TELKOM
NIKA Indonesia
n
Journal of
Electrical En
gineering
Vol. 12, No. 8, August 201
4, pp. 5751 ~ 5757
DOI: 10.115
9
1
/telkomni
ka.
v
12i8.627
3
5751
Re
cei
v
ed Ma
rch 3, 2
014;
Re
vised Ma
y 7, 2014; Acce
pted May 2
5
, 2014
Growing Neural Gas Based MPPT for Wind Generator
using DFIG
J. Priy
adarshini*, J. Karthika
Sri Krishn
a Col
l
eg
e of Engi
ne
erin
g and T
e
ch
nol
og
y, Kun
i
a
m
uthur, Coim
b
a
tore - 641
00
8
*Corres
p
o
ndi
n
g
author, e-ma
i
l
: darshu
3
8
9
1
@
gmai
l.com
A
b
st
r
a
ct
T
h
is pa
per pr
e
s
ents G
r
ow
ing
Neur
al G
a
s (G
NG
) base
d
a
max
i
mu
m p
o
w
e
r poi
nt trackin
g
(MPPT
)
techni
qu
e for
a hi
gh
perfor
m
ance w
i
n
d
g
e
n
e
rator us
ing
D
F
IG
.
It is used
in var
i
ab
le s
p
eed w
i
n
d
e
ner
gy
convers
i
on sy
stem. Here, tw
o back to
back conv
er
te
rs is used a
nd con
necte
d
to the stato
r
,
corresp
ond
in
gl
y F
O
C and VO
C is d
one
on
mac
h
i
ne a
nd s
upp
ly sid
e
co
n
v
erter. Consta
nt voltag
e ov
er
the
grid
is
obtai
ne
d thro
ug
h dc
li
nk vo
ltage.
F
o
r
Vari
abl
e s
pee
d w
i
nd
e
nergy
convers
i
on
sys
tem t
he
maxi
mu
m
pow
er poi
nt trackin
g
(MPPT
)
is a ve
ry imp
o
rtant requ
ire
m
e
n
t in ord
e
r to maxi
mi
z
e
th
e efficiency. H
e
r
e
Neur
al Netw
or
k has bee
n train
ed to lear
n
the turb
ine c
haracter
i
stic i.e torque vers
u
s
w
i
nd speed
and
mac
h
i
ne s
pee
d
.
It has be
en i
m
p
l
e
m
e
n
ted
to
obtai
n
max
i
mum pow
er
poi
n
t
tracking for v
a
ryin
g w
i
nd s
p
eed.
And fina
lly co
mparis
on has
be
en mad
e
w
i
th and w
i
thout gro
w
ing neur
al g
a
s
.
Ke
y
w
ords
:
MPPT, DFIG,
FOC, VOC,
GNG
Copy
right
©
2014 In
stitu
t
e o
f
Ad
van
ced
En
g
i
n
eerin
g and
Scien
ce. All
rig
h
ts reser
ve
d
.
1. Introduc
tion
Due to
ene
rg
y cri
s
is, u
s
e
of ren
e
wable
ene
rg
y for
el
ectri
c
ity gen
e
r
ation
ha
s in
crea
sed.
Ren
e
wable e
nergi
es have low
e
n
viron
m
ental
impa
ct
and thu
s
it is
the be
st sol
u
tion for
gro
w
in
g
deman
d. And the Wind
energy meet
s the increa
sed d
e
man
d
that could
not be offere
d by
conve
n
tional
method
s el
ectricity gen
erat
ion [1, 2].
Re
cent te
chn
o
lo
gy in wi
nd e
n
e
rgy
conve
r
si
on
system h
a
s le
d to cost redu
ction compa
r
ed to
that of non ren
e
wable
energy sou
r
ces.
In ca
se of g
r
i
d
co
nne
cted
system
s, inte
rf
ace i
s
req
u
ired to e
n
sure
good p
o
wer
quality.
The inte
rface
may con
s
i
s
t o
f
a power el
e
c
troni
c
conv
erter, trans
f
ormer, filter, etc [1]. They play a
very importan
t
role in mode
rn win
d
ene
rg
y conversion
system (WE
C
S).
WT
s are
cla
s
sified into tw
o main c
a
tegories
[3]:
a)
Fixed spe
ed
WT
s;
b)
Variabl
e sp
ee
d WT
s.
Fixed speed
WT
s a
r
e e
quipp
ed
with
indu
ct
ion g
enerator
(sq
u
irrel cage i
ndu
ction
gene
rato
r SCIG or woun
d
rotor ind
u
ctio
n ge
nerator
WRIG
)
dire
ctl
y
con
n
e
c
ted t
o
the
gri
d
a
n
d
a
cap
a
cito
r b
a
n
k
for re
active
po
wer comp
ensation. In I
ndu
ction
Gen
e
rato
r, syn
c
h
r
ono
u
s spee
d
is
fixed; this implies that WT
s ca
n obtain
maximu
m efficien
cy at one
wind sp
eed
only. As power
electroni
cs a
r
e not
used i
n
this
co
nfigu
r
ation, it is not
po
ssi
ble to
control
neithe
r
re
active
po
wer
con
s
um
ption
nor po
we
r qu
ality.
Figure 1. Block
Diag
ram o
f
WECS usin
g DFIG
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 8, August 2014: 575
1 –
5757
5752
Variabl
e spe
ed
WTs a
r
e
equi
pped
with a
n
ind
u
ction
or sy
nch
r
on
ou
s g
enerator
c
o
nn
ec
te
d
to
th
e
gr
id
thr
o
ug
h
a p
o
w
e
r
co
n
v
e
r
ter
.
T
he va
r
i
ab
le
s
pee
d
op
er
a
t
io
n
,
ma
de
po
ss
ible
by means of
power ele
c
tro
n
ics, allows WT
s to wo
rk
at the maximum conve
r
sio
n
efficien
cy over
a wid
e
rang
e
of win
d
spe
eds.
No
wad
a
y
s dou
bly fed
indu
ction
ge
nerato
r
s a
r
e
use
d
in
ca
se
of
variable
sp
e
ed win
d
en
ergy c
onve
r
si
o
n
system. Fi
gure
1 sh
ow
s the ba
si
c
block dia
g
ra
m o
f
WECS u
s
ing
DFIG.
Here the wi
n
d
turbin
e is d
i
rectly conn
e
c
ted
to do
ubl
y fed inductio
n gene
rato
r. Initially it
works as a m
o
tor, dra
w
in
g power from th
e grid i.
e 30%
powe
r
is dra
w
n from the g
r
id [4]. When
it
rea
c
he
s the
synchrono
us
speed it wo
rks as the gen
erator, su
pplyin
g
power to th
e grid i.e 10
0%
power is give
n to grid (30
%
is drawn and 70%
is
gene
rated
)
. The stato
r
of this gene
rat
o
r is
dire
ctly co
nn
ected to
the
grid. T
w
o
ba
ck to ba
ck converte
rs are
use
d
. Powe
r converte
rs
are
use
d
to
conv
ert eithe
r
a
c
t
o
d
c
o
r
d
c
to
ac. And
alo
n
g
with t
h
is
m
ppt techniqu
e
is
also
u
s
ed
[5].
MPPT are u
s
ed to extract
the maximum
availabl
e wi
nd po
wer. Do
ubly fed indu
ction ge
nerator
has vari
ou
s a
d
vantage
s. They are [6]:
a)
Red
u
ced cost
of the invert
er filters an
d EMI filters.
b)
Can o
perate in both su
per
sync
hro
nou
s
and supe
r sy
nch
r
on
ou
s m
ode.
c
)
Ba
c
k
to
ba
ck
c
o
n
v
er
te
r
us
ed
.
d)
At lower cost,
Power-facto
r
control
can b
e
impleme
n
te
d.
e)
Dynami
c
pe
rforma
nce and
controllability are go
od.
f)
Filters a
r
e u
s
ed to redu
ce
harm
oni
cs d
u
e
to converte
rs.
g)
System efficiency is imp
r
oved. And by
using IGB
T
inverters, approximatel
y 2-3%
efficien
cy improveme
n
t can
be obtaine
d.
2. Proposed
Metho
d
Figure 2 sho
w
s the blo
c
k diagram
of the prop
osed m
e
thod. He
re DFIG uses two back to
back conve
r
ters
co
nne
cte
d
to the rotor which is
the
n
con
n
e
c
ted
to the grid. And the stato
r
is
dire
ctly conn
ected
to th
e
grid. Fi
eld
ori
ented
co
ntrol
and
maximu
m po
we
r tracking
is do
ne
on
gene
rato
r si
d
e
converte
r.
Gro
w
ing
Neu
r
al G
a
s ba
se
d Maximum
Powe
rPoint T
r
ackin
g
i
s
u
s
ed to
track the
max
i
mum availa
b
l
e po
wer from
the win
d
. Vol
t
age o
r
iented
cont
rol i
s
do
ne on
gri
d
si
d
e
conve
r
ter.
Figure 2. Block
Diag
ram o
f
Proposed M
e
thod
3. DFIG Modeling
The gen
eral model for
wo
und roto
r ind
u
ction ma
chi
n
e is given a
s
follows [7]:
3.1. Voltage
Equ
a
tions
Stator Voltage Equation
s
:
V
qs
= p
λ
qs
+
λ
ds
ω
+ r
s
i
qs
(1)
V
ds
= p
λ
ds
–
λ
qs
ω
+ r
s
i
ds
(2)
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Gro
w
ing
Neu
r
al Ga
s Base
d MPPT for Wind G
ene
ra
tor usin
g DFI
G
(J. Pri
y
ad
a
r
shi
n
i)
5753
Rotor Volta
g
e
Equations:
V
qr
= p
λ
qr
+ (
ω
–
ω
)
λ
dr
+ r
r
i
qr
(3)
V
dr
= p
λ
dr
- (
ω
–
ω
)
λ
qr
+ r
r
i
dr
(4)
3.2. Po
w
e
r
Equa
tions
P
s
= 3/2(V
ds
i
ds
+ V
qs
i
qs
)
(5)
Q
s
= 3/2(V
qs
i
ds
– V
ds
i
qs
)
(6)
3.3. Torque
Equa
tion
T
e
= - (3/2
)(p/
2
)(
λ
ds
i
qs
–
λ
qs
i
ds
)
(7)
3.4.
Stator Flu
x
Linkage Equa
tions
λ
qs
= (L
ls
+ L
m
) i
qs
+ L
m
i
qr
(8)
λ
ds
= (L
ls
+ L
m
) i
ds
+ L
m
i
dr
(9)
3.5.
Rotor Flux L
i
nkage Equa
tions
λ
qr
= (L
lr
+ L
m
)i
qr
+ L
m
i
qs
(10)
λ
dr
= (L
ls
+ L
m
)i
dr
+ L
m
i
ds
(11)
4.
Contr
o
l Method
s
4.1.
FOC on G
e
n
e
rato
r Side Conv
erter
Field O
r
ie
nte
d
Control
te
chni
que
is a
dopted
control techniq
u
e
be
cau
s
e
of
its hi
gh
perfo
rman
ce
and also co
n
t
rols the a
c
tive and rea
c
tive powe
r
flows. Figu
re 3
shows the foc
control on ge
nerato
r
sid
e
conve
r
ter. He
re stat
o
r
pha
se currents a
r
e mea
s
u
r
ed
and is converted
to (d,q) syste
m
. Flux linkage cont
rol is
obtaine
d thro
ugh d-axis co
mpone
nt and
similarly sp
e
ed
control is obt
ained throug
h
q axis cont
ro
l. He
re, DFIG
works in the
dq refe
ren
c
e
frame.
Aligning the d
–axis of refe
rence fram
e to
be along the
stator flux linkage.
λ
qs
=
0
(12)
And hen
ce from (8):
i
qs
= -[
L
m
/(L
ls
+ L
m
)] i
qr
(13)
For
λ
ds
to re
main un
cha
n
ged at ze
ro, p
λ
ds
must be
zero. Substit
u
ting for p
λ
ds
using (1)
and (2
) will re
sult in:
V
ds
= r
s
i
ds
(14
)
Negl
ectin
g
st
ator
re
sista
n
ce will l
ead
to
Vds
= 0
an
d
sub
s
tituting i
n
(5
)
and
(6
)
will be
si
mplif
ied
as
follows
:
P
s
= 3/2(V
qs
i
qs
)
(15)
Q
s
= 3/2(V
qs
i
ds
)
(16)
The above e
quation
s
sho
w
that active and re
ac
tive
powers of the st
ator a
r
e controlle
d
indep
ende
ntly. PI controllers
alon
g wit
h
NN
ar
e used to co
ntrol
these
cu
rre
nts. The p
o
w
e
r
corre
s
p
ond
s t
o
the refere
n
c
e
spe
ed of t
he ma
chin
e i
s
obtai
ned th
rough th
e G
N
G ba
sed
MPPT
.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 8, August 2014: 575
1 –
5757
5754
Actual an
d re
feren
c
e
cu
rre
nts a
r
e
comp
ared
and
pr
o
duces th
e e
r
ror
signal
s a
n
d
it is ag
ain gi
ven
to PI controll
er. It produ
ce
s the dq volta
ge whi
c
h i
s
a
gain conve
r
te
d into
αβ
an
d fe
d
in
to
s
p
ac
e
vector Pul
s
e
Width Mo
dul
ation (PWM).
Output fr
om
this is given
as pul
se
s to machine
si
de
c
onverter [8].
Figure 3. FO
C Te
chniq
ue
on Gen
e
rato
r Side Conve
r
ter
4.2.
VOC on Grid
Side Conv
er
ter
Voltage ori
e
n
t
ed co
ntrol i
s
done o
n
g
r
id
side
conve
r
te
r. The
con
s
ta
nt voltage is
use
d
to
inject
po
wer to the
grid
fro
m
the m
a
chin
e. If voltage i
s
redu
ce
d in
the g
r
id, the
d
c
lin
k volta
g
e
is
given to the grid to maintain the co
nstant
voltag
e [9]. Here the three p
h
a
se
curre
n
ts are
measured an
d are
conve
r
ted in to dq co
-ordinate
syst
em. The refe
rence dc lin
k
voltage are
set,
the actu
al vol
t
age is m
e
a
s
ured
and
co
mpared. The
voltage eq
uat
ions i
n
syn
c
h
r
ono
usly
rotat
i
ng
dq-axi
s
refe
re
nce fra
m
e are [7]:
V
ds
= V
d1
– RI
ds
– L(d/dt)I
ds
+ L
ω
e
I
sq
(17)
V
qs
= V
q1
– RI
qs
– L(d/dt)I
qs
– L
ω
e
I
ds
(18)
Figure 4. VOC Te
chniq
ue
on Grid Sid
e
Conve
r
ter
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Gro
w
ing
Neu
r
al Ga
s Base
d MPPT for Wind G
ene
ra
tor usin
g DFI
G
(J. Pri
y
ad
a
r
shi
n
i)
5755
The d
-
axi
s
of
the referen
c
e fram
e i
s
ali
gned
with
th
e gri
d
voltag
e an
gula
r
p
o
s
ition
θ
e
.
Since the am
plitude of the grid voltage i
s
co
nsta
nt, V
qs
is zero a
nd
V
ds
is
c
ons
tant.
V =
V
ds
+
0
(19)
The active and reactive powe
r will
be proportional
t
o
i
Nd
and
i
Nq
resp
ectively. The g
r
id-
side tra
n
sfo
r
mer conn
ecti
on is sta
r
, the
conv
erte
r a
c
tive and rea
c
ti
ve powe
r
flow is:
P
s
= 3/2V
ds
I
ds
(20)
Q
s
= 3/2V
ds
I
qs
(21
)
The
cont
rol
schem
e h
a
s sl
ightly been
modified
by
addin
g
an
oth
e
r
cont
rol lo
o
p
for th
e
dc-li
n
k volta
g
e
, which out
puts the di
re
ct refe
ren
c
e
curre
n
t In voltage ori
ented
current
cont
rol
depe
nd
s on the dc lin
k voltage where
a
s in FO
C current co
ntrol
depen
ds on
the generator
referen
c
e sp
e
ed. The com
p
ared
sign
als
are given to
PI controlle
rs
and it is used
to generate the
approp
riate signal
s. Again
the dq
cu
rre
n
ts are co
nverted to
αβ
, it is given as
the input to the
PWM. Outp
ut from the P
W
M i
s
given
as the
pul
se
s to the gri
d
side converte
r [10]. In voltage
oriente
d
cu
rrent cont
rol de
pend
s on the
dc lin
k vo
ltag
e whe
r
ea
s in
FOC
curre
n
t control dep
en
ds
on the gen
erator refe
ren
c
e spe
ed.
5. Gro
w
i
ng
Neur
al Gas
Bas
e
d MPPT
In orde
r to in
crea
se the
out
put it is ne
ce
ssar
y to o
p
e
r
at
e the sy
stem
at the optimal
point.
So we
go fo
r Gro
w
in
g Ne
ural
Ga
s ba
sed MPPT. T
here
are two
main
cont
rol
s
in t
r
a
cki
ng
the
maximum p
o
w
er from
the
wind.
First
approa
ch
i
s
based o
n
torque a
nd
se
cond a
p
p
r
oa
ch is
based on the
spee
d of the machin
e. He
re the s
pee
d control metho
d
is ado
pted
and the ne
ural
netwo
rk ha
s
to be
spe
c
ifi
c
ally trai
ned
for a
parti
cu
l
a
r
cha
r
a
c
teri
stic
on
whi
c
h
it will b
e
u
s
ed.
Neu
r
al
netwo
rk
have th
re
e layers: inp
u
t, hidden,
a
nd outp
u
t lay
e
rs.
The
nu
mber of no
d
e
s in
each laye
r va
ries a
nd i
s
u
s
er-d
epe
nde
nt. The
G
N
G
network
ha
s bee
n traine
d
onlin
e a
n
d
then
use
d
offline. Duri
ng trai
nin
g
pha
se the i
nput to
this n
eural
network is win
d
spee
d and g
ene
ra
tor
spe
ed o
r
torq
ue as
sh
own
in Figure 5. And t
he out
put is the referen
c
e
gene
rator spee
d which
make
s the
wi
nd tu
rbine
to
ope
rate
at,
or
clo
s
e
to, the MPP [1
1], dep
endi
ng
u
pon
alg
o
rith
ms
use
d
by the
h
i
dden
layer a
nd the
traini
n
g
ph
ase of th
e ne
ural
net
work,
the
ope
rating p
o
int ge
ts
to the MPP.
The n
e
u
r
al
n
e
twork
ha
s t
o
be
pe
riodi
cally traine
d t
o
get th
e ex
act MPPT
as the
cha
r
a
c
teri
stics of a
wind t
u
rbin
e al
so
chang
e with ti
me. And no
w the inversio
n of this tu
rbi
ne
function
ha
s been
imple
m
ented
onlin
e by mea
s
u
r
ing th
e wi
n
d
free
sp
ee
d ba
sed
on
the
estimated to
rque an
d mea
s
ured ma
chi
n
e spe
ed [12].
Figure 5. Block
Diag
ram o
f
the GNG-ba
sed MPPT Al
gorithm
6. Results a
nd Analy
s
is
The voltage i
s
maintain
ed
con
s
tant over t
he grid say 400V. Figure 6 sho
w
s the voltage
over the
gri
d
. The a
c
tive
P and
rea
c
ti
ve Q po
we
r f
l
owin
g into t
he po
we
r g
r
i
d
are me
asu
r
ed
.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 8, August 2014: 575
1 –
5757
5756
They sho
w
th
at the rea
c
tive power Q is
neari
ng
to ze
ro and re
al po
wer i
s
abo
ut 25 KW. Figu
re 7
sho
w
s the compa
r
ison of
reference g
enerator
with
and with
out
GNG. And t
he re
sult
sho
w
s
spe
ed in
cre
a
s
e
s
in ca
se of
GNG net
wo
rk and the ma
ximum power is obtaine
d.
Figure 6. Voltage over the
Grid
Figure 7. Shows the
Comp
arison of Ref
e
ren
c
e G
ene
rator with an
d without G
N
G
7. Conclusio
n
This pap
er p
r
esents
the wind
tu
rbi
n
e
havi
ng
DFI
G
conne
cted
to the g
r
id
and the
maximum po
wer i
s
extra
c
ted from the
wind
sp
e
ed
usin
g neu
ral
netwo
rk. T
w
o
voltage so
urce
conve
r
ters a
r
e used. O
n
e
on the
gen
e
r
ator
sid
e
a
n
d
anoth
e
r
on
the gri
d
sid
e
. The g
r
id
-side
conve
r
ter i
s
controlle
d by a voltage orien
t
ed cont
ro
l. The d-axi
s
voltage compo
n
e
n
t is fixed with
grid volta
ge,
and th
e q
-
axis voltage
com
pon
ent
is
ze
ro. Th
e gen
erator
side
co
nvert
e
r i
s
controlled
by
a fiel
d o
r
ient
ed
cont
rol. T
he
spe
e
d
co
ntrol
of the
m
a
chi
ne i
s
don
e in
stead
of t
he
torque
cont
ro
l. The inform
ation of wind
spe
ed is ne
cessary to tra
c
k the maximu
m powe
r
ove
r
a
wide
spe
ed range. Th
us t
he GNG net
work
wh
i
c
h h
a
s be
en impl
emented tracks the m
a
xim
u
m
power.
Referen
ces
[1]
Blaa
bjer
g, F
Io
v.
W
i
nd p
o
w
e
r: a p
o
w
e
r so
urc
e
n
o
w
ena
bl
ed
by p
o
w
e
r el
ectronics
. Pr
oc of
9th Braz
ili
a
n
Po
w
e
r E
l
ectronics Confer
ence
COBEP
07, Blumenau S
anta
Catari
na,
Brazil.
2007.
ISBN
978-85-
991
95-0
2
-4.
[2]
Eil
y
a
n
Y B
i
tar,
Ram R
a
j
ago
pa
l, Pramo
d
P K
harg
onek
ar, K
a
mesh
w
a
r P
o
o
lla, Pr
avin
Var
a
i
y
a
.
Br
ingin
g
W
i
nd Ener
g
y
t
o
Market.
IEEE Transactions
on Power System
s
. 2
012; 2
7
(
3
).
[3]
Blaa
bjer
g, F
Iov, Z
Chen, R T
eod
orescu.
P
o
wer electronics
in re
newable energy system
s.
EPE-PEMC
200
6 Co
nferen
ce. Portoroz, Slove
n
ia. 2
006;
17. Numb
er: 06E
X14
0
5
C
, ISBN 1-42
44-0
4
4
9
-5.
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Gro
w
ing
Neu
r
al Ga
s Base
d MPPT for Wind G
ene
ra
tor usin
g DFI
G
(J. Pri
y
ad
a
r
shi
n
i)
5757
[4]
Cadirci I, Ermi
s M.
Double-
Output Inducti
on Gener
ator O
peratin
g at Sub-sync
h
ron
o
u
s and Su
per-
Synchro
n
o
u
s
Spee
ds: Stead
y-State Optimi
z
a
ti
on
an
d W
i
nd-En
ergy R
e
covery.
IEE Proc.B Electric
Po
w
e
r Ap
plic
ations. 19
92; 13
9(5): 429-
44
2.
[5]
Uctug MY, Eskandarzadeh I, Ince
H.
Mod
e
ling
a
nd O
u
tp
ut Pow
e
r Opti
mi
z
a
t
i
o
n
of
a
W
i
nd T
u
rb
in
e
Driven
Do
ub
le
Output Ind
u
cti
on Ge
nerat
or
. IEE Proc. Elec
tric Po
w
e
r
Ap
p
licatio
ns. 1
9
9
4
; 14
1(2): 3
3
-
38.
[6]
S Muller, M De
ike, RW
De Do
nker. Dou
b
l
y
F
E
D ind
u
ction
g
ener
ator s
y
ste
m
s for
w
i
n
d
tur
b
in
es.
IEEE
Ind. Appl. Mag.
, 2002; 8(3): 26
–33.
[7]
Gilsun
g B
y
eo
n, In K
w
o
n
P
a
rk, Gilsoo Ja
ng.
Mod
e
li
ng
and
Contro
l o
f
a Dou
b
l
y
-F
e
d
Inducti
o
n
Generator
(DF
IG) W
i
nd Po
wer Gen
e
ratio
n
S
y
stem f
o
r R
e
al-time S
i
mu
lat
i
ons.
J
ourn
a
l
of Electric
al
Engi
neer
in
g & T
e
chno
logy
. 2
010; 5(1): 6
1
-6
9.
[8]
C Har
i
ni,
N Kr
ishn
a Kum
a
ri,
Dr GS Ra
ju.
Analys
is of W
i
nd T
u
r
b
in
e Dr
i
v
en
Dou
b
ly F
ed In
ducti
o
n
Generator
. 1st Internatio
na
l C
onfere
n
ce o
n
Electrical E
nerg
y
S
y
stem. 201
1; 246-2
51.
[9]
Ritika Verma,
Amol Barve.
Vector contro
l o
f
a Grid Interfa
c
e
W
i
nd-
DF
IG Hybri
d
Syste
m
w
i
th SOF
C
F
uel cell.
Inter
natio
nal Jo
urn
a
l of Emergi
n
g
Rese
arch in
Manag
ement
&
T
e
chnol
og
y
ISSN: 2278-
935
9. 201
3; 96
-104.
[10]
Lun
a A, Lima
F
KA, Rodrí
guez P, W
a
tanabe EH, T
eod
orescu R.
Co
mpariso
n
of Po
w
e
r C
ontr
o
l
Strategies for
DF
IG W
i
nd
T
u
rbin
es.
IEEE tra
n
sactions.
200
8.
[11]
JS T
hongam,
P Bouc
hard,
H
Ezzaid
i, M Ou
hrouc
he.
Artific
i
al ne
ural netw
o
rk-bas
ed max
i
mu
m
p
o
w
e
r
poi
nt trackin
g
control f
o
r
varia
b
le s
p
e
e
d
w
i
nd
ener
g
y
convers
i
on
systems.
Proc
. 18th IEEE
Internatio
na
l C
onfere
n
ce o
n
Contro
l Appl
ica
t
ions. 200
9.
[12]
M Cirrinci
one,
M Pucci, G Vitale. Gro
w
i
ng N
eur
a
l
Gas (GNG)-Based ma
xi
mum po
w
e
r p
o
i
nt tracking
for high p
e
rfor
mance
w
i
nd g
e
nerator
w
i
t
h
an
inducti
on mac
h
in
e.
IEEE Ind. Appl. Mag.,
20
11; 47(2).
Evaluation Warning : The document was created with Spire.PDF for Python.