Int
ern
at
i
onal
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.
52
0
~
53
1
IS
S
N:
20
88
-
8694
,
DOI: 10
.11
591/
ij
peds
.
v12.i
1
.
pp52
0
-
53
1
520
Journ
al h
om
e
page
:
http:
//
ij
pe
ds
.i
aescore.c
om
Super
-
t
wisting SMC
for MPPT
a
nd gri
d
-
connect
ed WE
CS
based o
n SCIG
Bouhen
na A
m
ina
1
,
Kh
od
j
a Moha
mmed
2
,
Merabet
Boul
ou
ih
a
H
ouari
3
1
Instit
ute of
Sc
ience
and Technol
ogy,
Dep
art
m
ent of
E
lectr
i
ca
l
En
gine
er
ing, L
GID
D L
abor
a
tory
,
A
hme
d
Za
ban
a
Univer
sity
C
ent
e
r
of
Re
lizane,
R
e
li
z
ane
,
Alger
ia
2
Depa
rtment of
El
e
ct
ri
ca
l
Eng
in
ee
ring
,
Ahm
ed Z
aba
na
Univer
si
ty
Center
of
Reliz
a
ne,
R
el
i
za
n
e
,
Al
ger
ia
3
Depa
rtment of
El
e
ct
ri
ca
l
Eng
in
ee
ring
,
Na
ti
ona
l P
olyt
ec
hni
c
Sch
ool
of
Or
an
–
Maurice
Audin
,
E
NP
O
-
M
A,
Oran
,
Alger
i
a
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
J
ul
21
, 2
0
20
Re
vised
Jan
21
, 2021
Accepte
d
Fe
b
1
1
, 20
2
1
Thi
s
p
ape
r
prop
oses
a
high
-
ord
e
r
slid
ing
mode
c
ontrol
(HO
-
SMC)
with
the
super
-
twisti
ng
(
ST)
a
lgori
th
m
f
or
m
axi
mu
m
po
wer
poin
t
tracki
ng
(MP
PT)
and
grid
-
conn
ecte
d
wind
ene
rg
y
conve
rsion
s
ystem
(W
ECS),
base
d
on
squirre
l
-
c
age
in
duct
ion
g
ene
r
at
o
r
(SCIG
).
The
ma
in
f
ea
tur
es
of
thi
s
cont
ro
l
strat
egy
are
at
t
e
nuat
ion
o
f
the
c
hat
t
eri
ng
pheno
me
non
inh
ere
nt
in
first
-
orde
r
slidi
ng
mod
e
c
ontrol
and
i
ts
robustness
ag
ai
nst
ext
ern
al
and
in
te
rna
l
disturba
nc
es
en
count
er
ed
by
t
he
wind
powe
r
sys
te
m.
Th
e
s
im
ulation
is
ca
rri
ed
out
und
e
r
SimPow
erSystem
s
of
MA
TL
AB
/
Simul
ink
to
e
val
ua
te
th
e
per
forma
n
ce
an
d
eff
e
ct
iv
ene
ss
of
th
e
propos
ed
con
trol
compare
d
wi
th
conve
nt
iona
l
an
d
fuz
zy
log
ic
pr
oporti
onal
-
integr
al
(PI)
cont
ro
ll
e
rs
for
thre
e
diffe
ren
t
sce
nar
i
os
of
d
isturba
n
c
es,
a
fluctua
ti
ng
wind
spee
d
,
a
grid
vo
lt
ag
e
drop,
and
par
ametr
i
c
va
riati
ons.
Ke
yw
or
d
s
:
IRFO
C
M
PP
T
SCIG
ST alg
or
it
hm
Sim
p
ower
s
ys
te
ms
WECS
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
:
Bouhe
nn
a
A
mina
In
sti
tute
of
Sci
ence a
nd Tec
hnol
ogy,
De
pa
rtment
of
Ele
ct
rical
Enginee
rin
g
Lab
or
at
oi
re
de Génie
I
ndus
tri
el
et Dév
el
oppe
ment
Dura
ble (
L
GIDD)
Ahmed
Za
ban
a
Unive
rsity Ce
nter of
Re
li
zan
e
P.O Bo
x,
4800
0,
B
ourma
dia,
Re
li
zane
, A
l
ge
ria
Emai
l:
amina.
bouhe
nn
a
@c
u
-
r
el
iz
ane.d
z
1.
INTROD
U
CTION
Ele
ct
rici
ty
ge
ne
rati
on
from
th
e
kin
et
ic
e
nerg
y
of
t
he
wi
nd
i
s
the
mo
st
wide
sp
rea
d
re
ne
w
able
e
nerg
y
so
urce
i
n
the
world
.
This
de
velo
pm
e
nt
is
due
to
t
he
inc
re
ase
in
s
ource
dema
nd
f
or
cl
ean
a
nd
non
-
poll
uting
el
ect
rical
energy.
Howe
ver,
the
inte
gr
at
io
n
of
m
or
e
el
ect
rici
ty
pro
duct
ion
f
rom
wind
tur
bin
es
ma
y
cause
so
me
distu
rb
a
nc
es
with
a
negat
ive
impact
on
the
sta
bili
ty
of
t
he
gri
d
a
nd
the
qual
it
y
of
the
el
ect
rical
energ
y
pro
du
ce
d
[1].
Ther
e
f
or
e,
the
dev
el
opment
of
ne
w
c
ontrol
s
trat
egies
f
or
th
e
wind
powe
r
sy
ste
m
is
bec
omi
ng
a
necessit
y,
to
m
it
igate
the
pr
oble
ms
relat
ed
to
the
qual
it
y
of
this
ene
rgy
an
d
to
e
ns
ure
t
ha
t
the
WECS
re
mains
insensi
ble
to
in
te
rn
al
an
d
e
xte
rn
al
disturba
nc
es,
w
hich
de
grade
the
sy
ste
m
performa
nce
a
nd
ma
y
eve
n
le
ad
to
instabil
it
y.
T
he
pow
ers
co
ntr
ol
in W
ECS
is
c
omplex
since
it
is
relat
e
d
to
se
ver
al
v
a
riables su
c
h
as
wi
nd
s
peed
fluctuati
ons
,
pa
rametric
var
i
at
ion
s
of
the
i
nductio
n
gen
e
r
at
or
,
w
hich
ca
n
ma
ke
a
li
ne
ar
co
ntr
ol
ine
f
fecti
ve,
and
the
s
udde
n
dro
p
of
vo
lt
age
at
the
gri
d
since
it
pr
oduc
es
hi
gh
peak
currents
on
the
el
ect
rical
ci
rc
uits,
s
o
the
c
onve
rters
may
be
dama
ge
d.
To
s
olv
e
th
ese
pr
ob
le
ms
,
and
to
e
ns
ure
t
he
rob
us
tnes
s
of
the
syst
em
a
gains
t
var
i
ou
s
disturb
ances,
ma
ny
works
a
re
pro
po
s
ed
in
the
li
te
ratur
e
[
2
]
-
[
6]
.
T
he
resea
rchers
i
n
[5]
pro
po
s
e
a
direct
to
rque
c
on
t
ro
l
(D
TC
)
t
o
est
imat
e
the
tor
qu
e
a
nd
fl
ux
f
ollo
wing
th
e
desire
d
s
pee
d,
wh
il
e
the
in
te
gr
al
back
-
ste
ppin
g
con
t
ro
ll
ers
(I
B
C)
fo
r
the
c
ontrol
of
the
ma
chine
side
c
on
ver
te
r
(
M
SC)
in
WECS
is
use
d
to
regulat
e
the
s
peed
of
the
pe
rma
nen
t
ma
gnet
s
yn
c
hro
no
us
ge
ner
at
or
(
PSMG
)
in
[
6].
Amo
ng
the
di
ff
ere
nt
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
Super
-
tw
ist
ing
SMC fo
r MPP
T an
d grid
-
co
nnect
ed W
ECS
ba
s
ed
on SCI
G
(
B
ouhe
nna
A
mina)
521
nonlinea
r
co
nt
ro
l
sc
hemes
,
the
aut
hors
in
[
7
]
-
[
9]
ha
ve
int
rod
uced
sli
di
ng
m
od
e
c
ontr
ol
(S
M
C
),
wh
ic
h
co
uld
impro
ve
t
he
dy
namic
of
t
he
WECS.
T
he
S
M
C
is
c
on
si
dered
r
obus
t
agai
nst
exter
nal
an
d
internal
dist
urb
ances,
bu
t
it
pro
du
ce
s
undesira
ble
chatt
erin
g
an
d
do
es
not
sta
bi
li
ze
per
force
in
a
finite
ti
me.
Th
us
,
in
[9]
us
in
g
nonlinea
r
pe
rtu
rb
at
io
n
obse
rvers
f
or
W
ECS
base
d
on
doub
ly
fed
in
duct
io
n
ge
ner
at
or
(
D
FI
G
),
a
r
obus
t
SM
C
is
su
ggest
e
d
to
impro
ve
fa
ult
ride
-
thr
ough
(
FRT),
w
hile
a
high
-
orde
r
S
M
C,
wh
ic
h
c
omp
rises
sec
ond
-
orde
r
(S
O
-
S
M
C)
,
is
de
velo
ped
to
mit
igate
chat
te
ring,
hen
ce
reducin
g
mec
han
ic
al
str
esse
s
an
d
e
nsure
d
bette
r
performa
nces
[
10
]
-
[
14].
I
nde
ed,
a
ma
xim
um
powe
r
point
trackin
g
(
M
P
PT)
c
on
t
ro
l
ba
sed
on
vo
lt
ag
e
-
m
od
e
seco
nd
-
or
der
f
ast
te
rmin
al
S
M
C
(SO
-
FTS
M
C)
is
pro
posed
t
o
opti
mize
the
pe
rfo
r
mance
in
the
po
we
r
extracti
on
fro
m
P
M
S
G
ba
se
d
on
WECS
[
12]
.
W
her
eas
th
e
re
gula
ti
on
of
the
powe
r
pro
duced
by
the
ge
ner
at
or
for
a
DFI
G
bas
ed
wind
t
urbin
e
(
WT)
is
stu
di
ed
us
in
g
a
S
O
-
SM
C
[
13].
Usi
ng
an
H
O
-
S
MC
ap
plied
in
th
e
gri
d
side c
onver
te
r (GS
C
)
in
case
of grid
fa
ults, the F
RT
ab
il
it
y of a
DF
I
G base
d WT
is a
ssess
ed
in
[1
4]
.
The
s
up
e
r
-
t
wi
sti
ng
sli
ding
mode
co
ntr
ol
(S
T
-
S
M
C)
is
an
e
sp
eci
al
ca
se
of
the
HO
-
SM
C.
It
was
init
ia
te
d
first
i
n
[15]
f
or
the
s
ervo
-
c
ontr
ol
syst
em
of
relat
ive
de
gr
ee
eq
ual
to
one.
Its
mai
n
a
dv
a
ntag
e
is
that
it
needs
on
l
y
i
nformat
ion
on
t
he
sli
ding
s
urfa
ce
an
d
not
it
s
der
i
vative.
It
ha
s
bee
n
us
e
d
i
n
se
ve
ral
pu
blica
ti
on
s
to
imp
rove
t
he
con
t
ro
l
of
WE
CS
[16
]
-
[
19]
.
I
n
[
20],
the
S
T
sli
din
g
mode
di
rect
powe
r
co
ntr
ol
is
sugge
ste
d
to
con
t
ro
l
th
e
act
ive
an
d
reacti
ve
powers
of
Brushle
ss
-
DFI
G
(B
DFIG
)
w
it
ho
ut
us
in
g
a
n
in
ner
cu
rr
e
nt
loop
regulat
or
a
nd
ph
a
se
-
loc
ke
d
loop.
F
urt
herm
or
e
,
a
r
ob
us
t
S
T
-
S
M
C
f
or
W
ECS
is
desc
rib
ed
i
n
[
21]
to
opti
miz
e
the
aer
odynam
ic
torque
an
d
t
he
ext
racted
powe
r.
T
o
re
gu
la
te
the
DC
-
li
nk
volt
age,
es
pecial
ly
in
gri
d
fa
ult
conditi
on
s,
[22]
presents
t
he
ST
-
SM
C
for
the
gea
rless
high
-
i
ner
ti
a
P
M
S
G
-
base
d
w
ind
t
urbine.
T
he
same
appr
oach
with
M
PP
T
c
ontr
ol
of
WECS
ba
se
d
on
DFI
G
is
pro
ved
a
ble
t
o
co
ntr
ol
t
he
ac
ti
ve
power,
t
ra
ck
t
he
maxim
um
power
po
i
nt
a
nd
regulat
e
the
re
act
ive
power
[
23],
wh
i
le
i
n
[
24],
a
n
H
O
-
S
M
C
base
d
on
the
S
T
al
gorithm
of
t
he
BDF
IG
bas
ed
WEC
S,
is
pro
posed
for
th
e
sta
to
r
side
c
onve
rter
co
ntr
ol.
H
oweve
r,
t
he
gri
d
-
side
c
onve
rter
(G
SC
)
is
s
uper
vised
by
cl
assi
cal
PI
c
on
tr
ol.
The
integ
ral
S
T
-
S
M
C
(ISTS
M
C)
is
us
e
d
in
[
25]
to
el
imi
n
at
e chatt
erin
g
a
nd imp
r
ov
e
the
r
obus
t
ness
a
nd c
onve
rg
e
nce
of the i
nductio
n
m
ach
ine s
peed.
Com
par
e
d
wit
h
t
hese
rece
nt
ly
publishe
d
r
esearche
s,
this
pa
pe
r
pro
pos
es
a
ne
w
r
ob
us
t
c
ontrol
strat
egy
of
the
global
sy
ste
m
,
al
lowing
t
he
extracti
on
of
the
ma
ximum
pow
e
r
f
rom
th
e
wind
at
lo
w
and
medium wi
nd
sp
ee
d
ra
ng
e
s
, t
he
opti
miza
ti
on of the
act
ive
powe
r
tran
sfe
r t
o
the grid
with a unity
powe
r
facto
r
and
a
fi
xe
d
D
C
-
li
nk
volt
age.
U
nlike
the
pr
evio
us
ly
ci
te
d
works,
w
hich
us
e
d
di
ff
e
ren
t
con
t
ro
ls
f
or
MSC
an
d
GS
C,
t
he
no
v
e
lt
y
of
t
his
pa
pe
r
is
the
injec
ti
on
of
S
T
-
S
M
C
i
nto
t
he
ove
rall
con
t
ro
l
s
ys
te
m,
since
t
he
va
riable
sp
ee
d
WECS
base
d
on
SC
I
G
is
a
var
ia
ble
str
uctur
e
s
yst
em
an
d
subje
ct
to
ma
ny
c
onstrai
nts
,
diff
i
cult
to
con
t
ro
l
us
in
g
ge
ner
al
c
oncept
s.
I
n
a
ddit
ion,
t
he
ass
ocia
ti
on
of
the
WECS
base
d
SC
IG
be
came
ve
r
y
at
tr
act
ive
with
t
he
im
pro
veme
nt
of
po
wer
el
ect
ronic
s
an
d
l
ow
e
r
c
os
t
of
squir
rel
-
cage
machi
nes
,
w
hic
h
a
re
of
r
obus
t
const
ru
ct
io
n
a
nd
r
eq
uiring
le
ss
mai
ntena
nc
e.
T
he
co
ntri
bu
ti
on
of
the
pa
pe
r
is
to
imp
rove
great
ly
t
he
eff
ect
ive
ness
of
the
WEC
S
base
d
on
SC
IG
with
M
PP
T
co
ntr
ol
by
impleme
nting
the
ST
-
S
M
C
for
t
he
regulat
ion
of
t
he
in
duct
ion
ge
ner
at
or
sp
ee
d,
the
DC
-
li
nk
volt
age
as
well
as
the
direct
a
nd
qu
a
drat
ur
e
c
urren
ts
of
the
gri
d.
T
he
pro
pose
d
c
ontr
ol
strat
eg
y
ens
ur
es
t
he
re
du
ct
io
n
of
t
he
chatt
eri
ng
ph
enomen
on
i
nheren
t
i
n
first
-
order
S
MC
an
d
t
he
rob
us
tness
of
the
sy
ste
m
a
gain
st
exter
nal
a
nd
i
ntern
al
dist
urba
nces
s
uc
h
as
t
he
eff
ect
s
of f
l
uctuati
ng w
i
nd s
pe
ed,
gri
d fault
conditi
ons,
a
nd
p
a
rametric
vari
at
ion
s
.
The
pa
pe
r
is
orga
nized
as
fo
l
lows
:
Sect
io
n
2
desc
ribes
t
he
wi
nd
t
urbine
modeli
ng
with
the
M
PP
T
con
t
ro
l
a
nd
the
ind
i
rect
r
otor
fiel
d
-
or
ie
nted
con
t
ro
l
(I
RF
O
C)
of
SC
IG.
S
ect
ion
3
pr
ese
nt
s
the
im
pleme
ntati
on
of
ST
-
S
M
C.
S
ect
ion
4
s
hows
the
sim
ulati
on
res
ults
of
th
e
performa
nce
a
nd
the
e
ffec
ti
ve
ness
of
the
pr
opos
e
d
con
t
ro
l
strat
eg
y
c
ompare
d
with
t
he
c
onven
ti
onal
PI
con
t
ro
ll
er,
whic
h
is
t
he
m
os
t
popula
r
s
trat
egy
impleme
nted
i
n
man
y
in
du
st
rial
ap
plica
ti
on
s
a
nd
Fu
zz
y
Lo
gic
PI
co
nt
ro
ll
er
unde
r
SimPo
we
rSystems
of
M
A
TLAB
/
Sim
ulink.
Fi
nally,
in secti
on
5,
c
oncl
us
i
on
s
are
presente
d.
2.
PROP
OSE
D CO
NFIGU
R
ATIO
N
O
F T
HE SYST
EM
Var
ia
ble
s
peed
WT
co
nfi
gura
ti
on
base
d
on
SCIG
with
the
co
ntr
ol
s
ys
te
m
is
s
how
n
i
n
Fig
ur
e
1.
A
back
-
to
-
bac
k
(
BTB
)
co
nverte
rs
to
po
l
ogy
of
con
t
ro
l
base
d
on
I
GBTs
(
Ins
ulat
ed
Gate
Bi
po
la
r
Tra
ns
ist
ors)
i
n
the
wind
po
w
er
s
ys
te
m
base
d
on
SCI
G,
w
it
h
fe
w
el
ect
ronics
c
omp
on
e
nts,
prese
nts
a
te
ch
nical
a
dvantag
e
thr
ough
la
r
ge
c
on
t
ro
ll
abili
ty
c
apacit
y
[1].
A
three
-
blad
ed
w
ind
t
urbine
is
c
oupled
to
SCI
G
th
rou
gh
a
ge
arbo
x
wh
il
e the stat
or of SCI
G
is c
onnecte
d
to t
he gri
d
via tw
o
B
TB conve
rters.
O
n o
ne
side
, th
e ST
-
S
M
C re
gu
la
te
s
the
s
pee
d
of
S
CIG
by
adj
us
ti
ng
the
ge
ner
at
or
to
rque
to
tra
ck
t
he
sp
ee
d
re
fer
e
nce
val
ue
obta
ined
by
t
he
M
PP
T
con
t
ro
l.
T
hen,
the
IR
FO
C
of
Figure
2
ge
nerat
es
the
s
wi
tc
hi
ng
sig
nals
bas
ed
on
the
te
ch
nique
of
hy
ste
resi
s
current
re
gula
tor
that
re
gu
la
t
es
the
SC
IG
volt
age
a
nd
ach
ie
ves
t
he
maxi
mu
m
powe
r
e
xtracti
on.
O
n
ano
t
her
side,
t
he
SC
IG’s
ou
t
pu
t
is
co
nv
e
rted
into
D
C
by
M
SC,
w
hi
ch
c
harges
a
c
apacit
or
C.
Th
e
gri
d
side
obje
ct
ive
s
are
to
mainta
i
n
the
DC
-
li
nk
volt
age
on
the
c
apacit
or
fi
xed
to
en
sure
c
onne
ct
ion
t
o
the
el
ect
rical
gr
i
d
w
it
h
a
un
it
y
powe
r
fa
ct
or
an
d
ma
xi
mu
m
act
ive
powe
r
tra
ns
fe
r.
The
po
wer
fl
ow
betwe
en
the
gri
d
a
nd
t
he
GS
C,
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:2
088
-
8
694
In
t J
P
ow
Ele
c
&
D
ri S
ys
t,
V
ol
.
12
, N
o.
1
,
Ma
rch
20
21
:
52
0
–
53
1
522
dr
i
ven
by
the
P
W
M
te
c
hn
i
qu
e
,
is
re
gu
la
te
d
t
hro
ugh
t
he
d
-
q
axis
gri
d
c
urre
nts
a
nd
t
he
DC
-
li
nk
vo
lt
age
c
on
t
ro
l
base
d on the
S
T
-
S
M
C.
v
W
i
n
d
S
p
e
e
d
v
dc
M
S
C
G
S
C
Σ
S
C
I
G
W
i
n
d
D
C L
i
n
k
v
abc
i
abc
v
*
dc
i
*
q
=0
T
e
C
+
-
Ω
*
Ω
i
abc
_g
Figure
1. Wi
nd turbine
syst
em
b
ase
d o
n
SC
I
G wit
h
the
con
trol s
ys
te
m
2.1. M
od
el
ing
of t
he
w
in
d tu
rbine wi
th
M
PPT co
nt
r
ol
The me
cha
nica
l ou
t
pu
t
powe
r of t
he win
d
-
t
urbine
is
g
i
ven by the
(
1)
[26
]
-
[
28]:
=
2
2
(
,
)
3
(1)
w
he
re
is
the
c
oeffici
ent
of
powe
r
co
nversi
on
a
nd
is
t
he
wind
sp
ee
d
(m
/s).
To
m
odel
(
,
)
,
wh
ic
h
is
a
functi
on
of
t
he t
ip sp
ee
d rati
o
an
d
the
b
la
de pit
ch
a
ng
le
, the
(
2)
,
(3),
(4)
c
an be
us
e
d:
(
,
)
=
0
.
5176
(
116
−
0
.
4
−
5
)
21
+
0
.
0068
(2)
1
=
1
+
0
.
08
−
0
.
03
5
3
+
1
(3)
=
Ω
t
(4)
wh
e
re
Ω
t
(
rad
/
s
)
denotes
the
mec
han
ic
al
sp
eed
of
th
e
tur
bin
e.
The
ro
t
or
ra
diu
s
=
10
.
5
m
an
d
the
ai
r
densi
ty
=
1
.
225
Kg
/m
3
.
The
maxim
um
val
ue
of
,
(
ℳ
=
0
.
48
)
is
achiev
ed
f
or
=
0
an
d
the
opti
m
al
ti
p
sp
ee
d rati
o
ℴ
=
8
.
1
.
The ma
xim
um
powe
r
ℳ
from
th
e w
in
d
is
obtai
ned th
rou
gh th
e M
P
PT c
ontr
ol
as
(
5)
:
ℳ
=
ℴ
Ω
ℴ
3
(5)
w
he
r
e
Ω
ℴ
is t
he opt
imal
mech
a
nical
sp
ee
d
a
nd
ℴ
is an
opti
mal gai
n.
2.
2
.
I
ndi
rect
r
otor
fiel
d o
ri
e
nt
ed
contr
ol of
S
CIG
The
(
6)
an
d
(
7)
de
scrib
e
th
e
sta
tor
vo
lt
ag
es
co
mpo
ne
nts
(
,
)
in
the
tw
o
-
a
xis
d
-
q
re
fer
e
nc
e
fr
ame
, whe
re
and
denote re
s
pecti
vely
t
he
st
at
or
s
ynch
r
onous s
pee
d
a
nd
th
e ro
t
or an
gula
r spee
d,
an
d
re
pr
e
sent
t
he
s
ta
tor
c
urre
nts
c
ompone
nts
i
n
d
-
q
re
fer
e
nce
f
rame.
T
he
r
otor
flu
x
in
(9)
is
or
ie
nted
on
the
d
-
a
xis
su
c
h
as
=
a
nd
=
0
[
26]
.
is
t
he
el
ect
romag
netic
to
rque
of
the
SC
IG.
Th
e
blo
c
k
diag
ra
m
of I
RF
OC
of
S
CIG
is
prese
nted
in
Fig
ure
2.
=
+
−
(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
Super
-
tw
ist
ing
SMC fo
r MPP
T an
d grid
-
co
nnect
ed W
ECS
ba
s
ed
on SCI
G
(
B
ouhe
nna
A
mina)
523
=
+
+
+
(7)
=
;
=
+
p
Ω
;
̇
=
;
=
(8)
=
̇
+
(9)
=
1
.
5
(10)
The
SC
IG’s
pa
rameters
a
re:
s
ta
tor
re
sist
anc
e
=14.
85
mΩ,
r
otor
resist
a
nce
=
9
.
295
mΩ,
ro
t
or
le
akag
e
in
du
ct
ance
=
0
.
3027
mH,
sta
to
r
le
aka
ge
in
duct
ance
=
0
.
3027
mH
,
c
yc
li
c
mu
tual
in
duct
anc
e
=
10
.
46
mH,
dis
per
si
on
coe
ff
ic
ie
nt
=
0
.
055
,
inerti
a
=
3
.
1
K
g/m
2
,
visco
us
fr
ic
ti
on
c
oe
ff
ic
ie
nt
=
0
.
08
Nm/
(r
a
d/s), nu
mb
e
r of
pole
pai
rs
=
2
and the
ra
te
d
po
wer
=
149
.
2
kW [
26].
In the stea
dy st
at
e, (9)
bec
ome
s
(
11)
:
=
1
(11)
The
def
lu
xi
ng
blo
c
in
Fig
ur
e
2,
is
def
ine
d
by
(
12)
wh
e
re
denotes
the
no
minal
flu
x
value
an
d
Ω
n
is t
he nomi
nal
ro
t
or
sp
ee
d.
=
{
,
|
Ω
|
≤
Ω
n
Ω
n
|
Ω
|
,
|
Ω
|
>
Ω
n
(12)
F
l
u
x
C
a
l
c
u
l
a
t
i
o
n
C
a
l
c
u
l
a
t
i
o
n
t
o
H
y
s
t
e
r
e
s
i
s
C
u
r
r
e
n
t
R
e
g
u
l
a
t
o
r
To
r
q
u
e
C
o
n
t
r
o
l
2
D
e
f
l
u
x
i
n
g
R
o
t
o
r
S
p
e
e
d
i
abc
i
sd
i
sq
i
sd
i
sq
p
i
sd
c
C
a
l
c
u
l
a
t
i
o
n
T
e
T
e
C
a
l
c
u
l
a
t
i
o
n
a
b
c
t
o
d
q
t
r
a
n
s
f
o
r
m
a
t
i
o
n
e
q
.
(
9
)
e
q
.
(
8
)
e
q
.
(
1
1
)
e
q
.
(
1
2
)
e
q
.
(
2
2
)
θ
s
θ
s
θ
s
θ
s
Ω
Ω
r
r
d
q
t
o
a
b
c
t
r
a
n
sf
o
r
m
a
t
i
o
n
r
r
r
i
abc
_g
p
Ω
i
sd
c
i
sq
c
i
abc
c
i
sq
c
i
sq
c
i
sd
c
Figure
2. IRF
O
C
represe
ntati
on
of
SCI
G
3.
CONTR
OL S
TRATEGIE
S
OF THE
GENER
ATO
R A
ND CON
VER
TE
RS
3.1. Hi
gh
-
orde
r sli
ding
m
od
e
co
n
tr
ol w
it
h
t
he ST
algorith
m st
ep
s
The follo
wing
ste
ps
descr
i
be how t
he
al
go
rithm
w
orks.
Step
1:
Wr
it
e
t
he
no
nlinear
dyna
mics
of
the
sy
ste
m
as
f
ollow
s:
̇
=
(
,
,
)
w
her
e
,
,
are
res
pecti
vely
the ti
me, t
he
st
at
e v
aria
ble and the
in
pu
t
of t
he
s
ys
te
m.
Step
2:
Ch
oos
e
the
sli
ding
s
urface
=
−
∗
to
ach
ie
ve
the
co
ntr
ol
obje
ct
ive
by
f
ollow
i
ng
th
e
op
ti
mal
ref
e
ren
ce
∗
.
Step
3: D
e
fine
the f
ir
st de
rivat
ive
̇
.
Step
4: Writ
e the sec
ond de
rivati
ve
̈
in t
he
f
ollow
i
ng form
̈
=
Γ
(
,
)
+
Ψ
(
,
)
̇
(
)
.
Step
5:
Use
t
he
HO
-
S
M
C
with
the
S
T
al
go
rithm
unde
r
th
e
transiti
on
co
nd
it
io
ns
:
≤
|
Ψ
|
≤
,
|
Γ
|
≤
,
>
⁄
an
d
>
√
2
(
+
)
⁄
w
her
e
,
,
M
are positi
ves
c
onsta
nts [1
3
]
,
[
14
]
,
[
16
]
-
[
25]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:2
088
-
8
694
In
t J
P
ow
Ele
c
&
D
ri S
ys
t,
V
ol
.
12
, N
o.
1
,
Ma
rch
20
21
:
52
0
–
53
1
524
{
(
)
=
1
(
)
−
√
|
|
(
)
̇
1
(
)
=
−
∙
(
)
(13)
Step
6:
I
m
plement
(13
)
a
nd
ch
oose
by
si
mu
la
ti
ons
te
st
s,
the
co
ntr
ol
gai
ns
an
d
,
wh
ic
h
ver
i
fy
the
conditi
ons cit
ed
in
step
5.
This
co
ntr
ol
la
w
co
ntains
t
wo
te
rms.
T
he
firs
t
te
rm
is
a
disc
on
ti
nu
ous
f
un
c
ti
on
of
the sli
din
g
surface
,
and
the
oth
e
r
i
s
a
co
ntin
uous
functi
on
of
it
s
der
i
vative.
Th
is
al
gorithm
does
no
t
re
qu
ire
an
y
in
formati
o
n
on
the sli
ding s
urf
ace de
rivati
ve [1
3
]
,
[
14
]
,
[
24]
.
The
disco
ntin
uous
co
ntr
ol
in
put
̇
(
)
is
act
ing
on
̈
to
ma
ke
=
̇
=
0
in
a
finite
ti
me
with
(
)
a
con
ti
nu
ous
in
put.
T
her
e
fore,
the
chatt
erin
g
in
sy
ste
m
s
is
reduce
d,
pro
vid
in
g
hi
gh
prec
isi
on
a
nd
a
voidin
g
sign
ific
a
nt mec
han
ic
al
forces
wh
il
e mai
ntain
ing
t
he ro
bustn
ess of t
he
fi
rst
-
sli
din
g m
od
e
c
on
t
ro
l.
3.2.
I
mple
men
tation o
f ST
-
S
MC f
or the c
ontr
ol of
th
e
ge
nera
to
r
speed
The
pur
pose
of
this
s
ub
-
sect
ion
is
t
o
presen
t
the
st
rateg
y
of
the
HO
-
S
M
C
with
the
ST
a
l
gorithm
t
o
regulat
e
opti
mall
y
the
SCI
G’s
sp
ee
d.
T
he
sl
iding
va
riable
Ω
is
cho
se
n
t
o
a
chieve
t
he
co
nt
ro
l
obje
ct
ive
by
fo
ll
owin
g
a
n o
ptimal
r
e
fer
e
nc
e sp
ee
d
Ω
t
∗
pro
vide
d by the
MPP
T contr
ol.
Ω
t
=
Ω
t
−
Ω
t
∗
(14)
Ω
t
∗
=
ℴ
∙
(15)
The
der
i
vative
of the slidi
ng varia
ble is:
̇
Ω
t
=
Ω
̇
t
−
Ω
̇
t
∗
(16)
The deri
vative
of the s
pee
d
Ω
t
is o
btained
from
the mec
ha
nical
equati
on of t
he
w
in
d
t
urbine:
Ω
̇
t
=
1
(
−
−
∙
Ω
t
)
(17)
w
he
re
is t
he
t
urbi
ne
to
rque.
T
hen, the
d
e
riva
ti
ve
of the
sli
di
ng v
a
riable ca
n be
re
-
w
ritt
en
a
s:
̇
Ω
t
=
1
(
−
−
∙
Ω
t
)
−
Ω
̇
t
∗
(18)
The
sec
ond de
r
ivati
ve
of t
he
s
li
din
g su
rf
ace
̈
Ω
is as
(
19)
:
̈
Ω
t
=
Γ
Ω
t
+
Ψ
Ω
t
∙
̇
(19)
wh
e
re t
he
sm
ooth
f
un
ct
io
ns
a
re
def
ine
d b
y:
Γ
Ω
t
=
̇
−
∙
Ω
̇
t
−
Ω
̈
t
∗
;
Ψ
Ω
t
=
−
1
/
(20)
Fo
ll
owin
g
the
ste
p
4,
t
he
c
ontrol
in
pu
t
is
Ω
(
)
=
al
lowi
ng
t
he
im
pl
ementat
io
n
of
t
he
ST
-
S
M
C
i
n
the prese
nce
of uncertai
nties a
nd d
ist
urban
ce
s [
13
]
,
[
14
]
,
[
16]
w
it
h t
he foll
ow
i
ng tra
ns
it
ion
c
onditi
ons:
|
Γ
Ω
t
|
≤
C
Ω
;
Ω
≤
|
Ψ
Ω
t
|
≤
Ω
w
her
e
,
,
are positi
ve
c
on
sta
nt
s.
The
f
ollo
wing
equ
at
io
ns
held
:
Ω
t
=
Ω
t
∗
,
an
d
Ω
=
Ω
∗
since
Ω
=
G
∙
Ω
t
,
∀
>
wh
e
re
is
a
fin
it
e
ti
me an
d G is the mult
ipli
er of
the g
ea
r
box.
The
co
ntr
ol in
pu
t
can b
e e
xpresse
d
by (21)
wh
e
re
the
valu
es o
f
the contr
ol
gains
a
re
Ω
=
5000
,
Ω
=
3000
.
{
Ω
(
)
=
=
Ω
1
(
)
−
Ω
√
|
Ω
|
(
Ω
)
̇
Ω
1
(
)
=
−
Ω
∙
(
Ω
)
(21)
Fr
om
(10
),
t
he qu
a
drat
ur
e
c
urr
ent
ob
ta
in
ed
ind
i
rectl
y
f
r
om
the to
rque c
ontrol
, is:
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
Super
-
tw
ist
ing
SMC fo
r MPP
T an
d grid
-
co
nnect
ed W
ECS
ba
s
ed
on SCI
G
(
B
ouhe
nna
A
mina)
525
=
(
1
.
5
)
(22)
Ther
e
f
or
e,
the
eq
uatio
n
(22)
ex
plains
ho
w
the
pro
posed
con
t
ro
l
i
nput
of
(
21)
can
ac
hieve
the
main
obje
ct
ive
of
the
M
SC
c
on
t
ro
l.
Indee
d,
the
IRFO
C
,
r
epr
ese
nted
i
n
Figures
1
a
nd
2,
use
s
the
(
22
)
with
(11)
to
ge
nerat
e
the
s
witc
hi
ng
sig
nals
to
se
nd
to
M
SC,
th
rough
t
he
hyste
r
esi
s
cu
rr
e
nt
re
gu
la
to
r.
T
his
st
rateg
y
al
lows
t
he
re
gu
la
ti
on
of
SCI
G vo
lt
age
and
ext
racti
on of m
axi
mu
m
po
wer.
3.3.
I
mple
men
tation o
f ST
-
S
MC a
l
go
ri
th
m
for the c
ontro
l of the
DC
-
li
nk
volt
age
an
d
th
e
grid c
urre
nt
s
The
c
ontr
ol
ap
proac
h
on
the
gr
i
d
si
de
is
s
how
n
i
n
Fig
ur
e
3.
T
wo
ST
-
S
M
C
m
ana
ge
t
he
casca
de
d
loop.
T
he
oute
r
l
oop
re
gula
te
s
the
DC
-
li
nk
vo
lt
age
wh
il
e
the
i
nn
e
r
lo
op
adjusts
the
dir
ect
cu
rr
e
nt
of
t
he
gr
i
d
,
w
hich
co
ntr
ols
th
e
act
iv
e
po
wer
P
.
The
re
fer
e
nce
di
re
ct
cu
rr
e
n
t
∗
is
th
e
outp
ut
of
t
he
ou
te
r
ST
-
SM
C
.
The
qua
dr
at
ure
co
mpo
ne
nt
c
urren
t
,
wh
ic
h
con
t
ro
ls
the
re
act
ive
power
Q
is
al
so
regul
at
ed
by
a
S
T
-
SM
C.
The
ref
e
re
nce
qu
a
drat
ur
e
c
urren
t
∗
is
set
t
o
zero
to
ac
hie
ve
a
un
it
y
po
w
er
fact
or
an
d
the
DC
-
li
nk
vo
lt
a
ge
ref
e
ren
ce
is
set
to
10
00
V
.
Th
e
P
WM
us
es
the
outp
ut
of
th
e
cu
rr
e
nt
regul
at
or
s
an
d
co
mp
e
ns
at
ed
by
two
dec
ouplin
g
te
rms,
to
ge
ner
at
e
the
swit
chin
g
si
gnal
s
and
dr
i
ve
t
he
GS
C.
T
her
e
by,
the
pro
pose
d
con
t
ro
l
sy
ste
m
pr
oduc
es
the
m
axim
um
act
ive
po
wer
to
the
gr
i
d.
I
n
or
der
to
s
yn
c
hro
nize
t
he
WEC
S
t
o
t
he
gri
d
vo
lt
age
s,
a
ph
ase
-
loc
ked
lo
op
(
PLL
)
is
use
d
to
e
sti
mate
the
insta
ntan
eous
phase
.
The
pur
po
se
is
to
trans
form
t
he
gri
d
si
gn
al
s
i
n
the
d
-
q
re
fer
e
nc
e
fr
a
me
in
orde
r
to
s
ynch
r
on
iz
e
the
phase
of
this
re
fer
e
nce
fram
e
with
t
he
ro
ta
ti
ng
fiel
d
of
the
el
ect
rical
gr
i
d.
The
Par
k's
tra
nsfo
rmati
ons
al
low
a
dequate
c
on
t
ro
l
of
t
he
s
ys
te
m
on the
gri
d
si
de
[24
]
,
[
26].
i
q
i
d
v
dc
*
v
dc
Σ
Σ
Σ
P
Q
v
abc
i
abc
i
q
*
=0
i
d
*
-
-
+
+
-
+
-
L
f
+
-
R
f
v
dc
t
r
a
n
s
f
o
r
m
a
t
i
o
n
t
r
a
n
s
f
o
r
m
a
t
i
o
n
F
i
l
t
e
r
θ
v
gd
v
gq
θ
-
R
f
i
d
+
L
f
ω
i
q
+
v
d
G
S
C
Σ
Σ
-
R
f
i
q
-
L
f
ω
i
d
+
v
q
+
Figure
3. Sc
he
me of t
he
c
on
t
r
ol of
gr
i
d
c
urre
nts a
nd
DC
-
li
nk
vo
lt
age
The v
oltage e
quat
ions, ac
ross
the in
du
ct
or
in the d
-
q refe
rence
f
ra
me
[24
]
,
[
28]
, a
re as
(
23)
:
{
(
)
=
−
+
+
−
(
)
=
−
(
−
−
+
)
+
(23)
wh
e
re
the
filt
er
induct
a
nce
a
nd
resist
ance
a
re
res
pecti
vely
=
2
mH,
=
0
.
2
Ω;
,
repr
esent
the
gri
d
vo
lt
age
s
co
mpon
e
nts;
,
are
the
gri
d
c
urre
nts
com
pone
nts.
The
nominal
gr
i
d
volt
age
a
nd
fr
e
quenc
y
a
r
e
=
460
V
,
=
50
Hz
[
26].
Figure
4
s
how
s
the
sc
heme
f
or
re
gula
ti
ng
t
he
DC
-
li
nk
vo
lt
age
,
w
hich
va
ries acc
ordi
ng to
the
powe
r e
xch
a
nged
b
et
ween t
he
t
urbi
ne
a
nd the
gri
d.
Its
dyna
mic i
s
r
e
pr
ese
nted
by
(
24)
:
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:2
088
-
8
694
In
t J
P
ow
Ele
c
&
D
ri S
ys
t,
V
ol
.
12
, N
o.
1
,
Ma
rch
20
21
:
52
0
–
53
1
526
v
dc
i
dc
i
1
i
2
i
sa
i
sb
i
sc
M
S
C
G
S
C
P
u
l
s
e
s
1
P
u
l
s
e
s
2
R
f
L
f
v
a
v
b
v
c
Figure
4. Sc
he
me of t
he DC
-
l
ink
=
1
−
2
(24)
The
sli
ding
va
riables
=
−
∗
,
=
−
∗
and
=
−
∗
are
ch
os
en
t
o
achie
ve
the
con
t
ro
l o
bject
ives by
f
ollo
wing
opti
mal
re
fe
ren
ces
de
fine
d
resp
ect
ivel
y
by
the
DC
-
li
nk volt
age
∗
,
the
dir
ect
current
∗
an
d
t
he
quadrat
ure c
urren
t
∗
.
T
he
n,
t
he deri
vatives
of the slidi
ng varia
bles a
re:
{
̇
=
1
(
1
−
2
)
−
̇
∗
̇
=
(
1
−
+
−
1
)
−
∗
̇
=
(
1
−
−
−
1
)
−
∗
(25)
The p
ow
e
r bala
nce
giv
es
[24]
:
∙
2
=
∙
+
∙
(26)
The
sec
ond de
r
ivati
ves of
t
he
sli
din
g su
rf
ace
ob
ta
ine
d fro
m
(25) an
d (
26)
a
re a
s
f
ollow
s:
{
̈
=
Γ
+
Ψ
∙
̈
=
Γ
+
Ψ
∙
̇
̈
=
Γ
+
Ψ
∙
̇
(27)
w
he
re
t
he
sm
ooth
f
un
ct
io
ns
a
re
def
ine
d b
y:
{
Γ
=
̇
;
=
1
1
−
̇
∗
−
∙
;
Ψ
=
−
∙
Γ
=
̇
;
=
1
+
−
∗
−
;
Ψ
=
−
1
Γ
=
̇
;
=
1
−
∗
−
−
;
Ψ
=
−
1
(28)
The
S
T
-
S
M
C
i
s
impleme
nted
in
the
pr
e
senc
e
of
uncertai
nt
ie
s
and
distu
r
ba
nces
[13
]
,
[
14
]
,
[
16]
with
the foll
owin
g
t
ran
sit
io
n
c
ondi
ti
on
s:
|
Γ
|
≤
,
≤
|
Ψ
|
≤
,
|
Γ
|
≤
,
≤
|
Ψ
|
≤
,
|
Γ
|
≤
,
≤
|
Ψ
|
≤
w
her
e
,
,
,
,
,
,
,
,
are
po
sit
ives
const
ants.
Th
us
,
=
∗
,
∀
>
,
=
∗
,
∀
>
an
d
=
∗
,
∀
>
w
he
re
,
an
d
are f
i
nite t
imes.
T
he
c
on
tr
ols i
nputs
(
)
,
(
)
an
d
(
)
, r
e
presente
d
in
Fig
ur
e
3, are
as
fol
lows
:
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
Super
-
tw
ist
ing
SMC fo
r MPP
T an
d grid
-
co
nnect
ed W
ECS
ba
s
ed
on SCI
G
(
B
ouhe
nna
A
mina)
527
{
(
)
=
∗
(
)
=
1
(
)
−
√
|
|
(
)
̇
1
(
)
=
−
(
)
(
)
=
1
(
)
−
√
|
|
(
)
̇
1
(
)
=
−
(
)
(
)
=
1
(
)
−
√
|
|
(
)
̇
1
(
)
=
−
(
)
(29)
The val
ues
of t
he
c
on
t
ro
l
gain
s ar
e:
=
2
,
=
0
.
5
,
=
10
,
=
100
,
=
150
,
=
100
.
4.
RESU
LT
S
AND DI
SCUS
S
ION
The
sim
ulati
on
of
the
H
O
-
S
M
C
with
the
S
T
al
gorithm
f
or
the
WECS
ba
sed
on
SCI
G
is
performe
d
us
in
g
SimP
owerS
ys
te
ms
of
M
A
TLAB
/
Sim
ulink.
The
set
ti
ng
s
of
al
l
ST
-
SM
C
pa
ramete
rs,
w
hich
ver
if
y
t
he
conditi
ons
ci
te
d
in
ste
p
5,
a
re
ch
os
e
n
by
simulat
ion
s
te
sts.
More
ov
er
,
in
Fig
ur
e
3,
the
casca
de
lo
op
is
mana
ged
by
tw
o
S
T
-
S
M
C,
whic
h
c
on
sist
s
of
two
imb
ricat
e
loops
w
here
the
in
ner
lo
op
is
f
ast
er
tha
n
t
he
oute
r
loop.
T
he
str
uc
ture
of
P
I
c
ontrolle
rs
is
(
)
=
+
/
wh
e
re
=
5000
,
=
10
,
=
8
,
=
400
,
=
5
,
=
500
,
=
5
,
=
500
.
Th
ree
sce
na
r
ios
are
co
ns
id
ered
t
o
validat
e
the
perform
ance
and
the
feasi
bili
ty
of
the
ST
-
S
M
C
c
ompare
d
to
co
nve
ntion
a
l
and
f
uzz
y
l
ogic
(F
L
)
P
I
c
ontrolle
rs
.
F
uzz
y
Gain
-
Sche
du
li
ng
c
ontr
oller
use
d
for
M
SC
o
pe
rates
with
F
uz
zy
r
ules
a
nd
ex
plo
it
s
hu
man
knowle
dge
a
nd
exp
e
rience
in
c
on
t
ro
l
to
set
t
he
P
I
c
ontr
oller
par
a
mete
rs
[
29
],
w
hile
t
he
F
uz
zy
A
dap
ti
ve
PI
co
ntr
oller
a
pp
li
ed
for
GS
C,
a
re
an
a
sso
ci
at
io
n
of
c
onve
ntio
na
l
PI
co
ntr
oller
a
nd
fu
zz
y
lo
gic
c
on
t
ro
ll
er
(F
LC)
[
30]
.
It
ta
kes
adv
a
ntage
s
of
PI
c
ontr
oller
a
nd
the
ada
ptiv
e
capa
bili
ty
of
FLC.
T
he
fir
st
case
represe
nted
in
Fig
ur
e
5(a)
pro
po
ses
a
fl
uc
tuati
ng
wind,
operati
ng
unde
r
no
rmal
co
ndit
ion
s
w
hile
in
the
seco
nd
case,
re
pr
e
sent
ed
in
Figure
5
(
b),
a
su
dde
n
dr
op
i
n
gr
i
d
volt
age
of
30%
in
t
he
ti
me
inter
val
be
tween
2
an
d
4
s,
at
wind
no
minal
sp
ee
d
of
12
m
/s
is
simulat
ed
.
I
n
the
t
hir
d
case,
pa
ramet
r
ic
var
ia
ti
ons
of
the
SCIG
ar
e
introd
uced
a
t
wind
nominal s
pee
d.
(a)
(b)
Figure
5. C
urv
es of
fluctuati
ng
wind s
pee
d
a
nd g
rid
volt
age
drop,
(a
)
F
l
uctuati
ng w
i
nd s
pe
ed
, (
b)
G
rid
vo
lt
age
dr
op
4.1. Fl
uctua
ting
wind
speed
The
si
mu
la
ti
on
is
do
ne
i
n
6
s
,
and
the
res
ults
are
pr
ese
nted
i
n
Fi
gures
6
an
d
7.
Fig
ur
e
6(
a
)
s
hows
the
fast
dyna
mic
r
esp
on
se
of
t
he
sp
eed
a
nd
a
very
go
od
pe
r
forma
nce
f
or
t
he
traj
ect
ory
t
rack
i
ng,
with
high
accurac
y
des
pite
wind
s
peed
var
ia
ti
ons
for
t
he
ST
-
S
M
C
c
ompare
d
t
o
P
I
co
ntro
ll
er
s.
T
hat
mea
ns
t
he
error
betwee
n
t
he
a
ct
ual
a
nd
re
fere
nce
sp
ee
d
is
zero,
w
hich
e
ns
ure
d
t
he
i
de
al
trans
fer
of
t
he
maxi
mum
act
ive
powe
r
est
ablis
hed
by
t
he
M
P
P
T
co
ntr
ol
to
the
gr
i
d.
Fi
gure
6
(
b)
s
hows
t
he
ST
-
SM
C
pe
r
forma
nces
c
ompa
red
to
the
co
nvent
ion
al
PI
c
ontr
ollers
an
d
the
Fu
zz
y
ada
ptiv
e
PI
co
ntr
oller
.
Indee
d,
firstly
a
sm
oo
t
her
curve
is
ob
ta
ine
d
since
chat
te
rin
g
is
mit
igate
d
he
nce
reducin
g
mecha
nical
str
esses.
Sec
ondl
y,
DC
-
li
nk
vo
lt
age
rob
us
tness
a
ga
inst
wind
fl
uc
tuati
ng
is
e
nsure
d,
wh
ic
h
guara
nteed
t
hat
al
l
the
act
ive
power
is
i
ns
ta
ntly
trans
ferred
to
t
he
gr
i
d.
Thir
d,
the
lo
w
pea
k
va
lue
of
t
he
S
T
-
SM
C
is
cl
early
visible,
th
us
preser
ving
t
he
s
afety
of the elec
tric
a
l system.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:2
088
-
8
694
In
t J
P
ow
Ele
c
&
D
ri S
ys
t,
V
ol
.
12
, N
o.
1
,
Ma
rch
20
21
:
52
0
–
53
1
528
(a)
(b)
Figure
6. C
urv
es of the
g
e
ne
r
at
or
s
pee
d
a
nd
the D
C
-
li
nk vo
lt
age
in case
of
f
luct
uating wi
nd,
(a)
G
e
ner
at
or
sp
ee
d,
(b) DC
-
li
nk
vo
lt
ag
e
In
Fig
ur
e
7(
b)
,
a
zer
o
reacti
ve
powe
r
is
pro
vid
e
d
by
the
S
T
-
S
M
C,
re
gardless
of
rip
ples
pro
duced
b
y
PWM,
av
oi
ding,
the
refor
e
,
al
te
ring
the
qu
al
i
ty
of
the
gr
id
,
wh
il
e
t
he
t
wo
PI
c
ontr
ollers
exh
i
bited
fluct
uations
arou
nd
ze
ro.
U
sin
g
t
he
M
PPT
strat
egy,
the
wa
v
e
form
of
the
act
ive
po
wer
of
Figure
7(
a
),
de
li
ver
ed
t
o
the
gr
i
d
is
simi
la
r
to
the
wi
nd
sp
ee
d
cu
r
ve
of
Figure
5(
a
),
wh
ic
h
mea
ns
a
small
change i
n
th
e
wind in
duces a
sig
nificant c
ha
ng
e
in
t
he ou
t
put act
ive
powe
r.
(a)
(b)
Figure
7. Cu
rves of the
acti
ve
and
reacti
ve p
ow
e
rs
i
n
case
of f
l
uctuati
ng
w
ind
,
(a)
A
ct
ive
pow
er,
(b)
R
eact
iv
e pow
e
r
4.2. Gri
d
vo
l
t
ag
e
dr
op
The
sim
ulati
on
of
a
gr
i
d
vo
lt
age
dro
ps
up
to
30%
of
t
he
gr
i
d
nominal
volt
age
is
ca
rr
ie
d
ou
t
i
n
6
s
betwee
n
t
ϵ
[2
]
,
[
4]
s
.
T
he
res
ults
are
pr
ese
nt
ed
in
Fig
ur
es
8
and
9.
F
or
t
he
sp
ee
d
trac
king
of
Fig
ur
e
8
(a
)
,
the
ST
-
S
M
C
ca
n
track
t
he
M
PPT
op
ti
mal
s
peed
with
hi
gher
ac
cur
ac
y
a
nd
fast
dynamic
res
ponse
as
co
m
pa
red
t
o
the
co
nv
e
ntio
na
l
PI
an
d
the
Fu
zz
y
Lo
g
ic
PI
co
ntr
ollers
.
In
Fig
ur
es
8
(b),
9(
a
),
a
nd
9(b),
co
rr
es
pondin
g
resp
ect
ivel
y
t
o
the
DC
-
li
nk
vo
lt
age
,
t
he
a
ct
ive
po
wer
,
a
nd
the
reacti
ve
powe
r,
the
ST
-
S
M
C
pro
vi
des
a
smoothe
r
c
urv
e
co
mp
a
re
d
t
o
PI
co
ntr
ollers,
pa
rtic
ularly
at
ti
mes
of
su
dde
n
va
riat
ion
s
in
gr
i
d
volt
age
.
Fu
rt
h
er
more,
i
n
Fig
ur
e
8(b
)
the
pea
k
value
of
the
DC
-
li
nk
vo
lt
age
reac
he
s
1500
V
for
the
co
nv
e
ntio
na
l
PI
con
t
ro
ll
er,
wh
i
le
for
F
uzzy
a
dap
ti
ve
P
I
c
on
trolle
r
a
nd
ST
-
SM
C,
it
is
le
ss
tha
n
1250
V.
No
te
that
t
he
high
-
peak val
ue of
the
DC
-
li
nk
vo
l
ta
ge,
ca
us
e
d by the c
onve
nti
onal
P
I
c
on
tr
oll
er,
ca
n dama
ge
the c
onve
rters.
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
Super
-
tw
ist
ing
SMC fo
r MPP
T an
d grid
-
co
nnect
ed W
ECS
ba
s
ed
on SCI
G
(
B
ouhe
nna
A
mina)
529
(a)
(b)
Figure
8. C
urv
es of the
g
e
ne
r
at
or
s
pee
d
a
nd
the D
C
-
li
nk vo
lt
age
in case
of
drop i
n gr
id
volt
age,
(a)
G
e
ner
at
or
s
peed,
(b) DC
-
li
nk volt
age
(a)
(b)
Figure
9. Cu
rves of the
acti
ve
and
reacti
ve p
ow
e
rs
i
n
case
of
drop in
gri
d v
oltage,
(a)
A
ct
ive
pow
er, (b)
R
eact
iv
e pow
e
r
4.3. Par
ame
te
rs varia
tions
This
sce
nar
i
o
us
es
uncertai
nt
ie
s
includi
ng
va
riat
ion
s
i
n
the
ro
t
or
resist
anc
e,
the
i
ner
ti
al
mo
me
nt,
a
n
d
the
fr
ic
ti
on
c
oe
ff
ic
ie
nt
of
t
he
SCIG
up
to
80%
of
t
heir
no
minal
val
ues
be
tween
t
ϵ
[
0,
2]
s
.
T
he
sim
ul
at
ion
resu
lt
s
are
pres
ented
in
Fig
ur
e
s
10
a
nd
11.
A
s
show
n
in
Fi
gur
e
10
(a
),
in
a
ste
ady
sta
te
,
a
sp
ee
d
er
ror
re
mains
for
the
t
wo
PI
con
t
ro
ll
ers
,
unli
ke
th
e
ST
-
S
M
C.
Durin
g
t
he
tra
ns
ie
nt
re
sp
onse
,
the
ris
e
ti
me
of
t
he
sp
ee
d
towa
rd
s
the
refe
ren
ce
sp
ee
d,
f
or
the
c
onve
ntion
al
a
nd
F
uzz
y
P
I
c
ontrolle
r
s
increa
sed
co
ns
ide
rab
l
y
c
ompa
re
d
with
the
ST
-
S
M
C.
This
le
a
ds
to
sl
owness
i
n
the
power
tra
ns
fe
r
from
the
wind
tu
r
bi
ne
t
o
the
gr
id
as
s
how
n
in
Figures
11(a
)
a
nd
11
(b).
Mor
eov
e
r,
in
Fig
ure
10
(
b),
rep
et
it
ive
pea
k
val
ue
s
ap
pear
on
t
he
DC
-
li
nk
volt
age
for
bo
t
h
P
I
co
ntr
ol
le
rs,
w
hile
th
e
ST
-
S
M
C
ha
s
a
small
er
overshot
tha
n
th
e
co
nv
e
ntio
nal
PI
c
on
t
ro
ll
er
do
e
s
.
Ther
e
f
or
e,
t
he
P
I
c
ontr
ollers
ha
ve
poor
pe
rformance
duri
ng
tra
ns
ie
nt
operati
on
an
d
s
peed
i
naccurac
ie
s
in
ste
ady stat
e, c
ompa
red to t
he ST
-
S
M
C.
The
S
T
-
S
M
C
i
mp
le
me
ntati
on
on
the
overall
wind
power
syst
em
is
a
nota
bl
e
con
tri
bu
ti
on
an
d
a
new
aspect
of
t
he
obta
ine
d
res
ults.
T
hi
s
no
nl
inear
c
ontr
ol
ma
de
it
pos
sible
to
gu
a
ra
ntee
t
he
sa
me
go
od
performa
nces
in
the
th
ree
dis
tur
ban
ce
sce
na
rios,
w
hich
ar
e
fluct
uating
wind
s
pee
d,
gri
d
volt
age
dro
ps
a
nd
par
a
mete
rs va
r
ia
ti
on
s
of
t
he
S
CIG. T
hese
res
ults are,
the
refor
e
, non
-
obvi
ous t
o ob
ta
in
simult
ane
ously.
Evaluation Warning : The document was created with Spire.PDF for Python.