Int
ern
at
i
onal
Journ
al of Ele
ctrical
an
d
Co
mput
er
En
gin
eeri
ng
(IJ
E
C
E)
Vo
l.
10
,
No.
4
,
A
ugus
t
2020
,
pp. 335
0~33
57
IS
S
N: 20
88
-
8708
,
DOI: 10
.11
591/
ijece
.
v10
i
4
.
pp3350
-
33
57
3350
Journ
al h
om
e
page
:
http:
//
ij
ece.i
aesc
or
e.c
om/i
nd
ex
.ph
p/IJ
ECE
Optimi
zed pl
acem
ent of
mu
ltip
le FACTS d
ev
i
ce
s
u
sing PSO
and
CSA alg
or
ithm
s
Basana
goud
a Pat
i
,
S.
B.
Kar
ajgi
Depa
rtment
o
f
E
le
c
tri
c
al a
nd
Ele
ct
roni
cs
Engi
n
eering,
Shri
Dharm
asthala
Man
juna
th
eshwara
Col
le
g
e
of
Engi
ne
eri
ng
and
Technol
og
y
,
In
dia
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
N
ov
20, 201
8
Re
vised
Jan
2
4
,
20
20
Accepte
d
Fe
b 3
, 2
020
Thi
s
pape
r
is
an
attempt
to
deve
lop
a
m
ulti
-
facts
devi
c
e
pla
c
ementi
n
der
egulate
d
po
wer
s
y
stem
using
opti
m
iz
ation
al
gorit
hm
s.
Th
e
der
egu
la
t
ed
power
s
y
stem
is
the
re
ce
n
t
ne
e
d
in
the
power
distri
buti
on
as
i
t
has
m
an
y
inde
pend
ent
sel
l
ers
and
bu
y
ers
of
el
e
ct
ri
ci
t
y
.
Th
e
proble
m
of
der
e
gula
ti
on
is
the
qualit
y
of
the
power
di
strib
uti
on
as
m
an
y
sellers
a
re
invol
ved
.
The
place
m
ent
of
FA
CTS
devi
ce
s
provide
s
the
soluti
on
for
the
above
proble
m
.
Th
ere
are
r
ese
ar
che
s
ava
i
la
bl
e
for
m
ult
ipl
e
FA
CT
S
devi
ce
s
.
The
opti
m
izati
o
n
al
gorit
hm
s
li
ke
Parti
cle
Sw
arm
Optimiza
ti
on
(PS
O)
and
Cuckoo
Sear
c
h
Algorit
hm
(CSA
)
are
impleme
nte
d
to
pl
ac
e
t
he
m
ult
iple
FA
CTS
devi
ce
s
in
a
pow
er
s
y
stem.
MA
TL
AB
base
d
imple
m
ent
at
ion
is
ca
rri
ed
out
for
a
ppl
y
ing
Optimal
Pow
er
Flow
(OP
F)
with
var
iatio
n
in
th
e
bus
power
and
the
li
ne
reactance
par
amete
rs.
T
he
cost
func
ti
o
n
i
s
used
as
the
object
ive
fun
ct
ion
.
The
cost
r
educ
t
ion
of
FA
CTS
as
well
as
ge
ner
ation
b
y
pla
c
ement
of
diffe
ren
t
compens
at
ors
li
ke
,
Stat
i
c
Var
Com
pens
at
or
(SV
C),
Th
y
r
istor
Contr
oll
ed
Ser
ie
s
Co
m
pensa
tor
(TCSC)
and
Unifie
d
Pow
er
Flow
Control
le
r
(UP
F
C).
The
cos
t
c
a
lc
ul
at
ion
is
done
on
the
3
-
se
lle
r
sce
nar
io
.
The
I
EE
E
14
bu
s is
ta
k
en
h
ere a
s
3
-
seller
s
y
s
te
m
.
Ke
yw
or
d
s
:
CSA
Der
e
gula
te
d power
syst
em
Op
ti
m
al
p
la
cem
ent o
f
F
ACT
S
PSO
Copyright
©
202
0
Instit
ut
e
o
f Ad
vanc
ed
Engi
n
ee
r
ing
and
S
cienc
e
.
Al
l
rights
reserv
ed
.
Corres
pond
in
g
Aut
h
or
:
Ba
sanago
ud
a
P
at
il
,
Dep
a
rtm
ent o
f El
ect
rical
an
d
Ele
ct
ro
nics
E
nginee
rin
g,
Sh
ri
Dharm
ast
hala Ma
nj
un
at
hes
hw
a
ra C
ollege
of Engine
e
rin
g
a
nd Tech
nolo
gy
,
Dh
a
rwad
-
5 80002, Ka
r
nataka
, In
dia
.
Em
a
il
:
patil
.b
asana
gow
da@g
m
ai
l.co
m
1.
INTROD
U
CTION
The
w
or
ld
’s
e
le
ct
ric
powe
r
i
s
hea
vily
inter
connecte
d
f
or
econom
ic
reaso
n.
And
w
he
n
the
powe
r
trans
fer
i
ncr
ea
ses
the
c
onnect
ion
gro
ws
du
e
to
that
sec
ur
it
y
pro
blem
s
ta
ke
s
place.
T
he
se
cur
it
y
of
the
sy
stem
is
aff
ect
ed
w
he
n
the
la
r
ge
po
wer
tra
nsfe
r
is
done
th
r
ough
the
tra
ns
m
issi
o
n
li
ne
with
out
consi
der
i
ng
it
s
lim
it
s.
The
de
re
gu
la
ti
on
of
power
sy
stem
is o
ne
of
the i
m
po
rta
nt m
et
hods
in
po
we
r
syst
e
m
to
red
uce th
ese p
r
ob
l
e
m
s.
But
de
re
gu
la
ti
on
le
ads
to
po
wer
qual
it
y
prob
le
m
s.
For
i
m
pr
ov
in
g
pow
er
tra
nsfe
r,
F
A
CTS
de
vices
do
very
i
m
po
rtant
r
ole
[
1].
Se
ries
ca
pacit
or
s
wh
ic
h
is
va
riable,
unifie
d
po
wer
f
low
c
ontr
oller
s
(
UP
FC
)
a
nd
phase
sh
ifte
rs
can
be
util
iz
ed
[
2].
F
ACTS
de
vices
pro
vid
e
bette
r
con
t
ro
l
i
n
ste
a
dy
sta
te
an
d
i
n
dynam
ic
sta
te
[3,
4].
The
c
os
t
-
e
ff
ec
ti
ve
de
vices
are
series
ca
pa
ci
tors
w
hich
is
var
ia
ble
an
d
helps
i
n
m
ini
m
iz
ing
losses
[5,
6]
.
The
F
ACTS
dev
ic
es
a
re
c
os
tl
y
accor
ding
to
t
he
siz
e
of
it
.
If
t
he
s
iz
e
is
le
ss
the
cost
w
ould
r
edu
c
e
.
So
,
the
optim
al
locat
ion a
nd sizi
ng
bec
om
es i
m
po
rtant [7
-
9].
Ther
e
a
re
resea
rch
es
a
rtic
le
s
avail
able
on
optim
al
locat
ion
ba
sed
on
sen
sit
ivit
y
analy
sis
[1
0],
so
lvi
ng
econom
ic
load
disp
at
c
h
[
11]
,
congesti
on
m
a
nag
em
ent
us
in
g
F
ACTs
dev
i
ces
[12
-
14]
,
re
al
power
pe
rfo
rm
ance
ind
e
x
[15],
i
n
[
16
]
open
power
m
ark
et
an
al
ysi
s,
e
le
ct
ric
syst
em
ener
gy
[17],
t
he
a
ut
om
a
ti
c
con
ti
ngency
sel
ect
ion
[18],
Ele
ct
ric
ener
gy
syst
e
m
s
analy
sis
and
op
e
rati
on
[19],
Investi
gati
on
of
the
load
low
pro
blem
[2
0]
and
re
du
ci
ng
t
he
losse
s
w
he
n
co
ngest
io
n
is
not
present
[
21
-
24
]
.
T
he
pa
per
[25,
26]
sh
ows
the
eco
no
m
ic
disp
at
ch
s
ol
ution
m
et
ho
d
for
de
regulat
ed
en
vir
on
m
ent
.
The
s
ol
ution
te
c
hn
i
qu
e
s
sh
ow
n
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N: 20
88
-
8708
Op
ti
mize
d plac
emen
t
of
mu
lt
i
ple FACT
S dev
ic
es u
sin
g PS
O
and CSA
a
l
gorit
hm
s
(
B
asana
gouda P
ati
)
3351
in
[27,
28]
are
us
e
d
her
e
f
or
m
ul
ti
-
facts
devi
ce
place
m
ent.
This
pa
per
is
done
f
or
m
ini
m
i
zi
ng
the
total
cost
of
the
ge
ner
at
io
n
and
FA
CT
S
de
vices
(like
S
V
C,
TCS
C
&
U
PFC).
T
he
op
ti
m
al
locat
ion
and
siz
e
a
re
ide
ntifie
d.
Sect
ion
2
co
nsi
sts
Pr
ob
le
m
F
or
m
ulati
on
for
op
ti
m
a
l
locat
i
on
of
m
ulti
ple
FA
CTS
are
de
scribe
d.
Sect
ion
3
consi
st
of
P
roblem
so
luti
on
m
et
ho
ds
;
Sect
i
on
4
co
ns
ist
s
of
sim
ulati
on
resu
l
ts.
Fi
nally
,
a
concl
u
sio
n
about
the r
es
ults
of
sim
ula
ti
on
is
de
du
ce
d
i
n
Sect
i
on 5.
2.
PROBLE
M
F
ORMUL
ATI
ON
The
ge
ner
at
io
n
c
os
t
a
nd
th
e
cost
of
F
A
CTS
de
vices
are
the
m
ajo
r
eco
nom
ic
so
ur
ces
.
Her
e
i
n
the
opti
m
a
l
power
flo
w
t
he
c
os
t
of
ge
ner
at
i
on
m
ini
m
iz
at
io
n
a
nd
the
FA
C
Ts
de
vice
place
m
ent
with
m
i
nim
u
m
po
s
sible
or
optim
al
cost
has
t
o
be
ide
ntifie
d.
Bi
dd
i
ng
co
st
is
consi
de
red
a
s
the
the
rm
al
s
yst
e
m
cost
curve
s
o
the b
i
dd
i
ng c
ost
can be
repres
ented
a
s [2
5],
(
)
=
+
+
2
(1)
t
he
inc
rem
enta
l cost ca
n be re
pr
ese
nted
as
be
low,
(
)
=
+
2
(2)
d
ere
gula
te
d power
syst
em
o
pti
m
al
p
ow
e
r
fl
ow e
qu
at
io
n
is
gi
ven
belo
w,
:
∑
(
)
=
1
(3)
:
∑
=
(4)
<
<
,
[
1
,
]
(5)
w
he
n
∑
=
1
>
∑
=
1
=
,
-
no
feas
ible sol
ution,
w
he
n
∑
=
1
=
,
-
each
s
el
le
r
is co
ntrac
te
d
am
ou
nt
is a
t i
ts capaci
ty
low
er
lim
it
,
w
he
n
∑
=
1
<
an
d
∑
=
1
>
-
non
-
tri
vial ca
se.
Her
e
,
(
)
−
−
ℎ
,
,
−
−
,
−
ℎ
−
,
−
f
act
s d
e
vices c
os
ts
;
=
0
.
0015
2
−
0
.
713
+
153
.
75
(6)
=
0
.
0003
2
−
0
.
3051
+
127
.
38
(7)
=
0
.
0003
2
−
0
.
2691
+
188
.
2
(8)
her
e;
−
$
−
in
$
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
10
, No
.
4
,
A
ugus
t
2020
:
3350
-
3357
3352
−
$
−
$
−
−
−
C
on
si
der
i
ng th
e ab
ov
e
constr
ai
nts en
ti
re
co
s
t functi
on ca
n be
represe
nted as bel
ow [6]
.
=
∑
(
)
=
1
+
(9)
3.
SOLUTI
ON MET
HO
DS
Fo
r
the
prob
le
m
sh
own
i
n
(
9)
is
the
ob
j
ect
ive
f
unct
ion
t
o
so
lve
t
hat
m
an
y
te
chn
iq
ues
c
an
be
us
e
d.
Her
e
PS
O
al
gorithm
wh
ic
h
is
the
fa
ste
r
al
gor
it
h
m
and
the
CSA
al
gorithm
wh
ic
h
giv
es
guaran
te
e
d
resu
lt
s
ar
e
consi
der
e
d f
or
the s
olu
ti
on. T
he
al
go
rithm
e
xp
la
nati
on is
gi
ven
belo
w.
3.1
.
Part
i
cl
e
sw
ar
m o
p
timi
z
at
io
n
(P
SO
)
The
al
gorithm
is
form
ed
with
the
be
hav
i
or
of
insect
s
/
fish
on
it
s
beh
a
vior
of
f
ood
searc
hi
ng.
Ste
ps
of
al
gorithm
descr
ibe
d
giv
e
n below.
-
The Nsiz
e
of the s
war
m
, X
-
c
on
t
ro
l
var
ia
ble
(g
e
ner
at
e
d power
Pg)
a
re i
niti
al
iz
ed.
-
In
it
ia
l
popu
la
ti
on
of
P
g
is
give
n
as
within
t
he
power
li
m
i
t.
And
init
ia
l
velocit
y
of
the
sw
arm
par
ti
cl
es
(
V
j)
is t
aken as ze
ro.
-
Fo
r
eac
h
po
pula
ti
on
cal
culat
e
fuel
cost
(F)
an
d
fi
nd
vel
oc
it
ie
swith
giv
e
n
f
orm
ula
(10
).
a
nd
i
ncr
em
ent
the i
te
rati
on.
-
Each
par
ti
cl
e
is
pe
rsonal
best
(Pbest)
of
it
s
own
P
gval
ue.
The
n
the
X
value
w
hich
is
res
pons
i
ble
for
the
lowe
r
c
ost
value
is
ta
ke
n
as
global
best
(
Gb
e
st).
The
n
vel
oci
ty
functi
on
is
cal
culat
ed
us
i
ng
the foll
owin
g
e
qu
at
io
n,
(
)
=
(
−
1
)
+
1
1
[
−
(
−
1
)
]
+
2
2
[
−
(
−
1
)
]
(10)
wh
e
re
=
1
,
2
,
…
,
her
e
,
1
,
2
2
1
,
2
0
1
-
The
n
the
X val
ue
is
update
d wit
h
the
foll
ow
ing
e
quat
io
n
(
)
=
(
−
1
)
+
(
)
(11)
-
The
n go to ste
p
(c
), d
o
it
ti
ll
the st
op crite
ria
.
3.2
.
Cu
ck
oo
se
arc
h a
l
go
ri
th
m
(
CSA)
The
C
ucko
o
s
earch
al
gorith
m
is
based
on
the
cuc
koo
bi
r
d
on
be
ha
vior
of
it
s
br
e
edi
ng.
Th
e
c
ucko
o
bir
d
can
’t
buil
d
the
nest.
It
de
pends
on
the
ho
st
bir
d
nest
for
la
yi
ng
e
ggs
and
hatc
hing
it
.
But
host
bi
r
d
ne
st
no
t
al
lows
to
do
so.
It
m
ay
aban
do
n
the
nest
or
pu
s
hes
the
bi
rd
s
’
eg
gs
dow
n
.
But
cuc
koo
la
ys
egg
s
sim
ilar
t
o
the h
os
t
bir
d
a
nd
if
it
h
at
che
s
the
c
uc
koo
chi
cks
m
i
m
ic
s
the
sou
nd
of
t
he
host bir
d.
So, f
ind
i
ng
the
be
st nest
to
m
ake
su
r
viv
e
the
cuc
koo
bir
ds
m
akes
a
fine
search
that
is
r
epr
ese
nted
as
t
he
m
a
the
m
at
ical
equ
at
ion
ste
ps
are
fo
ll
owin
g.
-
The
i
niti
al
popula
ti
on
of
X va
riable in
n host
n
est
s is
r
a
ndom
ly
g
ener
at
ed
.
-
A
cuc
koo
is
sel
ect
ed
by
le
vy
ran
dom
distr
ibu
ti
on
an
d
ev
al
uated
the
ob
j
ect
ive
f
un
ct
io
n
for
al
l
the
ho
s
t
nests.
-
Ra
ndom
ly
se
l
ect
ed
nest
isc
om
par
ed
with
the
obj
ect
iv
e
wh
ic
h
is
ra
ndom
ly
sel
ect
e
d
an
d
cal
culat
ed.
If
t
he new
cu
c
koo fit
s the
n re
place t
he
o
l
d
c
ucko
o.
-
Rem
ai
nin
g nes
ts are a
band
oned wit
h
the
fra
ct
ion
of Pa a
nd
b
est
ones a
re sa
ved.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N: 20
88
-
8708
Op
ti
mize
d plac
emen
t
of
mu
lt
i
ple FACT
S dev
ic
es u
sin
g PS
O
and CSA
a
l
gorit
hm
s
(
B
asana
gouda P
ati
)
3353
-
Ra
nk the s
olu
ti
on
;
fin
d
t
he be
st cuc
koo.
-
In
c
rease t
he
it
erati
on and
go t
o
ste
p
sec
ond s
te
p.
-
Do it
ti
ll
ter
m
i
na
ti
on
The pr
opose
d
s
olu
t
ion
al
go
rithm
is d
escribe
d:
Step
1:
I
niti
al
i
ze
li
ne
and
bu
s
data
of
the
powe
r
syst
e
m
,
con
ti
nge
ncy
data,
al
l
con
st
raints,
an
d
PS
O/CSA
par
am
et
ers.
Step
2:
I
niti
al
iz
e
popu
la
ti
on
of
pa
rtic
le
s
with
rand
om
nu
m
ber
s
a
nd
vel
oc
it
ie
s/new
nest
represe
nting
F
ACTS
dev
ic
es
locati
on &
size
.
Step
3: Set it
er
at
ion
in
de
x
it
er
at
ion
=
0.
Step
4:
T
he
pa
rtic
le
carries
the
locat
io
n
a
nd
siz
e
of
F
AC
TS
dev
ic
es
up
dates
the
li
ne
-
data
at
the
rea
ct
ance
colum
n
an
d
i
n
bus
-
data
pow
er
in
j
ect
io
n
c
ol
um
n
.
Determ
i
ne
the
loa
d
le
ve
l
and
ou
t
pu
t
powe
r.
C
ondu
ct
OP
F
inco
rpor
at
in
g
FA
CTS
de
vice
s,
f
or
norm
al
a
nd
co
ntin
ge
nc
y
sta
te
s.
Com
pu
te
the
operati
ng
co
st
an
d
re
qu
i
red
dev
ic
es
capacit
ie
s f
or eac
h
sta
te
.
Step
5:
Ca
lc
ul
at
e
cost
with
FA
CTS
us
i
ng
op
e
rati
ng
co
sts
of
al
l
sta
te
s
and
t
heir
ass
oc
ia
te
d
pro
bab
il
it
ie
s
to
occur.
Cal
culat
e d
e
vices in
ves
t
m
ent co
st usi
ng
(
8
).
Step
6:
E
valu
at
e
the
va
lue
of
the
obj
ect
iv
e
f
un
ct
io
n
(
9)
sub
j
ect
to
al
l
the
c
onstrai
nt
s
(
4
&
5)
.
I
f
any
of
the
co
ns
trai
nt
vio
la
t
ion
pe
nal
ty
is
add
e
d
i
n
cost
.
The
cal
cul
at
ed
value
of
t
h
e
fitness
functi
on
is
se
r
ved
as
a fitness
v
al
ue of
a
par
ti
cl
e/
cu
ckoo.
Step
7:
E
ac
h
pa
rtic
le
obj
ect
iv
e
is
cal
culat
ed
with
t
he
per
s
on
al
best,
l
oca
l
best.
If
t
he
fitness
value
is
lowe
r
than
local
be
st,
set
this
value
as
the
current
local
best
,
an
d
save
the
par
t
ic
le
po
sit
ion
c
orres
pondin
g
to
this
local
b
est
valu
e.
Step
8:
Sele
ct
the
m
ini
m
u
m
value
of
local
best
f
r
om
al
l
par
ti
cl
es
to
be
t
he
c
urren
t
gl
obal
be
st,
Glob
al
best,
and rec
ord
t
he parti
cl
e posit
io
n
c
orres
pondin
g
to
this
Globa
l best
value
.
Step
9: Up
date
each
pa
rtic
le
vel
ocity
and
al
so
po
sit
io
n.
Step
10:
If
the
m
axi
m
u
m
num
ber
of
it
eratio
ns
is
reac
hed,
the
par
ti
cl
e/
cucko
o
associat
ed
with
the
cu
rr
e
nt
Global
best
is
the
optim
al
so
l
ution.
Ot
herwise,
set
it
erati
on
=
it
erati
on
+
1
an
d
goto
St
e
p
4.
A
nd
repea
t
ti
l
l
te
rm
inati
on
4.
RESU
LT
S
A
ND
DI
SCUS
S
ION
Test
syst
e
m
is
3
-
sel
le
r
syst
e
m
and
tw
o
s
olu
ti
on
al
gorith
m
s
are
us
ed
.Here
the
no
FAC
Ts
de
vice
s
resu
lt
s
a
re
the
conve
ntion
al
m
et
ho
ds.
T
hePSO
a
nd
CS
A
a
re
ta
ke
n
her
e
.
As
s
how
n
i
n
t
he
resu
lt
s
t
he
f
it
ness
value
of
PS
O
and
CS
A
in
[
28
]
,
it
var
ie
s
from
$8
340
to
8190.
As
it
is
econom
ic
load
dis
patch
th
e
loss
consi
der
at
io
n
a
lso
base
d
on
th
e
loss
m
at
rix.
Wh
e
n
the
sam
e
3
-
sel
le
r
syst
e
m
is
us
ed
i
n
th
e
opti
m
a
l
pow
er
flo
w
the
cost
of
t
he
gen
e
rati
on
re
du
ce
s
to
$
80
34.
4.
we
us
e
th
e
sa
m
e
3
-
sel
le
r
syst
e
m
as
the
te
st
syst
e
m
a
nd
we
i
m
ple
m
ent the f
act
s d
e
vices
w
it
h
inclusi
on of
inv
e
stm
ent co
st.
The
F
ACTS
de
vices
co
ns
i
de
red
he
re
are
S
VC,
TCSC
a
nd
U
PFC.
SV
C
and
UPFC
m
od
el
s
are
ta
ken
as
reacti
ve
po
wer
m
od
el
an
d
the TCSC
is
ta
ke
n
as r
eact
an
c
e
m
od
el
.
T
he
obj
ect
iv
e
f
unct
ion
disc
us
se
d
i
n
(
1)
is
ta
ken
as
fitne
ss
equ
at
io
n
w
it
h
vo
lt
age
li
m
it
and
powe
r
flo
w
const
ra
ints.
The
well
-
kn
own
m
e
ta
heu
risti
c
al
gorithm
cal
led
P
SO
a
nd
CS
A
al
gorithm
s
are
us
e
d
f
or
te
st
ing
the
fitness
functi
on
for
wi
t
hout
facts
de
vi
ces.
The
n
the
(
9)
is
us
e
d
f
or
te
st
ing
with
FA
C
TS
dev
ic
es.
I
Cdevices
va
riable
can
be
re
pl
aced
wit
h
eac
h
fact
s
dev
ic
e c
os
t e
quat
ion res
pecti
vely
. T
he
re
su
l
ts o
btained
are
discuss b
el
ow.
4.1
.
PSO
a
lg
orith
m
PSO
al
gorithm
as
exp
la
ine
d
i
n
the
s
ol
utio
n
te
chnolo
gy
sect
ion
the
M
ATL
AB
co
de
is
i
m
plem
e
te
d
to
s
olv
e
both
(
1)
and
(
2).
T
he
F
ig
ure
1
s
how
s
t
he
c
onve
rg
e
nc
e
grap
h
of
the
PSO
al
gorithm
f
or
with
ou
t
an
d
with
placem
ent
of
SV
C,
TCSC
a
nd
U
PFC.
F
rom
that
it
can
be
seen
that
t
he
UPFC
giv
es
r
edu
ce
d
cost
in
cl
ud
i
ng
the
cost
of
U
PFC.
Fig
ur
e
2
sh
ows
the
vo
lt
age
prof
il
e
of
NO
facts
de
vice
conditi
on,
SV
C
placed
,
TCSC
placed
a
nd
UPFC
placed.
T
he
per
f
orm
ance
of
votl
age
pr
of
il
e
is
bette
r
a
nd
TCSC
is
no
t
pe
rfo
rm
in
g
well
,
as
the
co
st
in
cr
eases.
Fig
ure
3
shows
t
he
power
ge
ner
at
e
d
at
gen
e
rato
r
nu
m
ber
1,
2,
3
,
6
and
8.
It
ca
n
be
see
n
from
F
igu
re
3
that
G3,
G
6
and
G
8
has
si
gn
i
ficant
re
duct
ion
in
gen
e
r
at
ed
total
pow
er
w
hen
t
he
F
ACTS
dev
ic
es
a
re
pla
ced.
Ta
ble
1
s
hows
the
ge
ne
rated
po
wer
i
n
IEEE
-
14
bus
syst
e
m
.
Table
2
sho
ws
the
lo
cat
ion
,
siz
e,
cost
an
d
loss
of
the
3
-
se
ll
er
syst
e
m
wit
h
PS
O
al
gorith
m
.
It
can
be
seen
from
[2
8]
the
cost
f
ro
m
$
8100
(appro
x.) to
$
7910.
4 wh
e
n u
sing UP
FC inc
lud
in
g
t
he
i
nv
e
st
m
ent co
st o
f UPFC.
4.2.
CSA
algorith
m
Figures
4
-
6
s
hows
th
e
resu
lt
s
ta
ken
from
C
SA
f
or
F
ACT
S
dev
ic
e
place
m
ent
and
T
abl
e
s
3
and
4
sh
ows
the
num
erical
r
es
ults.
Using CS
A
c
ost
is sti
ll
r
educ
ed
to
$ 79
07.5
with
UP
FC.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
10
, No
.
4
,
A
ugus
t
2020
:
3350
-
3357
3354
Figure
1.
Co
nverg
e
nce
gr
a
ph
of PS
O
al
gorithm
w
it
h
an
d w
it
ho
ut
SV
C
, T
CSC
and
UP
F
C
Figure
2.
V
oltage
prof
il
e
with
and w
it
ho
ut S
VC,
TCSC
and
UPFC
Figure
3. Ge
ne
rated
powe
r wit
h
an
d wit
hout
SV
C,
TCSC
and
UPFC
Table
1.
Ge
nerat
ed
po
wer i
n M
W
Gen
.
n
o
s
Gen
erate
d
po
wer
i
n
M
W
No
FACTS
SVC
TCSC
UPFC
G1
1
8
6
.7514
9
1
9
2
.454
1
9
1
.048
1
9
1
.687
G2
3
5
.82
0
4
0
5
3
6
.93
1
1
3
6
.11
2
3
7
.00
9
7
G3
4
4
.05
2
8
3
9
2
3
.91
3
1
2
0
.75
2
3
1
9
.88
0
6
G6
0
8
.20
8
1
4
9
.92
2
8
7
1
2
.39
8
0
8
G8
0
0
6
.29
4
4
4
0
Table
2.
L
ocati
on, s
iz
e,
co
st a
nd loss
of t
he 3
-
sel
le
r
syst
em
w
it
h
P
SO al
go
rithm
Locatio
n
Size
Total Co
st in
$
Los
s in
M
W
NO
FAC
TS
-
-
8
0
5
4
.4
7
.62
4
7
SVC
4
8
4
.27
M
VAR
7
9
3
1
.9
2
.50
6
1
TCSC
6
to 1
1
0
.75
oh
m
s
8977
5
.12
9
7
UPFC
13
2
7
.95
3
MVAR
7
9
1
0
.4
1
.97
5
4
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N: 20
88
-
8708
Op
ti
mize
d plac
emen
t
of
mu
lt
i
ple FACT
S dev
ic
es u
sin
g PS
O
and CSA
a
l
gorit
hm
s
(
B
asana
gouda P
ati
)
3355
Figure
4. Co
nverg
e
nce
gr
a
ph
of CSA
alg
or
it
hm
w
it
h
an
d w
it
ho
ut
SV
C
, T
CSC
and
UP
F
C
Figure
5. V
oltage
prof
il
e
with
and
with
out S
VC,
TCSC
and
UPFC
Figure
6. Ge
ne
rated
powe
r wit
h
an
d wit
hout
SV
C,
TCSC
and
UPFC
Table
3.
Ge
nerat
ed
po
wer i
n M
W
Gen
.
n
o
s
Gen
erate
d
po
wer
i
n
M
W
No
FACTS
SVC
TCSC
UPFC
G1
1
8
6
.8083
1
3
3
1
8
7
.7138
1
8
8
.8311
2
0
8
.7254
G2
3
5
.97
5
8
3
5
3
1
3
6
.09
2
1
3
3
4
.69
8
8
5
3
5
.30
8
5
4
G3
4
2
.57
0
6
6
5
3
1
2
0
.33
0
0
8
1
3
.26
9
1
4
1
.79
9
5
9
3
G6
0
1
6
.49
7
2
1
1
6
.97
1
3
8
1
6
.22
5
9
6
G8
1
.31
5
0
7
6
1
6
7
0
1
1
.00
9
3
1
0
Table
4.
L
ocati
on, s
iz
e,
co
st
a
nd loss
of
t
he 3
-
sel
le
r
syst
em
w
it
h
CS
A
al
gorithm
Locatio
n
Size
Total Co
st in
$
Los
s in
M
W
NO FAC
TS
-
-
8
0
5
4
.4
7
.66
9
9
SVC
13
2
6
.74
3
2
M
VAR
7
9
1
4
.5
1
.63
3
3
TCSC
6
to 1
1
1
pu
8
1
1
4
.8
5
.77
9
8
UPFC
13
2
8
.28
1
9
MVAR
7
9
0
7
.5
3
.05
9
5
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
10
, No
.
4
,
A
ugus
t
2020
:
3350
-
3357
3356
5.
CONCL
US
I
O
N
The
MATL
AB
i
m
ple
m
entat
ion
of
th
e
place
m
ent
of
m
ult
i
ple
FA
CTS
de
vices
on
the
I
EEE
14
bus
syst
e
m
and
th
e
res
ults
wer
e
infe
rr
e
d.
The
optim
iz
ation
al
gorithm
that
was
us
e
d
f
or
the
placem
ent
of
the
m
ult
iple
FA
CTS
dev
ic
es
include
d
P
SO
an
d
CS
A
al
go
rithm
.
The
res
u
lt
s
ob
t
ai
ned
f
r
om
the
CSA
i
m
ple
m
entat
io
n
outpe
rfor
m
ed
PS
O
al
gorit
hm
and
the
co
st
fu
nc
ti
on
reduced
value
whil
e
op
tim
iz
ing
us
in
g
the
CSA
al
gor
it
h
m
.
So
,
c
ompare
d
to
befo
r
e
placem
ent
a
nd
a
fter
place
m
ent
of
m
ulti
-
facts
de
vices
the
total
cost
of
ge
ner
at
ion
re
du
ces
eve
n
incl
ud
i
ng th
e FA
CT
S
dev
i
ce cost.
REFERE
NCE
S
[1]
S.
N.
Singh
and
A.
K.
David,
“Conge
stion
Mana
gement
b
y
Opt
imizi
ng
FA
CTS
Devic
e
Locat
io
n,
”
DRPT2000
.
Inte
rnational
Co
nfe
renc
e
on
Elec
tric
Ut
il
i
ty
D
ere
gulat
ion
and
R
estruct
uring
and
P
ower
Technol
og
i
es
.
Proc
ee
d
ings
(
Cat.
No.
00E
X3
82)
,
London,
U
K,
pp.
23
-
28,
20
00.
[2]
S.
N.
Singh,
“
Loc
ation
of
FA
CTS
devi
c
es
for
enh
ancing
powe
r
s
y
stems
’
sec
ur
ity
,
”
LESCOPE
01.
2001
Lar
ge
Engi
ne
ering
Syste
ms
Confe
renc
e
on
Powe
r
Engi
nee
ring.
Conf
ere
nce
Proceedi
ngs
.
Theme:
Powe
ring
Be
yond
2001
(
Cat.
No.
01e
x
49
0)
,
Hali
fax
,
NS
,
Cana
da
,
pp
.
162
-
166,
2001
.
[3]
N.
G.
Hingor
ani,
“
Flexi
ble AC t
r
ansm
ission,”
in
I
EE
E
Spectrum
,
vol.
30
,
no
.
4
,
pp
.
40
-
45
,
April
19
93.
[4]
L.
G
y
ug
y
i
,
“
A
unifi
ed
power
fl
ow
cont
rol
concept
for
fl
ex
ibl
e
AC
tra
nsm
issio
n
sy
st
ems
,
”
in
I
EE
Proceedi
ng
s
C
-
Gen
erati
on,
Tr
ansm
i
ss
ion
and
Distributi
on
,
v
ol.
139
,
no
.
4
,
pp
.
323
-
331
,
Jul
y
1
992.
[5]
E.
L
arse
n,
N.
Mi
ll
er
,
S.
Nilsson
and
S.
Li
ndgre
n
,
“
Bene
fit
s
of
GTO
Based
compensat
ion
s
y
st
ems
for
el
e
ct
ri
c
ut
i
l
i
t
y
appl
i
ca
t
ions,”
in
IEE
E
Tr
ansacti
o
ns on
Powe
r
Del
iv
ery
,
vo
l. 7, no.
4,
pp
.
2056
-
206
4,
Oct
.
1992
.
[6]
M.
Noroozi
an
and
G.
Anders
on,
“
Pow
er
flow
cont
rol
b
y
u
se
of
cont
rol
la
b
le
seri
es
components,
”
in
IE
E
E
Tr
ansacti
ons on Power
Del
iv
er
y
,
vol.
8
,
no
.
3
,
pp
.
1420
-
1429,
Jul
y
1993.
[7]
S.
Gerbe
x,
R.
C
her
kaoui,
and
A.
Germ
ond,
“
Opt
imal
loc
a
ti
on
of
m
ult
ity
pe
FA
CTS
devi
ce
s
in
a
p
ower
s
y
stem
b
y
m
ea
ns of
gen
et
i
c
al
gor
it
hm
s,”
in
IEE
E
Tr
ansacti
o
ns on
Powe
r
Sys
te
ms
,
vol
.
16
,
no
.
3
,
pp
.
537
-
544
,
Aug.
2001
.
[8]
N.
K.
Sharm
a,
A.
Ghos
h,
and
R.
Varm
a,
“
in
I
EE
E
Tr
ansacti
o
ns
on
Powe
r
Deli
v
ery
,
vol.
18
,
no.
3,
pp.
982
-
987,
Jul
y
2003
.
[9]
J.
G.
Singh,
S.
N.
Singh,
and
S.
C.
Srivasta
v
a,
“
Plac
ement
of
FA
CTS
cont
roll
e
rs
for
enha
nci
ng
power
sy
st
em
loa
dab
il
ity
,
”
200
6
IEEE Power I
ndia
Conf
ere
nce
,
New De
lhi,
pp
.
7
,
2006
.
[10]
P.
Pree
davi
ch
it
and
S.C.
Srivast
ava
,
“
Optimal
re
ac
t
ive
power
disp
at
ch
consid
eri
n
g
FA
CTS
devi
c
es,
”
1997
Fourt
h
Inte
rnational
Co
nfe
renc
e
on
Advance
s in
Powe
r Sy
stem
Control,
Operation
and
Manage
ment,
A
PSCOM
-
97.
(
Co
nf.
Publ
.
No.
450)
,
Hon
g
Kong,
vol
.
2,
pp.
6
20
-
625,
1997.
[11]
T.
T.
L
ie
and
W
.
Deng,
“
Opti
m
al
fle
xible
AC
tra
nsm
ission
sy
st
ems
(F
ACTS)
devi
c
es
al
locati
o
n,
”
Inte
rnat
iona
l
Journal
of
Elec
t
rical
Powe
r and
Ene
rgy
syst
ems
,
vol.
19
,
no
.
2
,
pp
.
125
-
134
,
Febu
a
r
y
1997
.
[12]
Fang,
R.
,
S.
,
David,
A
.
,
K.,
“
Tra
nsm
ission
Congesti
on
Mana
g
ement
in
a
n
Elec
tr
ic
i
t
y
M
ark
et,”
in
I
EEE
Tr
ansacti
ons on Power
Syst
ems
,
vol.
14
,
no
.
3
,
pp
.
877
-
883
,
Aug.
1999.
[13]
Acha
r
y
a
,
N.,
Mithul
an
at
h
an,
N
.,
“
Loc
ating
Seri
es
FA
CTS
devi
ce
s
for
Congest
ion
Mana
gemen
t
in
Dere
gul
at
ed
E
lectr
i
ci
t
y
M
ark
et
s,”
E
le
c
tric P
o
wer
Syste
ms
Re
s
earc
h
,
vo
l. 77, n
o.
3
-
4
,
pp
.
352
-
3
60,
Mar
ch
2007
.
[14]
Brosda,
J., Hand
schin.
E
,
“
Conge
stion
m
ana
geme
nt
m
et
hods wit
h
a
spec
i
al
considerat
ion
of
FA
CTS
-
devi
c
es
,
"
2001
IEE
E
Porto
Pow
er
Tech Proceed
ings (
Cat.
No.
01
EX
502)
,
Porto,
Portugal,
vol
.
1
,
p
p.
6
,
2001
.
[15]
A.
A.
Athamne
h,
W
.
J.
Lee,
“
Bene
fi
ts
of
FA
CTS
devi
c
es
for
power
ex
cha
ng
e
among
Jordan
ia
n
Int
erc
onn
ect
ion
with
oth
er
Count
rie
s,”
2006
I
EEE
Powe
r
Engi
n
ee
r
ing
Soc
ie
t
y
Gen
e
ral M
eeting
,
Mo
ntreal
,
Que.,
pp
.
7
,
2006
.
[16]
S.
N.
Singh,
A.
K.
David,
“
A
new
appr
oa
ch
for
pla
c
ement
of
F
ACTS
devi
ce
s
i
n
open
power
m
ark
et
s
,”
in
I
EE
E
Powe
r E
ng
ine
eri
ng
Revie
w
,
vo
l.
21,
no
.
9
,
pp
.
58
-
60,
Sept
.
2001
.
[17]
I.
O.
El
ger
d
and
H.
J.
Happ
,
“
El
e
ct
ri
c
Ene
rg
y
S
y
s
te
m
The
or
y
-
An
Introduc
ti
on
,
”
i
n
IEE
E
Tr
ansacti
ons
on
Syste
ms
,
Man,
and
Cybe
r
net
i
cs
,
vol
.
SM
C
-
2,
no
.
2
,
pp
.
29
6
-
297,
April
197
2.
[18]
G.
C.
Ejebe
and
B.
F.
W
oll
enbe
r
g,
“
Autom
at
ic
c
onti
ngen
c
y
se
lec
ti
on
,”
in
IE
EE
Tr
ansacti
ons
on
Powe
r
Apparatu
s
and
Syste
ms
,
vol
.
PA
S
-
98,
no.
1,
pp.
97
-
109
,
Jan
.
1979.
[19]
Antonio
Gom
ez
-
E
xposito,
Anto
nio
J.
Cone
jo,
and
Cla
ud
io
Ca
niz
ar
es,
“
Elec
tr
i
c
En
erg
y
S
y
st
e
m
s
Anal
y
sis
an
d
Opera
ti
on
,
”
CR
C
Press
,
New Y
ork,
2009
.
[20]
L.
L
.
Freris
and
A.
M.
Sass
on,
"Investi
g
ation
o
f
the
Lo
ad
Flo
w
Problem,"
in
Proce
ed
ings
of
the
Institution
o
f
El
e
ct
rica
l
Eng
in
e
ers
,
vol
.
115
,
n
o.
10
,
pp
.
1459
-
1470,
Octob
er
1
968.
[21]
S.
N.
Singh
and
S.
C.
Srivasta
va
,
“
Corre
ct
ive
a
ct
i
on
pla
nning
to
a
chi
ev
e
a
fea
sib
le
opti
m
um
po
wer
flow
soluti
on,
”
in
IE
E Proce
ed
i
ngs
-
Gene
rat
ion
,
Tr
ansm
ission and
Distributi
on
,
vol.
142
,
no
.
6
,
p
p.
576
-
582
,
Nov.
1995.
[22]
G.
D.
Ga
li
an
a
e
t
a
l,
“
As
sess
m
e
nt
and
cont
ro
l
o
f
the
impac
t
of
FA
CTS
devi
ce
s
on
power
s
y
st
e
m
per
form
anc
e
,”
in
IE
EE Tr
ansacti
ons on Power
S
yste
ms
,
vol
.
11
,
no.
4
,
pp
.
1931
-
1936,
Nov.
1996
.
[23]
R.
Raj
ar
aman,
F
.
Alvar
ado
,
A.
Mania
c
i,
R.
C
a
m
fie
ld
and
S.Jal
al
i
,
“
Dete
rm
ina
tion
of
loc
ation
a
nd
amount
of
serie
s
compensat
ion
to
inc
r
e
ase
power
tra
nsfer
ca
pa
bi
li
t
y
,
”
in
IEEE
Tr
ansacti
ons
on
Powe
r
Syste
ms
,
vol.
13,
no
.
2,
pp.
294
-
300
,
Ma
y
1998.
[24]
E.
J.
de
Oliv
ei
ra
and
J.
W.
M.
L
ima,
“
Allocati
on
of
FA
CTS
device
s
in
a
competitive
envi
ronm
en
t,
”
13th
PS
CC
,
pp.
1184
-
1190
,
1999.
[25]
R.
K.
Sw
ai
n,
P.
K.
Hota,
R.
Cha
kra
bar
t
y
,
“
An
Aucti
on
Based
Di
spatc
h
Algorit
h
m
for
Dere
gula
t
ed
Pow
er
Sy
ste
m
s
using
Di
ffe
ren
t
i
al
Evol
u
ti
on
T
e
chni
que
,
”
F
if
t
eenth
Nati
onal
Po
wer
Syste
ms
Con
fe
renc
e
(
NPSC)
,
IIT
Bom
ba
y
,
pp.
201
-
207
,
De
ce
m
ber
2008
.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N: 20
88
-
8708
Op
ti
mize
d plac
emen
t
of
mu
lt
i
ple FACT
S dev
ic
es u
sin
g PS
O
and CSA
a
l
gorit
hm
s
(
B
asana
gouda P
ati
)
3357
[26]
M.
C.
Bhuv
ane
s
wari
and
J.
Sax
e
na
(ed
s.)
,
“
Int
el
l
i
gent
and
Eff
icie
nt
Elec
tri
c
al
S
y
s
te
m
s,
”
L
ecture
Notes
in
El
e
ct
ri
c
al
Engi
ne
eri
ng
446
,
Springer
na
tu
re
Singapore Pvt L
td,
pp
.
167
-
174,
2018.
[27]
X.
Yang
and
S
uash
Deb,
“
Cuckoo
Sear
ch
via
Lé
v
y
f
li
ghts
,
"
2009
World
Con
gress
on
Natu
re
&
Bi
ologi
cal
ly
Inspired
Comput
ing
(
NaBIC)
,
Coim
bat
ore
,
pp.
21
0
-
214,
2009
.
[28]
Singire
su
S.
Ra
o,
“
Engi
neering
opti
m
iz
ation:
T
heor
y
and
Prac
t
ic
e
,
Fourth
Edit
ion,
”
John
W
il
l
e
y
&
Sons
,
Inc
.
,
2009.
BIOGR
AP
H
I
ES
OF
A
UTH
ORS
Mr
.
Basan
agou
da
Pati
l
Re
ceiv
ed
the
M.
T
ec
h
in
PES
from
BEC
Bagalkot
Karna
ta
k
ai
n
y
e
a
r
2010.
At
Present
He
is
Purs
uing
P
h.
D
(Pow
er
Sy
st
em)
from
S
DM
C
ET
Dharwad
&
Li
fe
Mem
ber
of
India
n
Societ
y
f
or
Te
chnica
l
Ed
uca
t
ion
(ISTE)
,
His
Resea
rch
In
te
rest
in
Pow
er
s
y
stem
&
Fact
s
Devic
es
Dr
.
S.
B
.
Ka
raj
gi
Recei
v
ed
th
e
M.E
in
R
EC
W
ara
nga
l
1987,
&
Ph.
D
from
NITK
Surathka
l
in
2014.
Presently
He
is
W
orking
asva
Profess
or
in
Depa
r
tment
of
EEE
SD
MCET
Dharwa
d
Karna
ta
k
a.
HIS
Resea
rch
Are
a
i
nte
rests
in
Pow
er
S
y
stem
Oper
at
ion
&
Distr
ib
uti
on
Gene
r
at
io
n,
Li
fe
Mem
ber
of
India
n
Soc
ie
t
y
T
ec
hni
ca
l
Edu
ca
t
i
on
(IS
TE
).
Evaluation Warning : The document was created with Spire.PDF for Python.