Internati
o
nal
Journal of P
o
wer Elect
roni
cs an
d
Drive
S
y
ste
m
(I
JPE
D
S)
V
o
l.
5, N
o
. 4
,
A
p
r
il
201
5, p
p
.
47
7
~
48
5
I
S
SN
: 208
8-8
6
9
4
4
77
Jo
urn
a
l
h
o
me
pa
ge
: h
ttp
://iaesjo
u
r
na
l.com/
o
n
lin
e/ind
e
x.ph
p
/
IJPEDS
Power Quality Improvement Usin
g Custom Power Devices in
Squirrel Cage In
duction Gen
e
rator Wind Farm
to Weak-Grid
Connection by using
Neuro-fuzzy Control
Kopella
S
a
i
T
e
ja, R.B. R.
pr
ak
ash
Departem
ent
of
Ele
c
tri
cal
and
E
l
ectron
i
cs
Eng
i
ne
ering,
K L
Univ
ers
i
t
y
Article Info
A
B
STRAC
T
Article histo
r
y:
Received
J
u
n 12, 2014
Rev
i
sed
No
v
18
, 20
14
Accepte
d Dec 5, 2014
W
i
nd farm
is co
nnect
ed to
the
g
r
id dir
ectly
.The
wind is not
cons
tant
vo
ltage
fluctu
ations o
c
cur at point of
co
mm
on coupling
(PCC) and WF terminal. To
overcome this
problem a new
compensa
tion
strateg
y
is used
. B
y
using
Cus
t
om
power devices
(UP
Q
C).It inj
ects
re
act
ive power a
t
P
CC. The
advant
ages
of U
P
QC are it
cons
i
s
ts
of both DVR and D-STATC
O
M. DVR is
connected in
ser
i
es to th
e lin
e an
d it injects in ph
ase voltage in
to
the lin
e
.D-
STATCOM is connected shunt
to the line
.The intern
al con
t
rol strateg
y
is
based on management of activ
e and r
eactiv
e
power in se
ries and shunt
converters of UP
QC. The power
exch
ainge is don
e b
y
using
DC-link.
Keyword:
DClin
k
Neuro-fuzzy l
o
gic control
Si
m
u
latio
n
SCIG
UP
QC
Copyright ©
201
5 Institut
e
o
f
Ad
vanced
Engin
eer
ing and S
c
i
e
nce.
All rights re
se
rve
d
.
Co
rresp
ond
i
ng
Autho
r
:
Kop
e
lla sai teja,
Depa
rtem
ent of Electrical a
n
d
El
ct
ro
ni
cs E
n
gi
nee
r
i
n
g,
K
L Un
iv
e
r
s
i
ty
.
Em
ail: kopellasaiteja@live.c
o
m
1.
INTRODUCTION
Th
e l
o
catio
n
o
f
g
e
n
e
ratio
n
facilities fo
r
wind
en
erg
y
i
s
d
e
term
in
ed
b
y
wi
n
d
en
erg
y
resou
r
ce
av
ailab
ility, o
f
ten
far
fro
m
h
i
g
h
vo
ltag
e
(HV) power
t
r
ansmissio
n
g
r
ids and
m
a
j
o
r con
s
u
m
p
tio
n centres.
In
case of facilities with
m
e
dium
power ratings, the
W
i
nd
Farm
is conne
cted thr
ough
m
e
dium
voltage (M
V)
d
i
str
i
bu
tio
n head
lin
es.
Al
so, i
s
wel
l
kn
o
w
n t
h
at
gi
ven t
h
e ra
nd
o
m
nat
u
re o
f
wi
n
d
res
o
urce
s, t
h
e wi
nd
fa
rm
generat
e
s
flu
c
tu
ating
electric p
o
wer. Th
ese
o
s
cillatio
n
s
h
a
v
e
a
n
e
g
a
tiv
e i
m
p
act o
n
stab
ility an
d
po
wer qu
ality in
electric
p
o
wer system
s
.
fu
rt
h
e
rm
o
r
e,
in
d
e
v
e
lop
m
en
t o
f
wind
reso
urces, tu
rb
in
es u
tilizin
g
sq
uirrel cag
e ind
u
ction
gene
rat
o
rs (
S
C
I
G
)
ha
ve
been
use
d
si
nce t
h
e
begi
nni
ng
s. T
h
e o
p
e
r
at
i
on
o
f
sq
ui
r
r
el
cage
i
n
d
u
ct
i
on
ge
n
e
rat
o
r
dem
a
nds react
ive powe
r,
ge
nerally provided from
th
e mains and/
or
by local ge
ne
ration i
n
capa
c
itor
b
a
nk
s [1
].
In the eve
n
t that changes
occ
u
r in
its
m
echanical speed, i.e. due to
wi
nd d
i
st
urba
nces wi
l
l
t
h
e W
I
N
D
FARM
active (reactive
)
power
injected
(
d
e
m
anded
)
i
n
t
o
t
h
e
po
wer
g
r
i
d
,
l
eadi
n
g t
o
va
r
i
at
i
ons
of
wi
n
d
fa
rm
Term
inal voltage
because of s
y
ste
m
im
pedance. T
h
is
power disturba
nces t
r
ansm
it
into the powe
r system
,
a
nd
can
p
r
od
u
ce a p
h
e
n
o
m
en
on
k
nown
as “flick
e
r”, wh
ich
co
nsists o
f
fluctu
atio
n
s
in
the illu
min
a
tio
n lev
e
l
cause
d by
v
o
l
t
a
ge va
ri
at
i
ons
.
Al
so, t
h
e n
o
r
m
a
l
ope
rat
i
on
of
W
i
n
d
Farm
i
s
im
pai
r
ed d
u
e t
o
suc
h
di
st
ur
ba
nces.
In pa
rticular
for the
case
of “
w
eak gr
id
s”, the i
m
p
act is ev
en
b
e
tter.
In ord
e
r t
o
red
u
c
e th
e vo
ltag
e
flu
c
tu
ations th
at m
a
y cause “
f
licker”
,
and im
prove
W
i
nd Fa
rm
t
e
rm
i
n
al
vol
t
a
ge re
gul
at
i
o
n,
several
re
sul
t
s
have
been
p
o
s
ed. T
h
e m
o
st
com
m
on one
i
s
t
o
rai
s
e t
h
e po
we
r
gri
d
, en
ha
nci
n
g t
h
e sh
o
r
t
ci
rcui
t
po
wer l
e
v
e
l
at
t
h
e poi
nt
of com
m
on cou
p
l
i
n
g p
o
i
n
t
of c
o
m
m
on coupl
i
n
g,
t
hus
re
duci
n
g
t
h
e i
m
pact
of
p
o
we
r
fl
uct
u
at
i
o
ns a
n
d
vol
t
a
ge
reg
u
l
a
t
i
o
n
p
r
ob
l
e
m
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-86
94
I
J
PED
S
Vo
l. 5
,
No
. 4
,
Ap
r
il 2
015
:
47
7
–
48
5
47
8
In
rece
nt
y
ear
s, t
h
e t
e
c
h
n
o
l
ogi
cal
devel
o
pm
ent
of
hi
g
h
po
wer
el
ect
ro
ni
cs
devi
ces
has l
e
d t
o
im
pl
em
ent
a
t
i
o
n o
f
el
ect
r
oni
c
equi
pm
ent
suited for electric
powe
r system
s, w
ith
ve
ry
fa
st resp
o
n
se c
o
m
p
ared
to
th
e line frequ
e
n
c
y. Th
ese activ
e co
m
p
en
sato
rs allow
groo
v
y
flex
i
b
ility
in
: I) con
t
ro
llin
g th
e
po
wer
flo
w
i
n
t
r
ansm
i
ssi
on s
y
st
em
s usi
n
g
F
l
exi
b
l
e
AC
Tra
n
sm
i
ssi
on Sy
st
em
(FAC
T
S
)
d
e
vi
ces, a
n
d I
I
)
enha
nci
n
g t
h
e
po
we
r
q
u
a
lity in
d
i
stri
b
u
tion
system
s
e
m
p
l
o
y
in
g
Custo
m
Po
wer Sy
ste
m
(CUPS)
dev
i
ces [2
]. Th
e u
s
e of th
ese activ
e
co
m
p
en
sato
rs t
o
im
p
r
ov
e in
teg
r
ation
o
f
w
i
nd
en
erg
y
in
w
e
ak
g
r
i
d
s is th
e
ap
pro
ach adopted
in
t
h
is work
.
I
n
t
h
i
s
pr
oject
we
anal
y
s
e a co
m
p
en
sat
i
on st
rat
e
gy
usi
n
g an
UP
QC
, f
o
r t
h
e
SC
IG
–base
d
W
i
n
d
Farm
connect
e
d
to
a weak
d
i
stri
b
u
tion
po
wer grid
.
Th
is
system
is taken
from
a real case [3].
Th
e
UPQC is
co
n
t
ro
lled to
reg
u
l
ate th
e
W
i
n
d
Farm
termi
n
al vo
ltag
e
, and
to
m
itig
ate h
a
rm
o
n
i
cs at
t
h
e
poi
nt
o
f
c
o
m
m
on co
upl
i
n
g
(
P
C
C
)
,
cau
sed
by
sy
st
em
l
o
a
d
c
h
an
ges
i
n
g
e
ne
rat
e
d
p
o
we
r
of
W
i
n
d
Farm
,
respectively. B
y
using UPQC
series convert
e
r in wind
far
m
vol
t
a
ge reg
u
l
at
i
on pr
ocess
was d
one
, by
v
o
l
t
a
ge
in
j
ection
“i
n
ph
ase” with
PCC
vo
ltag
e
.
Th
e shu
n
t
conv
erter is u
s
ed
to
filter th
e
Win
d
Farm
g
e
n
e
rated
p
o
wer to forb
id
v
o
ltag
e
flick
e
rs in
activ
e an
d
reactiv
e p
o
wer capab
ility. Th
e sh
aring
o
f
activ
e p
o
wer b
e
t
w
een
con
v
e
rters, is su
p
e
rv
ised
throug
h
t
h
e com
m
on D
C
l
i
nk.
2.
SYSTE
M
DESC
RIPTIO
N AN
D MO
DE
LLING
2.
1.
Sys
t
em Descri
ption
Fi
gu
re
1
de
pi
ct
s t
h
e
p
o
we
r sy
s
t
em
unde
r c
o
ns
i
d
erat
i
o
n i
n
t
h
i
s
st
u
d
y
.
Fi
gu
re
1.
Si
n
g
l
e
l
i
n
e di
a
g
ram
of
wi
nd
fa
rm
con
n
ect
ed
t
o
we
ek
gri
d
sy
st
em
The
W
i
nd Farm is co
m
posed by 36
wind
turb
in
es
u
s
ing
SC
IG, add
i
ng
up
t
o
21
.6
M
W
electric p
o
wer.
Each t
u
r
b
i
n
e
has
gi
ve
n
fi
xe
d
react
i
v
e c
o
m
p
ensat
i
on ca
paci
t
o
r
ba
n
k
s
(1
7
5
k
V
A
r
)
,
a
n
d i
s
c
o
nnect
e
d
t
o
t
h
e
po
we
r
gri
d
vi
a
63
0
K
V
A
0.
6
9
/
3
3
k
V
t
r
a
n
s
f
o
r
m
e
r. Thi
s
sy
st
em
i
s
C
a
rry
o
u
t
fr
om
, and
re
pr
esent
s
a
real
ca
se.
The
rat
i
o
bet
w
een
s
h
o
r
t
ci
r
c
ui
t
p
o
w
e
r a
n
d
rat
e
d
WI
N
D
F
A
R
M
po
wer
,
gi
ve
us
an i
d
e
a
of t
h
e
con
n
ect
i
o
n we
akne
ss. T
hus c
onsi
d
eri
ng t
h
at
t
h
e val
u
e o
f
s
h
o
r
t
ci
rcui
t
p
o
w
er i
n
M
V
6 i
s
SSC
≃
1
20M
V A thi
s
ratio ca
n
be cal
culated:
5
.
5
WF
SC
P
S
r
V
a
lu
e
s
of
r
<
19
ar
e
con
s
id
ere
d
as
a “
w
eak
gri
d
” c
o
nnection [2].
2.
2.
Turbine
Rotor and
Ass
o
ciated Dis
t
urbanc
es Model
Th
e po
wer
th
at
can
b
e
ob
tain
ed
fro
m
a
win
d
tu
rb
in
e,
is
expressed
b
y
:
p
C
V
R
P
3
2
2
1
Whe
r
e:
is air
d
e
nsity
R the ra
dius
of the s
w
e
p
t area
v t
h
e
wi
nd
s
p
e
e
d
CP the
powe
r c
o
efficient
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
PED
S
I
S
SN
:
208
8-8
6
9
4
Po
wer Qu
a
lity Imp
r
o
vemen
t Using
C
u
stom Po
wer Devices
in
S
q
u
i
rrel C
a
g
e
Indu
ctio
n
…
(Ko
p
e
lla
sa
i teja
)
47
9
Fo
r
t
h
e assu
m
e
d
tu
rb
ines (6
00kW
)
th
e
v
a
lu
es ar
e R = 3
1
.2
m ,
= 1.2
25
k
g
/
m
3
and C
P
ca
l
c
ul
at
i
o
n
is tak
e
n fro
m
[4
].
A co
m
p
lete
mo
d
e
l of th
e wi
n
d
farm
is o
b
t
ain
e
d
b
y
turb
i
n
e agg
r
eg
ation; th
is
m
ean
s th
at th
e who
l
e
wi
n
d
farm
can
be m
odel
l
e
d
b
y
onl
y
a
eq
ui
v
a
l
e
nt
wi
nd
t
u
rb
i
n
e,
wh
ose
p
o
w
er
ge
ne
rat
e
d
by
t
h
e
ari
t
h
m
e
ti
c sum
of each turbine
accordi
n
g to t
h
e e
quation
gi
ven bel
o
w:
36
.........
1
i
i
T
P
P
W
i
n
d
spee
d
v i
n
eq
n (
1
) ca
n di
ffe
r ar
o
u
n
d
i
t
s
avera
g
e val
u
e due t
o
vari
at
i
on i
n
t
h
e wi
nd
fl
ow
. Su
c
h
v
a
riation
s
can
b
e
classified
as rand
o
m
an
d
d
e
term
in
istic.
Th
e first are cau
s
ed
b
y
th
e sy
mmetry in
th
e win
d
f
l
ow
ob
serv
ed
b
y
th
e tu
rb
ine
blades
due to towe
r s
h
adow a
nd
due
to
th
e at
m
o
sp
h
e
r
i
c bou
nd
ar
y layer
,
w
h
ile
t
h
e l
a
t
t
e
r are r
a
nd
om
change
s kn
o
w
n a
s
t
u
rb
ul
ence
. F
o
r
ou
r a
n
al
y
s
i
s
, wi
n
d
fl
ow
var
i
at
i
on d
u
e t
o
s
u
p
p
o
r
t
st
ruct
u
r
e i
s
co
nsi
d
e
r
ed
, an
d m
odel
e
d by
a
si
nus
oi
dal
m
odul
at
i
o
n su
peri
m
posed t
o
t
h
e m
ean val
u
e of
v. The
fre
que
ncy
f
o
r t
h
i
s
m
odul
at
i
o
n
i
s
rotor
N
.
3
for t
h
e thre
e blade
d
wind t
u
rbine
,
while its distance
de
pe
nds
on t
h
e
geom
et
ry
of t
h
e t
o
we
r.
In
o
u
r case w
e
ha
ve co
nsi
d
e
r
e
d
a
m
ean wi
nd
spee
d of
1
2
m
/
s and t
h
e am
pl
i
t
ude
m
odul
at
i
on of
15
%.
The e
ffect
of the
bounda
ry la
yer can be i
gnore
d c
o
m
p
ared to those
produced
by
the s
h
a
d
ow e
f
fect
of
th
e tower i
n
al
m
o
st cases [3
]. It sh
ou
ld b
e
no
ted
t
h
at wh
ile th
e arit
h
m
et
ic
su
m
o
f
p
e
rt
u
r
b
a
tio
ns
o
ccurs
when
all tu
rb
in
es functio
n
syn
c
h
r
o
n
o
u
s
ly and
in
ph
ase, th
is is the case th
at h
a
s th
e g
r
eat im
p
a
ct o
n
th
e po
wer g
r
i
d
,
si
nce t
h
e
p
o
w
e
r
pul
sat
i
o
n
has
hi
g
h
am
pl
i
t
ude
. S
o
, t
u
r
b
i
n
e
ag
gre
g
at
i
o
n m
e
t
hod
i
s
val
i
d
.
The e
ffect
of the
bounda
ry la
yer can be i
gnore
d c
o
m
p
ared to those
produced
by
the s
h
a
d
ow e
f
fect
of
th
e tower i
n
al
m
o
st cases [3
]. It sh
ou
ld b
e
no
ted
t
h
at wh
ile th
e arit
h
m
et
ic
su
m
o
f
p
e
rt
u
r
b
a
tio
ns
o
ccurs
when
all tu
rb
in
es
fun
c
tio
n
sy
n
c
hono
u
s
ly and
in
ph
ase, th
is is the case th
at h
a
s th
e g
r
eat im
p
a
ct o
n
th
e power g
r
i
d
,
si
nce t
h
e
p
o
w
e
r
pul
sat
i
o
n
has
hi
g
h
am
pl
i
t
ude
. S
o
, t
u
r
b
i
n
e
ag
gre
g
at
i
o
n m
e
t
hod
i
s
val
i
d
.
3.
MODEL OF
INDUC
TION GENERATOR
Th
e m
o
d
e
l av
ailab
l
e in
Mat
l
ab
/Si
m
u
lin
k
Si
m
Po
wer Sy
ste
m
s l
i
b
r
ary th
e squ
i
rrel cag
e
in
du
ction
gene
rat
o
r i
s
us
ed.
It
c
onsi
s
t
s
of a
sec
o
n
d
–
o
r
der
m
ech
anical
m
odel and a
fourt
h
–or
de
r st
ate–space
electrical
m
odel
[5]
.
4.
D
Y
N
A
M
IC
CO
MPEN
SA
TOR MO
DE
L
The dy
nam
i
c com
p
ensat
i
on
o
f
v
o
l
t
a
ge cha
n
ges i
s
per
f
o
rmed
b
y
inj
ectin
g v
o
ltag
e
in
seri
es an
d
activ
e
& reactive power int
o
the MV
6 (PCC) busbar; this is accom
p
lished by us
ing a
n
UPQC [1]. In Figure
2 we
can see
the
bas
i
c single
line
diagram
of
t
h
is
com
p
ensator;
the im
pedances
and
bu
sb
ar
s nu
m
b
er
in
g is
r
e
f
e
rr
ed
to
Figur
e
1
.
Fi
gu
re
2(a
)
.
B
l
ock
di
a
g
ram
of
UP
QC
Fi
gu
re
2(
b
)
.
Ph
asor
di
a
g
ram
o
f
U
P
QC
The
o
p
erat
i
o
n
i
s
base
d
o
n
t
h
e
ge
nerat
i
o
n
o
f
t
h
ree
phase voltages, using powe
r
electronic
converters
either curre
nt
source type C
u
rrent
S
o
urce
In
vert
e
r
o
r
v
o
l
t
age so
urce t
y
pe V
o
l
t
a
ge S
o
urce
In
ve
rt
er.
Vol
t
a
ge
Source convert
e
rs are pre
f
e
r
red. Faster res
p
ons
e in the
syste
m
than CSI [1] and It has lowe
r DC link losses
.
Th
e sh
un
t co
nv
erter
of UPQC in
j
ectin
g current at PCC,
hear the se
ries
co
nv
erter gen
e
rates v
o
ltag
e
s between
U1
and
PCC, i
llu
strated
in
the p
h
a
sor d
i
ag
ra
m
o
f
Figu
re
3. An
im
p
o
r
tan
t
featu
r
e
of th
is co
m
p
en
sato
r i
s
th
e
o
p
e
ration
o
f
bo
th
VSI conv
erters sh
aring
th
e sam
e
DC–
b
u
s
, it en
ab
les th
e activ
e p
o
wer ex
chang
e
between
th
em
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-86
94
I
J
PED
S
Vo
l. 5
,
No
. 4
,
Ap
r
il 2
015
:
47
7
–
48
5
48
0
E
sh
u
c
(
t
)
I
sh
u
c
(t)
L
s
huc
(t)
R
sh
u
c
(
t
)
I
_gr
i
d
(
t
)
V
pc
c
(
t)
V
s
e
r
c
(t)
I
_
w
i
n
d_
fa
r
m
(t
)
U1
(t
)
Fi
gu
re
3.
P
o
we
r st
age
com
p
en
sat
i
on m
odel
A
C
si
de
Figu
re
4.
Serie
s
com
p
ensato
r
cont
roller
We ha
ve b
u
i
l
d
up a sim
u
l
a
t
i
on m
odel
fo
r t
h
e UP
QC
bas
e
d o
n
t
h
e i
d
ea
s chose
n
f
r
om
[6]
.
Si
nce
swi
t
c
hi
n
g
co
n
t
rol
of c
o
n
v
e
r
t
e
rs i
s
com
p
l
e
t
e
di
ffere
nt
of
t
h
i
s
wor
k
, a
nd c
onsi
d
eri
n
g t
h
at
hi
g
h
er
ord
e
r
h
a
r
m
o
n
i
cs
g
e
ner
a
ted
b
y
VSI
co
nv
er
ter
s
ar
e
o
u
t
si
d
e
th
e
b
a
nd
w
i
d
t
h
o
f
sign
if
ican
ce i
n
th
e si
m
u
latio
n
study, th
e
con
v
e
r
t
e
rs a
r
e
m
odel
l
e
d usi
n
g i
d
eal
co
nt
r
o
l
l
e
d v
o
l
t
a
ge
so
urces
. Fi
gu
re
4
sh
o
w
s t
h
e
ad
o
p
t
e
d m
odel
of
po
w
e
r
si
de of UP
QC
.
The c
ontrol of the UP
QC,
will be enforce
d
in a
rotating fram
e dq0
using Park’s tra
n
sform
a
tion as
gi
ve
n i
n
E
quat
i
on
(
3
&
4
)
.
2
1
2
1
2
1
)
3
2
cos(
)
3
2
cos(
)
cos(
)
3
2
sin(
)
3
2
si
n
(
)
sin(
3
2
T
c
b
a
q
d
f
f
f
T
f
f
f
.
0
Whe
r
e:
c
b
a
f
i
,
,
re
prese
n
t
s
pha
se
v
o
l
t
a
ge or
c
u
r
r
ent
s
0
,
,
q
d
f
i
re
pre
s
ents m
a
gnitudes
transform
e
d to t
h
e
dqo space.
Th
is tran
sformatio
n
ad
m
i
ts th
e alig
n
m
en
t o
f
a ro
ta
ting
referen
ce fram
e wit
h
th
e
po
sitiv
e seq
u
e
n
ce
of
the PCC
volt
a
ges space
ve
ctor. To attain this, a
re
fe
rence a
ngle
_
sy
nchronize
d
with the
PCC positive
sequence
funda
m
ental voltage space
v
ect
or is calculated
using a
Phase
Lo
c
k
ed L
o
op
(PLL) system
.
In thi
s
wo
rk
, a
n
“i
nst
a
nt
ane
o
u
s
po
we
r t
h
e
o
ry
”
base
d
PLL
has
bee
n
enf
o
rce
d
[
7
]
.
Und
e
r b
a
lan
c
e stead
y-state co
nd
itio
ns,
v
o
l
t
a
g
e
and
cu
rren
ts v
ect
o
r
s i
n
th
is syn
c
hro
nou
s referen
ce
fram
e
are con
s
tan
t
qu
an
tities.
Th
is strateg
y
is
u
s
efu
l
fo
r an
al
ysis an
d d
e
coup
led
con
t
ro
l.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
PED
S
I
S
SN
:
208
8-8
6
9
4
Po
wer Qu
a
lity Imp
r
o
vemen
t Using
C
u
stom Po
wer Devices
in
S
q
u
i
rrel C
a
g
e
Indu
ctio
n
…
(Ko
p
e
lla
sa
i teja
)
48
1
5.
UPQ
C
CO
NT
ROL
ST
R
A
TEGY
In
t
h
is p
a
p
e
r
we h
a
v
e
u
s
ed
th
e n
e
u
r
o
–
fuzzy lo
g
i
c con
t
ro
lling
strategy wh
ich is v
e
ry ad
v
a
n
c
ed
str
a
teg
y
now
a
d
a
ys.
Fi
gu
re 5.
S
h
u
n
t
com
p
ensat
o
r cont
rol
l
e
r usi
n
g neu
r
o –
f
u
zz
y
The powe
rs
shuc
P
and
shuc
Q
are calcu
lated
in th
e
ro
tatin
g
referen
c
e frame, as fo
llo
w:
)
(
.
)
(
.
2
3
)
(
)
(
.
)
(
.
2
3
)
(
t
I
t
V
t
Q
t
I
t
V
t
P
shuc
q
pcc
d
shuc
shuc
d
pcc
d
shuc
We
Igno
re
PC
C vo
ltag
e
v
a
riatio
n
,
t
h
e abo
v
e
equ
a
tio
ns can
b
e
written
as fo
llo
ws:
.
)
(
.
)
(
)
(
.
)
(
_
'
_
'
t
I
k
t
Q
t
I
k
t
P
shuc
q
q
shuc
shuc
d
p
shuc
Taki
n
g
i
n
c
o
n
s
i
d
erat
i
o
n t
h
at
t
h
e s
h
unt
co
n
v
ert
e
r
i
s
base
d
o
n
a
V
S
I
,
we
nee
d
t
o
ge
ner
a
t
e
adec
uat
e
v
o
ltag
e
s to
ob
tain
th
e cu
rren
ts in
eq
u
a
tion
.
Th
is is attain
ed
u
s
i
n
g
th
e
VSI m
o
d
e
l p
r
op
osed
in
[6
], lead
in
g
to
a
l
i
n
ear
rel
a
t
i
o
ns
hi
p bet
w
ee
n
t
h
e
co
nt
r
o
l
l
e
r vol
t
a
ges
a
n
d
ge
ne
rat
e
d p
o
we
r. T
h
e resul
t
a
nt
eq
uat
i
o
n
s
are:
)
(
.
)
(
)
(
.
)
(
*
_
'
'
_
*
'
'
t
E
k
t
Q
t
E
k
t
P
shuc
q
q
shuc
shuc
d
p
shuc
P an
d
Q
co
nt
r
o
l
l
o
op
s c
o
m
p
ri
se a
PI c
o
nt
r
o
l
l
e
r,
w
h
i
l
e
D
C
–b
us l
o
o
p
he
ar
we
use a
ne
ur
o
–
f
u
zzy
cont
rol
l
e
r,
I
n
g
e
neral
l
y
, i
n
t
h
e
pr
op
ose
d
sc
he
m
e
t
h
e UPQC
can be see
n
as
a
pow
er buf
f
e
r
, lev
e
lin
g
th
e
po
wer
in
j
ected
in
t
o
the po
wer system
g
r
id
. Th
e Fi
g
u
re
7
illu
strat
e
s a co
n
c
ep
tu
al
d
i
agram
o
f
t
h
is m
o
d
e
of
o
p
e
ratio
n
.
It
m
u
st be obs
erve
d that the
abse
nce of a
n
external
DC sou
r
ce in
th
e
UPQC bu
s, fo
rces to
m
a
in
tain
zero–a
v
era
g
e
powe
r in the st
ora
g
e elem
en
t
in
stalled
in
th
at b
u
s
.
Th
is is acco
m
p
lish
e
d
by a p
r
op
er d
e
si
g
n
o
f
DC vo
ltag
e
con
t
ro
ller.
Al
so,
i
t
i
s
nec
e
ssary
t
o
n
o
t
e
t
h
at
t
h
e
pr
op
o
s
ed sc
hem
e
cann
o
t
be i
m
pl
em
ent
e
d usi
n
g
ot
he
r C
U
PS
devi
ces l
i
k
e D
V
R
or D
–
St
at
c
o
m
.
The powe
r
bu
ffe
r co
nce
p
t
m
a
y
be im
p
l
em
ent
e
d usi
n
g
a DSt
a
t
c
om
, b
u
t
not
usi
n
g a D
V
R
.
On t
h
e ot
her
si
de, v
o
l
t
a
ge
r
e
gul
at
i
o
n d
u
r
i
ng
rel
a
t
i
v
el
y
l
a
rge
di
st
u
r
ba
n
ces, can
n
o
t
be
easi
l
y
u
s
ing
reactiv
e
p
o
wer
o
n
l
y from
DStatco
m
; i
n
th
is wo
rk
, a
DVR
d
e
v
i
ce is
m
o
re su
itab
l
e.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-86
94
I
J
PED
S
Vo
l. 5
,
No
. 4
,
Ap
r
il 2
015
:
47
7
–
48
5
48
2
Fi
gu
re 6.
P
o
we
r bu
ffe
r
c
once
p
t
Figure
7.
Active a
n
d
reactive
power
dem
a
nd at powe
r
gri
d
Th
e m
o
d
e
l
o
f
th
e power
syste
m
strateg
y
ill
u
s
trated
in
Figu
re
1
,
in
cl
u
d
i
ng
th
e con
t
ro
llers
with
the
cont
rol
sc
hem
e
det
a
i
l
e
d i
n
sect
i
on
II
I,
w
a
s i
m
pl
em
ent
e
d
usi
n
g M
a
t
l
a
b/
Si
m
u
l
i
nk
®
soft
ware
. N
u
m
e
rical
sim
u
l
a
t
i
ons w
e
re pe
rf
orm
e
d t
o
det
e
rm
i
n
e and t
h
e
n
c
o
m
p
ensat
e
vol
t
a
ge fl
uct
u
at
i
o
n d
u
e t
o
wi
n
d
p
o
w
er
vari
at
i
o
n, a
n
d
vol
t
a
ge
re
g
u
l
a
t
i
on
p
r
o
b
l
e
m
s
due
t
o
a
s
u
d
d
e
n
l
o
a
d
c
o
nn
ect
i
on.
T
h
e si
m
u
l
a
t
i
on
was c
o
n
duct
e
d
with
th
e fo
llowin
g
chrono
log
y
.
a)
At t = 0.0’’ t
h
e si
m
u
latio
n
starts with
t
h
e series
co
nv
erter an
d
t
h
e
DC-b
us v
o
ltag
e
con
t
rollers in
ope
rat
i
o
n.
b)
At t =
0
.
5
’
’ th
e to
wer sh
ad
ow
effect starts
c)
At
t
=
3.
0”
Q a
n
d
P c
o
nt
r
o
l
l
o
ops
d)
At t =
6.0” L
3
load is
connected.
e)
At t =
6.0 “
L3 load is
disconnected[8].
6.
COMPENSATION OF VO
LTAGE FL
UCTUATION
Si
m
u
latio
n
resu
lts fo
r 0 < t < 6 are shown in
Fi
g
.
8
.
At t = 0
.
5
′′
be
gi
n
s
t
h
e cy
cl
i
cal
po
w
e
r
pul
sat
i
o
n
pr
o
duce
d
by
t
h
e t
o
wer s
h
ad
o
w
effect
.
As w
a
s
m
e
nt
i
oned
,
t
h
e t
o
we
r sha
d
o
w
pr
o
duc
es va
r
i
at
i
on i
n
t
o
r
q
u
e
, and
h
e
n
ce i
n
th
e activ
e an
d
reactiv
e wind
farm
g
e
n
e
rated
p
o
wer. Fo
r
n
o
m
in
al win
d
sp
eed
co
nd
itio
n, th
e po
wer
fl
uct
u
at
i
o
n f
r
e
que
ncy
i
s
f = 3.4
H
z, an
d t
h
e a
m
pli
t
ude of
t
h
e
resul
t
i
ng
vol
t
a
ge vari
at
i
o
n at
PC
C
,
exp
r
esse
d as a
p
e
rcen
tag
e
is:
%
50
.
1
ra
te
d
U
U
7.
R
E
SU
LTS AN
D ANA
LY
SIS
Fi
gu
re 8 i
s
t
h
e
pcc vol
t
a
ge i
s
beha
vi
o
u
r i
s
s
h
o
w
n t
h
e u
ppe
r cur
v
e s
h
o
w
s
t
h
e vol
t
a
ge at
pcc w
h
en
UPQC is not
existing. The
middle cu
rve
shows t
h
e when the
UPQC
is co
n
n
ect
ed
t
o
t
h
e gri
d
by
usi
n
g
P
I
cont
roller. T
h
e
last curve in Figu
re 8 sh
o
w
s whe
n
U
P
QC
is connected to the gri
d
and he
ar we are usi
n
g ne
uro
–fuzzy logic c
ont
roller. T
h
ere is
a v
a
riation
in
th
e wav
e
form
s.
0
1
2
3
4
5
6
-3
-2
-1
0
1
x 1
0
7
Pg
ri
d
[W
]
Qgr
i
d [
VAr
]
A
c
t
i
v
e
a
nd r
e
a
c
t
i
v
e
po
w
e
r
de
m
a
nd a
t
pow
e
r
g
r
i
d
s
i
de
w
i
t
h
out
U
P
Q
C
0
1
2
3
4
5
6
-3
-2
-1
0
1
x 1
0
7
A
c
t
i
v
e
a
nd r
e
a
c
ti
v
e
pow
e
r
de
m
a
nd a
t
pow
e
r
g
r
i
d
s
i
de
w
i
th
U
P
Q
C
Pg
ri
d
[
W
]
Q
g
ri
d
[V
A
r
]
0
1
2
3
4
5
6
-2
-1
0
1
x 1
0
7
A
c
t
i
v
e
a
nd r
e
a
c
ti
v
e
pow
e
r
de
m
a
nd a
t
pow
e
r
g
r
i
d
s
i
de
w
i
th
U
P
Q
C
us
i
n
g
ne
u
r
o f
u
z
z
y
P
g
ri
d
[W
]
Qgr
i
d [
VAr
]
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
PED
S
I
S
SN
:
208
8-8
6
9
4
Po
wer Qu
a
lity Imp
r
o
vemen
t Using
C
u
stom Po
wer Devices
in
S
q
u
i
rrel C
a
g
e
Indu
ctio
n
…
(Ko
p
e
lla
sa
i teja
)
48
3
Fi
gu
re 8.
Pcc v
o
l
t
a
ge
In t
h
e
Fi
g
u
re
9
t
h
e sh
ows t
h
e
beha
vi
o
u
r
o
f
w
i
nd fa
rm
t
e
r
m
inal
v
o
l
t
a
ge. T
h
e up
per
wa
ve f
o
rm
sho
w
s
t
h
e beh
a
vi
ou
r whe
n
wi
nd
far
m
i
s
connect
ed
t
o
t
h
e gri
d
when UPQC is not connecte
d
.
Middle wa
ve form
is
whe
n
UPQC
i
s
con
n
ect
ed t
o
t
h
e g
r
i
d
a
nd t
h
e
cont
r
o
l
st
rat
e
g
y
used i
s
PI c
o
nt
r
o
l
l
e
r. Fi
nal
wave
fo
rm
of Fi
gu
re
9 s
h
ows
t
h
e
b
e
havi
ou
r
o
f
wi
nd
fa
rm
vol
t
a
g
e
co
nn
ect
ed t
o
t
h
e
gri
d
a
n
d t
h
e c
ont
rol
st
ra
t
e
gy
use
d
i
s
n
e
ur
o
–
fuzzy l
ogic c
o
ntroller.
Figu
re 9.
WF
t
e
rm
inal
voltag
e
Th
is
v
o
ltage fl
u
c
tu
ation
is seen
in Fi
g
u
re
9
for
0
.
5
< t < 3.
Tabl
e
1.
Ne
ur
o
Fuzzy
R
u
l
e
B
a
se
E(K
)
E
NB NM
NS
ZE
PS
PM
PB
NB NB
NB
NB
NB
NM
NS
ZE
NM
NB
NB
NB
NM
NS
ZE
PS
NS NB
NB
NM
NS
ZE
PS
PM
ZE
NB
NM
NS
ZE
PS
PM
PB
PS NM
NS
ZE
PS
PM
PB
PB
PM
NS
ZE
PS
PM
PB
PB
PB
PB ZE
PS
PM
PB
PB
PB
PB
In th
is
p
a
p
e
r t
h
e abo
v
e
ru
les are tak
e
n
fo
r
[9
]. In
t
h
e ab
ov
e t
a
b
l
e PB=p
o
s
iti
v
e
b
i
g
,
PM = po
sitiv
e m
e
d
i
u
m
, PS
= po
sitiv
e sm
al
l ZE = zero,
NS =
n
e
g
a
tiv
e small, NM = n
e
g
a
tiv
e m
e
d
i
u
m
, NB =
n
e
g
a
tive b
i
g
In t
h
e Fi
gu
re 1
0
t
h
e sh
o
w
s t
h
e beha
vi
o
u
r
of
wi
n
d
farm
t
e
rm
i
n
al
vol
t
a
ge and Pcc
v
o
l
t
a
g
e
. The u
p
p
er
wave
f
o
rm
sh
ows
t
h
e
be
ha
v
i
ou
r w
h
e
n
wi
n
d
f
a
rm
i
s
connected t
o
the
gri
d
whe
n
UPQC
is not
c
onnected.
0
5
10
15
-1
0
1
Vp
c
c
[
V
]
p
c
c
v
o
l
t
ag
e
w
i
t
h
o
u
t
U
P
Q
C
0
1
2
3
4
5
6
2.
6
2.
65
2.
7
x 1
0
4
P
c
c
vo
l
t
a
ge
w
i
t
h
U
P
Q
C
Vp
c
c
[
V
]
0
1
2
3
4
5
6
2.
64
2.
66
2.
68
x 1
0
4
P
C
C
v
o
l
t
a
g
e w
i
t
h
U
P
Q
C
u
s
i
n
g
n
u
ro
-
f
u
zzy
Tim
e
Vp
c
c
[
V
]
0
1
2
3
4
5
6
-1
-0
.
5
0
0.
5
1
V
p
cc[
v
]
W
F
te
r
m
i
n
a
l
v
o
lt
a
g
e
0
1
2
3
4
5
6
2.
6
5
5
2.
66
2.
6
6
5
x 1
0
4
W
F
te
r
m
i
n
a
l
v
o
lt
a
g
e
U
P
Q
C
V
p
cc[
v
]
0
1
2
3
4
5
6
2.
6
5
5
2.
66
2.
6
6
5
x 1
0
4
W
F
t
e
r
m
i
n
a
l
vo
l
t
a
ge
U
P
Q
C
u
s
i
n
g n
e
r
o
-
f
u
z
z
y
Ti
m
e
V
p
cc[
v]
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-86
94
I
J
PED
S
Vo
l. 5
,
No
. 4
,
Ap
r
il 2
015
:
47
7
–
48
5
48
4
m
i
ddl
e wave
f
o
rm
i
s
when
U
P
QC
i
s
co
nnec
t
ed t
o
t
h
e g
r
i
d
and t
h
e co
nt
r
o
l
st
rat
e
gy
use
d
i
s
PI co
nt
r
o
l
l
e
r.
Fi
nal
wave
f
o
rm
of
Fi
gu
re
10
sh
o
w
s t
h
e
be
ha
vi
ou
r
of
wi
n
d
fa
rm
vol
t
a
ge co
nnect
e
d
t
o
t
h
e
gri
d
a
nd t
h
e
cont
rol
st
rat
e
gy
u
s
ed
i
s
ne
ur
o
–
f
u
zzy
l
ogi
c c
o
nt
r
o
l
l
e
r.
Fi
gu
re
1
0
.
V
o
l
t
a
ge at
Pcc
a
n
d
WF
Figu
re 1
1
. Po
w
e
r of
ca
pacito
r in
DC
b
u
s
The ab
o
v
e Fi
gu
re 1
1
i
s
D
C
bus
vol
t
a
ge
at
UPQC
, i
n
up
per
wa
ve
fo
rm
t
h
ere i
s
no
UP
QC
i
s
co
nn
ected to
t
h
e grid. Th
ere is n
o
UPQC
n
o
DC
b
u
s
vo
ltag
e
so
it is 0. Th
e m
i
d
d
l
e wav
e
fo
rm
is UPQC is
con
n
ect
ed t
o
t
h
e g
r
i
d
t
h
ere i
s
vari
at
i
o
n.
We
have t
a
ken
va
r
i
at
i
on fr
om
t
i
m
e
i
n
t
e
rval
3,
so t
h
e
vari
at
i
o
n st
art
s
fr
om
3. T
h
e fi
n
a
l
wave
f
o
rm
i
s
al
so s
a
m
e
hear
we
use
neu
r
o
fuzzy
.
Fi
gu
re 1
2
. V
o
l
t
a
ge of
t
h
e
ca
pa
ci
t
o
r
i
n
t
h
e dc bus
0
1
2
3
4
5
6
7
8
-1
-0
.
5
0
0.
5
1
V[
v
]
v
o
l
t
a
g
e
at
W
F
an
d
PCC
U
P
Q
C
0
1
2
3
4
5
6
7
8
2.
5
2.
5
5
2.
6
2.
6
5
2.
7
x 1
0
4
v
o
l
t
a
g
e
at
W
F
an
d
PCC
U
P
Q
C
V[
v
]
0
1
2
3
4
5
6
7
8
2.
5
2.
6
2.
7
x 1
0
4
v
o
lta
g
e
a
t
W
F
a
nd P
C
C
U
P
Q
C
us
i
n
g
ne
r
o
-
f
uz
z
y
Tim
e
V[
v
]
0
1
2
3
4
5
6
-1
0
1
Pd
c
[
W
]
P
o
w
e
r
of
t
h
e
c
a
pa
s
i
to
r
i
n
th
e
D
C
-
b
us
w
i
th
out
U
P
Q
C
0
1
2
3
4
5
6
x 1
0
5
-5
0
5
x 1
0
6
P
o
we
r
o
f
t
h
e
c
a
p
a
s
i
t
o
r
i
n
t
h
e
DC
-
b
u
s
wi
t
h
UP
QC
Pd
c
[
W
]
0
1
2
3
4
5
6
x 1
0
5
-2
0
2
x 1
0
6
P
o
w
e
r
o
f
t
h
e cap
as
i
t
o
r
i
n
t
h
e D
C
-
b
u
s
w
i
t
h
U
P
Q
C
n
e
u
r
o
f
u
zz
y
TI
M
E
(
s
e
c
)
Pd
c
[
W
]
0
1
2
3
4
5
6
-1
0
1
Vd
c[
V
]
v
o
l
t
a
g
e
o
f
t
h
e
ca
p
aci
t
o
r i
n
t
h
e
D
C
b
u
s
w
i
t
h
o
u
t
U
P
Q
C
0
1
2
3
4
5
6
x 1
0
5
3
800
4
000
4
200
v
o
lta
g
e
of
the
c
a
pa
c
i
tor
i
n
the
D
C
bu
s
w
i
th
U
P
Q
C
Vd
c[V]
0
1
2
3
4
5
6
x1
0
5
3
800
4
000
4
200
v
o
lta
g
e
of
the
c
a
pa
c
i
tor
i
n
the
D
C
bus
w
i
th
U
P
Q
C
u
s
i
n
g
n
e
ur
o f
u
z
z
y
Tim
e
(
s
e
c
)
Vd
c[
V
]
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
PED
S
I
S
SN
:
208
8-8
6
9
4
Po
wer Qu
a
lity Imp
r
o
vemen
t Using
C
u
stom Po
wer Devices
in
S
q
u
i
rrel C
a
g
e
Indu
ctio
n
…
(Ko
p
e
lla
sa
i teja
)
48
5
Fig
u
r
e
13
.
Shun
t and
ser
i
es co
nv
er
ter
activ
e po
w
e
r
8.
CO
NCL
USI
O
N
In t
h
i
s
pa
per
,
a new com
p
en
sat
i
on st
rat
e
gy
was em
pl
oy
ed usi
n
g U
P
QC
com
p
ensat
o
r.
Whe
n
SC
I
G
base
d
wi
n
d
fa
rm
s conn
ect
ed
t
o
wee
k
gri
d
t
h
i
s
com
p
ensat
i
on
st
rat
e
gy
i
s
use
d
.
T
h
i
s
c
o
m
p
ensat
i
on st
r
a
t
e
gy
en
h
a
n
ces system
p
o
w
er qu
ality. Th
e sim
u
latio
n
resu
lts sho
w
th
e
go
od perfo
r
m
a
n
ce in
mit
i
g
a
tin
g
t
h
e
p
o
wer
fl
uct
u
at
i
o
ns
d
u
e
t
o
t
o
we
r s
h
ad
ow
ef
fect
an
d
vol
t
a
ge
re
g
u
l
a
t
i
on i
n
s
u
dde
n l
o
ad
co
n
d
i
t
i
ons
.
REFERE
NC
ES
[1]
A Ghosh,
G Ledwich.
Power Quality
Enh
a
nc
em
ent Us
ing Cus
t
om
P
o
wer Devi
ces
.
Kluwer Acad
em
ic P
ublis
h
e
r.
[2]
Z Saad-Saoud,
ML Lisboa, JB Ekan
ay
ake, N Jenkins and G Str
b
ac.
App
lica
tion
of STATCOM’s
to wind farms.
IEE
Proc. Gen
.
Trans
.
Distrib
.
1998
; 1
45(5).
[3]
P Rosas. D
y
nam
i
c inf
l
uen
ces of
wind power on the power s
y
s
t
em.
Technica
l report RISØR-140
8
.
Ørste
d
Institute
.
March 2003
.
[4]
T Burton, D Sh
arpe, N Jenk
ins, E Bossan
y
i. Wind En
erg
y
Han
dbook. John Wiley
& Sons, 20
01. ISBN 0-471-
48997-2.
[5]
P Kundur. Power S
y
st
em
Stabil
it
y
and Con
t
rol
.
McGraw-Hill, 1
994. ISBN 0-07-
035958-X
[6]
C Schauder, H Mehta.
Vector analysis and control of adv
anced
static VAR comp
ensators.
IEE P
R
OCEEDINGS
-C.
1993; 140(4).
[7]
MF Farias, PE Battaiotto
, MG Cendo
y
a
. Wind Far
m
to
Weak-Grid Connection using
UPQC Cu
stom Power Device,
IEEE
.
2010
[8]
M Vishnu vardhan, Dr P Sang
ameswa
raraju
.Using NFC and
modified CS
Algorithm based u
n
ified
power flo
w
condition
e
r for com
p
ensating p
o
wer qualit
y pro
b
lem
.
Internatio
nal Journal of Scien
tifi
c
&
Engi
neering Researc
h
.
2013; 4(9).
0
5
10
15
-5
0
5
x 1
0
6
s
hunt
a
nd s
e
r
i
e
s
c
onv
e
r
te
r
a
c
ti
v
e
-
pow
e
r
P[
W
]
0
5
10
15
-5
0
5
x 1
0
6
s
hun
t a
nd s
e
r
i
e
s
c
onv
e
r
te
r
a
c
t
i
v
e
-
pow
e
r
U
P
Q
C
P[
W
]
0
5
10
15
-5
0
5
x 1
0
6
s
hunt
a
n
d
s
e
r
i
e
s
c
onv
e
r
te
r
a
c
t
i
v
e
-
p
o
w
e
r
U
P
Q
C
us
i
n
g
ne
r
o
-
f
u
z
z
y
Ti
m
e
P[
W
]
Evaluation Warning : The document was created with Spire.PDF for Python.