TELKOM
NIKA Indonesia
n
Journal of
Electrical En
gineering
Vol.12, No.7, July 201
4, pp
. 4915 ~ 49
2
3
DOI: 10.115
9
1
/telkomni
ka.
v
12i7.399
3
4915
Re
cei
v
ed
Jul
y
24, 201
3; Revi
sed Ma
rch
15, 2014; Accepte
d
March
25, 2014
Sensitivity Fact
or Based Optimal Locat
ion of SVC on
Transmission Network
T. D. Sudhak
a
r
Dept of Electric
al an
d Electro
n
i
cs Engi
neer
in
g
St. Joseph’s C
o
lle
ge of En
gin
eeri
n
g
Che
nna
i, India
email: t_d
_
su
d
hakar@
ya
ho
o.co.in
A
b
st
r
a
ct
In mo
der
n day
s pow
er system ar
e bec
o
m
in
g hig
h
ly
unpr
e
d
ictab
l
e, so l
o
w
voltages a
n
d
black
o
u
t
are bec
o
m
in
g
common sce
n
a
rio. The volt
a
ge li
mits ar
e affected bec
aus
e the system i
s
operate
d
clo
s
e to
its stability li
mi
ts. To improve
the
voltag
e pr
ofile n
e
w
gen
e
r
ating p
l
a
n
ts have to be
dev
elo
ped, w
h
ich
ar
e
re
stri
cte
d
du
e to
e
c
on
om
i
c
a
n
en
vi
ronm
en
ta
l
lim
i
t
a
t
io
ns
. T
o
overco
me these
li
mitations
an
d use
th
e
existin
g
netw
o
rk to the maxi
mum, F
A
CT
S de
vices ar
e us
ed.
In this paper, SVC a F
A
CT
S
devic
e is used
to
improve
the v
o
ltage
profi
l
e
of
the n
e
tw
ork. To dec
id
e the
lo
cation
of
the S
V
C, voltag
e se
nsitivity a
nalys
i
s
meth
od is us
e
d
. F
A
CT
S devices are use
d
to compe
n
sa
te
both rea
l
an
d reactive p
o
w
e
r. T
o
demonstra
t
e
this IEEE 14 bus and 3
9
bus are consi
der
ed
. The compl
e
te
simu
latio
n
is d
one us
ing PSA
T version 2.1.6
in
Matlab 7.1
0
.0 (
R
20
10 a).
Ke
y
w
ords
:
se
nsitivity factor, SVC, transmis
s
ion n
e
tw
ork
Copy
right
©
2014 In
stitu
t
e o
f
Ad
van
ced
En
g
i
n
eerin
g and
Scien
ce. All
rig
h
t
s reser
ve
d
.
1. Introduc
tion
Power
syst
ems
co
mpo
nents mainl
y
con
s
i
s
t o
f
gene
rato
rs, tran
smissi
on lin
es,
transfo
rme
r
s, swit
che
s
, a
c
tive or pa
ssive com
pen
sat
o
rs
and l
oad
s. Power
syste
m
netwo
rks
a
r
e
compl
e
x syst
ems that a
r
e
nonline
a
r, n
on-stat
iona
ry, and p
r
one t
o
distu
r
ba
nces an
d fault
s
.
Reinfo
rceme
n
t of a p
o
wer
system
can be
a
c
co
mplish
ed
by improving t
he voltage
p
r
ofile,
increa
sing th
e transmissio
n cap
a
city an
d others.
The
ele
c
trical
ene
rgy
dem
a
nd in
crea
se
s
contin
uou
sly l
eadin
g
to
an
augme
n
ted
st
ress of
the tran
smi
s
sion li
ne
s hi
gher ri
sks fo
r faulted li
ne
s. The
exten
s
ion
of the t
r
an
smi
ssi
on
grid
need
ed to fu
rther gua
ra
ntee secure transmi
ssion
i
s
difficult fo
r enviro
n
ment
al and
politi
c
al
rea
s
on
s.
Th
e bla
c
kout
s i
n
different p
a
r
ts
of t
he worl
d in the last two
years hav
e sho
w
n that the
curre
n
t situati
on is
not satisfacto
ry and
a way to
in
crease tran
sfe
r
cap
ability an
d co
ntrolla
bility in
orde
r to e
n
su
re
se
cure po
wer tran
smi
s
sion
ha
s to b
e
found. An
option to a
c
hieve this i
s
t
he
utilization of flexible AC tr
ansmi
ssi
on sy
stem
s (FACT
S
).
The FACTS d
e
vice
s (Fl
e
xib
l
e AC Tran
sm
issi
on
Syste
m
s)
co
uld b
e
a mean
s to
carry o
u
t
this function
without the d
r
awba
cks of the electrom
ech
ani
cal de
vices
su
ch a
s
slo
w
ne
ss a
n
d
wear. FA
CTS
can im
prove the
stability
of network,
such as the transi
ent
and the
small
si
gnal
stability, and
can
red
u
ce the flow of h
eavily
loaded
lines a
nd
su
pport voltag
e
s
by co
ntrolli
ng
their pa
ram
e
ters i
n
cl
uding
seri
es im
ped
ance,
shu
n
t impeda
nce, current, and v
o
ltage an
d ph
ase
angle.
Controlling the power fl
ows in the
network leads to
redu
ce the fl
ow
of
heavily loaded
lines, in
crea
sed sy
stem lo
ad ability, less sy
stem
lo
ss and imp
r
ove
d
se
cu
rity of the sy
stem. T
he
increa
sed i
n
tere
st in the
s
e device
s
i
s
essent
ially d
ue to re
ce
ntly developme
n
t in high po
wer
electroni
cs th
at has ma
de t
hese dev
ices cost effe
ctive and in
crea
sed loadi
ng of
power
syste
m
s,
combi
ned
with dere
gulatio
n of powe
r
indu
stry.
On accou
n
t of consi
derable
co
sts of FACTS
devic
es
, it is
important to plac
e them in optimal loc
a
tion.
As voltage stability is one of the i
m
port
ant problems in the
power system,
FACTS
device
s
attra
c
ts the p
o
wer system e
ngi
neers to
so
lv
e this ki
nd of
probl
em. Th
e proj
ect aim
s
to
improve voltage stability margin by in
corporating SVC into power system.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 7, July 201
4: 4915 – 49
23
4916
Haa
d
i Saad
a
t
[2] discu
sse
d
about the i
m
pleme
n
tatio
n
of load flo
w
stu
d
ies
usi
ng NR meth
od.
Allen
J. Wo
od a
n
d
Bru
c
e F.
Wolle
n
berg,
propo
sed the
meth
od of
sen
s
itivity analysis and
sensitivity factor cal
c
ulation.
K.R Padiyar
[3] descri
bed
the vari
ou
s
FACT
S
controllers in
po
wer tran
smissi
on an
d
distrib
u
tion.
Narain G.
Hin
gora
n
i, La
szl
o Gyugy
i [4] discu
s
sed a
b
out und
ersta
nding th
e FA
CTS
device
s
pri
o
ri
tising SVC de
vice and its m
odelling.
R. Mohan Ma
thur and Raji
v K.Varma [5] carri
ed out a
basic
study of FACTS co
ntrolle
rs
along
with th
e re
active p
o
we
r
control
in the
ele
c
trical p
o
wer transmi
ssion
systems. Va
ri
ous
method
s of compen
satio
n
techni
que
s u
s
ed in t
he p
o
w
er
system
were al
so stu
d
i
ed.
In this pap
er,
the optimal locatio
n
of SVC, wi
th sp
e
c
ific cha
r
a
c
te
ristics i
s
foun
d. They
are lo
cate
d o
p
timally in order to maximi
ze the
se
cu
rity margin of t
he sy
stem in
terms
of bra
n
c
h
loadin
g
a
nd
voltage level
s
. Optim
a
l l
o
catio
n
of S
V
C in
the
g
i
ven sy
stem
is fo
und
u
s
ing
sen
s
itivity analysis with th
e help of sen
s
itivity fa
ctor values calcul
ated for the load bu
se
s in
th
e
system. T
he
bus with
high
est valu
e of
sen
s
itivity
factor with
re
sp
ect to
both
real an
d
rea
c
t
i
ve
power inje
cti
ons i
s
co
nsi
dere
d
as the
most se
nsiti
v
e bus in th
e netwo
rk a
n
d
is the opti
m
al
locatio
n
for the placement
of SVC.
To enha
nce the voltage profile and to increa
se
the tran
sfer
cap
a
b
ility of the system a
methodol
ogy
is p
r
op
ose
d
whi
c
h
de
als
with fi
nd
ing the
optimal lo
cation
of SVC in
the
transmissio
n
network by
calculating
the sen
s
it
ivity factor values u
s
ing v
o
ltage sensit
ivity
analysi
s
.
2. Proposed
M
e
thod
olog
y
No
w a d
a
y, sen
s
itivity analysis i
s
g
a
in
ing mo
re imp
o
rtan
ce in
practical po
we
r system
operation
s
. T
he p
o
wer op
erato
r
u
s
e
s
t
he
sen
s
itivity value
s
to
st
udy an
d m
o
n
i
tor the
sy
ste
m
behavio
ur an
d detect po
ssible pro
b
lem
s
in the network.
Voltage
sen
s
itivity
factor i
s
related to
vo
ltage sta
b
il
ity. Voltage instability is
mainly
asso
ciated
with reactive p
o
we
r imbala
n
c
e in a lo
cal
network or
a spe
c
ified b
u
s in a syste
m
whi
c
h is
call
ed the we
ak bus. Voltag
e sen
s
itiv
ity
analysi
s
can detect the weak b
u
ses in
the
power sy
ste
m
where the
voltage i
s
lo
w. Th
erefo
r
e
voltage
sen
s
i
t
iv
ity factor is used
sel
e
ct
the
optimal locations of reactiv
e
power supp
ort.
2.1.
Algorithm fo
r Calcula
t
in
g Voltage Sensitivit
y
Factor
Voltage sen
s
itivity factor [6] is the vari
ati
on of voltage stability i
ndex with re
spe
c
t to
cha
nge
s i
n
real a
nd
rea
c
tive power i
n
jectio
ns
at
a bu
s. Sen
s
i
t
ivity indices that relate t
h
e
cha
nge
s in
the voltage
st
ability index
with respe
c
t
to ch
ange
s i
n
inje
cted
active and
rea
c
tive
power at a l
oad bu
s a
r
e
derived from
the voltage
stability index formulatio
n. Voltage sta
b
i
lity
index at a lo
ad bu
s ide
n
tifies critical
b
u
se
s i.e. bu
ses
whi
c
h a
r
e
pron
e to vol
t
age collap
s
e
in
power sy
ste
m
.
Step 1: Calculate the voltage stability index
by usi
ng voltage e
quation. The
voltage
stability index
is given by,
(1)
Whe
r
e,
Step 2:
Cal
c
ulate
an
d
by
differenti
a
ting
eq
uatio
n (Equ
ation 1
)
with
re
spe
c
t
to
and
resp
ectiv
e
ly as sh
own:
(2)
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Sensitivit
y Fa
ctor Ba
sed O
p
tim
a
l Location of SVC on
Tran
sm
issio
n
… (T. D. Su
dha
kar)
4917
(sin
ce
co
sine
of
<< 1)
(3)
Step 3: Obtain the values
of elements.
From the inve
rse of loa
d
flow jacobia
n
matrix.
Step 4: Calcu
l
ate the first sensitiv
ity factor whi
c
h i
s
the cha
nge in
L
i
with res
p
ect to th
e
injecte
d
real
power P
i
in i
th
bus given by
:
(4)
Step 5: Express the a
bove
equatio
n in a matrix as bel
ow:
(5)
Step
6: Cal
c
u
l
ate
the se
co
nd sen
s
itivit
y factor which
i
s
the
chan
ge
in L
i
with respec
t to
the injecte
d
reactive po
we
r Q
i
in the i
th
bus given by:
(6)
Step 7: Express the
(equ
ation. 6) in a ma
trix as belo
w
:
(7)
Step 8: Equations (5
) and
(7) give the voltage
se
nsi
t
ivity factor value
s
with re
spe
c
t to
the injecte
d
real and rea
c
tive powe
r
at the i
th
bus re
spectively.
2.2.
Algorithm a
nd Flo
w
c
h
a
r
t for th
e Pro
posed Sch
e
me
The
pro
p
o
s
e
d
sch
e
me
is to en
han
ce
th
e voltage
p
r
o
f
ile of the
un
comp
en
sated
syste
m
by providin
g comp
en
satio
n
usin
g SVC
controlle
r. To
achi
eve this e
ffectively optimal location
s for
placi
ng the S
V
C have to b
e
found. The
r
efore for
fin
d
ing the opti
m
al locations of VAR sup
port
sen
s
itivity analysis i
s
ca
rri
ed out as di
scu
s
sed ea
rlie
r. The step by
step proce
d
u
r
e is a
s
follows.
Step 1: Creat
e the PSAT
model
of the
giv
en
system
and i
nput the
given li
ne
data and
bus data in the respective
blocks of the PSAT model.
Step 2: Run the power flo
w
pro
g
ra
m using
NR meth
o
d
and save th
e load flow re
sults for
the norm
a
l sy
stem.
Step 3: From the load flow re
sult
s, chec
k wh
ethe
r the system
is comp
ensated i.e.,
voltage mag
n
itude of all the bu
se
s are
within lim
its. If it is not compen
sate
d go to next step
otherwise pri
n
t the load flow re
sults.
Step
4: Run
the
matlab
coding
for se
n
s
it
ivity analysis by
providi
n
g line
data
a
nd b
u
s
data of the given system a
s
input
s and
save the se
n
s
itivity factor values for all
the load bu
se
s.
Step 5: Find the bu
s with the highe
st value of
sen
s
itivity
factor wit
h
respe
c
t to both real
and re
active
power inje
ctio
ns an
d de
clare that
bus a
s
the most sen
s
itive bus in t
he network.
Step 6: Th
erefore th
e m
o
st sen
s
itive b
u
s i
s
co
nsi
d
e
r
ed to
be
the
optimal l
o
ca
tion for
placi
ng the SVC. Model th
e SVC block
and conne
ct it
to the most
sen
s
itive bu
s in the netwo
rk in
parall
e
l in the PSAT model.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 7, July 201
4: 4915 – 49
23
4918
Step 7: Ru
n the po
we
r flo
w
p
r
og
ram u
s
ing NR meth
od for th
e ne
w mod
e
l with
SVC and
sav
e
t
he loa
d
f
l
ow re
sult
s.
Figure 1. Flowchart for the
Propo
sed Scheme
Start
Create
the PSAT
m
o
del
of the
give
n sy
ste
m
I
nput the line data
and bus data in PSAT
m
odel
Run the power
flo
w
pr
ogr
am
using
NR
m
e
thod for
the PSAT
m
odel
Save the load flow results for the syst
e
m
Rem
ove the sy
nch
r
onous con
d
enser
s
in the PSAT
M
odel
Save the new
m
odel without condens
er
s
Run the power
flo
w
pr
ogr
am
using
NR
m
e
thod
for
the new PSAT
m
odel
Print the load flow results
Run the
m
a
tlab
coding for sensitivity anal
ysis by
giving line data and bus data as input
s
Find out the bus with the highest valu
e of
sensitivity factor w
ith respect to both real
&
reactive power inje
ctions in the network
Place the S
V
C
m
o
del in the
m
o
st sen
s
itive bus
Check if
the
sy
ste
m
is
co
m
p
ensated
Yes
No
B
B
Stop
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Sensitivit
y Fa
ctor Ba
sed O
p
tim
a
l Location of SVC on
Tran
sm
issio
n
… (T. D. Su
dha
kar)
4919
Step 8: Com
pare
the l
oad
flow
re
sults
of
the sy
stem
whi
c
h i
s
un
compen
sate
d
and
with
SVC. Che
ck wheth
e
r the
system i
s
compen
sate
d
and the volt
age p
r
ofile o
f
the system
is
enha
nced. If
it is so print t
he re
sults ot
herwise
re
pe
at step 4 and
find the next most se
nsiti
v
e
bus i
n
the n
e
twork a
nd
rep
eat the ste
p
s
from 6
to
8. The flowch
art
of the propo
sed metho
d
olo
g
y
is given in Fig
u
re 1.
Therefore
by cal
c
ulatin
g the vo
ltage sen
s
itivity factor values fo
r all
the load b
u
ses, th
e
bus with the
highest valu
e of sensitivi
t
y factor
with resp
ect to both real and
rea
c
tive power
injectio
ns i
s
consi
dered to be the mo
st sensitive
bu
s
i
n
the system.
3.
Test Sy
stem and Res
u
lts
The
proposed scheme is
implemented in a st
andard IEEE 14
bus
and IEEE 39
bus
network
and
the results are discussed.
The
singl
e
line diagram of
standard IEEE 14 bus
netwo
rk i
s
sh
own in Fig
u
re
2.
A standard I
EEE 14 bus
network
i
s
considered and assu
mptions are m
ade to obtain
variou
s ill con
d
itions (i.e., voltage collap
s
e). In t
he assume
d netwo
rk sen
s
itivity
analysi
s
is do
ne
to find the optimal locatio
n
of FACTS co
ntrolle
rs.
Figure 2. IEEE 14 Bus Net
w
ork
Cons
idering the s
t
andard IEEE –
14 bus
network
[7] as
our tes
t
cas
e
, load flow analys
is
by NR method is done for the
network model in PSAT softwar
e. By analyzing the voltage
magnitud
e
from the re
sult
s (Ta
b
le 2
)
it is found t
hat
the system i
s
stabl
e.
In the next step,
th
e
con
den
se
rs locate
d in b
u
s
6
and
bu
s
8 in the
sy
st
em are remo
ved as shown in Fig
u
re
3
.
By
analyzi
ng th
e
voltage
mag
n
itude
(Vm) from lo
ad flo
w
results it vari
es i
n
th
e
ran
g
e from
0.89
654
(p.u) to 1.05
(p.u). Sen
s
iti
v
ity analysis is don
e us
i
n
g
matlab co
din
g
and sen
s
itivity
factor val
ues
for the load b
u
se
s are cal
c
ulated a
s
discussed a
nd ta
bulated in Ta
ble 1.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 7, July 201
4: 4915 – 49
23
4920
Figure 3. Modified 14 Bus
Network
Table 1. Sens
itivity Values
for IEEE 14
Bus
Net
w
ork
Bus
no
Normal s
y
stem
Without condensers
∂L
∂P
∂L
∂Q
∂L
∂P
∂L
∂Q
4 0.7489
0.0506
0.1032
0.4384
5 0.4987
0.2744
0.6394
0.0997
7 0.0969
0.3260
0.5720
0.4297
9 1.0138
0.5484
1.0841
1.2079
10 0.4698
0.6737
0.4688
0.8839
11 1.1782
0.7934
0.8412
1.5371
12 1.7732
0.2660
0.7494
0.6300
13 2.2847
0.1753
1.6339
0.5208
14 2.6178
0.2020
1.0699
0.8377
Based o
n
tab
l
e results the
sen
s
itivity factor whi
c
h is t
he ch
ang
e in L
i
with res
p
ect to th
e
injecte
d
re
al power P
i
, the
highe
st value
is at bu
s 13,
if co
mpe
n
sated at this b
u
s real p
o
we
r o
n
ly
will be com
p
ensated. This can be a
c
hi
eved by us
in
g distrib
u
ted
gene
rato
rs. Similarly on table
results th
e
se
nsitivity factor whi
c
h i
s
the
cha
nge i
n
L
i
with respec
t to the in
je
c
t
ed r
e
ac
tive
p
o
we
r
Q
i
, the highest value is at bus
11, if compensated at this bus
reactive power only will
be
comp
en
sated
.
This can be
achieve
d
by
usin
g ca
pa
citor ba
nks. Here SVC is u
s
e
d
whi
c
h
can
be
use
d
to comp
ensate both the paramete
r
s. Based o
n
whi
c
h it is fou
nd that the most se
nsitive
bus
in the net
wo
rk mo
del i
s
bu
s 9
sin
c
e
bu
s 9 have th
e
hi
gher value
of sen
s
itivity factor with
re
spe
c
t
to both re
al
and rea
c
tive power inj
e
ct
ions
com
p
a
r
ed to all oth
e
r loa
d
bu
se
s. The
r
efore
by
placi
ng SVC in that
bus it is found
that the voltage
s in
all t
he bu
se
s in
the netwo
rk got
comp
en
sated
which is give
n in Table 2. From t
he Ta
b
l
e 2, compa
r
i
ng the voltag
e magnitud
e
of
the
system wi
thout
co
nde
n
s
er and of
the system
with
SVC at bus
9 from the lo
ad flow
re
sult
s it
is clea
r that the voltage pr
ofile
of the system ha
s imp
r
oved an
d t
he system is co
mpen
sated. T
he
proposed scheme is impl
emented on a
standard
IEEE 14 bus net
work using P
SAT software and
the load flow
results obtai
n
ed veri
fy that the system i
s
com
pen
sa
t
ed by placi
n
g
the SVC in the
optimal location found by sensitivity factor analy
s
is.
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Sensitivit
y Fa
ctor Ba
sed O
p
tim
a
l Location of SVC on
Tran
sm
issio
n
… (T. D. Su
dha
kar)
4921
Table 2. Loa
d
Flow Re
sult
s
Bus no
Normal s
y
stem
Without condenser
With SVC
V
m
(p.u) V
a
(rad
)
V
m
(p.u) V
a
(rad
)
V
m
(p.u) V
a
(rad
)
1 1.05
0.000
1.05
0.0000
1.05
0.000
2 1.05
0.002
1.05
0.0021
1.05
0.050
3 1.01
-
0.140
0.96
-0.1377
1.00
-0.02
4 1.01
-
0.108
0.97
-0.1024
1.03
0.062
5 1.02
-
0.089
0.98
-0.0826
1.03
0.057
6 1.05
-
0.192
0.93
-0.1961
0.98
0.108
7 1.02
-
0.165
0.93
-0.1699
1.04
0.208
8 1.05
-
0.165
0.93
-0.1699
1.04
0.208
9 1.00
-
0.195
0.92
-0.2077
1.05
0.284
10 1.00
-
0.200
0.91
-0.2122
1.03
0.249
11 1.02
-
0.198
0.92
-0.2072
1.00
0.179
12 1.03
-
0.212
0.91
-0.2202
0.96
0.105
13 1.02
-
0.216
0.91
-0.2270
0.97
0.117
14 0.99
-
0.223
0.89
-0.2393
0.99
0.195
Figure 4. IEEE 39 Bus Net
w
ork
Figure 5. Modified 39 Bus
Network afte
r Removing G
enerators G3
and G6
Cons
idering the s
t
andard IEEE
39 bus
[8] as
s
h
own in Figure
4 networks
as
our tes
t
ca
se
s, load fl
ow an
alysi
s
by NR m
e
tho
d
is do
ne for
the netwo
r
k
model
s in PS
AT softwa
r
e. By
analyzi
ng the
voltage magnitude from the re
sults (T
able 4) it is found that the system
s are
compensated. In the
next step,
two generators i
n
the IEEE 39
bus
system
are removed as
sho
w
n i
n
Fig
u
re
5 an
d an
alyzing th
e voltage ma
gnit
ude
(Vm) fro
m
load flo
w
result
s it varie
s
in
the range from 0.878 p.u to 1.00 p.
u. for IEEE 39 bus network. Sens
itivity analysis i
s
done using
matlab co
din
g
and se
nsiti
v
ity
factor values for
the l
oad bu
se
s are cal
c
ulated
as tabulate
d in
Table 3I for 3
9
bus n
e
two
r
ks.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 7, July 201
4: 4915 – 49
23
4922
Table 3. Sens
itivity Values
for IEEE 39
Bus
Net
w
ork
Bus n
o
Normal s
y
ste
m
W
i
tho
u
t 2
gen
e
rators
15 0.2706
0.0702
0.3662
0.0296
16 0.0725
0.0371
0.4385
0.0644
17 0.1590
0.0565
0.0555
0.0440
18 0.3153
0.0786
0.0800
0.0669
19 0.4580
0.0339
0.2673
0.0333
20 0.1346
0.0426
0.5403
0.0360
21 0.2141
0.0544
0.2297
0.0654
22 0.1134
0.0373
0.3062
0.0416
23 0.1203
0.0453
0.1337
0.0416
24 0.0948
0.0525
0.2399
0.0673
25 0.0369
0.0447
0.0034
0.0417
26 0.0137
0.0742
0.0087
0.0738
27 0.0403
0.0818
0.4346
0.0422
28 0.5221
0.1068
0.2549
0.1047
29 0.1219
0.0564
0.0917
0.0573
Table 4. Loa
d
Flow Re
sult
s
Bus n
o
Normal s
y
ste
m
W
i
tho
u
t 2
gene
rators
With SVC
V
m
(p.u)
V
a
(ra
d)
V
m
(p.u)
V
a
(ra
d)
V
m
(p.u)
V
a
(rad
)
1 1.000
0.000
1.000
0.0000
1.000
0.000
2 1.011
-0.050
0.978
-0.4168
0.986
-0.409
3 0.983
-0.134
0.933
-0.6063
0.951
-0.593
4 0.949
-0.191
0.897
-0.7323
0.932
-0.712
5 0.948
-0.193
0.899
-0.7304
0.936
-0.710
6 0.950
-0.187
0.903
-0.7309
0.940
-0.711
7 0.940
-0.218
0.887
-0.7368
0.923
-0.716
8 0.940
-0.221
0.885
-0.7264
0.920
-0.707
9 0.985
-0.138
0.937
-0.4072
0.953
-0.404
10 0.958
-0.125
0.901
-0.7328
0.954
-0.714
11 0.954
-0.146
0.900
-0.7322
0.953
-0.713
12 0.935
-0.141
0.879
-0.7340
1.000
-0.716
13 0.956
-0.134
0.910
-0.7334
0.955
-0.715
14 0.956
-0.155
0.905
-0.7349
0.947
-0.715
15 0.965
-0.139
0.923
-0.7417
0.947
-0.722
16 0.985
-0.102
0.947
-0.7115
0.963
-0.693
17 0.987
-0.116
0.945
-0.6710
0.960
-0.654
18 0.984
-0.131
0.938
-0.6555
0.955
-0.640
19 0.989
-9e-05
0.976
-0.6067
0.981
-0.589
20 0.986
-0.017
0.979
-0.6246
0.982
-0.607
21 0.993
-0.057
0.950
-0.7250
0.964
-0.706
22 1.020
0.023
0.971
-0.6978
0.982
-0.679
23 1.019
0.022
0.983
-0.6797
0.992
-0.661
24 0.993
-0.100
0.957
-0.7242
0.972
-0.705
25 1.021
-0.031
0.985
-0.4271
0.992
-0.417
26 1.012
-0.068
0.976
-0.5422
0.985
-0.529
27 0.994
-0.112
0.952
-0.6243
0.965
-0.609
28 1.016
-0.002
0.998
-0.4747
1.002
-0.462
29 1.019
0.049
1.006
-0.4225
1.009
-0.410
30 1.048
-0.008
1.048
-0.3726
1.048
-0.365
31 0.982
-0.168
0.982
-0.7109
0.982
-0.692
32 0.983
0.014
0.901
-0.7328
0.954
-0.714
33 0.997
0.091
0.997
-0.5154
0.997
-0.498
34 1.012
0.073
1.012
-0.5339
1.012
-0.516
35 1.049
0.113
0.971
-0.6978
0.982
-0.679
36 1.064
0.162
1.064
-0.5351
1.064
-0.518
37 1.028
0.088
1.028
-0.3043
1.028
-0.295
38 1.027
0.172
1.027
-0.2979
1.027
-0.286
39 1.000
-0.082
1.000
-0.2128
1.000
-0.213
It is
found that the mos
t
s
e
ns
itive bus in
the IEEE 14
bus net
work is
bus nine
s
i
nc
e bus
nine h
a
ve the high
est val
ue of sen
s
itivity factor with
respe
c
t to b
o
th real
and
rea
c
tive po
wer
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Sensitivit
y Fa
ctor Ba
sed O
p
tim
a
l Location of SVC on
Tran
sm
issio
n
… (T. D. Su
dha
kar)
4923
injec
t
ions
c
o
mpared to all
other load bus
es
, s
i
m
ilarly
for IEEE 39 bus
network
the mos
t
s
e
ns
itive
bus i
s
foun
d to be bu
s 12.
Therefore
by placi
ng SVC i
n
the re
spe
c
ti
ve
buse
s
it is found that th
e
voltages i
n
al
l the bu
se
s i
n
the net
wo
rks got comp
en
sated
which i
s
given in
Tab
l
e 4. From th
e
table, com
p
a
r
ing the volta
ge mag
n
itud
e of the
sy
stems
without
con
den
se
r a
nd of the
systems
with SVC f
r
o
m
the lo
ad fl
ow
re
sults it is
clea
r that
the voltage
p
r
ofiles of the
system
s
ha
ve
improve
d
and
the system
s are comp
en
sated.
4. Conclu
sion
In this pa
per sen
s
itivity factor b
a
sed o
p
ti
mal locatio
n
of SVC is f
ound
wh
ere t
he be
st
location is inf
e
rred considering
the lo
sse
s
, investm
ent in the FACTS co
ntroll
ers a
nd al
so
the
voltage imp
r
o
v
ement value
s
. The
propo
sed
metho
d
i
s
an
effective
and
pra
c
tical
method fo
r t
h
e
allocation of FACTS co
ntrollers.
In the con
s
id
ered
test
ca
se bu
s 9 i
s
fo
u
nd to b
e
mo
st
sen
s
itive
with re
sp
ect to
b
o
th re
al
and re
active
powe
r
injections by trial and er
ror
method an
d
sensitivity analysi
s
meth
od.
Therefore
b
u
s
9
i
s
con
s
id
ered
a
s
th
e
optimal l
o
cati
on fo
r the
pl
acem
ent
of
SVC. Thu
s
t
he
optimal lo
cati
on for pl
aci
n
g the FACT
S device
s
(SVC) in the
netwo
rk
mo
del is fo
und
by
sen
s
itivity factor analysi
s
.
It is cle
a
rly e
v
ident from t
he result that effective pla
c
eme
n
t of FA
CTS devi
c
e
s
in prope
r
locatio
n
s ca
n si
gnificantl
y
improve
system
p
e
rfo
r
mance.
Thi
s
app
ro
ach could be a n
e
w
techni
que for
the installatio
n
of FACTS device
s
in the tran
smi
ssi
on system.
Referen
ces
[1]
Alle
n J W
ood, Bruce F
W
o
lle
nber
g. Po
w
e
r g
ener
ation, o
per
ation a
nd co
ntrol. 421-
43
3.
[2]
Haadi Saadat. Po
w
e
r sy
stem
analy
sis.
[3]
KR Padi
ya
r. F
a
cts controllers
i
n
po
w
e
r trans
mission
and d
i
stributio
n.
[4]
Narai
n
G Hing
o
ran
i
, Laszlo G
y
u
g
y
i. Un
derst
and
ing F
A
CT
S.
[5]
R Mohan Mat
hur, Rajiv K Varma. T
h
y
r
isto
r Based
F
A
CT
S Controll
ers for Electrical T
r
ansmissi
o
n
S
y
stems.
[6]
T
S
Abdel Sal
a
m, AY Chik
ha
ni, R H
a
ckam.
A ne
w
t
e
ch
ni
que f
o
r loss
re
ductio
n
us
ing
compe
n
satin
g
capac
itors a
p
p
lied
to
distrib
u
tion s
y
st
ems
w
i
th var
y
i
ng
loa
d
con
d
itio
ns.
IEEE Transactions on Power
Deliv
ery
. 199
4; 9(2): 819–
82
7.
[7]
R Ebr
ahim
pour, EK Abharian, SZ Moussav
i
, AA Mo
tie
Birj
and
i. T
r
ansient
Stabi
lit
y Asse
ssment of
a
Po
w
e
r S
y
stem
b
y
Mi
xtur
e of Exp
e
rts.
Internation
a
l Jour
na
l of Engin
eeri
ng (I
JE).
4(1): 93-104.
[8]
http://psdy
n
.
e
c
e
.
w
isc.edu/IEEE_benchm
arks
Evaluation Warning : The document was created with Spire.PDF for Python.