In
te
r
n
ation
a
l Jou
rn
al
o
f Po
we
r
Elec
tron
ic
s an
d
D
r
ive S
y
stem
(IJ
PED
S
)
Vol.
10, No.
1, Mar
ch 2019,
pp.
41~47
IS
S
N
: 2088-
86
94,
D
O
I
:
10.11
5
9
1
/ij
ped
s
.
v10
.
i
1.pp
4
1
-4
7
41
Jou
rn
a
l
h
o
me
pa
ge
:
ht
tp:
//i
a
e
score
.
com
/
j
o
u
r
na
l
s
/
i
n
d
e
x
.
p
hp/IJ
PED
S
Speed control of
uni
v
ersal motor
Omar
A
.
I
m
r
a
n,
Wisam N
a
j
m Al-Di
n
A
b
e
d
,
A
li
N
.
Jba
r
ah
D
e
partm
e
n
t
o
f Chem
ical
En
g
i
n
eeri
ng, Colleg
e
o
f
E
n
g
i
neeri
ng, Uni
versity o
f Diyal
a
, Iraq
Art
i
cl
e In
fo
ABSTRACT
A
r
tic
le hist
o
r
y
:
R
e
ce
i
v
e
d
Jan
2
6, 201
8
Re
vise
d O
c
t
3
1
,
2018
A
c
c
e
pte
d
D
ec 3,
201
8
Un
iv
ersal
Mo
tors
(
U
M
)
are
no
rm
a
lly
u
s
e
d
f
o
r
dri
v
in
g
port
a
ble
ap
pa
ra
tu
s
su
ch
a
s
hand
t
o
o
l
mach
in
es,
v
acuu
m
c
l
eaners
a
n
d
m
o
st
d
o
m
es
tic
a
pp
arat
us
.
Th
e
im
port
a
nce
of
U
M
i
s
d
u
e
t
o
its
o
w
n
ad
vant
ages
s
uch
as
h
igh
sta
r
tin
g
to
rqu
e
,
very
powerf
u
l
i
n
r
e
l
ati
o
n
to
its
s
m
a
ll
size,
h
avi
n
g
a
v
ariabl
e
s
p
eed;
and
low
e
r
co
st.
S
o
,
t
h
i
s
p
ap
er
f
o
c
us
o
n
UM
s
peed
c
o
n
t
r
o
l
u
n
d
e
r
variabl
e
lo
adi
ng
co
nd
it
ion
s
.
A
m
a
th
e
m
atical
m
o
d
el
f
or
U
M
i
s
d
esig
ned.
T
w
o
con
t
ro
llers
are
pro
p
o
s
ed
f
o
r
c
on
tro
l
l
i
ng
t
he
m
ot
or
s
peed,
ou
tp
u
t
ra
t
e
c
o
ntrol
l
e
r
a
nd
o
u
t
pu
t
re
se
t
c
o
n
t
rol
l
er.
Ant
Colony
O
ptimi
z
a
t
ion
(
A
C
O
)
i
s
pro
p
o
s
ed
f
o
r
t
u
n
in
g
t
h
e
con
t
ro
ll
e
r’s
p
aram
et
ers
du
e
to
i
ts
i
m
p
a
c
t
o
n
so
lvi
ng
diff
erent
op
ti
mizat
ion
p
r
ob
lems
.
It
p
o
s
s
e
ss
es
f
as
t
con
v
ergen
c
e
,
m
i
n
i
m
u
m
alg
o
rithm
param
e
t
e
rs
r
e
q
ui
red,
l
o
w
er
c
on
secutio
n
tim
e
an
d
g
i
ve
op
tim
al
resu
lts
w
i
t
h
o
u
t
n
eedi
n
g
l
a
rge
nu
m
b
er
o
f
it
erati
o
ns
.
Th
e
res
u
l
t
s
a
re
c
om
pa
red
a
n
d
disc
usse
d
ac
c
u
ra
te
ly
,
whic
h
sho
w
t
h
e
p
ro
po
se
d
tu
ning
t
e
c
h
ni
que
work
wel
l
a
n
d
g
iv
e
optim
al
r
es
ults
f
or b
ot
h
co
ntroll
ers.
K
eyw
ord
s
:
A
n
t co
lo
n
y
op
t
imiz
a
tio
n
Out
p
u
t
r
ate
co
ntr
o
l
l
er
O
u
tp
ut
r
e
s
et
c
on
t
r
o
l
l
e
r
Un
iv
e
r
sa
l
mo
to
r
Co
pyri
gh
t © 2
019 In
stit
u
t
e
of Advanced
En
gi
neeri
n
g
an
d
S
c
ien
ce.
All
rights
res
e
rv
ed.
Corres
pon
d
i
n
g
Au
th
or:
O
m
ar
A
.
Im
r
a
n,
D
e
pa
rtme
nt
o
f
Chem
ica
l
E
ng
ine
e
ri
ng,
Col
l
e
g
e
of
E
n
g
i
ne
er
in
g,
U
niv
e
rsit
y
of
D
iya
l
a
,
Ba
qu
ba
h,
D
iyala,
Ir
a
q.
Em
ail:om
a
rim
r
an5
3
@
ya
h
o
o
.
c
o
m
1.
I
N
TR
OD
U
C
TI
O
N
I
n
i
n
d
u
s
t
ry,
D
C
-m
otor
s
a
r
e
w
i
de
l
y
a
p
p
l
i
ca
bl
e
due
t
o
i
t
s
spe
e
d
c
an
b
e
a
d
j
u
s
t
ed
.
M
o
t
o
r-
s
p
e
e
d
c
o
n
t
ro
l
ca
n
be
d
o
n
e
i
n
d
i
f
fer
e
n
t
a
rrangme
n
ts
[
1].
The
u
n
ive
r
sal
mot
o
rs
(
U
M)
a
re
a
n
elec
t
r
i
c
r
ot
a
tin
g
ma
chi
n
e
tha
t
a
n
al
og
ou
s
to
a
d
i
r
e
c
t
cu
rre
n
t
(DC
)
m
o
t
o
r
s
bu
t
it
can
b
e
wo
rk
e
i
th
er
from
(DC)
s
our
ces
o
r
(A
C
)
s
ources
.
It
c
o
mb
in
es
s
o
m
e
ad
v
a
nt
a
g
es
l
i
k
e,
s
ma
l
l
e
r
si
ze
,
l
a
rg
e
st
a
r
t
i
ng
t
o
r
q
u
e,
h
ig
h
revo
lu
t
i
on
(
a
p
p
ro
x
3
0
,0
00
rp
m
)
a
nd
have
l
ow
e
r
c
osts.
G
e
ner
a
ll
y,d
i
ffe
r
en
t
h
o
m
e
a
pp
l
i
anc
e
s
a
re
pow
er
ed
by
UM,
such
a
s
electric
d
rills,
g
r
i
nders,
v
a
cu
u
m
c
l
e
a
n
ers,
s
a
w
s
et
c.[
2
],
[
3
]
.
UM
h
av
e
wi
d
e
sp
re
a
d
a
p
p
l
i
c
at
i
on.
I
t
s
c
ons
ump
t
i
o
n
e
n
ergy
of
t
he
i
npu
t
pow
er
is ver
y
l
ow
a
s co
pa
red
t
o
o
ther
t
y
p
e
s
.
S
o
, the r
eq
uire
m
e
n
ts be
c
om
e inc
r
ea
sin
g
l
y
higher for motors with
hi
gh
per
f
orm
a
nce
s
a
n
d
l
ow
-cos
t
c
o
ntro
lle
r.
A
lso,
r
ec
ent
l
y
,
S
mar
tH
om
e
sys
t
em
s
ha
ve
g
rea
t
a
t
t
e
n
t
i
on
i
n
th
e
con
t
ro
l
e
n
g
i
ne
erin
g.
F
urthe
r
m
o
re
i
n
the
S
m
a
r
tH
om
e
syste
m
,
U
M
a
r
e
expe
c
t
e
d
t
o
fi
nd
w
i
de
a
rea
a
p
pli
c
at
ion
s
[
3].
G
r
eat
p
art
of
t
he
r
ea
l-w
o
rl
d
o
p
tim
iz
at
ion
-
prob
l
e
ms
i
nc
l
ude
m
u
l
t
i-c
o
n
f
l
i
c
tin
g
o
b
j
e
c
t
iv
e
s
w
h
i
c
h
sho
u
l
d
b
e
r
eco
nc
ile
d
m
u
t
u
a
lly.
[4].
T
he
t
er
m
optim
iza
t
i
o
n
me
ans
d
i
s
c
ove
r
the
bes
t
s
olu
t
i
o
n
a
m
on
g
m
a
ny
pos
si
b
l
e
s
o
l
u
ti
ons
t
ha
t
a
r
e
a
v
ail
a
b
l
e
i
n
t
he
s
ea
rch
spac
e.
F
ea
s
ible
s
o
l
uti
o
ns
a
re
t
h
o
se
s
o
l
u
t
i
o
ns
t
hat
sa
t
i
sfy
a
l
l
op
tim
iza
t
i
o
n
pr
ob
l
e
m
c
onstra
i
nts.
I
n
t
h
e
o
p
timiz
a
t
ion
pro
b
lem
s
the
be
st
s
ol
u
t
i
o
n
co
u
l
d
be
m
i
n
im
i
z
ing
t
h
e
p
r
o
c
ess
co
st
o
r
ma
x
i
mi
zin
g
t
h
e
syst
em
e
ffi
ci
en
cy
.
In
a
ny
o
pt
i
m
i
z
a
t
i
on
p
ro
bl
e
m
,
a
sp
e
c
i
f
i
c
f
un
ct
io
n
i
s
t
o
be
minim
i
z
e
d
or
m
axim
ize
d
.
The
o
p
t
imiz
ed
f
unc
t
i
o
n
i
s
defi
ned
a
s
t
h
obje
c
tive
fu
nc
t
i
o
n
o
r
t
h
e
p
e
rform
a
nc
e
in
de
x
or
c
ost
func
t
i
o
n
.
T
h
e
op
tim
i
z
e
d
f
unc
t
i
o
n
is
a
q
uan
t
i
t
y
su
ch
a
s
co
st
,
si
ze
,
sh
ap
e
,
w
ei
ght
,
p
r
o
f
it
,
efficie
n
c
y
,
ou
t
p
u
t
,
an
d
so
o
n
[5].
R
ec
e
n
tly,
m
a
ny
re
sear
ch
p
ape
rs
f
ocus
o
n
ne
w
na
t
u
ral
insp
ire
d
o
p
tim
i
z
a
tio
n
t
e
c
hni
qu
e
ca
lle
d
an
t
col
o
ny
op
ti
mi
zati
o
n
(AC
O
)
t
e
c
h
n
i
q
u
e
.
Thi
s
o
p
t
i
mi
za
ti
on
t
ech
ni
qu
e
u
s
ed
f
o
r
s
ol
vi
ng
di
ffe
re
nt
o
p
t
im
izat
i
o
n
pr
ob
le
m
s
s
uc
cessfu
l
l
y
.
A
C
O
tec
h
nique
i
s
a
nove
l
me
t
a
heur
is
ti
c
s
t
r
a
te
gy
a
n
d
h
a
s
b
e
e
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN: 2088-
8694
I
nt
J
P
ow
Elec
& Dr
i
S
y
st, Vol. 10,
N
o.
1, Mar
c
h 2
0
1
9
:
41
–
47
42
effec
t
i
v
e
l
y
use
d
i
n
d
i
ffere
n
t
app
l
i
c
a
t
i
o
ns
e
s
p
ec
ia
l
l
y
o
p
tim
i
zat
i
o
n
pro
b
l
em
s.
A
CO
a
lgor
it
hm
i
mita
te
t
he
r
eal
a
n
ts co
l
on
i
e
s b
e
h
a
vio
r
i
n
f
ound
i
n
g
t
h
e sh
o
r
t
e
st
p
a
t
h
b
e
t
w
e
e
n
f
o
od
s
o
u
r
ces
a
nd its nests [
6
].
Rec
e
n
t
l
y,
m
any
re
sear
chers
foc
u
s
o
n
D
C
d
r
ives
c
o
n
t
r
o
l
,
due
t
o
g
o
o
d
t
or
que-spee
d
c
ha
ra
cterist
i
cs,
si
m
p
le
c
o
n
t
r
o
l
a
rra
ngem
e
nt
a
nd
ty
pe’
s
d
i
v
e
r
sity
t
ha
t
u
s
ed
i
n
d
i
fferent
a
pplications
.
Dif
f
e
rent
t
ypes
o
f
con
t
ro
l
l
ers
a
r
e
ap
pl
i
c
a
b
le
w
it
h
D
C
a
nd
A
C
dri
v
es
c
ontro
l
l
e
rs
r
a
n
g
e
f
ro
m
cl
assi
cal
P
ID
f
a
m
i
l
y
to
i
nt
ell
i
g
ent
con
t
ro
l
l
ers
[7].
T
he
P
ID
c
ont
rol
l
e
r
i
s
w
i
de
a
pp
l
i
c
a
b
l
e
in
i
nd
us
tr
i
a
l
ap
p
l
i
cati
o
ns
d
ue
t
o
its
s
im
p
l
e
str
u
ct
ure
,
ea
se
o
f
use
a
n
d
sim
p
l
e
s
im
ple
t
u
n
i
ng
m
e
t
h
ods,
bu
t
i
t
h
a
v
e
ma
i
n
d
i
sa
dv
anta
ge
w
hic
h
i
s
the
i
n
tro
d
u
ct
ion
o
f
bi
g-o
v
ersh
o
o
t
as
w
e
ll
gi
ve
o
sc
illa
t
i
o
n
i
n
sys
t
em
w
he
n
the
r
e
are
a
l
oa
d
di
s
t
urba
nc
e
e
s
pe
c
i
al
l
y
due
t
o
th
e
effec
t
s
of
p
ro
p
o
rt
io
na
l
an
d
d
e
riva
t
i
ve
k
ic
k
[8]
.
T
h
i
s
P
I
D
c
ontr
ol
ler
d
i
sa
dva
nta
g
e
s
d
e
n
ote
a
re
al
d
esi
g
n
i
n
g
pro
b
lem
in t
he
d
c
and a
c
d
rive
s con
t
rol
s
ys
t
e
m.
I
n
t
h
i
s
ar
tic
le,
the
pro
p
o
s
ed
s
ol
u
t
ion
foc
u
s
on
des
i
gn
i
n
g
O
u
tp
u
t
r
ate
c
o
n
t
ro
l
l
e
r
a
lso
ou
tpu
t
r
eset
con
t
ro
l
l
er
i
nste
a
d
o
f
us
in
g
PI
D
c
o
n
t
ro
ller
’
s
fa
mil
y
.
The
m
o
ti
va
t
i
on
o
f
u
s
e
the
s
e
co
n
t
ro
lle
rs
c
om
es
f
ro
m
the
fa
ct
t
ha
t
t
h
ese
c
o
n
t
r
o
ller
s
a
v
o
i
d
pro
p
o
r
tio
na
l
an
d
der
i
va
tive
k
ick
w
h
ic
h
le
a
d
s
to
r
e
d
uc
e
t
h
e
s
y
ste
m
ove
r
s
hoo
t
als
o
t
hese
c
on
t
r
ol
lers
a
re
l
ess
se
nsit
i
v
e
t
o
s
ys
tem
dis
t
urba
nc
e
w
h
i
ch
m
ak
e
th
e
m
b
et
t
e
r
t
o
u
se
w
it
h
sy
st
ems
tha
t
s
ub
je
c
t
ed
t
o
l
o
a
d
d
i
s
t
u
rb
anc
e
.
A
C
O
is
u
se
d
fo
r
tu
ni
n
g
t
he
pro
p
o
se
d
c
ontro
ller
s
o
p
tima
l
ly
t
o
impr
ov
e
i
t
s
perform
ance
.
ACO
tec
hni
q
u
e
h
a
s
s
im
p
l
e
s
e
a
r
ch
m
eth
o
d
t
h
a
t
can
c
o
v
e
r
th
e
sea
r
ch
s
p
ace
op
ti
ma
ll
y
a
l
s
o
i
t
h
as
low
e
r
alg
o
ri
th
m
pa
ra
m
e
t
e
rs as
w
ll as
i
t
avo
i
ds
e
n
t
ra
pp
e
d
i
n
loc
al
opti
m
a
.
La
rge
n
u
m
b
er
o
f
rea
s
ear
cher
s
gi
ve
g
r
eat
a
tte
nt
i
o
n
o
n
P
IDs
c
ontr
oll
e
r
i
n
t
he
d
c
an
d
a
c
-d
riv
e
c
o
n
t
rol
schem
e
s
base
d
vari
ous
o
p
tim
iz
at
ion
st
r
a
teg
i
es.
In
2
0
14
Ib
ra
him
e
t
.
A
l
.
,
[
9
]
p
r
e
s
e
n
t
t
u
n
i
n
g
m
e
t
h
o
d
f
o
r
P
I
D
con
t
ro
l
l
er
b
ase
d
B
F
and
P
S
O
t
ec
hni
q
u
es
f
or
c
on
t
r
oll
i
ng
dc
-
m
otor
.
In
201
5
Dieg
o
et. al
.,
[
1
0
]
d
i
s
c
u
s
s
t
h
e
dc-m
ot
or
c
ontr
o
l
i
n
r
o
b
o
t
a
r
m
u
si
ng
P
I
D
-
contro
l
l
er
b
ase
d
A
CO
.
I
n
20
1
6
S
um
an
a
n
d
G
iri
[11]
p
r
o
p
o
se
d
dc-
motor
s
p
ee
d
control
s
y
s
t
em
u
s
i
ng
P
ID
c
ont
r
ol
ler-base
d
G
A
.
I
n
20
16
A
b
d
u
l
amee
r
et
.
al
.
,
[
12]
p
re
sent
d
c-
motor
co
n
t
rol
system
b
a
s
e
d
P
ID
c
ontro
lle
r
tune
d
tra
d
iti
ona
l
l
y.
In
2
018
S
h
a
mse
l
din
et
. a
l
.
,
[
1
3
]
p
r
e
s
e
n
t
BL
D
C
M sp
e
e
d
c
ont
r
o
l
system
usi
ng n
o
n
l
i
nea
r
P
ID
c
ont
r
o
l
l
e
r
base
d
GA
.
2.
UNIVERSAL
M
O
T
OR MODEL
UM
i
s
un
co
mp
e
n
sa
t
e
d
-seri
e
s
mo
to
r.
I
ts
one
t
ype
o
f
series-comm
u
tat
i
on
m
achi
n
es.
I
t
c
a
n
ope
ra
te
eit
h
er
f
rom
dc
o
r
ac
s
ource
.
They
a
re
a
pp
lica
b
le
i
n
p
o
rt
able
a
p
p
a
rat
u
s
d
r
i
v
e
.
T
h
e
e
l
e
c
t
ri
c
a
n
d
dyna
mi
c
equa
t
i
o
n
s o
f
U
M a
r
e
[14]:
)
(
)
(
)
(
)
(
)
(
)
(
t
e
t
i
R
R
dt
t
di
L
L
t
V
a
a
f
a
a
f
(
1
)
)
(
).
(
.
)
(
1
t
i
t
K
t
e
a
m
(
2
)
)
(
)
(
)
(
)
(
2
t
K
dt
t
d
J
t
T
t
T
m
m
L
e
(
3
)
2
1
))
(
.(
)
(
t
i
K
t
T
a
e
(
4)
w
h
er
e;
e
i
s
t
h
e
rota
tio
na
l
e
l
ec
tro
mo
t
i
ve
f
orc
e
e
m
f
(
V
)
,
i
a
i
s
the
mo
tor
m
o
tor-
curr
ent
i
n
(
A
)
,
J
i
s
t
h
e
m
o
m
e
n
t
of
i
ner
tia
c
o
n
s
t
a
n
t
(
kg.m
)
,
K
1
,
K
2
i
s
the
motor
c
ons
tan
t
(
N
m
/A
),
L
a
, L
f
are
the
m
o
t
o
r
arm
a
ture
a
nd
fi
e
l
d
in
duc
ta
nce
(
H
),
R
a
, R
f
a
r
e
t
h
e
m
o
t
o
r
a
r
m
a
t
u
r
e
a
n
d
f
i
e
l
d
r
e
s
i
s
t
a
n
c
e
(
Ω),
T
L
is
t
he
l
oa
d
t
o
rque
(
Nm
)
,
T
e
i
s
elec
tr
oma
gne
t
i
c
torq
ue
(
Nm
)
,
ω
m
i
s
t
h
e
m
o
t
o
r
a
n
g
u
l
a
r
s
p
e
e
d
(
ra
d/
s
)
and
V
r
e
p
r
e
s
e
n
t
t
h
e
m
o
t
o
r
in
put
v
olta
ge
(V
)
.
3.
OUT
P
U
T
RATE CONTROLLE
R
A
N
D
O
UTPU
T RESET CONTRO
LLER
A
ny
s
y
s
t
em
i
s
kn
ow
n
t
o
h
av
e
o
u
t
p
ut
r
ate-
co
ntr
o
l
w
h
en
t
he
o
ut
p
u
t
g
e
n
era
tio
n
i
n
s
om
e
w
a
y
depe
n
d
s
o
n
th
e
ra
t
e
a
t
wh
i
c
h
outp
u
t
c
h
a
ng
e
s
.
Out
put
r
a
t
e
c
ont
ro
ll
er
c
an
b
e
ob
t
a
i
n
ed
b
y
f
e
e
d
ing
b
a
c
k
a
d
eri
v
ati
v
e
of
ou
tpu
t
s
i
g
nal
of
p
la
nt
a
n
d
c
om
pari
ng
i
t
w
i
t
h
p
ro
p
o
rt
io
na
l
err
o
r
s
ig
na
l
.
I
ts
i
ntro
duc
tio
n
ofte
n
inc
l
ud
es
t
h
e
cre
a
ti
on
of
a
s
up
ple
m
e
n
t
a
ry
l
o
op
t
h
at
l
ea
d
to
m
ul
til
o
o
p
s
ys
t
e
m.
I
t
is
u
sed
for
im
pr
o
v
in
g
t
h
e
sys
t
e
m
perform
ance
.
In
s
er
v
o
me
chan
ism
sys
t
em
,
tac
hoge
nera
tor
usua
l
l
y
p
r
o
v
i
des
ou
tpu
t
r
a
t
e
fe
e
dbac
k
.
Th
i
s
t
ype
o
f
c
o
nt
roll
e
r
p
rovi
d
e
s
hig
h
e
r
g
a
in
w
it
h
out
poo
rl
y
a
f
f
e
c
tin
g
th
e
d
a
mp
in
g
rat
i
o
tha
t
s
t
i
l
l
sat
i
s
f
y
t
h
e
dam
p
ing
ra
t
i
o
spec
ific
a
tio
ns
b
e
s
ides
t
he
s
t
e
ady
s
t
a
t
e
pe
rfo
r
m
a
nc
e
for
ste
p
i
n
pu
t
s
[
1
5
].
O
utpu
t
re
set
c
o
ntrollers
(
o
r
int
e
gral
Evaluation Warning : The document was created with Spire.PDF for Python.
Int J
P
o
w
El
e
c
&
D
ri S
yst
I
S
S
N
:
2088-
86
94
Spee
d
co
nt
rol of
u
n
i
v
e
rs
al
m
o
t
o
r
(O
m
a
r A.
Im
ran)
43
con
t
ro
l
l
ers)
a
re
u
se
d
t
o
d
e
c
re
ase
the
s
t
e
a
dy
sta
t
e
err
o
r
[16].
Ou
tput
r
e
s
e
t
c
on
t
r
ol
l
e
r
i
s
a
ch
i
e
ved
by
fe
ed
i
ngba
c
k
a
n
i
n
te
gra
l
o
f
out
p
u
t
p
l
a
n
t
si
g
n
al
a
n
d
c
om
pa
r
i
n
g
p
r
opo
rti
o
n
a
l
e
r
ro
r
si
gn
a
l
w
i
t
h
f
eedi
ng
b
ack
signa
l
.
F
ig
ur
e
1 and
F
i
g
u
re
2
show
t
he
o
ut
p
u
t
rate
a
nd
reset
co
n
t
ro
l
l
e
r
s r
e
spec
ti
ve
ly.
F
i
g
u
r
e
1
. O
utp
u
t r
a
te
c
o
n
tr
oll
e
r structure
F
ig
ur
e
2
. O
utp
u
t r
e
set
c
o
n
t
r
oller
struct
ure
4.
ANT COLO
NY
A
C
O
rec
e
ntl
y
p
r
o
p
o
se
d
b
y
D
o
rigo
e
t
a
l.
I
t
is
a
nov
e
l
p
op
u
l
at
i
o
n
-
b
ase
d
t
ec
hn
i
q
ue
u
se
d
for
op
tim
iza
t
i
o
n
p
r
ob
lem
s
s
ol
vi
n
g
.
I
t
m
im
i
c
s
rea
l
a
n
t
s
se
arc
h
in
g
be
hav
i
or
f
or
s
h
o
r
t
e
s
t
r
o
ut
e
fi
nd
ing
be
t
w
e
e
n
fo
o
d
c
en
tre
and
ne
s
t
[
17].
A
r
tificia
l
a
n
t
’
s
col
o
n
y
,
co
oper
a
tes
f
o
r
b
e
st
s
ol
u
tio
ns
f
i
ndi
n
g
,
w
h
ich
a
r
e
a
deve
l
o
p
i
ng
a
n
t
’s
c
o
o
p
era
tiv
e
c
o
mm
un
icat
io
n.
D
ue
t
o
t
h
e
s
i
m
ilar
i
t
i
e
s
w
i
t
h
nat
u
re
a
nt
c
o
l
on
ies,
A
CO
alg
o
ri
t
h
ms
a
r
e
r
ob
ust
an
d
a
d
apt
i
ve
a
nd
ca
n
be
a
ppl
ied
t
o
m
any
p
ro
bl
ems
n
e
ed
s
op
timi
z
a
ti
on
.
Th
e
ma
j
o
r
artific
ia
l
a
n
ts
f
e
a
ture
s
ar
e
c
o
py
from
its
n
a
t
ura
l
m
o
d
e
l
.
T
h
e
s
e
f
eatures
a
r
e
(1)
th
e
y
c
oo
p
e
rat
i
ng
i
ndiv
i
du
al
s
col
o
nie
s
,
(
2
)
by
de
p
o
s
i
t
i
ng
p
h
e
r
om
one
t
he
y
c
a
n
com
m
u
n
i
c
a
te
i
n
d
i
rec
tly
(
3)
b
ase
d
o
n
l
o
c
a
l
m
o
ves
se
que
n
c
e
for
fi
n
d
in
g
nea
r
est route
fr
o
m
be
g
i
n
in
g t
o
e
n
d
po
i
nt [1
8
].
A
n
ts
i
n
i
t
i
a
l
ly
d
i
s
c
over
t
h
e
a
r
ea
r
andom
l
y
t
ha
t
c
l
ose
d
t
he
ir
c
ol
o
ny.
D
urin
g
di
sco
v
er
ing
proc
ess,
a
nt
s
leav
i
ng
a
p
h
erom
one
p
a
t
h.
T
he
p
herom
o
ne
d
en
si
ty
r
e
l
ate
d
t
o
tra
i
l
le
n
g
t
h
a
nd
the
disc
ov
e
r
ed
f
o
o
d
s
o
u
rce
.
A
n
ant se
l
e
cts a
n
e
xa
ct ro
u
te
d
e
p
e
n
d
i
ng o
n
t
he
p
he
rmone
.
[19].
Step
1
: In
it
i
a
liz
ati
on
of
p
ara
m
ete
r
s
To c
alc
u
la
te
a
n
t
s t
our m
ax.
D
i
sta
n
c
e
use
fo
l
l
o
w
i
ng
eq
ua
ti
o
n
:
1
1
max
max
n
i
i
d
d
(
5
)
)
max(
u
r
d
i
(
6
)
Whe
r
e,
d
i
is the
nodes dist
a
n
ce.
u is u
nv
is
i
t
e
d
n
ode.
r i
s
p
r
ese
n
t
node
.
Step
2
: In
it
i
a
l pos
it
io
n ge
nera
t
i
o
n
Th
is
s
t
e
p il
lus
t
ra
t
e
s the
g
e
nera
tio
n
of r
and
o
m
f
irst pos
iti
on
f
or e
ac
h ant.
Step
3
:
Rule
of
transi
t
i
o
n
Th
is
s
t
e
p il
lus
t
ra
t
e
s the
ch
ose
n
pr
oba
b
ili
t
y
for
nex
t no
de
b
y a
n ant
(7):
K
T
ij
ij
ij
ij
ij
k
ij
T
j
i
t
t
t
t
t
P
k
,
:
)
(
)
(
)
(
)
(
)
(
(
7
)
wher
e
τ
ij
(t
)
: is the nodes p
he
rom
o
ne
trial.
η
ij
(t
)
: represent
t
he inver
se of
dis
t
ance.
T
k
:
i
s
the e
ffe
c
t
ua
te
d
pa
th.
S
t
ep
4
: U
p
dat
i
ng l
o
ca
l
pher
o
m
one
Thi
s
s
t
e
p
il
l
u
strat
e
s
th
e
ph
e
r
omo
n
e
u
pd
at
in
g
p
r
o
c
e
ss
f
o
r
e
ach
a
nt.
The
l
o
cal
p
h
e
rom
o
ne
u
p
d
a
t
i
n
g
i
s
unl
i
k
e
am
ong a
n
ts
d
u
e
to di
ffere
n
t
rou
t
e
t
a
ken b
y
a
nt
s. Ea
c
h
ant i
n
i
t
i
al
p
h
e
rom
o
ne is g
i
ve
n
b
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN: 2088-
8694
I
nt
J
P
ow
Elec
& Dr
i
S
y
st, Vol. 10,
N
o.
1, Mar
c
h 2
0
1
9
:
41
–
47
44
o
ij
ij
t
t
)
(
)
1
(
)
1
(
(
8
)
St
e
p
5
:
Ev
a
l
u
a
ti
on
of
c
o
st
funct
i
on
The
bes
t
cos
t func
t
i
o
n
r
ela
t
e
d
to sh
ortes
t
pa
t
h
w
h
ic
h r
e
late
d
t
o
h
i
ghe
r p
h
e
r
omone
de
n
si
t
y
.
S
t
ep
6
: U
p
dat
i
ng g
l
oba
l p
h
e
r
omone
The
p
h
erom
on
e le
vel
is gi
v
e
n
by:
)
(
)
(
)
1
(
)
1
(
t
t
t
ij
ij
ij
(
9
)
St
e
p
7
:
ACO al
g
o
r
i
t
h
m
t
e
r
mi
na
t
i
on
o
r s
t
op
pin
g
crit
eri
a
In
t
h
i
s
step
t
he
p
rogram
(
A
C
O
a
l
gor
i
t
hm
)
w
i
l
l
b
e
e
nde
d
w
h
en
t
he
m
ax.
Numbe
r
o
f
i
t
e
r
a
tions
i
s
ac
hie
v
ed
o
r
t
h
e
opt
i
m
a
l
so
l
u
tio
n
is at
t
a
i
n
e
d
wi
th
out
s
t
a
gn
at
io
ns o
f a
n
t
s
.
5.
S
I
M
U
LA
T
I
O
N
A
N
D
R
ES
UL
TS
Whe
n
us
i
ng tra
n
sfer func
t
i
o
n
t
o m
ode
l the
pl
ant, some
a
ppr
ox
i
m
a
t
i
o
n
s
ho
ul
d be
t
ak
i
ng p
l
ac
e suc
h
a
s
ig
nor
e
a
l
l
in
it
ia
l
c
o
n
d
i
t
i
o
n
s
for
p
l
a
n
t
m
ode
l.
T
h
i
s
a
p
pro
x
i
m
a
ti
o
n
w
il
l
g
i
ve
a
n
acc
e
p
ta
ble
r
e
su
l
t
f
or
p
la
nt
respo
n
se
i
n
c
o
n
t
r
o
l
st
u
d
i
e
s.
W
h
i
l
e
w
he
n
us
in
g
ma
t
h
e
m
a
t
i
c
a
l
m
ode
l
for
m
odel
lin
g
the
c
o
mp
lete
c
on
t
r
ol
syste
m
,
a
l
l
in
it
ial
c
o
n
d
i
t
i
o
n
s
a
re
c
onsi
d
ere
d
w
he
n
m
odel
l
i
ng
th
e
system
.
So,
the
o
b
ta
i
n
ed
r
esul
ts
a
re
m
or
e
ac
cura
t
e
a
n
d
t
h
e
m
athem
a
t
i
ca
l
m
odel
are
clo
s
e
t
o
a
c
t
ua
l
pla
n
t
b
eha
v
i
o
r
.
S
o,
i
n
t
h
i
s
a
r
t
ic
le
a
f
ocus
i
s
m
a
de
o
n
a
m
a
them
at
ica
l
m
ode
lli
ng
o
f
t
h
e
s
y
s
t
e
m
due
t
o
ad
va
n
t
a
g
es
o
f
thi
s
ap
proa
c
h
.
The
c
o
m
p
l
e
te
m
ode
l
o
f
t
he
U
M
is
d
es
ig
ne
d ba
s
e
d m
o
tor e
l
ec
tric a
nd
dy
n
a
m
ic
eq
u
a
t
i
o
n
s
(
se
e
se
c
t
i
on 2)
u
si
n
g
MA
TLA
B/S
I
M
U
L
I
N
K
to
ol
bo
x.
O
u
t
p
u
t
r
ate
a
nd
out
p
u
t
r
ese
t
c
on
tro
l
le
rs
a
re
d
esi
g
ned
a
l
s
o
f
or
cons
truc
ti
n
g
t
he
m
ot
or
c
l
o
se
d
l
o
o
p
c
ontr
o
l
system.
AC
O
strate
g
y
i
s
build
u
si
ng
m
a
t
l
a
b
m-f
i
l
e
a
nd
l
i
nke
d
wi
th
S
im
ul
i
nk
mo
de
l
o
f
m
otor
c
o
n
t
rol
sy
stem
.
Inte
gra
l
t
i
m
e
abs
o
lu
te
e
rr
or
i
s
use
d
a
s
per
f
o
r
ma
nce
i
n
d
i
ce
s.
U
n
i
v
er
sal
m
o
tor
use
d
p
ar
am
eters
m
e
nti
o
ned
in Ta
b
le
1.
Ta
b
l
e
1.
U
M pa
ram
e
te
rs
Unive
r
sa
l
m
o
tor
pa
r
a
m
e
t
e
r
s
S
ym
bol
U
nit
Va
lue
M
o
tor
P
o
we
r
P
hp
5
M
o
tor
input
v
olta
ge
V
V
220
r
o
t
o
r c
u
rre
nt
i
a
A
22
M
o
tor
s
p
e
e
d
ω
m
R
a
d/s
188.
4
roto
r
R
e
sista
n
ce
Ra
Ω
0
.6
Fie
l
d Re
si
st
a
n
c
e
R
f
Ω
0.
4
Ro
to
r
I
n
du
ct
a
n
c
e
L
a
H
0
.
0
3
F
i
e
l
d
Induc
t
a
n
c
e
L
f
H
0
.
002
M
o
m
e
nt
o
f
ine
r
tia
c
onsta
nt
J
K
g.m
2
0
.
0465
Vi
sc
ous
F
ri
c
tion
Coe
f
.
K2
N
m
s
/ra
d
0
.
005
M
o
tor
C
onsta
nt
K
1
N
m
/
A
0
.
027
The
A
C
O
a
l
g
o
r
ithm
par
a
m
e
ter
s
t
ha
t
use
d
f
or
t
u
n
i
n
g
the
out
pu
t
r
a
te
c
on
t
r
o
lle
r
a
n
d
ou
tpu
t
r
ese
t
con
t
ro
l
l
ers
a
r
e
lis
te
d
i
n
T
a
b
l
e
2
,
w
h
il
e
t
h
e
ob
ta
ine
d
t
u
n
e
d
c
on
trollers’
param
e
ters
b
ased
A
CO
t
e
c
h
ni
que
a
re
me
ntio
ne
d
in T
able
3.
T
a
b
l
e 2.
ACO alg
ori
t
hm
pa
r
am
eter
s
AC
O a
l
gor
i
t
hm
p
aram
et
e
r
s
O
u
tput
r
a
t
e
c
ontrol
l
e
r
R
e
s
e
t
c
ontrolle
r
Nu
m
b
er
o
f
an
t
s
(
m)
1
0
1
0
Nu
m
b
e
r
o
f
node
s
(n)
1000
00
1000
00
M
a
xi
m
u
m
ite
r
a
ti
on
(
t
m
a
x)
1
0
10
P
h
e
r
o
m
o
n
e
d
e
c
a
y
p
a
r
ame
t
er (
α)
0
.8
0
.8
p
h
e
r
o
m
o
n
e
r
el
at
i
v
e i
m
p
o
r
t
a
n
c
e
a
g
ai
n
s
t d
i
s
t
an
ce p
a
r
a
met
e
r
(
β
)
0
.
2
0.
2
H
e
urist
i
c
-
c
o
e
f
fic
i
ent
(
ρ
)
0.
7
0.
7
Ta
b
l
e
3.
The
obta
i
ned c
o
n
t
r
o
ller
s
’
pa
ram
e
ters
b
ased
A
CO
O
utput
r
a
t
e
c
ontrol
l
e
r
O
utput
r
e
s
e
t
c
ontrolle
r
K
1
16.
336
2
97.
11
44
K
2
0
.
3177
0
.
000
0
080
86
Evaluation Warning : The document was created with Spire.PDF for Python.
Int J
P
o
w
El
e
c
&
D
ri S
yst
I
S
S
N
:
2088-
86
94
Spee
d
co
nt
rol of
u
n
i
v
e
rs
al
m
o
t
o
r
(O
m
a
r A.
Im
ran)
45
The
F
i
t
n
e
s
s
func
tio
n
p
l
ot
f
or
t
un
i
ng
t
h
e
ou
t
p
u
t
r
a
t
e
co
ntr
o
l
l
e
r
a
n
d
o
utp
u
t
rese
t
c
ontr
o
l
l
ers
ar
e
ill
us
trate
d
i
n
F
i
g
u
re
3
a
n
d
F
igure
4
r
e
spec
ti
vel
y
.
F
r
om
t
he
t
w
o
f
it
ness
pl
ot
s,
it
is
c
lea
r
l
y
t
ha
t
a
f
t
e
r
th
e
seco
nd
or
t
h
i
r
d
itera
ti
on,
t
he
f
it
ness
f
u
n
c
tio
n
pl
o
t
i
s
de
cre
a
sed
s
h
ar
pl
y
.
I
t
i
s
o
b
v
io
us
t
h
a
t
t
h
e
co
st
f
un
ct
ion
p
l
ot
i
s
dec
r
ea
se
a
fter
t
he
s
e
c
o
n
d
o
r
thir
d
i
t
e
r
at
io
n
a
nd
a
ppr
ox
ima
t
e
l
y
s
t
ea
dy.
T
h
i
s
prove
t
he
f
as
t
c
o
nv
e
r
genc
e
abi
l
i
t
y
of
t
he
A
CO
a
lgor
it
hm
w
hic
h
l
ea
ds
t
o
o
b
t
a
in
t
he
o
pt
i
m
al
s
olu
t
i
o
n
s
wi
t
h
o
u
t
r
e
s
o
rte
d
t
o
i
n
cr
ease
t
h
e
nu
m
b
er
o
f
it
e
r
a
t
i
o
ns
w
h
i
c
h
l
eads
t
o
l
o
w
er
c
onse
c
u
t
i
on
t
i
me
o
f
tu
ni
ng
proc
ess.
W
h
i
l
e
o
t
h
er
o
pt
imiz
a
t
i
o
n
s
t
rateg
i
e
s
norm
a
l
l
y
nee
d
h
i
g
h
n
u
m
b
er
o
f
ite
rat
i
o
n
s
fo
r
con
v
e
r
ge.
F
r
om
a
ll
t
h
e
s
e
p
r
ove
s,
A
CO
h
ave
s
uperi
or
f
e
a
tures
tha
n
o
ther
o
pt
im
iza
tio
n stra
t
e
gi
e
s
.
F
i
gure
3. F
it
ne
ss
p
lot
for
t
u
ni
ng o
u
t
p
u
t
ra
t
e
F
i
gure
4. F
it
ne
ss
p
l
o
t
fo
r t
u
n
i
n
g
out
put
res
et
U
M
i
s
teste
d
b
y
ap
p
l
y
i
ng
differe
nt
l
oa
ds
t
o
show
t
he
r
o
b
u
s
t
n
ess
o
f
t
h
e
c
o
n
t
ro
lle
rs
t
h
a
t
t
u
ne
d
base
d
A
C
O
tec
h
n
i
qu
e
unde
r
d
i
ffer
ent
l
o
a
d
i
n
g
c
o
ndi
ti
o
n
s.
F
i
g
ure
5
a
n
d
F
i
g
u
re
6
s
how
s
a
c
o
mpa
r
is
on
i
n
s
pee
d
respo
n
ses
u
nde
r
d
i
ffer
en
t l
o
a
d
s usi
ng o
u
t
p
u
t
r
ate
and
o
u
tp
ut
r
e
set c
ontro
lle
rs re
s
pect
ive
l
y.
F
i
g
u
r
e
5
. UM
s
p
eed
u
s
i
ng o
u
t
pu
t ra
te c
on
t
r
o
ller
F
i
gure
6. UM
speed
u
si
n
g
o
u
t
pu
t re
set c
o
n
t
r
o
l
l
er
F
r
om
r
esult
s
it
is
c
l
e
arl
y
t
ha
t,
f
or
b
oth
t
une
d
c
o
n
t
ro
l
l
e
r
s
t
h
e
r
e
a
r
e
a
c
l
e
a
r
i
m
p
r
o
v
e
m
e
n
t
i
n
s
y
s
t
e
m
trans
i
en
t
a
n
d
s
t
ea
d
y
s
ta
te
p
er
form
ance
f
or
a
ll
l
o
a
d
i
n
g
c
o
n
d
it
io
ns.
Bo
th
t
une
d
c
o
n
t
ro
l
l
e
r
s
give
z
ero
p
e
r
cent
over
sh
oo
t
a
n
d
un
der
sh
oot
,
sli
g
h
t
r
i
s
e
t
i
m
e
a
nd
se
t
tli
n
g
time
an
d
f
a
s
t
r
e
s
pon
s
e
p
ro
du
c
e
f
o
r
a
l
l
l
o
ad
di
st
urba
nces.
F
i
gur
e
7
ill
u
s
t
r
ate
s
p
e
e
d
respo
n
se
c
om
p
a
ris
on
of
U
M
w
i
t
h
bo
t
h
c
on
tro
llers
u
n
d
er
di
ffe
re
nt l
oa
ds.
1
2
3
4
5
6
7
8
9
10
1.
21
45
1.
2
1
5
1.
21
55
1.
2
1
6
1.
21
65
1.
2
1
7
1.
21
75
x
1
0
4
I
t
er
at
i
o
n
F
i
t
n
e
ss
V
a
l
u
e
F
i
t
n
e
ss
P
l
o
t
1
2
3
4
5
6
7
8
9
10
0
0.
5
1
1.
5
2
2.
5
3
3.
5
4
4.
5
5
x
1
0
5
I
t
er
at
i
o
n
F
i
t
n
e
ss
V
a
l
u
e
F
i
t
nes
s
P
l
ot
0
0.
5
1
1.
5
2
2.
5
3
3.
5
4
4.
5
5
-5
0
0
50
100
150
200
Ti
m
e
(
s
)
A
n
g
u
l
a
r
s
p
e
ed
(
r
ad/
s
)
N
o
l
oad
Li
g
h
t
l
oad
H
a
lf
f
u
ll
lo
a
d
Fu
ll
lo
a
d
0
0.
5
1
1.
5
2
2.
5
3
3.
5
4
4.
5
5
-5
0
0
50
100
150
200
Ti
m
e
(
s
)
s
peed
(
r
ad/
s
)
no
l
o
ad
l
i
ght
l
oad
hal
f
f
u
l
l
l
oad
f
u
ll
lo
a
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N: 2
0
8
8
-
86
94
I
nt
J
P
ow
E
l
e
c
&
Dr
i
S
y
st,
Vol.
10,
N
o.
1
,
Mar
c
h
2
0
1
9
:
41
–
47
46
(a)
U
M
s
pee
d
r
e
s
p
ons
es
u
n
d
er
n
o
load
(
b)
U
M
s
pee
d
r
e
s
pons
e
s
u
n
d
er
lig
ht
l
oa
d
(c)
UM
sp
e
ed
respo
n
s
es u
nd
e
r
h
al
f
fu
ll
lo
a
d
(d
)
UM sp
e
ed
r
espo
n
s
es u
nd
er
f
ull lo
ad
F
i
gur
e
7.
C
ompa
r
i
son
o
f
U
M
spee
d
r
e
spo
n
se
s
und
er
d
i
f
fer
e
nt
l
oa
d
s
us
in
g
t
h
e
tw
o
co
ntr
o
l
l
e
r
s
F
r
om
F
igur
e
7,
it
is
o
bvi
o
u
s
l
y
tha
t
t
he
U
M
s
p
e
e
d
r
e
s
p
onse
s
w
i
t
h
t
h
e
t
w
o
c
o
n
t
r
o
l
l
e
r
s
a
r
e
a
ppr
ox
i
m
a
t
e
l
y
si
m
i
lar
wi
t
h
v
ery
sma
ll
err
o
r
f
o
r
a
ll
l
o
a
d
d
ist
u
r
ba
nce
s
.
The
tw
o
pr
o
pos
ed
c
o
n
t
r
oll
e
r
s
a
r
e
a
ppr
opr
ia
te
f
or
c
ontr
o
l
lin
g
sy
stem
s
sub
j
ec
te
d
to
l
oa
d
dist
ur
ban
ces.
6.
CONCLUSIONS
Th
is
p
a
p
er
f
oc
us
o
n
m
a
t
h
em
atica
l
m
o
d
el
d
e
s
i
g
n
i
ng
bec
a
u
se
i
t
s
m
o
r
e
r
e
a
li
ty
t
o
a
c
t
ua
l
p
l
an
t
wi
tho
u
t
r
e
sor
t
ed
t
o
i
g
n
o
r
e
a
ny
i
n
i
t
ia
l
con
d
i
t
i
on.
F
or
a
ppr
opr
i
a
te
U
M
sp
ee
d
co
ntr
o
l,
a
n
o
u
tp
ut
r
ate
c
o
ntr
o
l
l
er
a
nd
o
u
tp
ut
r
e
s
e
t
c
on
t
r
o
l
l
e
rs
a
re
u
sed
du
e
to
i
t
s
i
mp
ac
t
on
i
mp
r
o
vi
n
g
the
t
r
ans
i
en
t
a
n
d
s
t
ea
d
y
state
pe
rform
a
nc
e
o
f
the
sy
ste
m
.
Th
ese
co
ntr
o
l
l
er
s
ha
ve
s
ome
ad
va
n
t
a
g
es
o
v
e
r
ot
her
c
ont
rol
l
e
rs
s
u
c
h
si
mpl
e
c
on
st
r
u
c
tio
n
,
l
o
w
er
c
o
mple
x
i
ty,
i
m
pr
ove
t
he
s
y
s
tem
tr
a
n
sie
n
t
and
s
t
ea
d
y
s
t
a
te
p
er
fo
r
m
a
n
ce
a
nd
r
e
q
u
i
r
e
d
sim
p
ler
tu
nin
g
.
A
CO
t
e
chni
qu
e
i
s
u
se
d
fo
r
op
ti
mal
c
ont
rol
l
e
rs
t
uni
ng
t
o
g
e
t
b
e
st
s
y
s
t
em
p
e
r
for
m
a
n
c
e
.
A
m
ong
m
a
n
y
n
atur
a
l
in
sp
ir
ed
o
ptim
i
z
at
i
on
tec
h
n
i
q
u
es,
A
C
O
ha
v
e
m
any
adva
n
t
age
s
s
uc
h
a
s,
l
ow
a
lg
or
i
t
h
m
par
a
me
ter
s
r
equir
e
d
,
m
i
ni
m
u
m
iter
a
tio
ns
r
eq
uir
e
d,
f
ast
c
o
nver
g
e
n
c
e
a
n
d
l
ow
e
xe
cu
t
i
o
n
t
i
m
e
.
A
l
l
t
h
es
e
a
d
v
a
nt
ag
e
s
m
ak
e
t
h
i
s
t
e
chni
qu
e
mo
r
e
a
p
p
r
o
p
r
i
a
t
e
f
o
r
s
o
l
vi
ng
d
if
f
e
re
nt
o
pti
m
iza
t
i
o
n
p
r
o
b
l
e
m
s.
T
he
r
e
s
ul
ts
s
how
a
n
o
b
v
i
ou
s
im
pr
o
v
em
ent
in
s
yste
m
per
f
o
r
m
a
nce
by
r
e
duc
i
ng
r
i
se
t
i
m
e,
s
e
t
t
l
i
ng
t
i
me
a
nd
e
l
im
i
n
a
t
e
peak
over
s
h
o
o
t
f
or
di
ff
er
e
n
t
a
p
p
l
ie
d
l
o
ad
s
be
tw
e
e
n
n
o
-
l
oad
a
n
d
fu
ll-
l
o
a
d
.
The
syst
em
i
s
test
ed
w
it
h
l
o
a
d
s
c
h
an
ges
gr
adu
a
ll
y
a
t
di
ffere
n
t
t
i
m
e
s
as
w
e
l
l
a
s
a
t
c
h
a
n
g
i
n
g
t
he
r
efere
n
ce
s
p
e
e
d
fr
om
b
e
l
ow
r
at
ed
s
pe
ed
t
i
l
l
o
v
er
r
ate
d
s
pee
d
.
Th
e
U
M
w
i
t
h
c
o
ntr
o
l
l
er
s
ba
se
d
A
C
O
t
e
c
h
ni
q
u
e
show
s
a
n
o
p
t
im
al
p
er
f
o
r
ma
nc
e
i
m
p
r
ov
e
m
e
n
t
fo
r
d
i
ffe
r
e
n
t
l
o
a
di
ng
tes
t
s.
T
he
t
w
o
t
u
n
i
n
g
c
ont
r
o
l
l
er
s
g
i
ve
o
ptim
al
r
es
ul
t
s
f
or
c
o
n
t
ro
ll
ing
t
h
e
mo
tor
spee
d
u
n
d
er
v
ar
iou
s
loa
d
in
g a
p
p
l
ic
a
tio
ns.
REFERE
NC
E
S
[1
]
S
.
B
aisa,
B.
P
u
r
wahy
ud
i,
a
nd
K
.
Ku
sp
ijan
i
,
"
C
o
n
t
r
ol
S
tra
t
eg
y
for
PW
M
Vo
ltag
e
S
o
u
r
ce
Con
v
ert
e
r
Us
in
g
F
u
zzy
L
o
g
i
c
f
o
r
Ad
ju
s
t
ab
le
S
peed
D
C
M
o
tor,
"
In
t
e
rn
atio
nal
Jour
na
l
o
f
P
o
wer
E
l
ectronics
a
n
d
Dri
ve S
y
st
ems
(
I
J
P
EDS)
,
vo
l
.
8
, p
p.
51
-
58
, 20
1
7
.
[2]
A.
K
.
P.
P
a
n
k
a
j
S
a
hu
,
"
A
c
o
m
para
tive
a
na
lysis
b
e
twe
e
n
f
u
z
z
y
c
o
n
t
r
oll
e
r
an
d
pi
d
con
t
ro
ll
er
f
o
r
a
u
n
i
v
e
rs
al
m
o
t
or,"
A
s
i
an Jou
r
nal o
f
En
g
i
n
eeri
n
g
Re
s
e
ar
ch
,
v
o
l
.
1
,
pp.
0
1-0
4
,
201
3.
0
0.
5
1
1.
5
2
2.
5
3
3.
5
4
4.
5
5
-5
0
0
50
10
0
15
0
20
0
Ti
m
e
(
s
)
s
peed(r
ad/
s
)
o
u
t
p
u
t
r
at
e
c
o
nt
ro
l
l
e
r
r
e
se
t
co
n
t
r
o
l
l
e
r
0
0.
5
1
1.
5
2
2.
5
3
3.
5
4
4.
5
5
-5
0
0
50
100
150
200
Ti
m
e
(
s
)
sp
e
e
d
(
r
a
d
/
s)
out
pu
t
r
at
e
c
ont
r
o
l
l
e
r
re
s
e
t
c
o
n
t
ro
l
l
e
r
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t
J
P
o
w
Elec
&
D
r
i
S
y
st
I
S
S
N
:
2088-
86
94
Spe
e
d
c
o
ntro
l
of un
iv
ers
a
l
m
o
t
o
r (
O
m
a
r A.
I
m
ran)
47
[3
]
Y
.
J
i
a
y
u
an
,
L.
H
ong
xi
n,
Q
.
Yunxi
a,
Y
.
Xi
aoan
,
Y.
Y
u
a
n
,
G
.
A
i
p
in
g,
e
t
al.
,
"
F
u
zz
y
i
n
terf
eri
ng
co
nt
rol
of
u
n
i
vers
al
m
o
tor
f
o
r
el
ectro
m
a
g
n
et
ic
n
onlinearit
y,
"
p
p
.
519
-52
2
,
20
09.
[4]
A.
A
mbr
i
si,
M.
D
emagistris
,
and
R.
F
resa,
"Multi-obj
ecti
v
e
O
p
timi
zatio
n Ba
s
e
d
Des
i
gn
o
f
Hi
gh
E
ffi
cien
cy
D
C-DC
S
w
it
c
h
in
g
Convert
ers,
"
In
te
rn
atio
na
l J
o
u
r
na
l o
f
Po
we
r
Ele
c
tro
n
ic
s
an
d
Driv
e
Sy
ste
m
s
(
IJ
PE
D
S
),
v
ol
.
7,
p. 37
9
, 2
01
6.
[5
]
R.
K
. A
r
ora,
O
ptimi
zati
o
n
:
a
l
g
o
r
ith
m
s an
d ap
p
l
icat
ions: CRC
Pres
s
, 20
1
5
.
[
6
]
Z
.
C
.
S
.
S
.
H
l
a
i
n
g
a
n
d
M
.
A
.
K
h
i
n
e
,
"
A
n
a
n
t
c
o
l
o
n
y
o
p
t
i
m
i
z
a
tion
a
lg
or
ith
m
f
or
s
olv
i
ng
t
ra
v
e
ling
s
a
l
e
s
ma
n
problem,"
in
In
t
e
rn
atio
nal Co
n
f
erence o
n
In
fo
rm
a
t
ion Commu
ni
ca
t
i
on a
nd M
a
n
agemen
t
, pp
.
5
4-5
9
, 2
01
1.
[7
]
C.
C
op
ot,
C.
I.
M
ures
an,
and
R.
D
.
Keys
er,
"S
peed
a
nd
p
os
iti
on
c
o
n
t
r
ol
o
f
a
D
C
m
ot
or
u
si
ng
f
raction
a
l
ord
e
r
P
I-P
D
co
nt
rol,
"
pres
ented
at
t
he
3
r
d Int
e
rn
atio
na
l
Co
nferen
ce o
n
Fra
c
ti
on
al
Sign
als and Sys
t
em
s
-
F
SS
201
3,
G
hent,
Bel
g
ium
,
2
01
3.
[8
]
D
.
P
uang
do
wn
reon
g,
A
.
N
a
wi
k
a
v
a
tan,
a
n
d
C
.
T
h
amm
a
rat,
"
Op
tim
al
D
esign
o
f
I-PD
Controller
for
DC
M
ot
o
r
S
p
eed
C
o
n
tro
l
S
y
s
t
e
m
by
C
uck
oo
Search
,"
Pr
ocedi
a
Co
mp
ut
e
r
S
c
i
e
nce P
r
o
c
e
d
ia
Com
puter
Sci
e
nce
,
vo
l.
8
6,
pp
.
8
3
-
86
, 2
01
6.
[
9
]
H
.
E
.
A
.
I
b
r
a
h
i
m
,
F
.
N
.
H
a
s
s
a
n
,
a
n
d
A
.
O
.
S
h
o
m
e
r
,
"
O
p
t
i
m
a
l
P
I
D
c
o
n
t
r
o
l
o
f
a
b
r
u
s
h
l
e
s
s
D
C
m
o
t
o
r
u
s
i
n
g
P
S
O
a
n
d
BF
t
echn
i
qu
es,
"
A
i
n
Sh
am
s E
n
g
i
n
eerin
g
Jour
na
l A
i
n
Sh
am
s E
n
g
i
n
eerin
g
J
o
ur
na
l
,
vol.
5
,
p
p
.
391
-39
8
,
20
14.
[10
]
D
.
San
d
o
v
al,
I
.
S
o
t
o
,
P
.
Adasme,
E.
E
.
I
.
C
h
i
l
ean
C
on
feren
c
e
o
n
Elect
rical
,
and
T
.
C
o
m
m
u
n
i
catio
n,
"
Co
nt
rol
o
f
d
i
rect
c
urrent
mot
or
u
si
ng
Ant
C
o
lo
ny
opti
m
ization
,
"
pp
.
7
9-82,
201
5.
[1
1]
S
.
K
.
S
um
an,
V
.
K
.
Gi
ri,
I.
I
.
C.
o
.
E
n
g
i
n
eering,
a
nd
T
ech
n
o
l
ogy
,
"
S
p
eed
c
on
tro
l
o
f
DC
m
o
t
or
u
sin
g
o
pti
m
izatio
n
t
echn
i
qu
es b
ased
P
ID
Co
n
t
r
o
ller," p
p
. 5
81
-58
7
, 201
6.
[1
2]
A
.
Ab
dulam
eer,
M
.
S
u
l
a
i
m
a
n,
M
.
S.
M
.
A
r
a
s
,
an
d
D
.
S
aleem
,
"
T
uni
ng
m
eth
ods
o
f
P
I
D
co
ntro
ll
er
f
or
D
C
m
o
tor
s
p
eed
c
o
n
t
r
ol
,"
In
do
nes.
J.
Elec
t
r
i
c
al
Eng.
Com
p
u
t
.
S
c
i.
In
d
o
nesi
an
Jour
na
l o
f
El
ectr
i
ca
l
En
gi
neeri
n
g
and
Co
mpu
t
er
Sc
i
e
nce
,
vol
.
3
,
p
p.
3
4
3
-34
9
,
20
16.
[1
3]
M
.
A.
S
ham
s
e
l
din
,
M
.
A.
A
.
G
h
an
y,
a
nd
A
.
M
.
A
.
G
h
any
,
"
P
e
r
form
an
ce
s
t
u
dy
of
e
nh
anced
n
o
n
-l
in
ear
P
ID
c
on
tro
l
ap
pl
ied
o
n
b
ru
shless
D
C
m
o
t
o
r,
"
Int
e
rna
t
i
o
n
a
l
J
o
ur
na
l o
f
Power Electro
ni
cs an
d
Drive Sys
t
em
s
(
I
J
P
E
D
S
)
,
v
o
l
.
9
,
p
p
.
536
-54
5
,
2018.
[1
4]
A
.
S
.
Z
ei
n
El
D
in,
M
.
E
.
El
-S
hebiny
,
M
.
M
.
Kh
ater,
and
I.
I.
S
.
o.
I
.
E.
P
roceedi
n
g
s
o
f
,
"
M
i
cr
ocompu
ter
-
co
nt
roll
ed
uni
versal mo
t
o
r
,
"
v
ol.
2
, pp
.
6
53
-
6
5
8
v
o
l
.
2,
19
9
6
.
[15]
A
.
Kumar,
control
sys
tems vol
.
c
g
raw-
hill company. india
:
m
,
2
006.
[16
]
S
. Palan
i, Con
t
r
o
l
sys
tems en
g
in
eering
. New
D
el
hi: Ta
t
a
Mc
G
r
a
w
H
il
l
E
d
u
c
a
t
io
n,
2
0
1
0
.
[1
7]
P
.
S.
S
he
lo
kar,
V
.
K.
J
ay
araman
,
and
B.
D
.
K
u
lk
a
r
n
i
,
"
A
n
a
nt
c
ol
on
y
ap
proach
f
or
c
l
u
stering,
"
A
N
ALYTIC
A
CH
IMIC
A A
C
TA
,
vo
l.
5
09,
p
p
.
187
-19
5
,
200
4.
[1
8]
B
.
Nag
a
ra
j
,
N
.
Mu
rug
a
nan
t
h,
I
.
I.
C
.
o.
C
.
Contro
l,
a
nd
T
.
C
o
m
p
u
tin
g
,
"
A
co
m
p
arati
v
e
stu
dy
of
P
ID
c
o
n
t
r
oller
t
u
n
i
ng
u
s
i
ng
GA,
EP,
P
S
O
a
n
d
A
CO,
"
p
p.
3
05
-313,
201
0.
[1
9]
A
.
S
.
O
sh
aba,
A
.
S.
O
s
h
aba,
E
.
S
.
A
li,
E
.
S
.
A
l
i
,
and
S
.
M
.
Abd
El
az
im,
"Ant
C
o
l
ony
Op
timi
zat
ion
al
gori
thm
f
o
r
s
p
eed
c
o
n
tro
l
o
f
S
R
M
f
e
d
b
y
P
ho
to
vo
lt
aic
system
,"
J.
Elect
r. E
ng.
Jo
ur
na
l
o
f
E
l
ectrica
l
En
g
i
neer
ing
,
v
o
l
.
1
5,
p
p
.
55-6
3
,
2
0
15
BIOGRAPHI
E
S
OF
AUT
HORS
Om
ar
A
.
Im
ran
receiv
e
d
a
b
ach
elo
r
's
d
egree
in
e
lect
ro
ni
c
f
r
om
e
ng
in
eering
co
ll
ege-Di
yal
a
Un
iversit
y
i
n
2
0
06
a
n
d
r
ecei
v
e
d
a
m
a
st
er's
d
egree
i
n
e
lect
rical
e
ne
rg
y
fr
om
B
e
l
g
o
ro
d
Go
vernm
e
n
t
T
e
c
hn
ol
og
y
U
n
i
v
e
r
s
i
ty
R
u
ssia'
s
Fe
d
e
ral
in
2
0
1
3
.
A
rea
o
f
researc
h
i
n
t
eres
t
in
t
h
e
electri
c
p
o
w
e
r
eng
i
n
eering
.
H
e
h
a
s
m
o
re
t
han
s
c
ien
t
ifi
c
r
es
ea
rc
h
pu
blishe
d
in
i
nte
r
na
tion
a
l
jo
urnal
s
.
Em
a
il
:Om
a
rim
r
an
53@
yah
oo.
co
m
Wisam
Najm
A
L
-
D
i
n
Ab
ed
r
ec
eiv
e
d
a
b
ach
elo
r
's
d
egree
in
e
lect
ri
c
al
p
o
w
er
a
nd
m
achi
n
es
f
rom
eng
i
neeri
n
g
co
llege-Di
y
al
a
U
n
iversi
ty
i
n
20
05
an
d
rec
e
ived
a
m
a
st
er's
d
egree
in
e
l
ectrical
eng
i
neeri
n
g
/
p
o
w
e
r
f
r
om
t
h
e
U
ni
versity
o
f
T
echn
o
l
ogy
i
n
20
11
.
Area
o
f
r
es
ear
ch
i
nt
erest
i
n
t
he
electri
c
p
o
wer,
m
achi
n
ery
and
con
t
rol
en
gin
eerin
g
an
d
artif
i
c
i
a
l
in
te
l
l
i
g
ence
and
alg
o
rit
h
m
s
En
gin
eerin
g
Optimizati
on.
H
e
h
a
s
m
o
re
t
h
a
n
sci
e
ntif
ic
r
es
e
a
rch
pu
bl
ish
e
d
i
n
l
o
cal
a
nd
international jou
r
nals.
Em
a
il
:
wi
sam
_
a
l
ob
aid
ee@yah
oo.
com
Ali
N
.
J
barah
re
ce
i
v
ed
a
b
ach
elo
r
's
d
eg
ree
in
e
l
e
ct
rical
p
o
w
er
a
n
d
m
ach
in
es
f
rom
eng
i
neeri
n
g
col
l
ege-D
i
y
a
la
U
ni
versi
t
y
in
2
0
0
6
an
d
recei
ved
a
m
a
st
er's
d
eg
re
e
in
e
lect
rical
p
o
w
er
f
rom
th
e
Kazan
U
n
i
v
e
rs
ity
/
R
u
ssia
201
4.
A
rea
of
r
es
earch
i
nt
eres
t
in
t
h
e
el
ectric
po
wer
.
H
e
h
a
s
m
o
re
th
an s
cien
t
i
fi
c research
pu
b
l
i
sh
ed
i
n
internat
io
na
l
jo
urn
a
l
s
.
Em
a
il
:
a
lin
adhim
@
en
gi
neerin
g.
uod
iy
ala.
ed
u.i
q
Evaluation Warning : The document was created with Spire.PDF for Python.