Int
ern
at
i
o
n
al
Journ
al of
P
ower E
le
ctr
on
i
cs a
n
d
Drive
S
ystem
s
(
IJ
PEDS
)
Vo
l.
12
,
No.
1
,
M
a
r 202
1
, p
p.
44
1
~
45
2
IS
S
N:
20
88
-
8694
,
DOI: 10
.11
591/
ij
peds
.
v12.i
1
.
pp44
1
-
45
2
441
Journ
al h
om
e
page
:
http:
//
ij
pe
ds
.i
aescore.c
om
Improve
ment of
sliding m
ode po
wer cont
ro
l
applied t
o w
i
nd
system
bas
ed on dou
bly
-
f
ed ind
uction ge
nerat
or
Btissam
M
ajo
ut
, D
ouae
Ab
r
ah
mi
,
Yasmin
e Ihedr
an
e
, C
ha
kib
El
Bakkali
, Karim
M
ohammed
,
Badre B
os
s
oufi
LIMAS
La
bora
t
ory,
Facu
lt
y
of
S
ci
en
ce
s Dhar
E
l M
ahr
az
,
Sidi
Mohame
d
Ben
Abdellah
Univer
si
ty,
Fez,
Morocc
o
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
M
a
y
19
, 20
20
Re
vised
Dec
2
0
, 2
0
20
Accepte
d
Ja
n
19
, 2
0
2
1
In
thi
s
work,
we
are
i
nte
r
este
d
in
im
proving
the
p
erf
orma
n
ce
of
a
doubly
-
fed
induc
ti
on
g
ene
r
a
tor
(DF
IG)
b
ase
d
wind
sys
te
m
,
by
app
lyi
ng
a
sl
idi
ng
mode
cont
rol
stra
te
gy
.
Th
e
objecti
v
e
i
s
the
r
egul
a
ti
on
of
th
e
active
a
nd
re
active
power,
al
so
th
e
volt
ag
e
and
th
e
fre
quen
cy
of
the
signa
l
injected
int
o
th
e
distri
buti
on
n
etw
ork.
The
mod
el
proposed
for
th
e
cont
ro
l
is
base
d
on
the
slidi
ng
mod
e
t
e
chni
que
wi
th
pe
rform
ance
est
imators.
The
prop
osed
mode
l
was va
li
d
at
ed
by
a
si
mul
a
ti
on
on
MA
TL
AB/S
im
uli
nk.
Ke
yw
or
d
s
:
DF
I
G
M
A
TLAB/Si
m
ulink
M
PP
T
SM
C c
ontr
ol
Win
d powe
r g
ener
at
or
sy
ste
m
This
is an
open
acc
ess arti
cl
e
un
der
the
CC
BY
-
SA
l
ic
ense
.
Corres
pond
in
g
Aut
h
or
:
Bt
issam Majo
ut
LI
M
AS La
bor
at
ory
Faculty
of Scie
nces
D
har
El
M
a
hr
az
Sidi
M
ohame
d B
en A
bd
el
la
h
Un
i
ver
sit
y, Fe
z,
M
orocc
o
Emai
l:
badre
_isai@h
otmail
.c
om
1.
INTROD
U
CTION
Lat
el
y,
the
us
e
of
ren
e
wa
ble
energies
(
i.e
.
,
wind
an
d
s
olar
photov
oltai
c)
to
increase
i
n
an
inc
red
i
ble
way
t
ha
nk
s
t
o
the
scarcit
y
of
com
busti
bles.
Win
d
e
nerg
y
is
su
pp
os
e
d
to
be
the
best
i
n
te
rms
of
qual
it
y
an
d
pr
ic
e
[
1].
T
he
r
e
are
se
ver
al
re
search
st
ud
ie
s
about
t
he
wind
tur
bi
ne.
I
n
part
ic
ular,
t
he
one
s
with
as
ynch
r
onous
gen
e
rato
rs.
Althou
gh,
t
hey
ha
ve
a
lo
w
c
ost
an
d
simple
m
ai
ntena
nce
a
s
a
dv
a
ntage
but,
t
hey
requir
e
m
or
e
exp
e
ns
i
ve
eq
ui
pm
e
nt
an
d
c
omplex
co
ntr
ol.
Ther
e
f
or
e
,
in
the
rece
nt
yea
r
s,
the
wind
tu
r
bin
e
s
ys
te
m
m
ov
e
d
towa
rd
s
t
he
doubly
-
fe
d
i
nduc
ti
on
gen
e
rato
r
(
DFIG
)
mac
hin
e
wh
ic
h
ha
s
higher
qual
it
y
an
d
la
r
ge
r
powe
r
densi
t
y
[2].
F
ur
t
her
m
ore,
th
e
D
FIG
re
duces
the
mec
ha
nical
stress
by
rem
ov
i
ng
the
necessit
y
of
the
mu
lt
ipli
cat
or
wh
ic
h
imp
r
oves
the
sy
ste
m’
s
reli
abili
ty
[3]
an
d
dec
rease
s
the
mainte
na
nce
co
sts
by
di
rectl
y
couplin
g
th
e tu
rb
i
ne
a
nd the s
haf
ts
of the
g
e
ner
at
or
[
4]
.
Du
e t
o
the
h
i
gh varia
bili
ty
of the w
i
nd sp
ee
d,
it
’s dif
ficult
to obtai
n
a sati
sfacto
ry
pe
rformance
of the
Win
d
tu
rb
i
ne
Sy
ste
m
.
Re
cen
tl
y,
this
la
tt
er
is
desig
ned
t
o
extract
the
ma
ximum
po
wer
po
i
nt
(
M
PP
)
powe
r
from
the
wind
sp
eed
,
w
hich
is
commonl
y
known
as
t
he
maxim
um
power
po
i
nt
trac
king
(
M
P
PT)
strat
egy.
Diff
e
re
nt
met
hods
hav
e
been
dev
el
op
e
d
in
orde
r
t
o
mai
ntain
t
he
op
e
rati
ng
point
of
m
axi
mu
m
ef
fici
enc
y.
The
mo
st
wides
pre
ad
c
on
t
ro
l
stra
te
gy
is
the
opt
imum
pow
er/t
orq
ue
trac
king
,
w
he
re
the
use
of
the
pr
oport
i
on
al
and
i
nteg
ral
(PI)
c
on
tr
oller.
Howe
ver,
this
strat
egy
al
on
e
do
e
sn’t
reali
ze
a
bette
r
pe
rfo
rma
nce.
Hen
ce
,
there
are
oth
er
co
ntr
ol
meth
ods
s
uc
h
as
the
bac
kst
epp
i
ng,
f
uzz
y
lo
gic,
a
nd
sl
iding
m
od
e
c
ontr
ol
(SMC
)
[
5],
[
6].
This
pa
per
is
r
epatriat
ed
as
f
ollow
s:
Sect
io
n
2
presents
th
e
desc
riptio
n
of
the
wind
s
yst
em
(tu
r
bin
e,
DF
I
G,
inv
e
rter,
DC
-
bus,
a
nd
filt
er)
.
Sect
ion
3
disc
us
ses
t
he
pr
i
nc
iple
of
op
e
rati
on
of
th
e
sli
di
ng
m
od
e
co
m
man
d
as
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8
694
In
t J
P
ow
Ele
c
&
D
ri
S
ys
t,
V
ol
.
12
, N
o.
1
,
Ma
rch
20
21
:
44
1
–
45
2
442
well
as
it
s
a
pp
li
cat
ion
on
our
s
ys
te
m.
Sect
i
on
4
sho
ws
an
d
i
nter
pr
et
s
the
res
ults
of
t
he
simulat
ion.
Fin
al
ly,
a
con
cl
us
io
n
i
n Sect
ion
5.
2.
MO
DELIN
G
OF WI
N
D
S
Y
STE
M
B
AS
E
D
O
N DFI
G
The
c
onve
rsion
c
hain
i
nclu
de
s
in
se
ries
a
s
peed
m
ulti
plier
to
inc
rease
th
e
sp
ee
d
of
ro
ta
ti
on
to
ab
out
1500
r
pm,
a
doubly
fe
d
in
duct
ion
ge
ne
rator
(
DF
I
G)
o
pe
rati
ng
at
va
riable
sp
e
ed
,
T
hr
ee
-
phase
co
nv
e
rters
adjust
the
freq
uen
c
y
of
t
he
wind
tu
rb
i
ne
to
that
of
t
he
e
l
e
c
t
r
i
c
i
t
y
g
r
i
d
t
o
w
h
i
c
h
i
t
i
s
c
o
n
n
e
c
t
e
d
(
5
0
H
z
i
n
M
o
r
o
c
c
o
)
[7]
.
The
tran
sf
orm
at
ion
of
t
he
po
wer
of
the
aer
og
e
ne
rator
i
nto
ki
netic
e
nergy
the
n
i
nto
mec
han
ic
al
energ
y
of
r
otati
on
is
do
ne
in
t
wo
pa
rts:
at
th
e
tur
bin
e
r
otor
(primar
y
s
ha
ft)
,
w
hich
ca
ptu
r
es
pa
rt
of
t
he
ki
netic
energ
y
of
the
wind
prese
nt
to
co
nvert
it
into
m
echa
nical
energy
at
the
gen
e
rato
r
r
oto
r
(secon
dary
sh
aft
)
,
wh
ic
h ob
ta
in
s
mecha
nical
ene
rgy
a
nd c
onve
rts it
into
el
ect
r
ic
al
en
er
gy as s
how
n
in
Fig
ure
1
[
8]
.
Figure
1. A
rch
i
te
ct
ur
e
of
t
he
c
on
t
ro
l
2.1.
Wind
-
tu
rbine
mod
el
The mo
del
of the
tu
r
bin
e is
m
od
el
e
d from t
he
foll
ow
i
ng sys
te
m of e
qu
at
io
ns
(1)
-
(9)
[9]
,
[
10]:
3
.
.
.
2
1
v
S
P
i
n
c
i
d
e
n
t
=
(1)
3
).
,
(
.
.
.
2
1
v
C
S
P
p
e
x
t
r
a
c
t
e
d
=
(2)
v
R
t
.
=
(3)
593
.
0
27
16
)
,
(
m
a
x
=
p
C
(4)
.
.
.
1
.
.
)
,
(
6
1
4
3
2
1
5
c
e
c
c
A
c
c
C
A
c
p
+
−
−
=
−
(5)
3
1
0
3
5
.
0
.
08
.
0
1
1
+
−
+
=
A
(6)
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
P
ow Elec
& Dri S
ys
t
IS
S
N: 20
88
-
8
694
Impr
ovem
e
nt
of
sli
din
g m
ode
po
we
r
contr
ol
applied t
o
wi
nd syste
m based
on
…
(
Bti
ssam
Ma
j
ou
t
)
443
t
p
t
e
o
l
al
v
C
S
P
C
=
=
1
.
).
,
(
.
.
.
2
1
3
(7)
g
t
u
r
J
G
J
J
+
=
2
(8)
m
e
c
em
ar
m
e
c
m
e
c
f
C
C
C
dt
d
J
−
−
=
=
.
(9)
The
Fig
ur
e
2
s
hows
the
e
volu
ti
on
of
the
po
w
er
c
oeffici
ent
a
s
a
functi
on
of
λ
for
dif
fer
e
nt
values
of
β
[11
]
, [
12]
.
Figure
2.
Ev
ol
ution o
f
t
he
c
oe
ff
ic
ie
nt C
p
as
a f
un
ct
io
n of t
he
sp
eci
fic
sp
ee
d
λ
2.2.
Ma
xi
mi
za
tion o
f
p
ow
e
r c
on
t
rol w
i
thout sp
eed co
nt
r
ol
W
hile
th
e v
ari
at
ion
o
f
the w
i
nd
spe
ed
in
ste
ady
sta
te
is
lo
w
c
ompare
d
t
o
the
el
ect
rical
t
ime
co
ns
ta
nts
of
t
he
s
ys
te
m,
we
ass
um
e
th
at
the
sp
ee
d
of
r
otati
on
of
t
he
DF
I
G
is
fixed
an
d
ne
glect
ing
the
ef
fec
t
of
t
he
visco
us
to
rque
f
,
t
he
dy
nami
c
e
qu
at
io
n
of
the
tur
bin
e
bec
om
es
(12)
.
F
r
om
(12
)
we
ob
t
ai
n
t
he
sta
ti
c
e
qu
at
io
n
descr
i
bing
t
he
sta
ti
on
ar
y
sta
te
of
t
he
t
urbine
(
13)
.
T
he
ref
e
ren
c
e
el
ect
r
oma
gn
et
ic
t
orq
ue
is
deter
mine
d
fr
om
an
e
sti
mate
of
the
ae
rod
yn
a
m
ic
tor
que
giv
e
n
by
(
14)
an
d
w
e
obta
in
eq
uat
ion
(
15)
.
T
he
or
ie
ntati
on
an
gl
e
of
the
bla
des
β
i
s
assu
med
to
be
c
on
sta
nt
a
nd
the
est
imat
ed
s
peed
of
t
he
tu
rb
i
ne
is
cal
culat
ed
f
rom
the
mech
a
nical
s
pe
ed
(
16)
.
T
he
est
imat
ed
wind
s
pee
d
is
give
n
by
e
quat
io
n
(17
)
.
O
n
the
b
as
e
of
the
previ
ou
s
equ
at
io
ns,
we c
an
the
n w
rite
the equati
on
of the
ref
e
ren
ce
e
le
ct
ro
ma
gn
et
ic
coup
le
(
18)
[13]
,
[
14]
.
.
=
−
−
.
=
0
(10
)
=
=
(11)
=
1
2
.
.
(
,
)
.
.
.
2
.
3
(12)
=
(13)
=
(14)
=
.
(15)
_
=
.
.
5
.
_
(
)
.
Ω
2
2
.
3
.
3
(16)
2.3.
DFIG m
od
el
The
e
quat
ions
of the
DFI
M
in
the
ref
e
ren
ce
of Park a
re
wr
i
tt
en
as [1
5], [1
6]
:
−
Vo
lt
age
s at the
stat
or
:
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8
694
In
t J
P
ow
Ele
c
&
D
ri
S
ys
t,
V
ol
.
12
, N
o.
1
,
Ma
rch
20
21
:
44
1
–
45
2
444
{
=
.
+
−
.
=
.
+
+
.
(17)
−
Vo
lt
age
s at the
stat
or
:
{
=
.
+
−
.
=
.
+
+
.
(18)
With:
−
=
.
The ma
gnet
ic
equ
at
io
ns are
e
xpresse
d b
y
th
e flux e
xpressi
on
s
in
the
ref
e
r
ence
(d,
q)
[
3]
[17
].
−
Flux at
the
sta
tor:
{
=
.
+
.
=
.
+
.
(
19)
−
Flux at
the
r
otor:
{
=
.
+
.
=
.
+
.
(20)
With:
M
=M
sr
=M
rs
The
el
ect
r
om
a
gn
et
ic
to
r
qu
e
is ex
pr
e
ssed
as
a f
un
ct
io
n of t
he
curre
nts a
nd the
flo
ws by [
18]:
{
=
.
(
.
+
.
)
=
.
(
.
−
.
)
The fu
ndame
nt
al
eq
uati
on of
dynamics
is:
=
+
.
Ω
+
.
Ω
(21)
V
s(d,q)
,V
r(d,q)
: Sta
tor
a
nd rot
or volt
ages i
n
t
he refe
ren
ce
of
Park.
I
s(d,q),
I
r(d,q)
:
Stat
or
a
nd rot
or cu
rr
e
nts in
the
ref
e
ren
ce
of
Park.
s(d,q)
,
r(d,q)
:
Stat
or
a
nd rot
or f
lu
x
i
n
the
re
fer
e
nce
of Park.
The Fi
gure
3 p
resen
ts
the
model o
f
t
he DFIG mac
hin
e
on Si
mu
li
nk
Figure
3. D
FIG m
odel
sim
ulink
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
P
ow Elec
& Dri S
ys
t
IS
S
N: 20
88
-
8
694
Impr
ovem
e
nt
of
sli
din
g m
ode
po
we
r
contr
ol
applied t
o
wi
nd syste
m based
on
…
(
Bti
ssam
Ma
j
ou
t
)
445
2.4.
Model
of
th
e
t
wo
-
le
vel
volta
ge
in
vert
er
To
co
nnect
the
ge
ne
rator
wh
i
ch
unde
r
go
es
a
va
riable
s
pee
d
with
the
ne
twork
,
it
is
nec
essar
y
t
o
go
thr
ough
a
st
ag
e
of
powe
r
el
ect
ronics
in
orde
r
to
c
on
t
rol
the
power
i
nject
ed.
we
use
2
RSC
an
d
GSC
conve
rsion sta
ges, co
nverters
which c
onsist
s
of IGBTs
as s
how
n
in
Fig
ure
4
.
.
Figure
4. Dia
gram of t
he
tw
o
-
l
evel in
ver
te
r
The
Sa, Sb
, S
c
is t
he
sta
te
of
t
he uppe
r
s
witc
hes of ea
ch
ar
m of t
he
in
ve
rter.
−
The
e
xpressi
on of the
simple
vo
lt
age
s is
pr
e
sented
by t
he
f
ollow
i
ng s
ys
te
m [1
9]
:
{
=
1
3
(
−
)
=
1
3
(
2
.
−
−
)
=
1
3
(
−
)
=
1
3
(
2
.
−
−
)
=
1
3
(
−
)
=
1
3
(
2
.
−
−
)
(22)
−
The ma
trix
form o
f
sim
ple te
ns
io
ns
bec
om
e
s [20]:
[
]
=
1
3
[
2
−
1
−
1
−
1
2
−
1
−
1
−
1
2
]
.
[
]
(23)
we
ass
ociat
ed wit
h
Eac
h
a
rm
of the i
nv
e
rter
a b
ina
r
y
c
om
m
and v
al
ue
Si,
wh
e
re i =
a, b,
c:
[
]
=
2
[
]
(
24
)
we replace
(24
)
in
(23),
w
e
g
e
t:
[
]
=
6
.
[
2
−
1
−
1
−
1
2
−
1
−
1
−
1
2
]
.
[
]
(
25
)
The
sin
gle
vo
lt
ages
of
th
e
in
ve
rter
be
co
me
pro
portio
nal
to
t
he
sta
te
s
of
the
con
t
ro
l
qu
a
ntit
ie
s
of
the
switc
hes (Sa
, Sb, Sc)
.
2.5.
DC
-
bu
s m
od
el
The
DC
bus
as
sh
ow
n
in
Fi
gure
5
in
te
rcon
ne
ct
s
the
two
c
on
ver
te
r
s
of
the
wind
syst
em
(
RSC
and
GS
C
).
T
he
la
tt
er all
ow
s t
he
tran
s
f
e
r
o
f
p
o
w
e
r
b
e
t
w
e
e
n
t
w
o
s
o
u
r
c
e
s
a
t
d
i
f
f
e
r
e
n
t
f
r
e
q
u
e
n
c
i
e
s
.
I
t
i
s
m
o
d
e
l
e
d
by
(
26
)
[21].
{
=
∫
.
=
1
2
.
.
2
2
=
2
(
−
)
(26)
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8
694
In
t J
P
ow
Ele
c
&
D
ri
S
ys
t,
V
ol
.
12
, N
o.
1
,
Ma
rch
20
21
:
44
1
–
45
2
446
Figure
5. Dia
gram of t
he
c
on
ti
nuous
bus
2.6.
RL f
il
ter
mod
el
The
inte
rme
diate
filt
er
us
e
d
is
of
t
he
RL
t
ype.
The
cu
rr
e
nts
passe
d
betwee
n
the
GS
C
c
on
ver
te
r
an
d
the
net
work
ar
e
imp
os
ed
by
t
he
c
oils
co
ns
ti
tuti
ng
t
he
lo
w
pass
filt
er.
T
he
ex
pr
essi
ons
of
the
volt
ages
at
the
te
rmin
al
s
of th
e filt
ers
in
the t
he par
k refe
rent
ia
l are present
ed by (
27
).
{
=
−
.
−
.
+
.
.
=
−
.
−
.
−
.
.
+
(27)
3.
SLIDI
NG M
ODE
CONTR
OLL
ER
The
basic
idea
of
sli
ding
mode
c
ontr
ol
design
is
first
to
at
tract
the
sta
te
s
of
the
s
ys
te
m
to
a
s
uitabl
y
sel
ect
ed
reg
i
on,
a
nd
t
hen
to
de
sign
a
co
ntr
ol
la
w
that
will
al
way
s
keep
t
he
s
ys
te
m
in
th
at
re
gion.
Wh
e
re
t
he
desig
n of t
he sl
iding m
od
e
c
ontr
ol alg
or
i
th
m
is d
e
fine
d by three
compleme
ntar
y
ste
ps
[22]
:
3.1.
Choice
of slidi
ng
s
urf
ace
Fo
r
a
non
-
li
ne
ar
s
ys
te
m
pr
es
ented
in
the
f
ollow
i
ng
form
(
28
),
where
A
(
x,
t)
a
nd
B
(x,t)
a
re
tw
o
con
ti
nu
ous a
nd uncertai
n n
on
li
near
fu
nctio
ns assu
me
d
to
be b
ounded
[
23]
.
(
)
̇
=
(
,
)
+
(
,
)
(
)
;
∊
,
∊
(28)
(
)
=
(
+
)
−
1
∗
(
)
(29)
(
)
=
−
(30)
=
[
,
̇
,
…
.
−
1
]
;
=
[
,
̇
,
…
…
]
3.2.
Conv
er
gence
an
d
exi
stence
cond
i
tion
s
To
ma
ke
t
he
s
urface
at
tract
iv
e
an
d
in
var
ia
nt
,
we
retu
rn
e
d
to
the
sec
ond
theo
rem
of
L
YAPU
N
OV
wh
e
re
the
scal
ar
f
un
ct
io
n
is
def
i
ned
posit
ive
by
(
31
)
.
T
he
de
rivati
ve
of
this
functi
on
gi
ves
(
3
2
),
and
t
o
gu
a
ra
ntee
the
existe
nce
of
the
sli
din
g
m
ode,
wh
e
re
the
sli
di
ng
var
ia
ble
S
(
x,
t)
te
nds
to
w
ard
s
ze
ro,
it
suffices
to ensu
re
that
(3
2
)
is
de
fine
d neg
at
ive
(3
3
).
(
)
=
1
2
.
(
)
2
(31)
(
)
̇
=
(
)
(
)
̇
(32)
(
)
(
)
̇
<
0
(33)
3.3.
Det
ermi
n
ati
on o
f the l
aw of
co
n
tr
ol
The
co
ntr
ol
la
w
is
def
i
ne
d
by
the
relat
io
n
(
34
),
With:
u
+
a
nd
u
-
are
c
onti
nuous
f
unct
ions
(
35
)
w
here
u
-
≠
u
+
.
T
he
c
o
nt
ro
l
by
sli
din
g
modes
is
c
ompose
d
of
t
w
o
te
rms:
u
eq
:
the
eq
uiv
al
e
nt
con
t
ro
l
vecto
r,
u
n
:
The
sta
bili
zi
ng
c
omman
d
[24]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
P
ow Elec
& Dri S
ys
t
IS
S
N: 20
88
-
8
694
Impr
ovem
e
nt
of
sli
din
g m
ode
po
we
r
contr
ol
applied t
o
wi
nd syste
m based
on
…
(
Bti
ssam
Ma
j
ou
t
)
447
=
+
(34)
=
{
+
(
)
(
,
)
>
0
−
(
)
(
,
)
<
0
(35)
3.3.1.
Det
ermi
n
ati
on o
f the e
quiv
alent
co
m
ma
n
d u
eq
:
(
)
̇
=
=
.
(36)
(
)
̇
=
=
(
(
,
)
+
(
,
)
(
+
)
)
(37)
(
)
̇
=
=
(
(
,
)
+
(
,
)
)
+
(
,
)
(38)
(
)
=
(
−
(
,
)
)
(
(
,
)
)
−
1
(
39)
3.3.2.
Det
ermi
n
ati
on o
f the
ba
sic
discon
tinu
ou
s
co
mm
and u
n
:
The
si
mp
le
st
di
scon
ti
nu
ou
s
c
om
ma
nd
u
n
is
giv
e
n
by
(
40
)
,
wh
e
re
K
is
the
com
ma
nd
gai
n.
This
ty
pe
of
c
on
tr
ol
has
a
dra
wb
ac
k
kn
own
by
“C
H
A
TTERI
NG
”
.
T
o
s
olve
t
his
prob
le
m
in
this
case
we
rep
la
c
ed
t
he
"SI
G
N" f
un
ct
i
on by t
he
"S
A
T" f
un
ct
io
n
[
25]
:
=
.
(
(
)
)
(40)
(
)
=
{
(
)
=
1
>
(
)
=
−
1
<
(
)
=
|
|
<
(41)
4.
APPLI
CA
TI
ON OF THE
SLIDI
NG M
ODE
COM
M
AND
TO
TH
E DFI
G
4.1.
Contr
ol
of
th
e
co
n
ver
ter
on
th
e
DFI
G
(
RSC) si
de
an
d
on
t
he
netw
ork
s
ide (G
SC)
Con
si
der
i
ng
th
e
sli
ding
s
urfa
ce
pro
pose
d
by
S
LO
TI
NE
(
42
),
F
or
n=
1;
the
sli
ding
sur
face
of
the
act
ive
an
d
reac
ti
ve
powe
r
is g
iven
by
(
43
)
, wher
e
P
sref
an
d
Q
sref
are
t
he
refe
ren
ces
of
sta
t
or
pow
e
rs
(acti
ve
a
nd
reacti
v
e)
of
DFIG a
nd
Q
fref
a
nd P
fref
a
re t
he
r
efere
nces
of po
wer
s
(react
ive
and act
ive)
of
RL fil
te
r
[
26]
.
{
(
)
=
1
=
−
(
)
=
2
=
−
(
)
=
3
=
−
(
)
=
4
=
−
(42)
{
(
)
̇
=
̇
1
=
̇
−
̇
(
)
̇
=
̇
2
=
̇
−
̇
(
)
̇
=
̇
3
=
̇
−
̇
(
)
̇
=
̇
4
=
̇
−
̇
(43)
with:
{
̇
=
−
.
̇
̇
=
2
.
−
.
̇
̇
=
.
−
.
.
+
.
̇
=
.
−
.
.
−
.
−
.
.
.
.
.
(44)
We
rep
la
ce
ea
ch
te
r
m
by
it
s
expressi
on
gi
ve
n
by
(
44
)
,
the
de
rivati
ve
of
t
he
sli
di
ng
s
urf
ace
bec
om
es
as (
45
)
a
nd
(
46
)
[
27]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8
694
In
t J
P
ow
Ele
c
&
D
ri
S
ys
t,
V
ol
.
12
, N
o.
1
,
Ma
rch
20
21
:
44
1
–
45
2
448
{
̇
1
=
̇
+
.
(
+
.
−
.
−
−
.
.
.
)
̇
2
=
̇
+
.
(
+
.
−
.
+
)
(45)
{
̇
3
=
̇
+
.
+
(
+
)
+
.
.
−
2
̇
4
=
̇
−
.
−
(
+
)
+
.
.
(46)
Accor
ding
to
t
he
sli
di
ng
m
od
e
an
d
t
he
perm
anen
t
m
od
e
,
w
e
ha
ve
(
47
)
,
(
48
)
.
T
he
e
xpres
sion
of
t
he
equ
i
valent
Ve
q
c
omman
d
be
comes
(
51)
a
nd
(
52)
.
T
he
sta
bili
zi
ng
c
omman
d
is
give
n
by
(
51
)
a
nd
(
52
)
.
F
inall
y,
t
he
e
xpressi
on
of
t
he
total
o
r
der (
V
rd
, V
rq
)
a
nd (V
fd
, V
fq
) bec
om
es
(
53
)
a
nd
(
54
)
[
27]
.
{
1
,
2
=
0
̇
1
,
2
=
0
=
=
0
(47)
{
e
3
,
4
=
0
e
̇
3
,
4
=
0
V
f
dn
=
V
f
qn
=
0
(48)
{
V
rdeq
=
−
L
r
.
L
s
.
σ
M
.
V
s
Q
̇
sr
e
f
+
R
r
.
I
rd
−
ω
r
.
L
r
.
σ
.
I
rq
V
rq
e
q
=
−
L
r
.
L
s
.
σ
M
.
V
s
P
̇
sr
e
f
+
R
r
.
I
rq
+
ω
r
.
L
r
.
σ
.
I
rd
+
ω
r
V
s
M
L
s
.
ω
s
(49)
{
V
f
deq
=
L
f
V
s
Q
̇
sr
e
f
−
R
f
.
I
df
+
L
f
.
ω
s
.
I
qf
V
f
qeq
=
L
f
V
s
P
̇
sr
e
f
−
R
f
.
I
qf
−
L
f
.
ω
s
.
I
df
+
V
s
(50)
{
V
rdn
=
K
rdn
.
Sa
t
(
e
1
)
V
rqn
=
K
rqn
.
Sa
t
(
e
2
)
(51)
{
V
f
dn
=
K
f
dn
.
Sa
t
(
e
4
)
V
f
qn
=
K
f
qn
.
Sa
t
(
e
3
)
(52)
{
V
rd
=
−
L
r
.
L
s
.
σ
M
.
V
s
.
Q
̇
sr
e
f
+
R
r
.
I
rd
−
ω
r
.
L
r
.
σ
.
I
rq
+
K
d
s
a
t
(
e
2
)
V
rq
=
−
L
r
.
L
s
.
σ
M
.
V
s
.
P
̇
sr
e
f
+
R
r
.
I
rq
+
ω
r
.
L
r
.
σ
.
I
rd
+
ω
r
.
M
.
V
s
L
s
.
ω
s
+
K
q
s
a
t
(
e
1
)
(53)
{
V
fd
=
−
L
f
V
s
.
Q
̇
f
ref
−
R
f
.
I
df
+
ω
s
.
L
f
.
I
qf
+
K
df
n
s
a
t
(
e
4
)
V
fq
=
−
L
f
V
s
.
P
̇
f
ref
−
R
f
.
I
qf
−
ω
s
.
L
f
.
I
df
+
V
s
+
K
qf
n
sat
(
e
3
)
(54)
4.2.
Simul
at
i
on
&
resul
s
To
ver
if
y
the
performa
nce
a
nd
sta
bili
ty
of
the
co
ntr
ol
sy
s
te
m
by
S
M
C
c
on
t
ro
l,
t
he
D
FIG
is
subje
ct
to
tw
o
r
obus
t
ne
ss
te
sts
as
s
hown
i
n
Fi
gure
6
(the
T
rac
king
and
Re
gula
ti
on
Test
s
f
or
S
MC
and
the
rob
ust
ness
te
sts reg
a
rd
i
ng
the v
a
riat
ion p
aramete
rs
).
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
P
ow Elec
& Dri S
ys
t
IS
S
N: 20
88
-
8
694
Impr
ovem
e
nt
of
sli
din
g m
ode
po
we
r
contr
ol
applied t
o
wi
nd syste
m based
on
…
(
Bti
ssam
Ma
j
ou
t
)
449
Figure
6. S
M
C
contr
ol appli
ed
to
DFI
G win
d
tu
r
bin
e s
ys
te
m
4.3.
Tr
ackin
g and
regula
tio
n for
SMC
In
t
his
te
st,
we
consi
dered
t
he
aerod
yn
a
mic
powe
r
acco
r
din
g
t
o
the
M
P
P
T
as
a
re
fer
e
nc
e
of
act
i
ve
powe
r,
a
nd
zer
o
as
r
e
fer
e
nce
f
or r
eact
ive
po
wer.
4.3.1 Tes
t wit
h const
ant s
pe
ed (
r
ung spee
d)
:
Figure
7
s
how
s
the
re
su
lt
s
obta
ined
for
the
ap
plica
ti
on
of
the
co
ntr
ol
by
Sli
ding
m
ode
to
a
wi
nd
powe
r
s
ys
te
m
at
the b
a
se
of
t
he DFIG.
(a)
(b)
(c)
(d)
0
5
10
15
20
25
30
-
1
.
5
-1
-
0
.
5
0
0
.
5
1
1
.
5
x
1
0
4
T
i
m
e
s
[
s
]
T
h
e
A
c
t
i
v
e
S
t
a
t
o
r
P
o
w
e
r
P
s
[
W
]
P
s
m
e
s
P
s
r
e
f
0
0
.
1
0
.
2
-1
0
1
x
1
0
4
X
:
0
.
0
5
2
2
Y
:
-
1
9
.
4
9
9
.
9
8
10
1
0
.
0
2
-
4
0
0
0
-
2
0
0
0
0
1
9
.
9
9
5
20
2
0
.
0
0
5
-
1
0
0
0
0
-
5
0
0
0
0
0
5
10
15
20
25
30
-1
-
0
.
8
-
0
.
6
-
0
.
4
-
0
.
2
0
0
.
2
0
.
4
0
.
6
0
.
8
1
x
1
0
4
T
i
m
e
s
[
s
]
T
h
e
R
e
a
c
t
i
v
e
S
t
a
t
o
r
P
o
w
e
r
Q
s
[
V
A
R
]
Q
s
m
e
s
Q
s
r
e
f
0
0
.
0
5
0
.
1
0
.
1
5
-
1
5
0
0
0
-
1
0
0
0
0
-
5
0
0
0
0
5
0
0
0
X
:
0
.
0
4
0
5
Y
:
0
0
5
10
15
20
25
30
-
2
0
0
-
1
5
0
-
1
0
0
-
5
0
0
50
1
0
0
1
5
0
2
0
0
T
i
m
e
s
[
s
]
T
h
e
R
o
t
o
r
C
u
r
r
e
n
t
I
r
q
[
A
]
I
r
q
5
10
15
20
25
30
-
3
0
0
-
2
0
0
-
1
0
0
0
1
0
0
2
0
0
3
0
0
4
0
0
5
0
0
T
i
m
e
s
[
s
]
T
h
e
R
o
t
o
r
C
u
r
r
e
n
t
I
r
d
[
A
]
I
r
d
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8
694
In
t J
P
ow
Ele
c
&
D
ri
S
ys
t,
V
ol
.
12
, N
o.
1
,
Ma
rch
20
21
:
44
1
–
45
2
450
(e)
(f)
Figure
7. Re
su
l
ts of the c
ontr
ol
b
y sl
idin
g mo
de of
the
DFIG
for
r
ung
s
pee
d;
(
a) acti
ve power
p
s
,
(
b) re
a
ct
ive
powe
r
Q
s,
(
c)
d
-
q
sta
to
r
c
urre
nt,
(
d)
dq roto
r c
urren
t,
(
e
)
a
bc
stat
or
c
urre
nt,
(
f
)
a
bc
r
otor c
urre
nt
Accor
ding
to
t
hese
re
su
lt
s,
t
he
act
ive
powe
r
s
Ps
Fig
ure
7(
a)
an
d
re
act
ive
Qs
Fi
gure
7(
b)
f
ollow
t
he
ref
e
ren
ce
.
The
sta
tor
cu
rr
e
nts
Ir
Fi
gure
7(c)
and
t
he
r
oto
r
c
urren
ts
Fi
gure
7(d)
a
re
of
go
od
qual
it
y
a
nd
f
ollow
the
gi
ven
instr
uction.
I
n
Fig
ures
7(
e
)
an
d
(f)
the
sta
t
or
a
nd
r
otor
c
urr
ent
s
are
si
nu
s
oid
a
l
with
a
fr
e
qu
e
ncy
o
f
50
Hz,
a
TH
D
le
ss
tha
n
5%
,
wh
ic
h
im
plies
that
the
wi
nd
powe
r
sy
ste
m
re
sp
ect
s
t
he
c
on
diti
on
s
of
co
nn
ect
ion
with the
elec
tric
al
g
ri
d.
4.3.2. Tes
t wit
h varia
ble spe
ed:
Durin
g
this
te
s
t
the
wind
pro
file
il
lustrate
d
in
Fig
ur
e
8
wa
s
ap
plied
to
t
he
DFIG
.
Acc
or
ding
to
t
he
curves
il
lustrate
d
i
n
th
e
Fi
gure
8
we
noti
ce
a
good
be
havi
or
of
the
mac
hin
e
in
sp
it
e
of
the
var
ia
ti
on
of
t
he
wind,
w
her
e
t
he
ge
ner
at
or
f
ollows
the
re
fe
ren
ces
of
the
powe
rs
with
out
ov
e
rs
hoot
a
nd
with
a
n
al
m
os
t
zer
o
error.
A
nd
the
el
ect
ro
ma
gnet
ic
torque
of
th
e
DFIG
va
ries
accor
ding
t
o
t
he
wind
s
peed,
an
d
pro
portio
nal
to
the
act
ive
sta
to
r
powe
r
ge
ner
a
te
d.
We
can
noti
ce
that
in
s
pite
of
the
va
riat
ion
s
of
the
wind,
the
sta
tor
c
urre
nt
Is
-
a
bc
remai
n
s
sin
usoidal
with
a
fi
xed
f
re
quency
50Hz
eq
ui
valent
to
that
of
the
net
wor
k.
The
DC
bu
s
volt
age
sh
ows
that
it
f
ollows
it
s
ref
e
r
ence
value
qu
i
ckly
with
out
overs
hootin
g
wi
th
a
small
sta
ti
c
er
ror.
T
he
sli
p
value
g
is
ne
gative th
is i
mp
li
es that t
he fu
nctio
ning
of the
DFIG
is
in hy
po
-
s
yn
c
hr
onous.
(a)
(b)
(c)
(d)
0
5
10
15
20
25
30
-
1
0
0
-
8
0
-
6
0
-
4
0
-
2
0
0
20
40
60
80
1
0
0
T
i
m
e
s
[
s
]
T
h
e
S
t
a
t
o
r
C
u
r
r
e
n
t
s
I
s
-
a
b
c
[
A
}
9
.
9
5
10
1
0
.
0
5
-
1
0
0
10
1
9
.
9
6
1
9
.
9
8
20
2
0
.
0
2
2
0
.
0
4
-
2
0
0
20
I
s
a
I
s
b
I
s
c
0
5
10
15
20
25
30
-
1
0
0
-
8
0
-
6
0
-
4
0
-
2
0
0
20
40
60
80
1
0
0
T
i
m
e
s
[
s
]
T
h
e
R
o
t
o
r
C
u
r
r
e
n
t
I
r
-
a
b
c
[
A
]
9
.
9
5
10
1
0
.
0
5
-
1
0
0
10
1
9
.
9
5
20
2
0
.
0
5
-
2
0
0
20
I
r
a
I
r
b
I
r
c
0
5
10
15
20
25
30
0
2
4
6
8
10
12
T
i
m
e
s
[
s
]
W
i
n
d
S
p
e
e
d
[
m
/
s
]
w
i
n
d
S
p
e
e
d
0
5
10
15
20
25
30
-
1
2
0
0
0
-
1
0
0
0
0
-
8
0
0
0
-
6
0
0
0
-
4
0
0
0
-
2
0
0
0
0
2
0
0
0
4
0
0
0
6
0
0
0
8
0
0
0
T
i
m
e
s
[
s
]
T
h
e
A
c
t
i
v
e
S
t
a
t
o
r
P
o
w
e
r
P
s
[
W
]
0
0
.
0
5
0
.
1
-1
0
1
x
1
0
4
X
:
0
.
0
5
2
4
Y
:
-
1
2
0
9
P
s
m
e
s
P
s
r
e
f
0
5
10
15
20
25
30
-
1
0
0
-
5
0
0
50
1
0
0
1
5
0
T
i
m
e
s
[
s
]
T
h
e
R
o
t
o
r
C
u
r
r
e
n
t
I
r
q
[
A
]
I
r
q
0
5
10
15
20
25
30
-1
-
0
.
8
-
0
.
6
-
0
.
4
-
0
.
2
0
0
.
2
0
.
4
0
.
6
0
.
8
1
x
1
0
4
T
i
m
e
s
[
s
]
T
h
e
R
e
a
c
t
i
v
e
S
t
a
t
o
r
P
o
w
e
r
Q
s
[
V
A
R
]
Q
s
m
e
s
Q
s
r
e
f
0
0
.
0
5
0
.
1
-
3
0
0
0
-
2
0
0
0
-
1
0
0
0
0
1
0
0
0
X
:
0
.
0
4
0
9
Y
:
0
Evaluation Warning : The document was created with Spire.PDF for Python.