
b)
can
be
de
s
c
ri
be
d
by
t
h
e
f
o
l
l
o
wi
ng
eq
uat
i
on:
T
3,
f
o
r
C
arrier1
E
2,
f
orC
arrier2
E
1
1,f
or
Carrier3
E
2
0,f
or
Carrier4
E
3
1,f
o
r
Carrier5
E
4
2,
f
o
r
Carrier6
E
5
3,
f
o
r
Carrier6
(9)
The T
stat
si
g
n
a
l g
e
n
e
rated
b
y
th
e CFTC
b
l
o
c
k w
ill b
e
u
s
ed
b
y
th
e looku
p tab
l
e
b
l
o
c
k in
ord
e
r t
o
choose a
p
propriate voltage
vector
according to the insta
n
taneous
torque
dem
a
nd. The
rate of T
stat
alte
rn
ation
(frequ
en
cy) from
in
crease
to
decrease
t
o
rq
u
e
d
e
m
a
n
d
will
be sam
e
as th
e frequ
e
n
c
y of t
h
e carrier
wav
e
s. Th
e
fre
que
ncy
(
6
.
2
5k
Hz
) and
pea
k
-t
o-
peak m
a
gni
t
ude
(Tri
p-p
)
value of each t
r
iangle wave (t
he six carrie
r
wave
s)
was set to
a same v
a
lu
e b
u
t
th
e carrier waveform
s were o
f
fset b
y
th
e m
a
g
n
itud
e
of Tri
p-
p
. Th
is ph
enomen
a is
cl
earl
y
descri
b
e
by
t
h
e Fi
gu
r
e
4
w
h
ere
i
t
sh
ows
rel
a
t
i
o
ns
h
i
p bet
w
ee
n t
h
e
PI
com
p
ensat
e
d t
o
r
que
er
r
o
r
(E
pi
),
six carrier wa
ves and T
stat
si
gnal
w
h
i
c
h i
s
i
nvol
ved i
n
e
quat
i
o
n 9
.
In t
h
i
s
sim
u
l
a
t
i
on t
h
e E
pi
is
regu
latin
g
around
carrier 2 (C2)
wh
er
e
co
rr
e
s
p
ond
ing
T
stat
sig
n
a
l for
m
e
d
i
u
m
v
o
ltag
e
v
ector will
b
e
g
e
n
e
rated.
(a)
(b
)
Fi
gu
re
3.
(a
) s
h
ows
t
h
e
o
r
i
g
i
n
al
st
ruct
u
r
e
of
C
onst
a
nt
F
r
eq
u
e
ncy
T
o
r
que
C
ont
rol
l
e
r
(C
FT
C
)
an
d
(
b
)
sh
o
w
s t
h
e
m
odi
fi
ed st
r
u
ct
ure
C
o
nst
a
nt
F
r
eq
ue
ncy
To
r
q
ue C
ont
r
o
l
l
e
r
(
C
FTC
)
fo
r
3-l
e
vel
D
T
C
s
h
em
e.
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
Opt
i
m
al
PI
P
a
ra
met
e
r Tu
ni
n
g
St
r
a
t
e
gy t
o
I
m
prove
C
onst
ant
Sw
i
t
c
hi
n
g
…
(R. Su
nd
ram
)
42
6
The sel
ect
i
o
n
of t
h
e o
p
t
i
m
al
vol
t
a
ge
vect
or
depe
n
d
s
on t
h
e
dem
a
nd nee
d
e
d
i
n
or
de
r t
o
r
e
gul
at
e t
h
e
estim
a
ted torque accordi
ng t
o
the refe
re
nce
torque val
u
e. Voltage
vector with hi
gher magnitude can s
u
pport
hi
g
h
er t
o
r
q
ue
d
e
m
a
nd a
n
d
hi
g
h
sp
eed
o
p
erat
i
o
n
w
h
erea
s
vol
t
a
ge vect
or
wi
t
h
l
o
wer
m
a
gni
t
ude
can
su
p
p
o
r
t
l
o
w
to
rq
u
e
d
e
m
a
n
d
an
d
l
o
w sp
eed o
p
e
ration
.
Sm
aller to
rqu
e
sl
op
e will red
u
ce
in
v
e
rter switchin
g
freq
u
e
n
c
y.
Th
is
ad
v
a
n
t
ag
e
will co
n
t
ribu
te in
min
i
mizin
g
the to
rq
u
e
ripp
le an
d
switch
i
ng lo
sses. For exa
m
p
l
e in
th
e case of
forward
to
rqu
e
o
p
e
ration
,
Carrier 1
will g
e
nerate lo
ng
es
t mag
n
itu
d
e
of
vo
ltag
e
v
ect
o
r
(h
igh
sp
eed
o
p
e
ratio
n)
i
n
whi
c
h fol
l
o
wed
by
C
a
rri
er
2 for m
e
di
um
vect
o
r
(m
edi
u
m
speed ope
rat
i
on
) and C
a
r
r
i
e
r 3 fo
r l
o
we
st
vect
o
r
(
l
ow
sp
eed op
er
atio
n)
.
The
a
b
s
o
l
u
t
e
s
l
ope val
u
e of
T
pi
signal m
u
st be
lesser t
h
a
n
the a
b
s
o
lute s
l
ope
val
u
e
of the ca
rrier in
or
der
t
o
re
gul
a
t
e t
h
e t
o
rq
ue
o
u
t
p
ut
al
o
n
g
t
h
e
refe
re
nce
val
u
e. T
h
i
s
i
s
possi
bl
e by
t
h
e
pr
o
p
o
r
t
i
o
nal
gai
n
of t
h
e
PI con
t
ro
ller. To
arch
i
v
e th
e ab
ov
e con
d
ition,
th
e fo
llo
wi
n
g
eq
u
a
tion
m
u
st b
e
fu
lfilled
:
2
_
(9)
and also
2
_
|
|
(10
)
The
is p
o
s
itive slo
p
e
equ
a
tio
n
and
th
e
is n
e
g
a
tiv
e
slop
e eq
u
a
tion
.
Th
e selectio
n
of op
ti
m
a
l
K
tp
val
u
e i
s
b
a
sed
o
n
t
h
e s
m
al
l
e
st
val
u
e
pr
o
duce
d
t
h
e
equat
i
o
n
(
9
)
a
n
d
(
1
0).
T
h
e s
u
p
p
o
rt
i
n
g e
q
u
a
t
i
ons i
n
o
r
d
e
r to
so
l
v
e t
h
e ab
ov
e equ
a
t
i
o
n
are
g
i
v
e
n
by th
e (11
)
un
til (1
5).
(11
)
1
1
(12
)
1
(13
)
1
(14
)
3
4
(15
)
The m
i
nim
u
m
sam
p
l
i
ng peri
o
d
ca
n be
o
b
t
a
i
n
i
n
t
h
e e
xpe
ri
m
e
nt
al
set
up
b
y
usi
n
g
DSP
A
C
E
i
n
t
e
rface
car
d DS110
4 i
s
20
µs. Th
e sam
p
l
i
n
g
tim
e o
f
th
e Matlab
si
m
u
la
tio
n
is set to
2
0
µ
s. Th
e min
i
m
u
m sa
mp
ling
requ
ired
to
mak
e
low reso
lu
tio
n
triang
le wav
e
carrier
is 8
.
Th
e 8
step
p
e
r-cycle wi
th
20
µs sam
p
l
e
will
p
r
od
u
c
e carr
i
er f
r
e
qu
en
cy
of
62
50H
z.
Th
e Tab
l
e 1 sh
ow
s
t
h
e pa
ram
e
ter of
3-phase induction m
achine
us
ed in
sim
u
l
a
t
i
on an
d
ex
peri
m
e
nt
.
Tabl
e
1.
In
d
u
ct
i
on M
a
c
h
i
n
e
a
n
d
co
nt
r
o
l
para
m
e
t
e
rs
Induction Machi
n
e
Pa
ra
m
e
ter
Value
Rated power
,
P
1.
1 kW
Rated speed
,
2800 r
p
m
Stator resistance
,
6.
1
Ω
Rotor resistance,
6.
2293
Ω
Stator
self inductance,
0.
4797
9 m
H
Rotor
self inductan
ce,
0.
4797
9 m
H
M
u
tual inductance,
0.
4634 m
H
Nu
m
b
er
s of pole pair
s,
P
2
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-86
94
I
J
PED
S
Vo
l. 7,
No
.
2,
Ju
ne 20
16
:
422
–
4
30
42
7
Fi
gu
re
4.
Si
m
u
l
a
t
i
on res
u
l
t
us
i
ng
6
-
l
e
vel
ca
rr
i
e
r wa
ve
wi
t
h
c
o
r
r
es
po
n
d
i
n
g
o
u
t
p
ut
si
g
n
al
.
(a
) re
fere
nce t
o
r
que
with
estim
ated
to
rq
u
e
ou
tpu
t
,
(b)
6
-
lev
e
l carrier wa
v
e
with
t
h
e PI co
m
p
en
sated
torqu
e
erro
r sign
al (E
pi
),
(c)
7
-
l
e
vel
e
q
u
i
v
al
ent
t
o
r
que
d
e
m
a
nd st
at
us
(
T
st
at
) ge
ner
a
t
e
d
by
t
h
e m
odi
fi
ed C
F
TC
bl
oc
k
4.
PROP
OSE
D
METHO
D
O
F
HIG
H
S
A
MPLI
NG
AP
PLICATI
O
N
I
N
3-LEVEL
CF
TC
For t
h
e
pr
o
pos
ed m
e
t
hod
, t
h
e
C
F
TC
cont
r
o
l
l
e
r was im
pl
em
ent
e
d i
n
3-l
e
vel
DTC
sy
st
em
of 3-
phase
i
n
d
u
ct
i
on m
a
chi
n
e wi
t
h
hi
g
h
er sam
p
l
i
ng fre
que
ncy
an
d
st
udi
es t
h
e effect
o
f
sam
p
li
ng t
i
m
e
. Si
nce t
h
e
li
mitatio
n
o
f
h
a
rdware
on
ly p
r
ov
id
es d
i
g
ital sa
m
p
lin
g
o
p
tion
,
th
ere i
s
an
i
n
ev
itab
l
e p
r
esen
ts
of
d
i
g
ital
sam
p
l
i
ng err
o
r
.
I
n
ge
neral
,
di
gi
t
a
l
err
o
r
ca
n
be cl
assi
fi
e
d
i
n
t
w
o
part
s,
w
h
i
c
h i
s
sam
p
l
i
ng an
d
res
o
l
u
t
i
o
n
.
T
h
e
sam
p
lin
g
is a
p
r
o
cess conver
t
s a co
n
tinuo
u
s
tim
e sig
n
a
l to
a d
i
scr
e
te ti
m
e
sig
n
a
l w
ith
a d
e
f
i
n
e
d
ti
m
e
r
e
so
l
u
tio
n in
w
h
ich
is know
n
as th
e sam
p
l
i
n
g
r
a
te,
u
s
ually ex
pr
essin
g
i
n
H
e
r
t
z (
H
z)
o
r
sam
p
les p
e
r
seco
nd
.The
sa
m
p
li
ng rat
e
i
m
port
a
nt
f
o
r t
h
e re
pr
o
d
u
c
t
i
o
n o
f
t
h
e
si
g
n
a
l
depe
n
d
s o
n
t
h
e sha
r
p
n
ess
of t
h
e
flu
c
tu
ation
s
of
th
e sign
al
b
e
ing
sam
p
led
and
p
r
o
cessed
i
n
a
co
n
t
ro
ller. Fi
gu
re 5 sh
ows an illu
stratio
n
o
f
a sine
wave
wi
t
h
t
h
e
fre
que
ncy
o
f
8
Hz t
h
at
i
s
sam
p
l
e
d at
100
Hz 2
5
Hz a
n
d 10
Hz.
As s
h
o
w
n t
h
e 1
0
0
H
z
sam
p
l
i
ng
ab
le to
rep
r
o
duce th
e
o
r
i
g
in
al
sin
e
wav
e
b
e
tt
er th
an
t
h
e
o
t
her two
sam
p
les.
Un
de
rsam
pl
i
ng i
s
kn
o
w
n
a
s
al
i
a
si
ng
w
h
ere t
h
e
o
r
ig
inal sin
e
wav
e
is d
i
storted
to
th
e
po
in
t
un
rec
o
g
n
i
zabl
e
. Al
i
a
si
ng ca
n be o
v
er com
e
by
t
h
e N
yqu
ist sa
m
p
lin
g
ru
le wh
ere it states th
at th
e sa
mp
ling
rate
m
u
st be at least twice the
freque
n
cy of t
h
e sine
wa
ve (
f
<fs/
2). M
o
re t
h
e hi
g
h
e
r
sam
p
l
i
ng t
i
m
e, bett
er t
h
e
result will be
produce
d
.
The
hardware e
xperim
e
nt wa
s carried out usi
n
g the
DS
pace
DS1004 whe
r
e
it is
g
e
n
e
r
a
lly u
s
e th
e m
a
tlab
si
m
u
l
i
n
k
m
o
d
e
l b
l
o
c
k
o
f
simu
latio
n
to
run as th
e alg
o
r
i
t
h
m
o
f
th
e DS1004
cont
rol
l
e
r.
A si
m
u
li
nk m
odel
wi
t
h
ba
si
c 3l
e
v
el
DTC
al
go
r
i
t
h
m
can o
n
l
y
ru
n
up t
o
2
2
.
2
22
k
H
z (
4
5µs
sam
p
l
e
t
i
m
e
). The
3-l
e
vel
C
F
TC
ba
sed
DTC
si
m
u
l
i
nk m
odel
o
n
l
y
abl
e
t
o
r
u
n
upt
o
50µs
sam
p
l
e
t
i
m
e
. Inc
r
e
a
se i
n
sam
p
lin
g
ti
m
e
in
terv
al
will cau
se th
e con
t
ro
ller t
o
lack
o
f
accuracy and
m
o
re pron
e
to
h
a
v
e
o
v
e
rsho
o
t
or
u
n
d
e
r
s
hoo
t pro
b
l
em
s. Th
e
nu
m
b
er
s of
b
its in
vo
lv
e i
n
th
e ADC
conv
ersio
n
and
leng
th o
f
th
e al
g
o
rith
m
o
f
si
m
u
link m
o
del cannot
be
reduce
d
due t
o
accuracy a
n
d
reliability problem
s
. So in
order to inc
r
e
a
se the
sam
p
lin
g
frequen
c
y, th
e on
ly o
p
tion
left is to
conv
ert
th
e en
tire sim
u
lin
k
b
l
o
c
k
in
to
simp
ler C prog
rammin
g
l
a
ng
uage
. T
h
e com
m
a
nd p
r
o
m
pt
wi
ll
be us
ed t
o
u
p
l
o
ad t
h
e C
pr
o
g
ram
m
i
ng fi
l
e
i
n
t
o
t
h
e DS
1
0
0
4
co
nt
r
o
l
l
e
r.
Th
ere are to
tal
3
files requ
ire in
ord
e
r to
u
p
lo
ad
p
r
og
ra
m
wh
ich
are t
h
e .c file, .trc file
an
d
.sd
f
file. Th
e C
program
m
i
ng option will abl
e
to sim
p
lif
y the total algorit
hm
without m
a
ny
unnecessa
ry pre
d
efi
n
e values as
t
h
e si
m
u
l
i
nk m
odel
.
Thi
s
a
dva
nt
age c
o
nt
r
i
but
es t
o
t
h
e s
a
m
p
l
i
ng s
p
eed
of t
h
e D
S
1
0
0
04 c
o
nt
r
o
l
l
e
r i
n
w
h
i
c
h
can
ru
n
u
p
t
o
5
0
kHz
(
2
0µs sa
m
p
l
e
tim
e).
1
Torque status
2
Carri
er wa
ve
w
i
t
h
E
p
i
-3
-2
-1
0
1
2
3
To
r
q
ue
(p
.u
)
0.
7
(a)
(b
)
(c)
T
sta
t
us
E
T
esti
m
a
ted
T
reff
eren
ce
C6
C5
C4
C3
C2
C1
(a)
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
Opt
i
m
al
PI
P
a
ra
met
e
r Tu
ni
n
g
St
r
a
t
e
gy t
o
I
m
prove
C
onst
ant
Sw
i
t
c
hi
n
g
…
(R. Su
nd
ram
)
42
8
Fig
u
re
5
.
Sim
u
latio
n
resu
lt sho
w
s effect
o
f
differen
t
sam
p
lin
g ti
m
e
co
m
p
are to th
e
o
r
i
g
inal sig
n
a
l
Pre
v
iously there was only one
set of kp and
ki is
used for 2-level DTC syst
e
m
because there is only
si
ngl
e m
a
gni
t
ude
o
f
vol
t
a
g
e
vect
o
r
was
use
d
t
h
ro
u
g
h
o
u
t
t
h
e
o
p
erat
i
o
n
.
Si
nce t
h
e
3-l
e
vel
D
T
C
sy
st
em
pr
o
v
i
d
es 3
-
l
e
v
e
l
m
a
gni
t
ude of v
o
l
t
a
ge vec
t
or, t
h
e
r
e sh
ou
l
d
be 3 set
ok
kp an
d ki
t
o
be use
d
i
n
3 d
i
ffere
n
t
o
p
e
rating
conditio
n
in
o
r
d
e
r
to
prov
id
e
b
e
tter to
rqu
e
regulatio
n
at ev
ery
p
o
i
n
t
of op
eratin
g
cond
ition
.
For
ev
ery
o
p
e
ration
co
nd
itio
n
t
h
ere will b
e
3
d
i
fferen
t op
tim
a
l
to
rq
u
e
error st
atu
s
es will b
e
selected
to
redu
ce th
e
to
rq
u
e
ripp
le.
First, wh
ich
i
s
lo
w-sp
eed
op
eration
wh
ere to
rqu
e
1
-
0
will b
e
selected
. Secon
d
, wh
ich
is
Med
i
u
m
-sp
eed o
p
e
ratio
n
wh
ere to
rq
u
e
2
-
1
will b
e
select
e
d
. Last will b
e
h
i
g
h
-sp
eed
o
p
eratio
n
wh
ere to
rque
3
-
1
will b
e
sel
ected
. Th
e Table 2
shows th
e p
a
ram
e
ter o
f
CFTC con
t
ro
ll
er fo
r
3
-
lev
e
l DTC system
o
b
tain
ed
usi
n
g t
h
e
para
m
e
t
e
rs and
eq
u
a
t
i
on
gi
ve
n
pre
v
i
o
usl
y
.
Tabl
e 2. Param
e
t
e
rs
o
f
C
F
TC
C
ont
r
o
l
l
e
r fo
r 3-Le
vel
DTC
Sy
st
em
Induction m
o
tor para
m
e
te
rs
A 3826
60
3800
4
0.
0672
K 3325
7
(low speed)
0.
0031
(low speed)
1181.
5
(
m
ediu
m
speed)
0.
0005
279
4
(
m
ediu
m
speed)
202.
02
11
(h
ig
h
sp
eed
)
0.
0003
961
3
(h
ig
h
sp
eed
)
151.
58
1
5.
SIM
U
LATI
O
N
AN
D
E
X
PE
RIME
NTAL
RESULT
A c
o
m
p
lete experim
e
ntal set-up ha
s
been re
alized to
veri
fy the
f
easibility of the
propose
d
m
e
thod i
n
t
e
rm
s of ha
rd
ware i
m
pl
em
ent
a
t
i
on.
T
h
e e
xpe
ri
m
e
nt
al
set
-
u
p
co
nsi
s
t
s
o
f
D
S
P
A
C
E
11
04
co
nt
r
o
l
l
e
r,
har
d
war
e
setu
p of
3-
level CH
MI and
a 1.1 HP, t
w
o-
po
le indu
ction
m
o
to
r
.
Th
e
co
n
t
r
o
l al
g
o
r
ith
m
is i
m
p
l
e
m
en
ted
on
D
S
PA
CE
1
104 w
ith
sam
p
lin
g p
e
r
i
o
d
of
2
0
µs. Figu
re .6
sho
w
s th
e exp
e
r
i
men
t
al r
e
su
lt of
d
i
f
f
e
r
e
n
t
sam
p
lin
g
t
i
m
e
of 3l
evel
DTC
wi
t
h
t
h
e t
o
r
que
out
put
r
i
ppl
e o
f
(a
)1
6.
45
%, (
b
)
1
3.
71
%, an
d (c
)1
0.
9
7
%. T
h
i
s
sh
o
w
s t
h
at
th
e torqu
e
ou
t rip
p
l
e ab
le to red
u
c
e b
y
i
n
creasi
n
g
th
e sam
p
lin
g frequ
e
n
c
y of th
e con
t
ro
ller.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-86
94
I
J
PED
S
Vo
l. 7,
No
.
2,
Ju
ne 20
16
:
422
–
4
30
42
9
(a)
(b
)
(c)
Fi
gu
re
6.
Ex
pe
ri
m
e
nt
al
resul
t
of
3
-
l
e
vel
DTC
wi
t
h
sam
p
l
i
ng t
i
m
e
of (a
)
5
0
µ
s, (
b
)
2
0
µs
, a
n
d
(c
)
20µs
wi
t
h
r
e
du
ced
b
a
ndwith
Tor
q
ue out
p
ut
FFT d
i
ag
ra
m
(a)
(b
)
(c)
Fi
gu
re
7.
Ex
pe
ri
m
e
nt
al
resul
t
of
t
o
r
q
ue
out
p
u
t
an
d
FFT
di
a
g
ram
of
st
at
or
cur
r
ent
(a)
2
-
l
e
vel
D
T
C
,
(b
)
2-l
e
vel
C
F
TC
base
d
DTC
,
an
d
(c)
p
r
op
o
s
ed m
e
t
hod
o
f
3-l
e
vel
C
F
TC
base
d
DTC
Thi
s
e
xpe
ri
m
e
nt
res
u
l
t
wa
s c
a
rri
ed
o
u
t
usi
n
g
20µs
sam
p
l
i
ng t
i
m
e
Fi
gu
re 7 s
h
ow
s t
h
e expe
ri
m
e
nt
al
resul
t
o
f
o
u
t
p
ut
t
o
r
que
ri
ppl
e a
nd t
h
e F
F
T di
ag
ram
ofof t
h
e st
at
o
r
cur
r
ent
i
n
or
de
r t
o
anal
y
s
i
s
h
a
rm
oni
cs and
swi
t
c
hi
ng
fr
equ
e
n
c
y betw
een th
e co
nv
en
tion
a
l D
T
C and
CFTC
base
d DTC
.
T
h
e u
p
p
er
res
u
l
t
(i
) s
h
o
w
t
o
r
q
ue ri
ppl
es
o
f
(
a
) 1
7
.
1
3%,
(b
)
14
.0
2%
, an
d
(
c
) 7
.
7
9
%
.
Th
e
up
pe
r
resu
lt (ii) sh
ows th
e FFT d
i
agram
,
wh
ere th
e switch
i
ng
f
r
eq
uency
2-l
e
vel
c
o
n
v
e
n
t
i
onal
D
T
C
was n
o
t
u
n
i
form
and
ha
ve hi
gh
er TH
D c
o
m
p
are t
o
C
F
TC
c
ont
rol
l
e
r
base
d
DTC
w
h
er
e b
o
t
h
2-l
e
vel
an
d 3
-
l
e
vel
have
sam
e
switching
fre
quency a
n
d less
er T
H
D. Im
plementation
of
CFTC controller in DTC
she
m
e shows si
gnificant
im
pro
v
em
ent
in r
e
d
u
ci
n
g
t
h
e
t
o
r
que
ri
p
p
l
e
a
n
d
al
so
by
pr
o
v
i
d
i
n
g c
o
nst
a
n
t
swi
t
c
hi
n
g
fre
que
ncy
base
d
on
t
h
e
carrier freque
ncy.
6.
CO
NCL
USI
O
N
Thi
s
pa
per
has
prese
n
t
e
d t
h
e
im
pl
em
ent
a
t
i
o
n o
f
co
nst
a
nt
s
w
i
t
c
hi
n
g
m
e
t
h
od i
n
c
o
n
v
e
n
t
i
onal
t
h
e 3
-
l
e
vel
DTC
sy
st
em
. Thi
s
sy
stem
has t
o
t
a
l
num
ber of
18 a
c
t
i
v
e and
2 ze
ro
vol
t
a
ge
vec
t
ors. T
h
e p
r
op
ose
d
schem
e
was si
m
u
l
a
t
e
d and c
o
m
p
ared
wi
t
h
C
F
TC
a
n
d
co
nv
en
tio
n
a
l meth
od
using
Matlab
/
Si
m
u
li
n
k
. Th
e
p
r
op
o
s
ed
m
e
th
o
d
was also tested
and
v
e
rified
its feasib
ility
u
s
i
n
g a co
m
p
l
e
te ex
p
e
rim
e
n
t
al set-up
.
AC
KN
OWLE
DG
MENT
Th
e au
th
o
r
s wo
u
l
d
lik
e to
th
a
n
k
th
e Min
i
str
y
o
f
Ed
u
catio
n
(MOE) an
d
Un
iv
ersiti Tek
n
i
k
a
l Mal
a
ys
ia
M
e
l
a
ka (UTeM
)
for
pr
ovi
di
ng t
h
e
research
grant
FR
G
S
/
2
01
3/
FKE/
TK0
2
/
02/
1/
F0
01
59
for t
h
i
s
researc
h
.
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
Opt
i
m
al
PI
P
a
ra
met
e
r Tu
ni
n
g
St
r
a
t
e
gy t
o
I
m
prove
C
onst
ant
Sw
i
t
c
hi
n
g
…
(R. Su
nd
ram
)
43
0
REFERE
NC
ES
[1]
Takah
a
shi and
T. Noguchi, "A New Quic
k-Respo
n
se and High-Ef
ficiency
Contro
l
Strateg
y
of
an In
duction Motor"
,
Industr
y
Applications, I
EEE
Tr
an
sactions on
, vo
l.
IA-22, pp. 820-8
27, 1986
.
[2]
T. Noguchi, M.
Yamamoto, S. Kondo, a
nd I. Tak
a
hashi, "Enlarging switching fre
quency
in
direct torque-con
trolled
inverter b
y
mean
s of dith
ering",
I
ndustr
y
Applications, IE
EE Tr
an
sactions on
, vo
l.
35, pp
. 1358-13
66, 1999
.
[3]
N.
R.
N.
Idris and A.
H.
M.
Yatim,
"Direct torque cont
rol of
indu
ction machin
es with constant sw
itching fr
equen
c
y
and redu
ced
torq
ue ripp
le", Indus
trial Electronics
, IEEE
Transactions on,
vol. 51
,
pp. 758-767
, 20
04.
[4]
Jidin, N. Idris,
N. Rumz
i, M. Yatim, A. Halim,
T. Sutikno,
and M.
Elbuluk, "Ex
t
ending switch
i
n
g
frequency
for
torque ripple reduction utilizing
a cons
tant freq
u
ency
torque
co
ntroller in
dtc o
f
induction motors", Journal of
Power Electron
ics, vol. 11
, pp
. 1
48-155, 2011
.
[5]
M. Sanila, "Direct Torque Contr
o
l of induction motor w
ith constant switching fr
eque
ncy
"
, in Power Electronics
,
Drives and
Ener
g
y
S
y
stems (PEDES), 2012 I
E
EE
International
Conference on
,
2012, pp
. 1-6
.
[6]
Z. Yu, J. Zhenh
u
a, and Y. Xunwei, "Indirect field-orie
nted con
t
rol of inductio
n machines based on sy
n
e
rgetic
control th
eor
y
",
in
Power and Energy Society G
e
neral Meeting -
Con
version and Delivery o
f
Electrical
Energy in
the 21st Century, 2008
IEEE
, 20
08, pp
. 1-7
.
[7]
Casadei, D., Serr
a, G.,
and Ta
ni,
A. Analy
t
ical in
vestigation of
to
rque
and
flux rip
p
le in
DTC schemes for inductio
n
motors. 23rd International Conf
erence on Industrial
Elec
tronics, Control and Instru
mentation
,
1997. IECON 97.
1997.
[8]
Rodriguez, J.; B
e
rnet, S.; Steimer, P.
K.; Lizama, I.E., "A
Survey on Neutr
a
l-Point-Clamped Inv
e
rters",
Industrial
Electronics, I
E
EE Tr
ansactions o
n
, vol.57
,
no
.7,
pp.2219,2230
, J
u
ly
2010
[9]
Yen, G.W.,
et al. (2014). "Proposed Voltag
e
Vector to Optim
ize
Efficiency
of Dir
e
ct To
rqu
e
Contr
o
l", Intern
ational
Journal of Power Electron
i
cs
and
Drive S
y
stems
4(4): 578.
[10]
Malinowski, M., et
al
., A Survey
on
Cascad
ed
Multilev
e
l In
v
e
r
t
ers. Industr
ial
E
l
ectronics, IE
EE
Transactions on
,
2010. 57(7): p. 2
197-2206.
[11]
Zuber, M.
Z. R., et
al. (2015). "
I
mproved Torqu
e
Cont
ro
l Perfor
m
ance in
Direct To
rque Control using Optimal
Switching Vecto
r
s", Intern
ation
a
l Journal of
Po
wer Electron
i
cs
and
Drive S
y
stems
5(3): 441.
BIOGRAP
HI
ES OF
AUTH
ORS
R.
Su
ndr
am
was born in 1989
in Penang, Malay
s
ia. He
received the B.Eng. degree (Hons) in
Electri
cal
Engin
eering from
Universiti Teknik
a
l Ma
lay
s
ia Melak
a
, Malay
s
ia
in 2
012 and now h
e
is
currentl
y
purs
u
ing the M
.
S
c
degree in P
o
wer
Electron
i
cs
and
Drive. His
area
s
of res
earch
inter
e
s
t
in
clud
e
Direct
Torqu
e
C
ontrol of
M
u
lti
P
h
as
e S
y
s
t
em
and
P
o
wer El
ec
troni
cs
.
Au
z
a
ni Jidin
rece
ived
the B
.
Eng.
degr
ees
,
M
.
Eng.
degr
ees
and Ph.D. d
e
g
r
ee
in power
electroni
cs and
drives from
Universiti
Teknolog
i Malay
s
i
a
, Joh
o
r Bahru, M
a
lay
s
ia, in
2002
,
2004 and 2011
, respectiv
ely
.
He is curren
t
ly a
Lectur
er with the Dep
a
rtm
e
nt of Power
Ele
c
troni
cs and Drives, Facult
y
of
Elec
tric
al En
gineer
ing, Unive
r
siti Tekn
ikal M
a
la
ysi
a
Melak
a
,
Durian Tunggal, Malay
s
ia. His re
search
interests
includ
e the f
i
eld of power electronics, motor
drive s
y
stems, f
i
eld-programmable g
a
te
arr
a
y,
an
d DSP applic
atio
ns.
Tole
Sutikno
is an exp
e
rt
in th
e
field of
power
electro
n
i
cs, industrial electro
n
i
cs &
informatics,
embedded s
y
stems and electric drives. Since 2001
he has b
een a lecturer
in Electrical
Engineering Department, Univ
er
sitas Ahmad D
a
hlan (UAD), Indonesia. He is
an Associate
Professor at the
above Univ
ersity
sin
ce 2008
. H
e
receiv
e
d his B
.
Eng., M.Eng
.
an
d Ph.D. degr
ee
in Electrical En
gineer
ing from
Diponegoro Univer
sity
, Gadjah Mada Univ
ersity
and Universiti
Teknologi Malay
s
ia, respectively
. H
e
is
an Editor-
in-Chief of TELKOMNIKA
Telecommunication Computing
Elec
troni
cs
and
Control,
and
a
l
s
o
acts
as
an
Editor
in s
o
m
e
intern
ation
a
l
jou
r
nals in
el
ectr
i
c
a
l and
com
pute
r
engin
eering
,
p
o
wer el
ectron
i
cs
, m
o
tor driv
e
s
y
s
t
em
s
and F
P
GA applic
ations
areas
.
Evaluation Warning : The document was created with Spire.PDF for Python.