Int
ern
at
i
onal
Journ
al of
R
obot
ic
s
and
Autom
ati
on (I
JRA)
Vo
l.
9
,
No.
1
,
Ma
rch
20
20,
pp.
1
~
5
IS
S
N:
20
89
-
4856,
DOI: 10
.11
591/
i
jra
.
v9
i
1
.
pp
1
-
5
1
Journ
al h
om
e
page
:
http:
//
ij
ra
.iae
score
.com
Mini
mization of
re
al p
ower loss b
y enhanc
ed teachi
ng
learnin
g based
op
timizati
on
algorit
hm
K.
Le
nin
Depa
rtment
o
f
E
EE
Prasad
V.
Potluri
Sid
dhar
th
a Instit
ute of Te
ch
nolog
y
,
Ind
ia
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
Dec
07
, 201
7
Re
vised
Oc
t
06
, 2
01
9
Accepte
d
Oct
31
, 201
9
Thi
s
pap
er
pr
ese
nts
an
Enh
anc
e
d
Teac
hing
-
Lear
ning
-
Based
Opt
i
m
iz
at
ion
(ET
LBO)
al
go
rit
hm
for
sol
ving
r
ea
c
ti
v
e
power
flow
probl
em.
Te
a
chi
ng
-
l
ea
rn
i
ng
proc
ess
is
an
itera
ti
ve
pro
ce
s
s
where
in
th
e
c
onti
nuous
int
er
ac
t
ion
t
ake
s
pla
c
e
for
the
tr
ansfe
r
of
knowl
edge
.
Movem
ent
s
of
trial
soluti
ons
will
i
nvesti
gate
th
e
i
nte
rna
lly
fin
al
stage
s.
Up
gra
dat
ion
of
the
al
gori
thm
has
bee
n
done
t
hrough
b
y
adding
weight
in
t
he
l
ea
rne
r
val
ues.
Projecte
d
ET
LBO
al
go
rit
hm
has
bee
n
te
sted
in
st
and
a
rd
IE
EE
57,
118
bus s
y
s
tem
s a
nd
power
lo
ss
has
bee
n
red
u
ce
d
eff
i
ci
en
tly
.
Ke
yw
or
d
s
:
En
han
ce
d
te
ac
hing lear
ning
Op
ti
m
al
r
eact
i
ve powe
r
Transm
issi
on
loss
This
is an
open
acc
ess arti
cl
e
un
der
the
CC
B
Y
-
SA
l
ic
ense
.
Corres
pond
in
g
Aut
h
or
:
K.
Leni
n
,
Dep
a
rtm
ent o
f EEE
,
Pr
asa
d V. Potl
ur
i
Sid
dh
a
rtha Instit
ute
of Tec
hnology,
Kanu
ru, V
i
j
ay
awad
a
, An
dhra
Pr
a
des
h,
I
nd
ia
.
Em
a
il
:
gk
le
nin@gm
ai
l.co
m
1.
INTROD
U
CTION
Op
ti
m
al
reactiv
e
powe
r
dis
pa
tc
h
pro
blem
i
s
one
of
the
di
ff
ic
ult
opti
m
izati
on
pro
blem
s
in
po
wer
syst
e
m
s
&
va
ri
ou
s
m
at
he
m
at
i
cal
te
chn
i
ques
[1
-
9
]
ha
ve
bee
n
util
iz
ed
to
s
ol
ve
the
pro
blem
.
Re
centl
y
m
any
ty
pes
of
E
vo
l
ut
ion
ary
al
gorithm
s
[1
0
-
15]
ha
ve
been
use
d
t
o
so
l
ve
th
e
re
act
ive
powe
r
pro
blem
.
This
pa
per
pr
ese
nts
a
n
En
han
ce
d
Teac
hin
g
-
Lea
rn
i
ng
-
B
ased
O
pti
m
iz
a
ti
on
(ETLB
O)
al
gorithm
for
so
l
ving
reacti
ve
powe
r
flo
w
pro
blem
.
Ba
sic
Teac
hing
-
Lea
rn
i
ng
-
Ba
sed
O
pti
m
iz
at
ion
[
16]
s
uccessfull
y
so
lve
d
va
ri
ou
s
op
ti
m
iz
ation
pro
blem
s.
In
this
pro
j
ect
e
d
w
ork
ne
w
l
earn
e
r
values
the
par
t
of
it
s
pr
e
vious
va
lue
is
co
ns
ide
red
and
it
ha
s
bee
n
decide
d
by
a
wei
gh
t
facto
r
“
wf’
.
D
ur
i
ng
the
ea
rly
sta
ge
s
of
t
he
sea
r
ch
Indivi
du
al
s
are
enc
oura
ge
d
t
o
sam
ple
dive
rs
e
zo
nes
of
the
exp
l
or
at
io
n
s
pa
c
e.
P
r
oj
ect
e
d
ETLBO
al
gorit
h
m
has bee
n
te
ste
d i
n
sta
nd
a
rd I
E
EE 57,
118 b
us
syst
e
m
s an
d re
al
p
owe
r
los
s
ha
s b
ee
n red
uce
d.
2.
PROBLE
M
F
ORMUL
ATI
ON
Re
du
ct
io
n
real
powe
r
loss
is
t
he
key goal
of the
w
ork
a
nd
the
ob
j
ect
ive f
unct
ion
has
bee
n
w
ritt
en
as foll
ows
(1)
:
F
=
P
L
=
∑
g
k
k
∈
N
br
(
V
i
2
+
V
j
2
−
2
V
i
V
j
cos
θ
ij
)
(1)
Vo
lt
age
d
e
viati
on m
at
he
m
atic
al
ly
w
ritt
en
as
(2
-
3)
,
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2089
-
4856
I
nt
J
R
ob
&
A
uto
m
,
Vo
l.
9
, No
.
1
,
Ma
rch 2
020
:
1
–
5
2
F
=
P
L
+
ω
v
×
Voltag
e
Deviat
io
n
(2)
=
∑
|
−
|
=
(3)
Con
st
raint (E
qual
it
y)
(4)
:
P
G
=
P
D
+
P
L
(4)
Con
st
raints
(In
equ
al
it
y)
(5
-
9):
P
g
sla
ck
min
≤
P
g
sla
c
k
≤
P
g
sla
ck
ma
x
(5)
Q
gi
min
≤
Q
gi
≤
Q
gi
max
,
i
∈
N
g
(6)
V
i
min
≤
V
i
≤
V
i
max
,
i
∈
N
(7)
T
i
min
≤
T
i
≤
T
i
ma
x
,
i
∈
N
T
(8)
Q
c
min
≤
Q
c
≤
Q
C
max
,
i
∈
N
C
(9)
3.
ENHAN
CED
TE
AC
HI
NG
-
LE
ARNING
-
BASED
OPTI
MIZ
ATION
ALGO
RITH
M
Ba
sic
Teachin
g
-
Lea
r
ning
-
Ba
sed
Op
ti
m
iz
ati
on
Al
gorithm
consi
st
of
fi
rst
pa
rt
“Teac
her
Ph
as
e”
and
t
he
seco
nd
“Learne
r
P
has
e”.
Lear
ning
f
r
om
the
te
acher
is
the
“Teac
he
r
Ph
a
se”
m
ea
ns
an
d
le
ar
ning
thr
ough
the
int
eracti
on
bet
we
en
le
ar
ners
is
t
he
“Lea
rner P
ha
se”.
I
n
searc
h
sp
ace b
ounded
the p
op
ulati
on
Y
is arb
it
ra
rily
in
it
ia
li
zed b
y
(1
0
-
11)
:
(
,
)
0
=
+
×
(
−
)
(10)
(
)
=
[
(
,
1
)
,
(
,
2
)
,
(
,
3
)
,
.
.
,
(
,
)
,
.
.
,
(
,
)
]
(11)
3.1.
Te
ac
her
p
ha
se
At g
e
ne
rati
on
g
the
m
ean p
ar
a
m
et
er
E
g
of ea
ch
s
ubj
ect
le
a
r
ner
s
in
t
he
cl
as
s
is gi
ven as
(12)
:
=
[
1
,
2
,
.
.
,
,
…
,
]
(12)
A new
-
fa
ngle
d set
of im
pr
ove
d
le
ar
ners is a
dded
to
t
he
e
xisti
ng
popula
ti
on
of lea
rn
e
rs
(
13
-
14)
.
(
)
=
(
)
+
×
(
ℎ
−
)
(13)
=
[
1
+
(
0
.
1
)
{
2
−
1
}
]
(14)
3.
2
.
Le
ar
ner
pha
se
Kno
wled
ge of
the lea
r
ner is i
m
pr
ov
e
d by
(
15)
,
(
)
=
{
(
)
+
×
(
(
)
−
(
)
)
(
(
)
)
<
(
(
)
)
(
)
+
×
(
(
)
−
(
)
)
ℎ
(
15)
3.3.
Al
go
ri
thm
t
ermi
n
at
i
on
Af
te
r
MAX
IT
c
onditi
ons sati
sf
ie
d
the al
gorith
m
is t
er
m
inate
d.
Value
of t
he
w
ei
ght
factor r
edu
ce
d
li
near
ly
w
it
h
ti
m
e fr
om
a m
axi
m
u
m
to
a m
ini
m
u
m
v
al
ue
by
(16)
,
=
−
(
−
max
)
∗
(16)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
R
ob
&
A
uto
m
IS
S
N:
20
89
-
4856
Mi
nimiza
ti
on
of
rea
l
po
we
r
lo
ss b
y
enh
an
ce
d t
each
i
ng lear
ni
ng
base
d op
ti
miz
atio
n… (
K.
Lenin
)
3
En
han
ce
d
le
a
r
ner
s
in
t
he
te
ac
her p
hase ca
n be
(17)
,
(
)
=
∗
(
)
+
∗
(
ℎ
−
)
(
17)
And
i
n
le
ar
ner
ph
a
se a set
of i
m
pr
ov
e
d
l
ear
ne
rs
a
re
(
18)
,
(
)
=
{
∗
(
)
+
×
(
(
)
−
(
)
)
(
(
)
)
<
(
(
)
)
∗
(
)
+
×
(
(
)
−
(
)
)
ℎ
(18)
4.
SIMULATI
O
N RESULTS
At
first
En
ha
nc
ed
Teac
hin
g
-
L
earn
i
ng
-
Ba
sed
Op
ti
m
iz
ation
(
ETLBO
)
al
gor
it
h
m
has
be
e
n
te
ste
d
in
sta
nd
a
rd
IE
EE
-
57
bus
power
syst
e
m
.
18
,
25
and
53
a
re
the
reacti
ve
power
com
pen
sat
io
n
bu
s
es.
PV
buse
s
are
2,
3,
6,
8,
9
a
nd
12
a
nd
bus
1
is
sla
ck
-
bus.
In
Ta
ble
1
T
he
s
ys
tem
var
ia
ble
lim
i
ts
a
re
giv
e
n
.
The preli
m
inary conditi
ons fo
r
the
I
EE
E
-
57
bu
s
po
wer sy
stem
are
gi
ven a
s foll
ow
s:
P
load
=
12
.
126 p
.u
.
Q
load
=
3.0
64
p
.
u.
The
t
otal i
niti
al
g
ene
rati
ons a
nd po
wer l
os
se
s ar
e
obta
ined
as foll
ows:
∑
= 12.
478 p
.u.
∑
=
3.316
5 p.u.
P
loss
= 0
.
25886
p.u. Q
loss
=
-
1.2081 p.u
.
Table
2
s
hows
the co
m
par
is
on of
op
ti
m
u
m
r
esults.
Ta
ble
3
s
hows
t
he vari
ous syst
em
co
nt
ro
l
var
ia
bles.
Table
1.
Var
ia
ble
lim
it
s
Rea
c
ti
ve
power gene
ra
ti
on
li
m
it
s
Bus no
1
2
3
6
8
9
12
Qgm
in
-
1.
4
-
.
015
-
.
02
-
0.
04
-
1.
3
-
0.
03
-
0.
4
Qgm
ax
1
0.
3
0.
4
0.
21
1
0.
04
1.
50
Volta
ge
and
ta
p
sett
ing
li
m
it
s
vgm
in
Vgm
ax
vpqm
in
Vpqm
ax
tkmin
tkmax
0.
9
1.
0
0.
91
1.
05
0.
9
1.
0
Shunt Ca
pa
ci
tor
Li
m
it
s
Bus no
18
25
53
Qcm
in
0
0
0
Qcm
ax
1
0
5.
2
6.
1
Table
2
. C
om
par
iso
n resu
lt
s
S.No.
Op
ti
m
izatio
n
alg
o
r
i
th
m
Fin
est so
lu
tio
n
Po
o
rest so
lu
tio
n
No
r
m
al
so
lu
tio
n
1
NLP
[
1
7
]
0
.25
9
0
2
0
.30
8
5
4
0
.27
8
5
8
2
CGA [
1
7
]
0
.25
2
4
4
0
.27
5
0
7
0
.26
2
9
3
3
AGA [1
7
]
0
.24
5
6
4
0
.26
6
7
1
0
.25
1
2
7
4
PSO
-
w
[
1
7
]
0
.24
2
7
0
0
.26
1
5
2
0
.24
7
2
5
5
PSO
-
cf
[
1
7
]
0
.24
2
8
0
0
.26
0
3
2
0
.24
6
9
8
6
CLPSO [
1
7
]
0
.24
5
1
5
0
.24
7
8
0
0
.24
6
7
3
7
SPSO
-
0
7
[
1
7
]
0
.24
4
3
0
0
.25
4
5
7
0
.24
7
5
2
8
L
-
D
E
[
1
7
]
0
.27
8
1
2
0
.41
9
0
9
0
.33
1
7
7
9
L
-
SACP
-
DE
[
1
7
]
0
.27
9
1
5
0
.36
9
7
8
0
.31
0
3
2
10
L
-
SaD
E
[
1
7
]
0
.24
2
6
7
0
.24
3
9
1
0
.24
3
1
1
11
SOA [
1
7
]
0
.24
2
6
5
0
.24
2
8
0
0
.24
2
7
0
12
LM
[
1
8
]
0
.24
8
4
0
.29
2
2
0
.26
4
1
13
MBEP1 [
1
8
]
0
.24
7
4
0
.28
4
8
0
.26
4
3
14
MBEP2 [
1
8
]
0
.24
8
2
0
.28
3
0
.25
9
2
15
BES1
0
0
[
1
8
]
0
.24
3
8
0
.26
3
0
.25
4
1
16
BES2
0
0
[
1
8
]
0
.34
1
7
0
.24
8
6
0
.24
4
3
17
Prop
o
sed
E
TL
BO
0
.22
0
4
8
0
.23
0
1
2
0
.22
2
8
2
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2089
-
4856
I
nt
J
R
ob
&
A
uto
m
,
Vo
l.
9
, No
.
1
,
Ma
rch 2
020
:
1
–
5
4
Table
3
. C
on
t
r
ol v
a
riables
ob
ta
ined
a
fter
optim
iz
ation
Co
n
trol
Variables
ET
LB
O
V1
1
.1
V2
1
.03
5
0
V3
1
.03
4
0
V6
1
.02
8
0
V8
1
.02
0
0
V9
1
.00
9
0
V1
2
1
.01
6
0
Qc1
8
0
.06
6
2
0
Qc2
5
0
.20
0
0
Qc5
3
0
.04
7
1
0
T4
-
18
1
.00
9
0
T21
-
20
1
.04
6
0
T24
-
25
0
.86
4
0
T24
-
26
0
.87
2
0
T7
-
29
1.
0500
T34
-
32
0
.87
0
0
T11
-
41
1
.01
2
0
T15
-
45
1
.03
0
0
T14
-
46
0
.91
0
0
T10
-
51
1
.02
0
0
T13
-
49
1
.06
0
0
T11
-
43
0
.91
0
0
T40
-
56
0
.90
0
0
T39
-
57
0
.95
0
0
T9
-
55
0
.95
0
0
The
n
E
nhance
d
Te
achi
ng
-
Le
arn
i
ng
-
Ba
sed
Op
ti
m
iz
ation
(
ETLBO
)
al
gor
it
h
m
has
bee
n
te
ste
d
in
sta
nd
a
rd
IE
E
E
118
-
bus
te
st
sy
stem
[
19
].
T
he
syst
e
m
has
54
generato
r
bu
s
es,
64
loa
d
bus
es,
18
6
bra
nches
and
9
of
the
m
are
with
th
e
ta
p
set
ti
ng
t
ran
s
f
or
m
ers.
T
he
li
m
i
ts
of
volt
age
on
ge
ne
rator
buses
a
re
0.95
-
1.1
pe
r
-
un
it
.,
an
d
on
lo
ad
buses
are
0.95
-
1.05
per
-
un
it
.
T
he
l
im
i
t
of
tr
ans
f
or
m
e
r
rate
is
0.9
-
1.1,
with the
ch
a
ng
es step
of 0.0
25.
W
it
h
t
he
c
ha
ng
e
in
ste
p of
0.01
t
he
li
m
it
at
i
on
s
of
reacti
ve powe
r
s
ource a
re
li
ste
d
in
Ta
ble
4
.
The
sta
ti
sti
cal
com
par
ison
res
ults
of
50
tria
l
runs
ha
ve
be
en
li
st
in
Ta
ble
5
an
d
the
re
s
ults
cl
early
sh
ow
t
he
bette
r
perf
or
m
ance
of
pro
posed
E
nha
nced
Teachi
ng
-
Lea
r
ning
-
Ba
sed
Op
ti
m
iz
ation
(ETLB
O
)
al
gori
thm
in r
ed
ucin
g
the
r
eal
powe
r
loss
.
Table
4.
Lim
itati
on
of r
eact
iv
e pow
e
r
s
ource
s
BUS
5
34
37
44
45
46
48
QCMAX
0
14
0
10
10
10
15
QCMIN
-
40
0
-
25
0
0
0
0
BUS
74
79
82
83
105
107
110
QCMAX
12
20
20
10
20
6
6
QCMIN
0
0
0
0
0
0
0
Table
5.
C
om
par
iso
n resu
lt
s
Activ
e po
wer
lo
ss
(M
W
)
BBO
[
2
0
]
IL
SBB
O/st
rategy1
[
2
0
]
IL
SBB
O/st
rategy1
[
2
0
]
Prop
o
sed
ET
LBO
Min
1
2
8
.77
1
2
6
.98
1
2
4
.78
1
1
6
.120
Max
1
3
2
.64
1
3
7
.34
1
3
2
.39
1
2
0
.340
Av
erage
1
3
0
.21
1
3
0
.37
1
2
9
.22
1
1
7
.040
5.
CONCL
US
I
O
N
In
t
his
w
ork
En
han
ce
d
Tea
chin
g
-
Lea
r
ning
-
Ba
se
d
Op
ti
m
iz
at
ion
(ET
LBO)
al
gorith
m
so
lved
the
opti
m
a
l
reacti
ve
powe
r
pro
blem
.
A
pa
ram
et
er
cal
led
as
“
weig
ht”
has
bee
n
incl
ud
e
d
i
n
th
e
ba
sic
Teachin
g
-
Lear
ning
b
ase
d
al
gorithm
.
The
perform
ance
of
t
he
pro
pose
d
En
ha
nced
T
eachin
g
-
L
ear
nin
g
-
Ba
sed
O
ptim
izati
on
(
ETLBO
)
al
gorithm
has
bee
n
ha
s
bee
n
te
ste
d
in
sta
nd
a
r
d
IE
EE
57
,118
bus
syst
e
m
s
and real
powe
r
loss
c
onside
ra
bly re
du
ce
d.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
R
ob
&
A
uto
m
IS
S
N:
20
89
-
4856
Mi
nimiza
ti
on
of
rea
l
po
we
r
lo
ss b
y
enh
an
ce
d t
each
i
ng lear
ni
ng
base
d op
ti
miz
atio
n… (
K.
Lenin
)
5
REFERE
NC
ES
[1]
O.
Alsa
c
and
B.
Sco
tt,
“
Opti
m
al
lo
ad
flow
with
st
ea
d
y
sta
t
e
se
cur
i
t
y
,
”
I
E
EE
Tr
ansa
ction.
PA
S
-
1973,
pp.
745
-
751
.
[2]
Le
e
K
.
Y
.
,
Paru
Y
.
M
.
,
Oritz
J
.
L
.
,
“
A
un
it
ed
a
pproa
ch
to
opt
i
m
al
r
ea
l
and
re
a
ct
iv
e
pow
er
dis
pat
ch
,
”
I
EE
E
Tra
nsac
ti
ons on
power
Apparatu
s
and
s
y
s
te
m
s 19
85:
PA
S
-
104
:
1
147
-
1153
.
[3]
A.
Montic
e
ll
i
,
M.
V.
F
.
Pere
i
ra,
and
S
.
Gra
nvil
le,
“
Secur
i
t
y
constr
ai
n
ed
o
pti
m
al
power
fl
ow
with
post
cont
i
ng
ency
co
rre
ctive
r
esc
he
duli
ng
,
”
IE
EE
Tra
nsa
ct
ions
on
Pow
er
S
y
stems
:
PW
R
S
-
2,
No.
1,
pp.
175
-
182.
,
19
87.
[4]
Dee
b
N
.
,
Shahi
dehpur
S.
M
,
Li
ne
ar
r
eact
iv
e
power
op
ti
m
izati
on
in
a
l
arg
e
power
ne
twork
using
the
dec
om
positi
on
a
pproa
ch. I
E
EE T
ran
sac
t
ions o
n
p
ower
s
y
st
em 199
0:
5(2):
428
-
435
.
[5]
E.
Hobs
on
,
’Net
work
consra
ine
d
react
iv
e
power
cont
ro
l
using
linear
progr
amm
ing,
‘
I
EEE
Tr
a
nsac
ti
ons
on
power
s
y
st
ems
PA
S
-
99
(4)
,
pp
8
68=877,
1980
.
[6]
K.
Y
L
ee
,
Y
.
M
Park
,
and
J.
L
.
Orit
z,
“
Fuel
–
c
ost
opt
imiza
t
ion
for
both
r
eal
an
d
re
active
power
dispa
tc
h
es”
,
IEE
Proc
;
131C
,
(3),
pp
.
85
-
93.
[7]
M.
K.
Mango
li,
and
K.
Y.
L
ee
,
“
Optimal
r
e
al
and
r
ea
c
ti
v
e
power
cont
rol
using
l
ine
a
r
p
rogra
m
m
ing
,
”
El
e
ct
r.
Pow
er
S
y
st.Re
s,
Vol
.
26,
p
p.
1
-
10,
1993.
[8]
S.
Granvi
l
le
.
“
Optimal
r
ea
c
ti
ve
d
ispat
ch
through
i
nte
rior
point
m
ethods
,
”
IE
EE
Trans
ac
ti
ons
on
Po
wer
S
y
stem,
vol/
issue: 9(1),
p
p.
136
–
146
,
ht
tp
:/
/dx.
do
i.
org
/10.1109
/59.
317548
,
1994.
[9]
Apara
jita
Mukh
erj
e
e,
Viv
eka
na
nda
Mukhe
rjee,
"S
olut
ion
of
op
t
imal
r
eact
iv
e
po
wer
dispa
tc
h
b
y
ch
aot
i
c
kri
ll
her
d
a
lgori
thm
,
"
IET
Gen
er.
Tr
an
sm
.
Distrib,
Vol
.
9,
Iss
ue.
15
,
pp
.
2351
–
2362,
201
5.
[10]
Hu,
Z
.
,
W
ang
,
X.
&
T
a
y
lor
,
G.
Stoch
asti
c
opt
i
m
a
l
re
active
po
wer
dispa
tc
h:
Fo
rm
ula
ti
on
and
soluti
on
m
et
hod.
El
e
ct
r. Pow
er En
erg
y
S
y
s
t.,
vol
.
3
2,
pp
.
615
-
621
.
2010.
[11]
Maha
letchumi
A/P
Morgan
,
Nor
Rul
Hasm
a
Abdulla
h
,
Moh
d
Herwan
Sul
aim
an,
Mahfuz
ah
Mus
ta
fa
and
Rosdi
y
an
a
Sam
ad,
“
Com
puta
tional
in
te
l
li
gen
c
e
te
chni
que
for
stat
i
c
VA
R
co
m
pensa
tor
(SV
C)
insta
llat
ion
conside
ring
m
ul
ti
-
contingen
ci
es
(N
-
m
)”,
ARP
N
Journal
of
E
ngine
er
ing
and
Applie
d
Scie
n
c
es
VO
L.
10,
NO
.
22,
D
e
ce
m
b
er
2015.
[12]
Mohd
Herwan
Sulai
m
an,
Zu
ria
ni
Mus
ta
ff
a,
Ham
dan
Dani
y
al,
Mohd
Ruslli
m
Moham
ed
and
Om
ar
Aliman,
“
Solvin
g
Optimal
Re
ac
t
ive
Pow
er
Planni
ng
Probl
em
Util
izing
Natur
e
Inspire
d
Com
puti
ng
Te
chn
ique
s”
,
ARP
N
Journal
of
Engi
ne
eri
ng
and
Ap
pli
ed
Scie
n
ce
s
VO
L.
10,
NO
.
21
,
p
p.
9779
-
9785.
N
ovember
2015
.
[13]
Mohd
Herwan
S
ula
iman,
W
ong
Lo
Ing
,
Zuri
an
i
Mus
ta
ffa
and
M
ohd
Ruslli
m
Mo
hamed,
“
Gre
y
W
olf
Optimiz
er
for
Solving
Ec
o
nom
ic
Dispat
ch
Problem
with
Valve
-
Lo
adi
ng
Eff
ects”,
ARP
N
Journal
of
Eng
ine
er
ing
and
Applie
d
Sc
ie
nc
e
s VO
L.
10,
NO
.
21,
,
pp
.
9796
-
9
801,
NO
VEMBER
2015.
[14]
Pandia
ra
ja
n,
K.
&
Babu
la
l
,
C
.
K.,
“
Fuzz
y
har
m
ony
sea
r
ch
algorithm
base
d
o
pti
m
al
power
fl
ow
for
power
s
y
stem sec
ur
ity
enha
nc
ement”.
I
nte
rna
ti
ona
l
Jour
nal
El
e
ct
r
ic
Pow
er
En
erg
y
S
y
st.
,
vol.
78
,
pp
.
72
-
7
9.
2016
.
[15]
Mus
ta
ffa
,
Z.,
Su
la
iman,
M.H.
,
Yus
of,
Y.,
K
amarul
z
aman,
S.F
.
,
“
A
novel
h
y
br
i
d
m
et
ahe
ur
isti
c
al
gorit
hm
for
short
te
rm
load
fore
c
asti
ng”
,
Inte
rna
ti
ona
l
Journal
of
Sim
ula
ti
on
:
S
y
st
ems
,
Scie
nc
e
and
Te
chno
log
y
,
Volum
e
17,
Iss
u
e
41,
Pages 6.1
-
6.
6.
2017.
[16]
R.
V.
Rao
,
V.
J.
Savsani
and
D.
P.
Vakh
ari
a
,
“
T
ea
ch
ing
-
Le
arn
i
ng
-
Based
Optim
iz
a
ti
on:
A
Nove
l
Method
for
Con
-
stra
ine
d
Mec
hanica
l
Des
ign
Optimi
za
t
io
n
Problems
,
”
C
om
pute
r
-
Aided
Design,
Vol
.
43
,
No.
1
,
2011,
pp.
303
-
315.
do
i:
10.
1016/
j.ca
d
.
2
010.
12.
015
.
[17]
Chaohua
Dai
,
W
ei
rong
Ch
en,
Yunfang
Zhu
,
a
nd
Xuexi
a
Zh
an
g,
“
Seek
er
optim
iz
at
ion
al
gori
t
hm
for
op
ti
m
al
rea
c
ti
ve
power
d
ispat
ch
,
”
IEEE T
ran
s.
Pow
er
S
y
st
ems
,
Vol.
24
,
N
o.
3
,
Augus
t
200
9,
pp
.
1218
-
123
1.
[18]
J.
R.
Gom
es
and
0.
R.
Sa
ave
dr
a,
“
Optimal
rea
c
tive
power
dispatch
using
evol
u
tio
nar
y
computati
on:
Extended
al
gorit
hm
s,”
IE
E
Proc.
-
Gen
er.
Tr
ansm
.
Distrib.. V
ol.
146
,
No.
6.
Nov.
1999.
[19]
IEE
E
,
“
Th
e
IEEE
30
-
bus
te
st
s
y
stem
and
t
he
IE
EE
1
18
-
te
st
s
y
ste
m
”,
(1993),
htt
p://ww
w.e
e
.
washington.edu/
tr
sea
rch
/pst
ca
/
.
[20]
Jiangt
ao
C
ao,
Fu
li
W
ang
and
Pin
g
Li
,
“
An
Im
prove
d
Bioge
ogr
ap
h
y
-
bas
ed
Optimi
za
t
ion
Algorit
h
m
fo
r
Optima
l
Rea
c
ti
ve
Pow
er F
low”
Inte
rn
at
io
nal
Journa
l
of
C
ontrol
and
Auto
m
at
ion
Vol.
7
,
N
o.
3
(2014)
,
pp
.
1
61
-
176.
Evaluation Warning : The document was created with Spire.PDF for Python.