Int
ern
at
i
onal
Journ
al of
P
ower E
le
ctr
on
i
cs a
n
d
Drive
S
ystem
(I
J
PE
D
S
)
Vo
l.
11
,
No.
3
,
Septem
be
r 2020
, pp.
15
8
8
~
1
59
5
IS
S
N:
20
88
-
8694
,
DOI: 10
.11
591/
ij
ped
s
.
v11.i
3
.
pp
15
8
8
-
1
59
5
1588
Journ
al h
om
e
page
:
http:
//
ij
pe
ds
.i
aescore.c
om
T
he enh
anceme
nt of po
wer
qualit
y for
t
he
distri
bution syst
em
via dy
namic volt
age r
esto
rer
Ali B
as
im
M
ohamme
d, M
ohd
Aifaa M
ohd Ari
ff
Facul
ty
of Electr
ic
a
l
and
E
le
c
tron
ic
Engi
ne
eri
ng
,
Univer
sity Tun H
uss
ei
n
Onn Mal
aysia (UTHM),
Mala
ysia
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
Feb
20
, 20
20
Re
vised
A
pr
4
,
20
20
Accepte
d
Apr
1
6
, 20
20
Thi
s
pap
er
rep
r
ese
nts
a
new
co
nfigura
t
ion
of
t
he
dyna
mi
c
voltage
r
estore
r
consists
of
appr
oxim
ate
cl
assic
a
l
slidi
ng
mod
e
d
iffe
ren
ti
a
tor
(AC
SM
D)
with
the
termin
al
sli
ding
mode
con
trol
ler
(TSMC)
as
th
e
nonli
n
ea
r
slid
ing
var
ia
b
le
.
In
thi
s
study,
the
pro
posed
struc
ture
of
th
e
DV
R
is
uti
lized
to
ma
intain
th
e
m
a
gnit
ude
of
the
lo
ad
vol
ta
g
e
at
a
c
onstant
v
al
ue
,
m
ai
nt
ai
n
the
sys
te
m
t
ot
al
har
moni
c
distortion
(THD),
boost
t
he
robustness
pr
oper
ty
and
mi
nimize
th
e
st
e
ady
-
state
err
or.
The
powe
r
qu
al
i
ty
has
recei
v
ed
more
int
e
rest
due
to
th
e
im
p
lementati
on
of
var
ious
industrial
d
evi
c
es
and
cr
it
i
c
al
loa
ds
at
the
distr
ibut
ion
side.
Now
ada
ys,
t
he
ma
in
chall
e
nges
in
power
q
ual
it
y
in
the
sys
te
m
are
voltage
sags/sw
el
ls,
har
moni
cs
and
voltage
i
mba
l
a
nce
.
Var
ious
devi
c
es
ar
e
ut
il
i
z
ed
to
addr
ess
th
e
se
challe
ng
es.
T
he
dyna
mi
c
voltage
r
estore
r
is
one
of
the
se
d
evi
c
es.
It
is
con
nec
t
ed
in
ser
ie
s
with
the
d
istri
bu
ti
on
sys
te
m
and
inject
s
a
pr
oper
voltage
magnitude
to
main
ta
in
the
vol
ta
g
e
loa
d
a
t
th
e
consta
nt
val
u
e.
I
n
thi
s
p
ape
r
,
the
DV
R
mode
l
wi
th
th
e
AS
MF
an
d
TSMC
is
im
plemented
in
using
MA
TLA
B/Sim
uli
nk.
The
propos
ed
cont
roller
is
eva
lu
at
ed
using
t
he
stand
a
rd
vo
ltage
sag
ind
ic
es
.
Ke
yw
or
d
s
:
Dynamic
volt
age
resto
rer
Terminal
sli
ding m
ode
Power q
ualit
y
Total
h
a
rm
onic
d
ist
ort
ion
Vo
lt
age
sa
g/swe
ll
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
:
M
oh
d Aifaa
Mohd
Ar
i
ff
,
Faculty
of Elec
tric
al
an
d El
ect
ronic E
ng
i
neeri
ng
,
Un
i
ver
sit
y T
un Hussein
On
n Malaysia
(
UT
HM),
86400 Pa
rit R
aj
a, Joh
or
,
M
al
aysia
.
Emai
l:
aifaa@
uth
m
.edu.
my
1.
INTROD
U
CTION
In
c
reasin
g
t
he
low
qual
it
y
of
loads
an
d
s
our
ces
de
gr
a
des
the
powe
r
qu
al
i
ty
in
both
el
ect
rical
util
it
ie
s
and
c
us
to
mers
.
This
mea
ns
th
at
the
el
ect
rici
ty
s
uppl
y
sy
ste
m
ma
y
ha
ve
unwa
nted
f
luctu
at
ion
s
an
d
dist
or
ti
ons,
especial
ly
i
n
it
s
f
reque
ncy
a
nd
volt
age
[1,
2].
Powe
r
qual
it
y
has
tw
o
vie
wpoints;
from
the
sid
e
of
the
su
ppl
y
of
el
ect
rical
po
wer
with
the
s
pecified
sta
nda
rd
s
an
d
from
t
he
u
ser
e
nd.
It
means
that
el
e
ct
ric
powe
r
ha
s
to
be
delivere
d
t
o
t
he
load
a
nd
eq
uip
me
nt
with
out
an
y
dis
ruption.
T
he
powe
r
qual
it
y
is
facing
a
wide
ra
nge
of
disturba
nces
li
ke
har
m
onic
s,
sag/swell
,
noise
,
inter
ruptio
n,
vo
lt
a
ge
imb
al
ance,
a
nd
t
r
ansients.
I
n
order
t
o
address
t
hese
chall
enges,
a
dynamic
vo
lt
a
ge
rest
or
e
r
(
D
VR)
ma
y
be
ut
il
iz
ed
to
keep
issues
f
or
t
he
power
qu
al
it
y
s
uc
h
as
sagg
i
ng
/s
well
ing
of
volt
ages
and
oth
e
r
dist
urba
nces
at
a
minimu
m
val
ue
to
protect
ap
pliances
[3
-
7].
The
D
V
R
has
been
use
d
in
pract
ic
e
to
mit
igate
the
vo
lt
a
ge
qual
it
y
issue
in
th
e
powe
r
distri
buti
on
sy
ste
m.
The
D
VR
is
instal
le
d
bet
wee
n
the
gr
i
d
an
d
t
he
sensiti
ve
loa
ds
thr
ough
a
tra
ns
f
ormer
to
in
je
ct
the
require
d vo
lt
a
ge
magnit
ude t
o com
pe
ns
at
e f
or the
volt
age
di
ff
e
ren
ce
[8
–
10
].
Con
tr
oll
ing
a
DV
R
requires
a
soph
ist
ic
at
ed
co
ntr
ol
s
ys
te
m
su
c
h
as
sli
ding
m
ode
co
ntr
ol.
I
n
thi
s
pap
e
r,
t
he
sli
di
ng
m
ode
c
on
t
ro
l
is
use
d
t
o
con
t
ro
l
the
D
VR
in
re
gula
ti
ng
t
he
volt
age
in
the
s
ys
te
m
.
The
sli
din
g
m
od
e
con
t
ro
l
is
a
de
cent
br
a
nch
of
co
ntr
oller
dea
li
ng
wi
th
nonl
inear
a
ppli
cat
ion
s
.
Its
respo
nse
is
qu
ic
k,
eas
y
to
impleme
nt,
an
d
r
obus
t
t
ow
a
r
ds
the
va
riat
ion
of
op
e
rati
ng
par
a
mete
rs.
I
n
[11],
quasi
Z
s
ource
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
Th
e e
nhance
m
ent o
f
powe
r
qualit
y for
the
di
stri
bu
ti
on syst
em
vi
a dy
nami
c volta
ge
…
(
A
li
Ba
sim
Mo
ha
mm
e
d
)
1589
inv
e
rter
ba
sed
sing
le
-
phase
D
VR
is
pr
ese
nte
d.
I
n
the
pa
per,
the
DV
R
w
orks
r
obus
tl
y
a
nd
re
du
ce
s
the
s
ys
te
m
total
harmo
nic
disto
rtion
(T
HD)
at
the
lo
ad
side
.
T
he
method
al
s
o
minimi
zes
the
require
ment
of
e
ne
rg
y
stora
ge.
T
he
s
li
din
g
m
od
e
c
on
t
ro
l
-
base
d
DV
R
with
12
-
switc
h
VSI
is
re
ported
in
[
12].
I
n
this
pa
per,
t
he
ref
e
ren
ce
vo
lt
a
ge
is
gen
e
rated
by
util
iz
ing
th
e
ad
a
ptive
no
tc
h
filt
er.
The
n,
a
sli
ding
m
od
e
co
ntr
oller
bas
ed
f
or
a
three
-
phase
DV
R
is
re
port
ed
in
[
13]
to
r
egu
la
te
the
volt
age
at
the
distribu
ti
on
loa
d.
The
a
uthors
in
this
pap
e
r
util
iz
e
a
rec
ur
si
ve
le
as
t
square
al
gori
thm
t
o
gen
e
ra
te
the
r
efe
ren
c
e
sync
hron
iz
at
ion
.
M
os
t
of
t
hese
pap
e
rs
util
iz
e
a
li
near
sli
ding
va
riable
with
a
li
nea
r
di
fferentia
to
r
to
obta
in
the
e
rro
r
functi
on
der
iv
at
ive
in
pro
vid
in
g
a
c
orres
pondin
g
con
t
ro
l
act
io
n.
The
a
pp
li
cat
ion
of
t
he
li
ne
ar
dif
fer
e
ntiat
or
is
li
mit
ed
as
it
is
sensiti
ve
to
wa
r
d
the
prese
nce
of noise t
he
me
asur
e
d
i
nput si
gn
al
.
This
pa
pe
r
pre
sents
a
new
D
VR
meth
od
us
ing
a
n
a
pproxi
mate
cl
assic
al
sli
din
g
mode
di
ff
ere
ntiat
or
with
a
nonline
ar
sli
di
ng
var
ia
ble
to
ov
e
rc
ome
the
dra
wb
ac
k
of
t
he
li
near
di
ff
ere
ntiat
or.
T
he
nonlinea
r
sli
din
g
var
ia
ble
is
ba
s
ed
on
the
a
pp
li
cat
ion
of
the
te
rmin
al
sli
ding
mode
co
ntr
oller
t
o
mainta
in
t
he
ma
gn
it
ude
of
the
load
vo
lt
a
ge
at
the
co
ns
ta
nt
va
lue,
im
prov
e
t
he
s
ys
te
m
T
H
D,
boos
t
the
r
obus
t
ness
pro
pe
rty
a
nd
minimi
ze
the
ste
ady
-
sta
te
error
of
the
c
ontr
oller
outp
ut.
T
he
pro
pose
d
m
et
hod
of
D
VR
is
app
li
ed
a
nd
inv
est
igate
by
us
in
g
M
A
TLAB/Si
m
ulink.
The
struct
ur
e
of
this
pa
per
c
on
sist
s
of
f
our
sect
ions.
t
he
method
ology
r
epr
ese
nts
a
s
a
n
ap
pro
ximate
sli
din
g
m
ode
di
ff
ere
ntiat
or
an
d
nonlinea
r
sli
ding
var
ia
ble
a
s
a
te
r
m
inal
sli
ding
mode,
the
D
VR
st
ru
ct
ure,
the
pro
po
se
d
s
ys
te
m
of
the
D
VR
,
the
cal
culat
io
n
of
vo
lt
a
ge
sa
g
a
nd
t
he
in
di
ces
of
volt
age
sag
a
re
disc
us
s
ed
in
Sect
ion
2,
the
resu
lt
s
a
nd
dis
cussion
with
t
he
performa
nc
e
evaluati
on
a
r
e
pr
e
sente
d
in
Sect
ion
3.
Sec
ti
on
4
discusse
s the
c
on
cl
ud
i
ng r
e
m
ark
s
of t
his stu
dy.
2.
METHO
D
OL
OGY
2.1.
A
ppr
oxim
at
e
classic
al
sli
ding
m
od
e
differen
tia
to
r
In
this
meth
od,
t
he
chatt
eri
ng
at
ti
tud
e
in
the
sli
ding
mode
c
ontr
oller
a
nd
obser
ve
r
is
us
e
d
to
disti
nguish
the
cha
racteri
sti
c
in
the
meas
ur
e
ment
t
o
prov
i
de
the
a
ppr
opria
te
con
t
ro
l
act
i
on.
A
n
a
ppr
ox
i
mati
on
cl
assic
al
sli
din
g
m
od
e
dif
fer
e
ntiat
or
is
us
ed
to
dif
fer
e
ntiat
e
this
at
tribu
te
,
al
lowing
t
he
measu
reme
nt
of
t
he
ulti
mate
bo
und
of
the
er
ror
be
tween
the
act
ua
l
an
d
est
imat
ed
value
.
I
n
a
dd
it
ion
,
t
he
meth
od
al
lo
ws
the
use
r
t
o
con
t
ro
l
this
bo
und
by
cha
nging
t
he
est
imat
or
par
a
mete
rs
[14,
15]
.
The
a
ppr
ox
imat
e
cl
assic
al
sli
din
g
mode
diff
e
re
ntiat
or
i
s
util
iz
ed
to
est
imat
e
error
si
gnal
de
rivati
ves
in
the
pro
po
se
d
s
ys
te
m
to
minimi
ze
the
e
rro
r
to
a
lowe
r ban
d
as
foll
ow
:
=
+
(1)
̇
=
−
2
∗
tan
−
1
(
)
(2)
Wh
e
re
is
ob
se
rv
e
r
sli
ding
va
riable,
is
the
e
rror
in
put
si
gnal
,
x
is
th
e
obs
erv
e
r
dyna
mic,
̇
is
the
der
i
vativ
e
of
the
obser
ve
r
dyna
mic,
an
d
a
re
the
sli
di
ng
m
ode
dif
f
eren
ti
at
or
desi
gn
par
a
mete
rs,
res
pecti
vel
y.
The
value
of
will
be
sel
ect
ed
to
e
ns
ure
that
goe
s
to
zer
o
wh
e
n
>
|
̇
|
.
Let
the
est
im
at
ion
of
e
rror
be
come
t
he
ou
t
pu
t
of the
foll
ow
i
ng lo
w p
ass f
il
te
r.
̇
+
=
2
∗
t
an
−
1
(
)
(3)
W
he
re
is a ti
m
e co
ns
ta
nt
,
an
d
is t
he
outp
ut
of LPF
, th
e
AC
SMD
be
as
f
ollows
;
̇
=
1
(
−
+
2
∗
tan
−
1
(
)
)
(4)
2.2.
N
on
li
ne
ar
sli
ding v
aria
ble
2.2.1.
Te
rmi
n
al slidi
ng mo
d
e
The
ap
plica
ti
on
of
t
he
te
rmi
nal
sli
di
ng
mode
c
ontr
oller
i
s
re
porte
d
i
n
[
16
–
18]
.
I
n
thi
s
pap
e
r,
the
con
ce
pt
of
a
te
rmin
al
sli
di
ng
var
ia
ble
with
a
nonlinea
r
te
rm
is
util
iz
ed
to
s
olv
e
t
he
pro
ble
ms
of
the
finite
ti
me
error
co
nc
ours
e
. T
he
te
r
minal
sli
ding
var
ia
ble is d
e
fine
d
as
fo
ll
ow:
=
̇
+
/
(5)
Fr
om
(
5),
is
de
fine
d
as
the
sli
din
g
va
riable
and
̇
is
def
ine
d
as
the
ACS
MD
out
pu
t.
,
an
d
are
the
sli
ding
va
r
ia
ble
par
a
mete
rs.
T
he
par
a
m
et
ers
(
β
,
an
d
)
m
us
t
be
posit
ive
odd
i
nteg
ers
[
19]
a
nd
[
20]
.
Additi
on
al
l
y,
>
.
In ord
e
r
t
o
est
imat
e the
ulti
mate
b
ou
nd
on t
he
e
rror
, where
(
)
=
/
.
̇
=
{
−
∗
(
)
+
}
∗
sign
(
)
=
−
|
|
+
∗
sign
(
)
(6)
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
.
11
, N
o.
3
,
Se
ptembe
r
2020
:
15
8
8
–
1
59
5
1590
In
(
6)
̇
is
the
der
i
vative
of
the
L
ya
punov
f
un
ct
io
n
,
(
)
is
th
e
f
un
ct
io
n
of
error,
sign
(
)
is
the
sign
al
f
un
ct
io
n,
λ
is
t
he
sli
di
ng
var
ia
ble
par
a
mete
r.
T
o
est
imat
e
the
ulti
m
at
e
boun
d
on
the
e
rror
(
)
,
(
6)
is
reformulat
ed
in (7
)
as
foll
ows:
̇
=
{
−
∗
(
)
+
}
∗
sign
(
)
≤
−
|
|
/
+
(7)
The
in
(
7)
re
pr
ese
nts
a
po
s
it
ive
co
ns
ta
nt.
Acc
ordin
gly,
the
ulti
mate
bound
on
the
error
is
determi
ned
usi
ng the
f
ollow
i
ng e
qu
at
io
n:
|
(
)
|
≥
(
)
/
,
as
→
∞
(8)
With
a
s
uitable
sel
ect
ion
f
or
p
and
r
,
the
ulti
mate
bound
on
the
erro
r
will
be
small
er
t
han
the
li
nea
r
sli
din
g va
riabl
e.
2.3.
The i
n
dice
s of
vo
l
tage sag
This
sect
ion
di
scusses
the
i
ndic
es
util
iz
ed
in
this
pa
pe
r
t
o
descr
i
be
the
qual
it
y
a
nd
reli
abili
ty
of
the
netw
ork.
The
s
e
ind
ic
es
are
util
iz
ed
to
re
present
t
he
pow
er
s
ys
te
m
be
ha
viour
to
ai
d
t
he
op
e
rato
r
de
ci
sion
-
making
pr
oces
s.
T
he
Det
ro
it
E
dison
sa
g
s
cor
e
(
SS),
t
he
volt
age
sa
g
l
os
t
e
nerg
y
i
ndex
(VSLE
I),
a
nd
the
vo
lt
age
sa
g
e
ne
rgy (
)
are
bri
efly discu
ssed
a
s foll
ow
s:
2.3
.1
.
De
tr
oit E
dison
s
ag sc
ore
(S
S
)
The
detai
l
of
t
his
i
nd
e
x
ca
n
be
obta
ine
d
i
n
[
21]
a
nd
[22
]
.
T
he
e
qu
at
io
n
to
cal
culat
e
this
i
nd
e
x
is
sh
ow
n
in
(9
)
:
=
1
−
+
+
3
100%
(9)
V
A
,
V
B
,
a
nd
V
C
ind
ic
at
e
the
ph
a
se
volt
ages
,
res
pecti
vely
.
The
value
of
SS
is
prefe
rr
e
d
w
he
n
it
is
cl
os
er t
o
ze
ro
, whic
h rep
rese
nt
s a
bette
r rec
overi
ng volt
age
fo
ll
owin
g
t
he
c
ompen
sat
ion.
2.3.2.
V
olt
age
sag
l
os
t e
ner
gy
index
(VSL
EI)
The
detai
l
of
VS
LE
I
ca
n
be
re
ferred
to
in
[21]
a
nd
[
22]
.
This
i
nd
e
x
cal
culat
es
the
lost
ene
rgy
(
W)
durin
g
t
he vo
lt
age sa
g occ
urr
ence
.
It is cal
c
ulate
d usin
g
t
he
foll
ow
i
ng
eq
uation:
=
[
1
−
]
3
.
14
(10)
is
def
in
ed
as
the
nominal
volt
age
,
V
is
de
fine
d
as
the
phase
volt
age
,
a
nd
T
is
def
i
ne
d
as
the
ti
me d
uri
ng t
he
sag i
n
a
mil
li
second
,
r
es
pecti
vely.
2.3.3
.
V
olt
age
sag
en
ergy
(
):
The
is
cal
culat
ed
us
in
g
t
he
e
quat
ion s
how
n
a
s foll
ow
s:
=
∫
{
1
−
(
(
)
)
2
}
0
(11)
wh
e
re
the
(
)
is
def
ine
d
as
the
volt
age
ma
gnit
ude
at
ti
me
,
an
d
is
def
i
ned
a
s
the
du
rati
on
ti
me
of
the
sag
,
resp
ect
ivel
y
[22]
.
3.
RESU
LT
S
AND
DI
S
C
USS
ION
3.1
.
The
sim
u
lation
an
d
res
ults
of
th
e
syst
em unde
r s
tudy
Figure
1
s
how
s
the
te
st
syst
em
m
od
el
us
e
d
to
eval
uate
th
e
perf
or
ma
nce
of
the
pro
po
s
ed
met
hod.
The
pa
rameter
s
of
t
he
te
st
syst
em
m
odel
ca
n
be
obta
ine
d
in
[
23]
.
T
he
te
st
sy
ste
m
m
od
el
is
m
od
el
e
d
us
in
g
the
MATL
AB/
Simuli
nk
s
of
t
war
e
.
T
he
pro
po
s
ed
meth
od
discusse
d
i
n
S
ect
ion
2
is
a
ppli
ed
to
the
te
st
sy
ste
m
model
to
e
valu
at
e
it
s
pe
rform
ance
unde
r
sev
eral
ty
pes
of
di
sturb
a
nces
in
the
s
ys
te
m.
T
he
te
st
s
ys
te
m
model
consi
sts o
f
tw
o fee
der
s
w
it
h d
iffer
e
nt ty
pes o
f
loa
d.
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
Th
e e
nhance
m
ent o
f
powe
r
qualit
y for
the
di
stri
bu
ti
on syst
em
vi
a dy
nami
c volta
ge
…
(
A
li
Ba
sim
Mo
ha
mm
e
d
)
1591
Figure
1
–
T
he
test
sy
ste
m
m
od
el
util
i
zed fo
r
the
stu
dy
3.1.1
.
Thre
e
-
pha
se
sh
ort ci
rcui
t
In
this
stu
dy,
the
s
ys
te
m
is
su
bject
e
d
t
o
a
thr
ee
-
ph
a
se
f
ault
at
=
0
.
05
s.
T
he
fau
lt
is
ap
plied
a
t
Feede
r
1
f
or
0.
13
s
.
Fi
gure
2
s
hows
t
he
respo
ns
e
of
the
s
ys
t
em,
incl
ud
i
ng
the
DV
R,
with
the
pro
pose
d
method
in
mit
igati
ng
t
he
distu
r
ban
ce
in
the
sy
ste
m
.
Figure
2(a)
s
hows
t
he
thr
ee
-
ph
a
se
volt
age
in
Fee
der
1
f
ollow
i
ng
the
fau
lt
i
n
t
he
syst
em.
It
sho
ws
t
hat
th
e
vo
l
ta
ge
at
al
l
t
hr
e
e
phases
re
duc
ed
t
o
ze
r
o
duri
ng
the
a
pp
li
cat
ion
of
fau
lt
.
F
ig
ure
2
(
b
)
s
hows
the
volt
age
at
the
a
djacent
fee
der,
Fee
der
2
f
ollo
wing
t
his
fa
ult.
T
he
res
ult
in
di
cat
es
that
the
vo
lt
ag
e
of
al
l
three
ph
a
ses
drops
to
65
%
of
it
s
nominal
value
.
T
he
DV
R
sens
e
s
this
c
onditi
on
a
n
d
injec
t
s
the
r
eq
uire
d
vo
lt
age
mag
nitud
e
to
c
ompen
sat
e
f
o
r
this
dif
fer
e
nce
.
Fig
ur
e
2
(c)
re
pr
ese
nts
t
he
volt
age
com
pensat
ion
injec
te
d
by
th
e
pro
pose
d
D
VR
.
As
a
res
ult,
the
volt
ag
e
am
plit
ud
e
af
te
r
the
co
mp
e
ns
at
io
n
impro
ved
t
o
0.998
1,
0.998
,
0.
9981
in
P
has
e
A,
B,
a
nd
C
,
as
show
n
in
F
ig
ure
2
(
d)
,
res
pect
ively
.
In
ad
d
it
ion
t
o
the
vo
lt
age
ma
gn
it
ude,
the
T
HD
at
the
loa
d
be
fore
t
he
c
ompe
ns
at
io
n
is
13.62
.
F
ollo
wing
the
co
mp
e
nsa
ti
on
,
the
T
H
D
im
pro
ves
to
1.44
.
T
he
res
ult
i
mp
li
es
t
hat
t
he
pr
opos
e
d
D
VR
is
able
to
mit
igate
t
he
vo
lt
age
disturba
nce
ca
us
e
d
by
a
bala
nced
thre
e
-
ph
a
se
fa
ult.
The
pr
opos
e
d
met
hod
is
able
to
mainta
in
the
loa
d
vo
lt
age
and re
duce the
THD at
the
ad
j
acent fee
de
r
f
ol
lowing a
fa
ult i
n
the
sy
ste
m
.
(a)
(b)
(c)
(d)
Figure
2.
The
r
esults f
or
the
t
hr
ee
-
phase
sho
rt circuit
: (a
)
t
hree
-
ph
a
se s
hort
circuit i
n
F
ee
de
r
1
, (b
)
t
he
un
c
ompe
ns
at
e
d
loa
d v
oltage,
(
c)
the
volt
age
inject
by DVR
, (d)
the c
ompe
ns
at
e loa
d v
oltage
.
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
.
11
, N
o.
3
,
Se
ptembe
r
2020
:
15
8
8
–
1
59
5
1592
3.1.2
.
D
oubl
e
-
li
ne
-
to
-
groun
d fault
In
t
his
case,
a
double
-
li
ne
-
to
-
gro
und
fa
ult
of
Ph
a
se
A
an
d
B
is
a
pp
li
ed
to
the
s
ys
te
m
at
=
0
.
1
s
.
The
fa
ult
is
a
ppli
ed
f
or
0.
1s
.
Figure
3(a)
re
pr
ese
nts
the
lo
ad
volt
age
at
Feede
r
1.
T
he
resu
lt
sho
ws
t
hat
the
vo
lt
age
at
phas
e
A
a
nd
B
is
r
edu
ce
d
t
o
zer
o,
w
hile
the
vo
lt
age
at
ph
ase
C
is
reduce
d
sli
gh
tl
y
as
c
ompa
red
t
o
the
nomi
nal
va
lue.
Fi
gure
3(b
)
s
hows
the
load
volt
age
measu
red
at
F
eeder
2,
wh
ic
h
is
th
e
same
vo
lt
age
viewe
d
by
the
pro
po
se
d
D
VR
.
T
he
r
esults
s
how
t
hat
the
volt
age
at
phase
A,
B
a
nd
C
a
r
e
87.63%,
86.
31%,
and
93.23%
as
co
mp
a
re
d
t
o
t
he
nominal
vo
l
ta
ge,
res
pecti
ve
ly.
The
n,
the
pro
po
se
d
D
VR
injec
ts
t
he
re
quire
d
vo
lt
age
t
o
mit
igate
t
he
volt
age
distu
rbance
.
Fi
gure
3(c)
r
epr
ese
nts
the
injec
te
d
volt
ag
e
f
rom
the
pr
opose
d
DV
R.
Fig
ure
3(d)
sho
ws
th
e
load
volt
age
at
Feede
r
2
a
fter
the
co
mp
e
ns
at
io
n.
T
he
r
esults
sho
w
th
at
the
pro
po
se
d
D
V
R
is
able
to
resto
re
the
volt
age
f
ollo
wing
t
he
fau
lt
t
ha
t
occ
urred
at
the
adjace
nt
feed
e
r.
Con
se
quently
,
the
TH
D
at
Fe
eder
2
is
al
so
i
mpro
ve
d,
bein
g
4.61
be
fore
t
he
c
omp
en
sat
ion
an
d
im
pro
ve
d
to
0.93 f
ollo
wing
the compe
ns
at
i
on.
(a)
(b)
(c)
(d)
Figure
3.
The
r
esults f
or the
double
-
li
ne
-
to
-
gro
und fa
ul
t
: (a)
double
li
ne
to gr
ound
i
n
F
ee
de
r
1
, (b
)
t
he
un
c
ompe
ns
at
e
d
loa
d v
oltage,
(
c)
the
volt
age
inject
by
DVR
, (d)
the c
ompe
ns
at
e loa
d v
oltage
.
3.1.3
.
Sin
gle li
ne to
ground
fa
ul
t
This
sect
io
n
di
scusses
the
pe
r
forma
nce
of
th
e
pro
posed
me
thod
in
the
e
ve
nt
of
a
sin
gle
-
ph
a
se
fa
ult
occurre
d
i
n
th
e
sy
ste
m
.
This
case
is
cru
ci
a
l
to
this
stu
dy
because
a
sin
gl
e
-
phase
fa
ult
is
the
most
co
mmo
n
fau
lt
that
occ
ur
red
in
pract
ic
e.
I
n
t
his
stu
dy,
a
sin
gle
li
ne
to
gro
und
fa
ult
a
t
phase
C
is
c
onside
red.
T
he
f
ault
is
app
li
ed
at
t=
0.
05
s
a
nd
cl
eare
d
at
t=
0.18
s
.
Si
mil
ar
to
ot
her
aforeme
ntio
ne
d
cases,
t
he
fa
ult
is
app
li
ed
a
t
Feeder
1.
Fi
gure
4
s
hows
the
resu
lt
s
ob
ta
i
ned
from
this
stu
dy.
Fi
gure
4(
a
)
re
pre
sen
ts
the
l
oad
vo
lt
age
at
Fee
der
1.
The
res
ult
in
di
cat
es
that
onl
y
phase
C
is
reduce
d
t
o
ze
r
o
w
hen
the
fa
ult
occ
urre
d.
Ph
ase
A
a
nd
B
only
exp
e
rience
d
a
sli
gh
t
re
du
ct
io
n
i
n
volt
age
duri
ng
this
per
i
od.
Fig
ur
e
4(b
)
s
how
s
t
he
l
oa
d
volt
age
at
F
eeder
2.
The
fig
ur
e
sp
e
ci
fies
the
vo
lt
a
ge
ma
gnit
ud
e
of
Ph
a
se
A
,
B,
an
d
C
is
90
.
7
%
,
9
4
.
86
%
,
67
.
46
%
of
t
he
nominal
vo
lt
age
,
resp
e
ct
ively.
C
on
se
qu
e
ntly,
the
pr
opos
e
d
DV
R
detect
s
an
d
i
nject
s
the
re
quir
ed
vo
lt
age
s
hown
i
n
Figure
4(
c
)
t
o
com
pensat
e
f
or
these
di
ff
e
re
nces.
As
a
re
s
ul
t,
the
volt
ag
e
is
rest
or
e
d
t
o
it
s
no
minal
value
a
s
sh
ow
n
in
Fig
ure 4(d). A
ls
o,
it
is no
te
d
that t
he
THD at the l
oad
is im
pro
ve
d
after
the c
ompen
sat
ion. T
he THD
befor
e
and a
fte
r
the
compe
ns
a
ti
on
a
re
6.23 a
nd 1.18,
r
es
pec
ti
vely.
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
Th
e e
nhance
m
ent o
f
powe
r
qualit
y for
the
di
stri
bu
ti
on syst
em
vi
a dy
nami
c volta
ge
…
(
A
li
Ba
sim
Mo
ha
mm
e
d
)
1593
(a)
(b)
(c)
(d)
Figure
4.
The
r
esults f
or sin
gl
e li
ne
to
gro
un
d fault
DV
R:
(
a) s
in
gle li
ne
to gr
ound i
n
fee
der
1
, (b
)
the
un
c
ompe
ns
at
e
d
loa
d v
oltage,
(
c)
the
volt
age
inject
by DVR
, (d)
the c
ompe
ns
at
e loa
d v
oltage.
3.1.4
.
V
olt
age
imbal
an
ce
d
In
this
case,
an
im
balanc
e
d
volt
age
is
ap
plied
to
t
he
sy
ste
m
at
t=
0.0
5
s
unti
l
t=
0.185
s
.
The
dist
urba
nc
e
is
app
li
ed
to
Feede
r
2.
Figure
5
s
ho
ws
the
r
esult
ob
ta
ine
d
for
this
stu
dy.
Fig
ur
e
5(
a
)
il
lustrate
s
the
l
oad
volt
age
be
fore
the
co
mp
e
ns
at
io
n.
In
this
operati
ng
sit
ua
ti
on
,
t
he
loa
d
vo
lt
age
of
pha
se
A,
B,
a
nd
C
a
re
decr
ease
d
to
72.
69%,
52.
18
%,
a
nd
79.
24%
of
the
nom
inal
volt
age
,
r
especti
vely
.
F
ollow
i
ng
this
disturba
nc
e,
t
he
pro
pose
d
D
VR
detect
s
a
nd
injec
ts
th
e
need
e
d
volt
a
ge
ma
gn
it
ude
t
o
re
gu
la
te
the
vo
lt
age
at
the
loa
d
side
.
T
he
i
nject
ed
vo
lt
age
is
s
ho
wn
in
Fig
ur
e
5(b
).
C
onseq
ue
nt
ly,
the
co
mp
e
ns
at
ed
loa
d
volt
age
is
sh
ow
n
in
Fig
ure
5(c)
.
I
n
t
he
fig
ur
e,
the
l
oa
d
vo
lt
age
of
phase
A,
B,
an
d
C
are
99.86%,
99.
73
%
,
a
nd
99.82%
of
the
nomina
l
vo
lt
ag
e,
res
pecti
vely
.
Acc
ordi
ng
t
o
the
IE
C
6100
0
-
3
-
13,
the
im
bal
anced
fact
or
s
hould
no
t
excee
d
2%
[
24]
.
In
thi
s
stu
dy,
the
imbala
nc
e
factor
is
11.
98
be
fore
t
he
c
omp
ensati
on
a
nd
r
edu
ce
d
to
a
pproximat
e
ly
zer
o
after
th
e
co
mp
e
ns
at
io
n.
I
n
a
dd
it
io
n,
t
he
total
harmo
nic
distor
ti
on
a
t
the
loa
d
volt
age
i
s
al
so
imp
rove
d;
the
TH
D
is
equ
al
t
o
11.
25
be
fore
the
com
pensat
ion
and
re
du
ce
d
to
1.4
4
f
ollow
i
ng
the compe
ns
at
i
on.
(a)
(b)
(c)
Figure
5.
S
im
ul
at
ion
r
es
ults
f
or volt
age im
ba
la
nce: (a
)
the
un
c
ompe
ns
at
e
d
loa
d v
oltage,
(b)
t
he vo
lt
a
ge
injec
t
ed
by DV
R, (c
)
the
com
pensat
e loa
d v
oltage.
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
.
11
, N
o.
3
,
Se
ptembe
r
2020
:
15
8
8
–
1
59
5
1594
3.2
.
P
erf
orm
ance
evalu
at
i
on
Table
1
summ
arizes
the
perf
ormance
of
th
e
pro
posed
w
ork
i
n
mit
igati
ng
the
volt
age
disturba
nce.
The
perf
or
ma
nc
e
is
eval
uated
us
in
g
t
he
i
nd
i
ces
disc
us
se
d
i
n
Sect
io
n
2.3.
The
res
ults
ta
bula
te
d
in
the
ta
ble
represe
nts
the
performa
nce
of
the
ACS
M
D
with
TS
M
C
m
et
hod
in
te
r
ms
of
S
S
,
VS
LE
I
,
and
E
.
T
he
res
ul
ts
sh
ow
t
hat
the
pro
po
se
d
D
VR
imp
roves
the
qual
it
y
of
the
vo
lt
ag
e
in
te
rms
of
S
S,
V
SLE
I
,
a
nd
E
.
This
im
plies
t
ha
t
ACS
MD
with
TS
M
C
is
a
bl
e
to
mit
igate
the
s
ys
te
m
vo
lt
age
i
n
t
he
pr
e
s
ence
of
th
ree
-
phase,
double
li
ne
t
o gro
und
,
and t
he
sin
gle li
ne
to
gro
und fault.
Table
1.
V
oltage sa
g
i
nd
ic
es
for A
C
S
M
D
wi
th TS
M
C
Con
se
quently
,
the
performa
nc
e
of
t
he
pro
pose
d
met
hod
is
co
mp
a
re
d
with
the
meth
od
repor
te
d
i
n
[23]
.
The
met
hod
in
[
23
]
ut
il
iz
es
a
li
near
sli
ding
va
riab
le
with
a
li
ne
ar
dif
fer
e
ntiat
or
to
c
ompen
s
at
e
f
or
the
volt
age
disturbance
i
n
t
he
s
ys
te
m.
Ta
ble
2
s
hows
t
he
co
mpa
rati
ve
analysis
i
n
te
rms
of
i
nteg
r
al
ti
me
abs
olu
te
er
ror
(I
T
AE
).
T
he
de
ta
il
of
IT
AE
i
s
obta
ined
in
[
25]
.
Table
2
disp
la
y
s
t
he
perf
ormance
c
omp
ariso
n
betwee
n
t
hese
two
meth
ods.
The
res
ults
s
how
that
the
pr
opos
e
d
meth
od
outpe
rfo
rms
the
meth
od
repor
te
d
i
n
[23]
in all
oper
at
ing
sit
uatio
ns co
ns
ide
red in t
his stu
dy in
ter
ms
of
IT
AE.
Table
2.
I
ll
us
tr
at
e the c
on
tra
st
betwee
n
[
23]
with
ACS
M
D
wit
h
T
SM
C
Op
erating
con
d
itio
n
s
IT
A
E(
DC
)
[23
]
IT
A
E
(
DC)
Three
-
p
h
ase sh
o
rt
circuit
1
.31
2
0
.15
2
6
Do
u
b
le
-
lin
e
-
to
-
g
rou
n
d
f
au
lt
1
.33
5
0
.11
1
9
Sin
g
le
-
lin
e
-
to
-
g
rou
n
d
f
au
lt
1
.51
9
0
.12
5
4
Vo
ltag
e im
b
alan
ce
d
1
.55
7
0
.27
2
1
4.
CONCL
US
I
O
N
Con
cl
us
ively
,
a
novel
DV
R
volt
age
c
ompen
sat
ion
met
hod
base
d
on
ACS
M
D
with
TS
MC
,
has
bee
n
pro
po
se
d.
T
he
method
is
ba
sed
on
a
nonl
inear
di
ff
e
ren
t
ia
tor
to
obta
in
the
error
fun
ct
ion
de
rivati
ve
in
pro
vid
in
g
a
co
r
respo
nd
i
ng
c
ontr
ol
act
io
n.
T
he
meth
od
is
a
pp
li
ed
t
o
a
te
st
s
ys
te
m
m
od
el
unde
r
va
rio
us
vo
lt
age
disturba
nce
sc
enar
i
os
t
o
e
val
uate
it
s
perf
ormance
in
mit
ig
at
ing
t
he
volt
age
distu
rb
a
nce.
The
resu
lt
s
show
that
the
DV
R
base
d
on
ACS
MD
and
T
SM
C
ca
n
im
pro
ve
the
vo
lt
age
fo
ll
o
wi
ng
a
th
ree
-
ph
a
se
fa
ult,
a
double
li
ne
to
gro
und
fa
ult,
a
sin
gle
li
ne
f
ault
an
d
a
vo
lt
age
im
balance
conditi
on.
T
he
study
has
s
ho
wn
t
hat
the
pro
po
s
ed
method
has
imp
rove
d
t
he
vo
lt
age
,
TH
D,
SS
,
V
SLEI
,
and
"E"
_v
s
of
the
s
ys
te
m
after
the
volt
age
com
pensat
ion.
Con
se
quently
, t
he
pe
rformanc
e o
f
the
meth
od is compa
red
with a linear sl
iding
va
riable
with a
li
near
dif
fer
e
nt
ia
tor
meth
od.
The
D
VR
base
d
on
ACS
MD
and
T
SM
C
ou
t
performs
the
li
near
sli
ding
va
riable
method
in
te
r
m
of
I
TA
E
i
n
com
pensat
ing
t
he
volt
age
un
de
r
va
ri
ous
t
yp
e
s
of
volt
age
disturbance
occ
urre
d
i
n
the s
ys
te
m.
REFERE
NCE
S
[1]
R.
A.
Kumar
,
G.
S.
Kumar
,
B.
K.
Kumar
,
and
M.
K.
Mishra,
“
Compe
nsati
on
o
f
volt
ag
e
sags
a
nd
har
monics
wit
h
phase
-
jum
ps
thr
ough
DV
R
with
mi
nim
u
m
VA
ra
ti
ng
using
Par
ti
c
le
Sw
arm
Optimization,”
in
2009
World
Congress
on
Nature & Bio
logi
cally
Inspire
d
Computing
(N
aBIC)
,
2009
.
[2]
Rauf,
A.
M.,
an
d
Khadkik
ar,
V.
,
“An
En
h
anced
voltage
sag
co
mpe
nsati
on
sch
e
me
for
dyn
am
i
c
voltage
resto
r
er
,
”
IEE
E
Tr
a
nsacti
o
ns on
Industrial Elect
ronics
,
vol
.
62,
no.
5,
pp.
26
83
–
2692,
201
5
.
[3]
M.
B.
M.
Ro
zla
n,
A.
F.
Zobaa,
and
S.
H.
E.
A
.
Alee
m
,
“
The
op
t
im
isation
of
st
an
d
-
al
one
hyb
rid
r
ene
wabl
e
en
erg
y
sys
te
ms using
H
OM
ER,
”
Int
erna
ti
onal Re
vi
ew
of
El
e
ct
rica
l
Eng
in
ee
ring
,
vol
.
6
,
n
o.
4
,
p
.
1802
-
181
0
,
2011
.
[4]
A.
B
.
Moha
mme
d,
M
.
A
.
M
.
Ar
if
f,
and
S.
N.
Ramli,
“Power
qu
al
ity
i
mprove
m
ent using
dyna
mi
c
v
olt
ag
e
r
estore
r
in
el
e
ct
ri
ca
l
distri
b
uti
on
sys
te
m:
an
over
vie
w,
”
Ind
onesian
Journal
of
Elec
tri
cal
E
ngine
ering
and
Computer
Sci
en
ce
(IJ
EE
CS)
,
vo
l. 1
7,
no
.
1
,
pp
.
86
–
93,
2020
.
[5]
H.
Haf
ez
i
and
R.
Far
anda,
“D
ynam
i
c
vo
lt
ag
e
co
nd
it
ion
er:
A
new
con
ce
pt
f
or
smar
t
low
-
v
olt
ag
e
d
istri
but
i
on
sys
te
m
s,”
I
EEE
Tr
ansacti
ons on Power
E
le
c
troni
cs
,
vol
.
33
,
no
.
9
,
pp
.
7582
–
7590
,
2018.
[6]
Moham
me
d
,
J.
A.
K.,
Hus
sein,
A.
A.,
and
Al
-
Sakini,
S.
R.
,
“Vo
lt
ag
e
disturba
n
c
e
mi
t
igation
in
I
raq
’s
low
volt
ag
e
distri
buti
on
sys
t
em
,
”
Indone
sian
Journal
o
f
Elec
t
rical
Engi
n
ee
rin
g
and
Computer
Sci
en
ce
(IJ
EE
CS
)
,
vo
l.
17
,
no.
1
,
Vo
ltag
e sag
in
d
ices
VSLE
I
(pu
)
Vo
ltag
e sag
energ
y
(
E
VS
)
(pu
)
Detroit Edis
o
n
sag
sco
re
(
SS
)
(pu
)
Fau
lt typ
e
Befo
re
After
Befo
re
Afte
r
Befo
re
After
Three
-
p
h
ase
fault
1
4
.33
6
7
1
.17
7
4
×
10
-
6
4
7
.56
6
0
0
.00
1
5
0
.34
9
2
0
.00
1
9
Do
u
b
le
-
lin
e
-
to
-
g
rou
n
d
f
au
lt
0
.35
6
8
1
.00
1
3
×
10
-
6
3
.86
2
7
0
.00
1
2
0
.10
9
4
0
.00
2
0
Sin
g
le
-
lin
e
-
to
-
g
rou
n
d
f
au
lt
3
.91
4
3
1
.61
1
3
×
10
-
6
1
5
.23
2
9
0
.00
1
7
0
.15
6
6
0
.00
2
0
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
Th
e e
nhance
m
ent o
f
powe
r
qualit
y for
the
di
stri
bu
ti
on syst
em
vi
a dy
nami
c volta
ge
…
(
A
li
Ba
sim
Mo
ha
mm
e
d
)
1595
pp.
47
–
60
,
2019
.
[7]
Abed,
A.
H
.
,
Ra
hebi
,
J.,
Saji
r
,
H
.
,
and
Far
za
mn
i
a,
A.
“Prot
ec
t
io
n
of
sensiti
v
e
lo
ads
from
vo
lt
ag
es
fluctuations
i
n
Ira
qi
grids
by
D
VR,”
In
2017
I
EE
E
2nd
Inte
rn
ati
onal
Conf
ere
nce
on
Aut
omati
c
Control
and
I
nte
lligent
Syst
e
ms
(I2CACIS)
,
pp.
1
44
–
149,
2017
.
[8]
A.
Pa
kh
ari
a
and
M.
Gupt,
“Dyna
mic
vo
lt
ag
e
r
estore
r
for
co
m
pensa
ti
on
of
vo
lt
ag
e
sag
and
sw
el
l:
A
literatur
e
rev
ie
w,
”
In
te
rna
ti
onal Journal
o
f
Adv
an
ce
s
in En
gine
ering
&
Te
c
hnology
,
vol
.
4
,
no.
1
,
p
.
347
,
20
12.
[9]
V.
Deshpande
C
hinm
ay
and
V.
Deshpande
Chaitanya,
“Opt
im
u
m
design
of
dyn
am
i
c
voltage
r
estore
r
for
vol
ta
g
e
sag
mi
ti
g
ation
i
n
distri
buti
on
n
et
work,”
Inte
rna
ti
onal
J
ournal
o
f
Powe
r
E
le
c
tronic
and
Dr
iv
e
S
yste
m
(IJ
P
EDS)
,
vol.
10
,
no
.
3
,
p
p
.
1364
-
1372
,
20
19.
[10]
D.
Dan
al
akshm
i
,
S.
Bug
at
a
,
J.
Kohila
,
“A
cont
ro
l
str
at
egy
on
po
wer
qua
li
ty
im
pr
ovem
en
t
in
cons
ume
r
side
using
custom
power
d
evi
c
e,”
Indone
si
an
Journal
of
E
le
c
tric
al
Engi
ne
ering
and
Computer
Scienc
e
(IJ
EE
CS)
,
vo
l.
15
,
no.
1
,
pp
.
80
-
87
,
2019.
[11]
L.
P.
Vasud
evan
and
V.
Prasad
,
“Per
form
anc
e
enha
nc
em
en
t
o
f
a
dyna
mi
c
vol
t
age
restor
e
r,
”
T
urkish
Journal
of
El
e
ct
rica
l
Eng
in
ee
ring
&
Compu
te
r Sc
ie
nc
es
,
vol
.
25
,
no
.
3
,
pp
.
2
293
–
2307,
2017
.
[12]
S.
Bir
ic
ik
,
H.
Komurc
ugil,
N
.
D.
Tuy
en,
an
d
M.
Basu,
“P
rote
c
ti
o
n
of
se
nsiti
ve
lo
ads
us
ing
slid
ing
mo
de
cont
rolled
three
-
phase
DV
R
wi
th
ada
pt
ive
not
c
h
filter
,
”
IE
EE
Tr
ansacti
ons
on
Industrial
Elec
tronic
s
,
vol.
66
,
no.
7
,
pp
.
5465
–
5475,
2018
.
[13]
A.
J.
Sguare
z
i
F
il
ho,
D.
A.
Fern
ande
s,
J.
H
.
Suá
rez
,
F.
F.
Cost
a,
and
J
.
A.
T
.
Alt
una,
“Recursive
le
ast
square
s
an
d
slidi
ng
mod
e
c
ontrol
for
vol
tage
com
p
ensa
tio
n
of
three
-
phas
e
loa
ds,
”
Journal
of
Control
,
Aut
omation
and
El
e
ct
rica
l
S
ystem
s
,
vol. 29, no.
6,
pp
.
769
–
777
,
2018.
[14]
S.
A.
AL
-
Sa
marrai
e
and
oth
ers,
“A
cha
t
te
r
ing
f
r
ee
sl
idi
ng
mod
e
observ
er
with
application
to
DC
mot
or
spe
e
d
cont
rol
,
”
in
2018
Thir
d
Scientific
Confe
renc
e
of
E
l
ec
tri
cal
Engi
n
eer
ing
(SCEE
)
,
20
18.
[15]
Y.
Shtessel
,
C.
E
dwards,
L
.
Frid
ma
n,
A. Le
v
ant,
Sli
ding
mod
e co
ntrol
and
observ
ati
o
n
,
Birkhä
use
r
,
Springer
,
2014
.
[16]
M.
Zh
ihong,
A.
P.
Pap
li
nski
,
a
nd
H.
R.
Wu
,
“
A
robust
MIM
O
te
rm
inal
sl
iding
mode
cont
ro
l
sche
me
for
rig
id
roboti
c
ma
n
ipul
a
tors,
”
I
EEE
Tr
ansacti
ons on Auto
matic
Con
trol
,
v
ol.
39
,
no
.
12
,
pp
.
2464
–
2469
,
19
94.
[17]
M.
Zhi
hong
an
d
X.
Yu,
“A
da
pti
ve
te
r
mi
na
l
s
li
ding
mode
tracki
ng
cont
rol
fo
r
rigi
d
robotic
ma
nipulators
wi
th
unce
rt
ai
n
dyn
a
mi
cs,
”
JS
ME
I
nte
rnational
Jo
urnal
Serie
s
C
Me
chan
ic
al
S
yste
ms
,
Ma
chi
n
e
E
le
ments
an
d
Manufac
turing
,
vol.
40
,
no
.
3
,
pp
.
493
–
502
,
1997
.
[18]
S.
T
.
Venka
ta
r
a
ma
n
and
S.
Gula
ti
,
“Con
trol
of
n
onli
ne
ar
sys
tems
using
t
erm
in
al
slidi
ng
mod
es,
”
i
n
1992
Ame
ri
ca
n
Control
Confe
r
e
nce
,
1992.
[19]
L.
Frid
ma
n
,
J.
A.
Moreno
,
B
.
Ba
ndyopadhya
y,
a
nd
S.
Ka
ma
l
,
Rece
nt
advan
ce
s
in
slidi
ng
mod
es:
From
C
ontrol
t
o
int
el
li
gen
t
me
ch
atroni
cs,
spring
e
r
,
vol
.
24
,
2015
.
[2
0]
R.
Fridm
an,
L.,
Moreno,
J.
,
and
Iria
rt
e,
“Sl
idi
ng
mode
s
aft
er
th
e
first
d
ec
ad
e
of
t
he
21st
c
ent
ury
,
”
Le
ct
ure
note
s
i
n
cont
rol an
d
inf
o
rm
ati
on
scie
n
ce
s
,
vol
.
412
,
pp
.
11
3
–
149,
2011
.
[21]
“IE
E
E
guid
e
for
vol
ta
g
e
sag
ind
i
c
es,
”
IEEE
Std
15
64,
pp
.
1
–
55
,
20
14.
[22]
A.
M.
Sae
ed
,
Aleem,
S.
H.
A.
Alee
m
,
A.
M.
I
bra
him,
M.
E.
Bal
ci,
and
E
.
A
.
E
l
-
Za
h
ab
,
“Po
wer
conditioni
n
g
using
dynam
i
c
volt
ag
e
restor
ers
under
diff
ere
nt
volt
ag
e
sag
typ
e
s,”
Journal
of
a
dvanc
ed
rese
arc
h
,
vol.
7
,
no.
1
,
pp.
95
–
103
,
201
6
.
[23]
A.
I
.
Om
ar,
S.
H
.
E.
A.
Ale
em,
E
.
E.
A.
E
l
-
Za
h
ab,
M.
Algablawy,
and
Z.
M.
Ali
,
“
An
im
prov
ed
ap
proa
ch
for
robust
cont
rol
of
dyna
mi
c
vo
lt
ag
e
rest
ore
r
and
powe
r
q
ual
it
y
enha
n
cem
ent
using gr
asshopper
optimizat
i
o
n
al
gori
thm,”
I
SA
transacti
ons
,
vo
l
.
95
,
pp
.
110
–
12
9,
2019
.
[24]
I.
E
.
C
.
(IE
C)
,
“E
l
ec
tro
ma
gn
et
i
c
co
mpa
t
ibi
l
it
y
(EMC)
-
Part
3
-
6:
Li
m
it
s
–
As
sess
me
nt
of
emiss
ion
li
m
it
s
for
th
e
conne
c
ti
on
of
di
storti
ng
inst
allati
ons t
o
MV
,
HV
and
EHV p
ower
sys
te
ms,”
B
asic
EM
C
publicatio
n,
ed
.
2,
,
2008
.
[25]
H.
K.
Shaker
,
H.
E.
Zoghby,
M.
E.
B
ahgat,
and
A.
M.
Abd
el
-
Ghany,
“Adv
anc
ed
C
ontrol
t
ec
hniqu
es
for
a
n
int
er
conne
c
te
d
mul
ti
ar
ea
pow
e
r
sys
te
m
for
load
fre
quen
cy
cont
rol,
”
in
2019
21st
Int
ernati
onal
Middl
e
East
Po
wer
Syste
ms
Confe
re
nce
(M
EP
CON)
,
pp.
710
–
715
,
20
19
.
Evaluation Warning : The document was created with Spire.PDF for Python.