TELKOM
NIKA
, Vol. 13, No. 4, Dece
mb
er 201
5, pp. 1194
~1
203
ISSN: 1693-6
930,
accredited
A
by DIKTI, De
cree No: 58/DIK
T
I/Kep/2013
DOI
:
10.12928/TELKOMNIKA.v13i4.2362
1194
Re
cei
v
ed
Jul
y
27, 201
5; Revi
sed O
c
tob
e
r 13, 201
5; Acce
pted O
c
t
ober 2
7
, 201
5
Modelling on Tracking Test Condition of Polymer
Nanocomposite using Finite Element Simulation
Fatin Liy
a
na
Muhamedin, MAM Piah a
nd Nordian
a
Azlin Othma
n
Institute of Hig
h Voltag
e an
d High C
u
rre
nt,
F
a
cult
y
of Elec
trical Eng
i
ne
eri
ng,
Univers
i
tiT
e
knolo
g
i Mal
a
ysia,
8131
0, UT
M Johor Ba
hru, Jo
hor, Mala
ys
ia
e-mail: fen
d
i@f
k
e.utm.m
y
A
b
st
r
a
ct
Electrical track
i
ng is a for
m
at
ion pr
ocess of
a
perman
ent cond
uctin
g
pat
h across the i
n
sul
a
ting
mater
i
al
d
ue to
surface
eros
io
n u
nder
hi
gh v
o
ltag
e stre
ss.
T
he existi
ng
of
leak
ag
e curr
e
n
t (LC)
on th
e
w
e
t
conta
m
i
nate
d
mater
i
al surfac
e causes
the
gen
eratio
n of surface disc
ha
r
ges that resul
t
ed in the materi
a
l
degr
adati
on. T
he
effects of el
ectric
fiel
d distributi
on an
d
cur
r
ent
d
ensit
y
on
LLDP
E
-Natur
al R
u
b
ber
ble
n
d
s
mater
i
al w
e
re i
n
vestig
ated us
i
ng finite
ele
m
e
n
t meth
od (F
EM) analys
is. In this pap
er, a variety of physic
a
l
para
m
eters pa
rticularly c
onta
m
i
n
a
n
t flow
rate, various
a
p
p
l
ied vo
ltag
es, mater
i
al
prop
er
ties of per
mitti
vity
and c
o
n
ductivi
ty w
e
re studie
d
w
hen n
a
n
o
fil
l
er is a
d
d
ed to
LLDPE-
Natur
a
l ru
bber
bl
en
d. T
he si
mu
lati
o
n
w
o
rks usin
g F
E
M softw
are of Quickfiel
d
w
a
s ap
pli
ed to
th
e trackin
g
test
cond
ition
of IE
C 60
58
7 stan
d
a
rd.
T
he resu
lts sh
ow
that the ele
c
tric fi
eld d
i
stri
butio
ns are cr
iti
c
al o
n
the
e
d
g
e
s of
conta
m
i
n
ant sol
u
tion
pat
h at
hig
her vo
ltag
e
level.
T
he c
u
rrent de
nsity
a
nd e
l
ectric fie
l
d distrib
u
tio
n
i
s
increas
e
w
i
th hi
gher
ap
pli
ed
voltag
e. T
he p
o
ly
mer n
anoc
o
m
p
o
site w
i
th 1
-
5 % of
nan
ofil
ler exh
i
bits a g
ood res
i
stanc
e
to tracking an
d
erosi
on test.
Ke
y
w
ords
: F
i
n
i
te Ele
m
e
n
t Method, electric fi
eld, curre
nt de
nsity, surface d
i
schar
ges an
d, IEC 6058
7
Copy
right
©
2015 Un
ive
r
sita
s Ah
mad
Dah
l
an
. All rig
h
t
s r
ese
rved
.
1. Introduc
tion
The wid
e
usa
ge of polymer as in
sulatio
n
material
s in
high voltage
(HV) eq
uipm
ent has
led to further investigation
in their pe
rfo
r
man
c
e
s
, mai
n
ly in term of electri
c
al a
s
pect. The a
g
i
ng
of polymer d
ue to e
n
viron
m
ental
stre
sses
can
ca
us
e d
e
g
r
a
da
tio
n
o
f
in
s
u
la
to
r
po
lyme
r
thr
o
ug
h
surfa
c
e
tra
cki
ng ph
enom
e
na.Insul
a
ting
surfa
c
e
are e
x
pose
d
to e
n
vironme
n
tal stresse
s
su
ch
as
contami
nant,
UV ray
stre
ss, pollution an
d low
seve
rity fogco
ndition
s [1, 2]. Lea
kage
curre
n
t (LC)
exists on th
e
insulato
r surface when
a certai
n volt
ag
e gra
d
ient is
sufficie
n
t to cause the L
C
to
flow un
der
a
wet contam
inant conditi
on.The
cont
i
nuou
s flo
w
of LC
will h
eat the in
sul
a
tor
surfa
c
e
and
th
e accu
mulate
d heat
dissip
ation in
narro
w path
on th
e insulato
r su
rface
eventu
a
lly
formed a d
r
y
band a
r
ci
ng
at the surfa
c
e of the insu
l
a
tor. Dry ba
n
d
arcing o
c
cu
rs at the lo
west
surfa
c
e
resi
stance
when
non-unifo
rm
water laye
r i
s
fo
rmed
du
e to hyd
r
o
p
h
obicity lo
ss.
The
hydrop
hobi
cit
y
feature in the materi
als
can redu
ce t
he on
set of LC by increa
sing the failu
re
time.Hydro
ph
obicity loss coul
d increa
se t
he devel
opment of L
C
and redu
ce the insulat
o
r
resi
stan
ce
he
nce
co
ntribut
e in de
gra
dat
ion [3]. Mean
while,
surfa
c
e disch
a
rg
es
occur wh
en t
h
e
air g
ap rea
c
h
ed the
criti
c
al
flashove
r
vol
t
age a
c
ross t
he d
r
y ban
d. Carboni
ze
d track an
d e
r
o
s
i
on
will be form
ed
when an arc
burn
s t
he insul
a
tor
material
due
to surface discharges. The
carboni
ze
d track that d
e
v
eloped p
a
thway bet
we
en two el
ectrode
s event
ually will ca
use
ins
u
lation failure to the s
y
stem.
To
stu
d
y
the electri
c
al pe
rforma
nces of
pol
ymeri
c
in
sulating mate
ri
als, L
C
me
asurem
ent
are u
s
ed a
s
the tools to in
dicate the det
erio
ration
of the material
s [4, 5].The online monitori
ng
of
LC ha
s bee
n develope
d
by prev
ious rese
arche
r
s and this LC are acquire
d throug
hout
the
duratio
n of six-hou
r [6-8]; Indeed, the o
b
tained L
C
a
r
e pro
p
o
r
tion
al to degrad
a
t
ion of polymer
material
s [9,
10]. The i
n
flu
ence of
cont
aminant
co
nd
uctivity wa
s
studied
usin
g t
h
ree
conditio
n
s
whi
c
h are coastal
with condu
ctivity of
4746 µS/cm, industri
a
l
(818µS/cm) and NH
4
Cl (550
µS/cm) and
the re
sult sho
w
s
that
in
du
strial c
ontami
n
ant give
s a
smallest
LCval
ue of
327.6
mA
with the le
sser d
egradatio
n [11]. The
study of hy
dro
phobi
city loss in term of
L
C
waveform
wa
s
con
d
u
c
ted ini
[12] sin
c
e th
e hydropho
bi
city loss of
th
e materi
als af
fect the mate
rial’s re
si
stan
ce
towards
surface tracking and eros
ion. Hydrophobi
city w
ill be destroyed by the
presence of local
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 13, No
. 4, Decem
b
e
r
2015 : 119
4 – 1203
1195
high ele
c
tri
c
field and wh
en hydro
pho
bicity abilit
y of the materi
al loss, the tende
ncy of the
surfa
c
e in
sul
a
tor to degra
de rise [13]. In the si
mula
tion analysi
s
of electri
c
al
stre
sse
s
on the
surfa
c
e
of in
sulato
r,the in
vestigation
of
ele
c
tric
field
distri
bution
arou
nd
wate
r dro
p
let o
r
water
films a
r
e
con
ducte
d by
se
veral
re
sea
r
chers [1
4,
15].
Thei
r
re
sult
sho
w
sthat th
e conta
c
t a
n
g
l
e of
water d
r
opl
et
in the
shed
m
odel
whi
c
h
ap
plied volt
a
ge i
s
p
e
rpen
dicul
a
r to
the
in
sul
a
tor
su
rface i
s
negle
c
ted a
s
the maximum stren
g
th o
f
electri
c
field are foun
d to be app
eared on top of
the
dropl
ets [16].
However, th
e conta
c
t an
g
l
e of water
droplets affe
ct the ele
c
tri
c
field distri
bution
in
the she
a
th
model
whe
r
e
voltage is
applie
dalon
g
the insul
a
to
r su
rfa
c
e [17
]. The maximum
electri
c
fiel
d i
s
lo
we
st
whe
n
the
dista
n
ce bet
we
e
n
two water dropl
ets
redu
ce
an
d in
cre
a
si
ng
of
a
water droplet
on the in
sul
a
tor
surfa
c
e [
18]. Whe
n
th
e co
ntact a
n
g
le of water
dropl
et is lo
w, it
forms
wate
r l
a
y
e
r o
n
the i
n
sul
a
tor
su
rface. In
tracki
ng an
d e
r
o
s
i
on resi
stan
ce
experim
ent,
the
contami
nant
solutio
n
laye
r are
inten
ded
to be
flown
a
t
certai
n flo
w
rate o
n
the
in
sulato
r
su
rface.
This can initiate the dry band a
r
ci
ng whe
n
t
he co
ntaminant la
yer is dry n
ear the grou
nd
electrode
wh
en the voltage is sup
p
lied.
Coro
na dish
arge
s emmi
si
on occu
red a
t
the tips of the
dropl
ets whe
r
ea
s
the dry
band arcin
g
occu
re
d
d
u
ring
wet
co
ndition
s an
d
wate
r dropl
ets
deform
a
tion a
ffected the L
C
wavefo
rm
obtaine
d from
the current d
ensity in the simulation [13]
.
The previo
us rese
arch in
the field simu
lation is
con
d
u
c
ted m
a
inly on the partial
discha
rge, p
h
enemo
na
with a focus
of the void
an
d
water
drople
t
as the p
a
ra
meters [19, 2
0
].
There is little
attempt to investigate th
e field simul
a
tion in the i
n
clin
ed pla
n
e
tracking te
st in
related to
su
rface tra
c
king
studie
s
. The
r
efore, t
here
is a po
ssiblit
yto condu
ct this field
work to
encounte
r
th
e correlatio
n
betwe
en th
e f
i
eld
simulatio
n
an
d a
c
tual
experim
ent. If the
app
roa
c
h of
the field
simu
lation in IPT t
e
st
could
foreca
st the
re
sult of the exp
e
rime
nt of IPT test, then
the
field simulati
on wo
rk
can
be used a
s
another opt
ion in investi
gating the surface tra
cki
ng
resi
stan
ce. In
this p
ape
r, the ele
c
tri
c
fie
l
d,
voltage a
nd current
de
nsity dist
ribut
ion a
r
e
studi
ed
usin
g Finite
Element
software o
n
L
L
DPE-Natu
ra
l Rubbe
r
m
a
teri
als with and
without nan
ofiller.
The an
alysi
s
wa
s cond
uct
ed at vario
u
s
applie
d voltages
und
er
we
t contamin
ant
con
d
ition
s
. The
simulatio
n
works
we
re ap
plied on th
e
test co
nf
iguration of IEC
6058
7 st
an
d
a
rd trackin
g
and
ero
s
ion
test.
The
paramet
ers u
s
ed
to i
n
vestigate th
e electri
c
field a
nd cu
rre
nt
de
nsity
di
stributi
on
of insul
a
tor
surfa
c
e
are
applie
d volta
ge, ele
c
tr
i
c
condu
ctivity and pe
rmittivities of in
sul
a
ting
sampl
e
and
contamina
n
t solution.
2. Rese
arch
Metho
d
2.1. FEM Simulations
The finite el
e
m
ent comme
rcial
softwa
r
e
of Quic
kfield
wa
s u
s
ed
to i
n
vestigate th
e cu
rrent
den
sity, voltage an
d ele
c
tri
c
field distributio
n
on
the surfa
c
e
of insulatin
g
sampl
e
s.
The
simulatio
n
was mo
delle
d
according to
the test
confi
guratio
n of IEC
605
87
standa
rd; the t
e
st
method
s
use
d
for evalu
a
ting
re
sista
n
ce
to tra
c
ki
ng
a
nd e
r
o
s
ion.
T
he
simulatio
n
wa
s drawn i
n
plane pa
rallel
2D model cl
ass with the sampl
e
and e
l
ectro
de confi
guratio
n is sh
own in Figu
re
1.
A re
ctang
ular sp
ecim
en
wi
th a
size of
5
0
mm x
120
mm an
d thi
c
kne
s
s of
6 m
m
was u
s
ed
as
s
a
mple to be tes
t
ed.
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
Modellin
g on
Tra
cki
ng Te
st
Conditio
n
of Polym
e
r Na
n
o
com
p
o
s
ite u
s
ing Finite …
(Fatin L.M.)
1196
Bo
t
t
o
m
El
e
c
t
r
o
d
e
(G
r
o
u
n
d)
T
o
p E
l
ec
t
r
od
e
(H
V
)
F
i
l
t
e
r
-
p
ap
e
r
p
a
d
u
n
de
r
t
o
p
el
e
c
t
r
ode
C
ont
a
m
i
n
an
t
T
e
s
t
in
s
u
la
t
i
n
g
sp
e
c
i
m
e
n
50
m
m
Figure 1. Test speci
m
en wi
th the electro
des
To inve
stigat
e the effe
ct
of contamin
ant
on th
e i
n
sul
a
tor
su
rf
ace, t
w
o p
a
t
t
erns of
contami
nant
solutio
n
we
re
dra
w
n a
s
de
picted in
Fig
u
re 2. Thi
s
p
a
ttern is
cho
s
en ba
sed
on
the
picture of the conta
m
ina
n
t solution a
l
ong the
in
sulator
surfa
c
e captu
r
ed f
r
om the a
c
tual
experim
ent. There are two pattern
s of
contami
nant
solution flo
w
are d
r
a
w
n i
n
2D in
a pl
ane
parall
e
l which are patte
rn A and
Pattern B. Pa
tte
rn A i
s
the
straig
ht an
d
narro
w path
of
contami
nant
solutio
n
and
the pattern B
is bend
an
d
wide path of
contamin
ant
solution.An
AC
voltage su
ppl
ied of 2.5, 3.5, 4.5 and 6.
0 kV wa
s a
p
p
lied to the i
n
sul
a
ting mat
e
rial
s of LL
DPE-
Natural Ru
bb
er blen
ds wit
hout nanofille
r. The conta
m
inant flow rate is adju
s
te
d according t
o
the
applie
d volta
ge a
s
stated
in IEC 6
0
5
87 sta
nda
rd.
The to
p el
ectro
de i
s
conne
cted to
HV
electrode
whi
l
e gro
und
ed
at bottom el
ectro
de. In
this
study, on
ly the insulat
i
ng sample
a
nd
electrode
are
con
s
id
ere
d
while
other
a
c
cesso
r
ie
s a
r
e negl
ecte
d. Table 1
sh
o
w
s th
e ap
pli
e
d
voltage and
contamina
n
t flow rate a
c
co
rding to
IEC 60587
stand
ard that used in
the models.
(a)
(b)
Figure 2. 2D
parall
e
l mode
l of different patte
rns
of con
t
aminant sol
u
tion, (a) Patte
rn A:
straig
ht and n
a
rrow p
a
th of contami
nant
solu
tio
n
and,
(b) Pattern B: bend an
d wi
der path of
contami
nant solutio
n
HV
Electrod
e
Sample
surface
Ground
Electrod
e
Path of
contami
nant
solution
Air
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 13, No
. 4, Decem
b
e
r
2015 : 119
4 – 1203
1197
Table 1. Te
st para
m
eters [21]
Test voltage (kV)
Preferr
ed test voltage for met
hod
1 (kV)
Contaminant flo
w
rate
(ml/min)
1.0 to 1.75
-
0.075
2.0 to 2.75
2.5
0.15
3.0 to 3.75
3.5
0.30
4.0 to 4.75
4.5
0.60
5.0 to 6.0
-
0.90
It is importa
nt to state that the mate
rials p
r
o
perti
es pa
rticul
arl
y
electri
c
co
ndu
ctivity,
relative pe
rm
ittivity of
the sample, co
ntam
inant so
lution and ai
r are requi
re
d for simul
a
tion
purpose. In t
h
is
s
t
udy, the relative permittivit
y
of
the
air
wa
s fixed
to 1, the
co
nd
uctivity of the
air
is 2 x 10
-4
Sm
-1
[13] andt
he rel
a
tive p
e
rmittivity of
contam
i
nant
solutio
n
is
81
[14]. In the FEM
simulatio
n
, the pa
ram
e
te
r u
s
ed to v
a
ry the
cont
aminant flo
w
rate i
s
the
con
d
u
c
tivity of
contami
nant
solutio
n
. The
co
ndu
ctivity of co
nt
amina
n
t sol
u
tion
was va
ried
correspon
dingly
to
the volume
o
f
conta
m
inant
sol
u
tion
ba
sed o
n
conta
m
inant flo
w
rate a
s
sho
w
n
in T
able
2.
The
volume of co
ntaminant
sol
u
tion wa
s obt
ained from
e
a
ch
conta
m
in
ant flow rate
sho
w
n in T
a
b
l
e 1
and the cond
uctivity of contaminant wa
s cal
c
ul
at
ed by using eq
u
a
tion (1
). The resi
stan
ce
of
contami
nant
solutio
n
is
calcul
ated u
s
i
ng equ
at
ion
(2) by in
se
rting the resistivity equal to
3.95
Ω.
m
as stated in
IEC 60587
stand
ard. Th
e resi
stivity
of each
cont
aminant volu
me wa
s
cal
c
ulate
d
fo
r the
second
time to
obt
ain el
ec
tri
c
condu
ctivity by usin
g e
qua
tion (2
)
and
(1)
r
e
spec
tively.
1
(
1
)
l
R
A
(
2
)
whe
r
e
R
,
,
l
and
A
are
the
re
sist
ance, resi
stiv
ity,
length
of contami
nant solutio
n
a
nd area
o
f
contami
nant solutio
n
,
resp
ectively.
Table 2. Elect
r
ic
cond
uctivity of contamin
ant solutio
n
Volume of conta
m
inant (cm
3
)
0.15
0.30
0.60
0.90
Area of contamin
ant (cm
2
)
0.70
0.77
1.18
1.24
Resistance (k
Ω
)
65.83
32.92
16.46
10.97
Resistiv
ity
(
Ω
.m
)
92.16
50.7
38.85
27.2
Conductivity
(S/
m
)
0.011
0.020
0.026
0.037
2.2. Insulatin
g
Samples Properti
e
s
The fo
rmul
ated the
r
mo
pl
astic ela
s
to
mer
materi
al
co
mpo
s
e
d
of Line
ar Lo
w-Den
s
ity
Polyethylene with
Natu
ral
Rubber
(LLDPE/NR) fille
d and unfilled with di
fferent percentage of
silicone oxid
e
(SiO
2
)
is
p
r
es
e
n
t
ed
in th
is
w
o
rk
.
T
h
e
c
o
nd
uc
tivity o
f
e
a
c
h c
o
mp
os
itio
n w
a
s
obtaine
d fro
m
the con
d
u
ctivity analysis p
a
rtic
ula
r
ly polarizatio
n and de
pol
arization curren
t
(PDC)
mea
s
urem
ent [22].
The
ca
pa
citance valu
e
was
obtaine
d f
r
om m
e
a
s
u
r
e
m
ent of
sam
p
le
usin
g L
C
R
meter
betwe
en the
up
per and
bottom
ele
c
trod
e. T
able
3
sho
w
s
com
poun
d
and
desi
gnation
of the sampl
e
s with thei
r material
pro
pertie
s
. For
FEM simulati
on, the materials
prop
ertie
s
of
con
d
u
c
tivity and pe
rmittivity of sa
mpl
e
material i
s
re
quire
d an
d th
e inform
ation
can
be o
b
taine
d
i
n
Ta
ble
3. Th
e cond
uctivity of e
a
ch
sam
p
le
wa
s
cal
c
u
l
ated u
s
in
g e
quation
(3)
a
n
d
permittivity was calculated
usin
g equ
atio
n (4).
()
()
ro
pd
p
oo
it
i
t
CU
(
3
)
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
Modellin
g on
Tra
cki
ng Te
st
Conditio
n
of Polym
e
r Na
n
o
com
p
o
s
ite u
s
ing Finite …
(Fatin L.M.)
1198
whe
r
e
o
is free
spa
c
e di
ele
c
tric
con
s
tant (
8.854 x 10
-12
F/m),
o
C
is
c
a
pac
itanc
e
of material,
and
o
U
is applie
d voltage (10
00 V).
r
o
Cd
A
(
4
)
whe
r
e
o
is free
spa
c
e diel
ect
r
ic
con
s
tant (
8.854 x 10
-12
F/m),
C
is ca
pa
citan
c
e of ma
terial,
d
is thickness of sample mat
e
rial (1mm) and
A
is area of sampl
e
(5
0.2
7
cm
2
).
Table 3. Co
m
poun
d de
sign
ation and mat
e
rial p
r
op
ertie
s
T
e
st sample
Composition %
w
t
Designation
Material Properti
es
LLDPE NR
Nanofiller
Capacitance
(pF)
Relative
Perm
ittivity
Conductivity
(S/m)
Unfilled
LLDPE+NR
80 20
0
P0
120.67
2.71
2.944E-12
LLDPE + NR
+ SiO
2
80 20
1
A1
113.64
2.55
4.759E-13
LLDPE + NR
+ SiO
2
80 20
3
A3
106.63
2.36
4.356E-13
LLDPE + NR
+ SiO
2
80 20
5
A5
106.66
2.40
3.859E-13
LLDPE + NR
+ SiO
2
80 20
7
A7
118.60
2.66
8.055E-12
3. Results a
nd Analy
s
is
3.1. Electric Field and Cu
rrent
Densi
t
y
Anal
y
s
is
The p
r
o
b
lem
type cho
s
e
n
in thi
s
wo
rk
wa
s AC condu
ction
an
alysis to a
n
a
l
yse the
distrib
u
tion of
electri
c
field
cau
s
ed by AC voltages
i
n
insulato
r m
a
terial
s. Vari
ation of the field
with respect
to time is
assume
d to
be si
nu
soid
al
. For AC
co
ndu
ction p
r
o
b
lems, th
e field
simulato
r solves lo
cal an
d integral q
uanti
t
ie
s is represented in the followin
g
equ
a
t
ion:
i
0
U
(
5
)
whe
r
e, ele
c
tri
c
co
ndu
ctivity,
σ
and com
p
onent of ele
c
tric pe
rmittivity,
ε
are con
s
ta
nts within e
a
ch
block of the model. The
complex vecto
r
of electri
c
field inten
s
ity is cal
c
ul
ated u
s
ing e
quatio
n
6;
d
Eg
r
a
U
(
6
)
And equatio
n
for compl
e
x vector of a
c
tive current de
nsity is expan
ded to
J
E
(
7
)
The ele
c
tri
c
f
i
eld dist
ributi
on an
d voltage di
stributio
n of the insu
lator surfa
c
e
for both
contami
natio
n flow patterns A and B are
sho
w
n re
sp
e
c
tively in Figure 3(a
)
-3
(d). Th
e ele
c
tric
field and
current den
sity value were me
asu
r
ed
alon
g
the conta
m
in
ant solutio
n
p
a
th in red li
ne
as
sho
w
n in
Fig
u
re 3
(
a
)
-3(d
).
Non
-
unifo
rm
voltage dist
ri
bution
was
o
b
se
rved in Fi
gure
3 (b
) a
n
d
(d
)
whe
r
e
the
distribution
of vo
ltage g
r
ad
uall
y
decre
as
es from
HV
ele
c
trode to
the
ground
ele
c
tro
d
e
.
Mean
while, from the dist
rib
u
tion of ele
c
tric field,
the el
ectri
c
field di
stributio
n see
m
s hig
her
at a
certai
n region
esp
e
cially at
t
he highe
r current den
sity
patha
s sh
ow
n in Figu
re 3
(a)
and
(c). It is
believed that
high cu
rrent
density ca
n
cau
s
e dr
y
-
b
and an
d eve
n
tually lead to arci
ng eve
n
ts
across d
r
y-b
and re
gion.T
he cu
rrent d
ensity is
the
density of leakage
curre
n
t flowing in
the
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
9
30
TELKOM
NIKA
Vol. 13, No
. 4, Decem
b
e
r
2015 : 119
4 – 1203
1199
certai
n cro
s
s-se
ctional
area of the
contami
nation sol
u
tion
cond
uctin
g
film.The surface
discha
rge
s
a
r
e mu
ch rel
a
ted to the activity of
the d
r
y-ba
nd arcin
g
that occurred in a ce
rta
i
n
regio
n
con
s
ta
ntly cau
s
e
s
b
y
leakage
cu
rrent.Hig
he
r el
ectri
c
stre
ngt
h di
stributio
n
appe
ars at th
e
narro
w an
d e
dge
sha
pe
of co
ntamina
n
t sol
u
tion. At
this
regi
on, th
e ioni
zation
p
r
ocess
may b
e
o
c
c
u
rr
ed
as
th
e
e
l
e
c
tr
ic field
in
te
ns
ity is
c
o
nc
en
tr
ated
at
the edge
s. The ele
c
tric
fi
eld
di
strib
u
tio
n
is more critical in the narro
w path of the
contami
nant
as sho
w
n in
Figure 3 (a
).
St
r
e
n
g
th
E(
10
5
V/
m)
RM
S
Ma
g
n
i
t
u
d
e
1
.
320
1
.
118
1
.
056
0
.
924
0.
7
9
2
0
.
660
0.
528
0.
3
9
6
0
.
264
0
.
132
0.
00
0
(a)
Vo
l
t
a
g
e
U(
V)
RMS
Va
l
u
e
2
470
22
23
19
76
17
29
1
482
12
35
988
74
1
494
247
0
(b)
St
r
e
n
g
th
E(
10
5
V/
m)
RM
S
Ma
g
n
i
t
u
d
e
1
.
740
1
.
566
1
.
392
1
.
218
1.
0
4
4
0
.
870
0.
696
0.
5
2
2
0
.
348
0
.
174
0.
00
0
(c)
Vo
l
t
a
g
e
U(
V)
RM
S
Va
l
u
e
2470
222
3
197
6
172
9
1482
123
5
98
8
741
494
247
0
(d
)
Figure 3. (a)
Electri
c
field distrib
u
tion fo
r Pa
ttern A, (b) Voltage di
stribution for P
a
ttern A,
(c) Elec
tric
field dis
t
ribution for Pattern
B, (d) Voltage
dis
t
ribution for Pattern B
Grap
h of cu
rrent den
sity and ele
c
tri
c
field
distri
buti
on alon
g the
insulato
r surface a
r
e
rep
r
e
s
ente
d
in Figure 4 a
nd Figu
re 5,
respe
c
tive
ly. As sh
own, the cu
rre
nt den
sity and ele
c
t
r
ic
field are in
creasi
ng towards the g
r
ou
n
d
elect
r
ode f
o
r na
rro
w
co
ntaminant pa
th as illust
rat
e
d in
Figure 4
(a
)
and Fi
gure 5
(
a).At both
p
a
tterns A
an
d B, the ave
r
age
an
dmax
imum value
of
curre
n
t den
si
ty and ele
c
tri
c
field in
crea
se
whe
n
the
applie
d voltage in
cre
a
ses.
The maximu
m
value of
cu
rrent
den
sity for P
a
ttern
A a
r
e
0.0
7
6 A/cm
2
, 0
.
192A/cm
2
, 0.323A/c
m
2
, a
nd
0.614A/cm
2
f
o
r 2.5
kV, 3
.
5 kV, 4.5
kV and
6.0
kV, respe
c
tively. The simil
a
r o
u
tco
m
e
s
of
increme
n
t in the maximum
value of current den
sity
with wide
r pa
th were
withi
n
expectation
s.
The m
a
ximu
m value
of current de
nsit
y and el
ect
r
i
c
field
i
s
obt
a
ined at
a di
stan
ce of
3.5
cm
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
Modellin
g on
Tra
cki
ng Te
st
Conditio
n
of Polym
e
r Na
n
o
com
p
o
s
ite u
s
ing Finite …
(Fatin L.M.)
1200
measured fro
m
the HV
ele
c
trod
e for na
rrow an
d st
rai
ght co
ntamin
ant path.
It is notice
d
that
for
each applie
d voltage, high
est value is o
b
se
rved in
th
e narrow a
n
d
straight cont
aminant path.
For
Pattern B, the maximum v
a
lue of
curre
n
t den
sity an
d ele
c
tric fiel
d we
re fou
n
d
at a dista
n
ce of
1.7 cm
mea
s
ured
from th
e
HV ele
c
trod
e. The m
a
ximum value
reco
rde
d
for current de
nsit
y for
Pattern B a
r
e
0.066 A/
cm
2
, 0.167A/cm
2
, 0.287A/cm
2
, and 0.537A/cm
2
for 2.5
kV
, 3.5 kV, 4.5
kV
and 6.0
kV, respe
c
tively.The hig
her
curre
n
t den
si
ty and elect
r
ic field a
r
e reco
rde
d
at the
bendi
ng-sh
ap
ed of
conta
m
inant flo
w
for the
ben
d
and
wide
pat
h of contami
nant solution
as
depi
cted in Fi
gure 4
(
b
)
and
5(b).
The re
sult
s
sho
w
that in
cre
a
si
ng in
applie
d voltage ca
uses a
n
increa
se
of both
maximum
current d
e
n
s
ity and el
ect
r
ic fi
eld. It is b
e
li
eved that hi
g
her
appli
ed v
o
ltage m
a
y g
i
ve
more en
ergy for elect
r
on to get depo
sited on the so
l
i
d insulato
r surface. The p
r
esen
ce of this
cha
r
ge in
cre
a
se
s the su
rf
ace cond
ucti
vity and leads
to the incre
a
sin
g
of discharg
e
magnit
ude.
The la
rge
current d
e
n
s
ity causes
dry-ba
nd to b
e
formed. Thi
s
co
ntinuou
s p
r
o
c
ess will
devel
op
arci
ng spa
r
ks and ca
uses
degradatio
n p
r
ocess of t
he
insul
a
ting surface called
ca
rbon track.
Dis
t
an
c
e
fr
om h
i
gh
v
o
lta
g
e
el
ec
t
r
od
e to
gr
ou
n
d
el
ec
tr
od
e (c
m)
0
123
45
C
u
r
r
en
t
D
ens
i
t
y (
A
/
c
m
2
)
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
2.5 k
V
3.5
k
V
4.5
k
V
6.0
k
V
(a)
Dist
ance f
r
om hi
gh v
o
lt
ag
e elect
r
ode
t
o
ground e
l
ect
r
ode (cm)
0
123
45
Cu
r
r
e
n
t
De
n
s
i
t
y
(
A
/
c
m
2
)
0.
0
0.
1
0.
2
0.
3
0.
4
0.
5
0.
6
2.
5 kV
3.
5 kV
4.
5 kV
6.
0 kV
(b)
Figure 4. (a)
Curre
n
t den
si
ty for Pattern A, (b) Cu
rrent
density for Pattern B
Dis
t
anc
e fr
om h
i
gh
voltage elec
tr
ode to
ground
elect
r
ode
(cm
)
01
23
45
E
l
ec
t
r
i
c
f
i
e
l
d
(
k
V
/
cm
)
0.
0
0.
2
0.
4
0.
6
0.
8
1.
0
1.
2
1.
4
1.
6
1.
8
2.
5 k
V
3.
5 k
V
4.
5 k
V
6.
0 k
V
(a)
Di
sta
n
ce f
r
om hi
gh v
o
lt
ag
e el
ect
r
o
de t
o
g
r
o
und e
l
ect
r
ode
(
c
m)
0
12345
E
l
ect
r
i
c
f
i
el
d (
k
V
/
cm
)
0.
0
0.
2
0.
4
0.
6
0.
8
1.
0
1.
2
1.
4
1.
6
2.
5 k
V
3.
5 k
V
4.
5 k
V
6.
0 k
V
(b)
Figure 5. (a)
Elec
tric
field for for Pattern
A, (b) Elec
tric field for Patt
ern B
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 13, No
. 4, Decem
b
e
r
2015 : 119
4 – 1203
1201
As mentio
ne
d ea
rlier, th
e
avera
ge val
ue of
curre
n
t den
sity and
elect
r
ic fiel
d
for both
con
d
ition
s
incre
a
sed with
the incre
a
se
d of
applied
voltage. The averag
e cu
rrent den
sity and
averag
e ele
c
t
r
ic field that p
r
esented i
n
T
able 4
sh
o
w
s that there is
not a big ga
p
of differen
c
e
s
.
Ho
wever, the
standa
rd d
e
v
iation (SD) for the curr
ent
density and
electri
c
field
shows a different
pattern. T
he
SD for the
el
ectri
c
field
in
the wi
der an
d
ben
d p
a
th of
co
ntamina
n
t
solutio
n
sho
w
s a
small differen
t
range of 0.2
1
to 0.36. Meanwhile,
a qu
ite large diffe
rent of the SD ra
nge of 0.
17
to 1.0 report
ed in the electri
c
field of
straig
ht and narro
w of co
ntaminant sol
u
tion. The small
stand
ard
devi
a
tion value
cl
ose to
ze
ro i
ndicates th
e
rang
es
of me
asu
r
ed
data
are n
e
a
r
to the
mean val
ue.
Thus, th
e da
ta evaluate from the
b
end
and
wid
e
pa
th of co
ntami
nant a
r
e m
o
re
likely to
be
co
nsi
s
tent a
nd
doe
s n
o
t diffe
r
with e
a
ch
v
a
lue.
Unli
ke t
he el
ectri
c
field of
strai
ght a
n
d
narro
w of co
ntaminant so
lution, the ra
nge of
mea
s
ured d
a
ta are in a wide
rang
e and n
o
t
uniformly
as
at so
me p
o
int
the valu
e a
r
e
mu
ch m
o
re
l
a
rge
o
r
small
e
r tha
n
the
m
ean val
ue. T
he
SD for curren
t density for both co
ntami
nant sol
u
tion
pattern
s is in
the ran
ge of 0.0.2 to 0.37 for
straig
ht and
narro
w of co
ntami
nant sol
u
tion while 0.
04 to 0.36 for bend a
nd wide conta
m
in
ant
solutio
n
. Thu
s
, the rang
e
value me
asu
r
ed in th
e
current den
sity fo
r both
ci
rc
um
st
an
ce
s is
cl
o
s
e
to the averag
e value and n
o
t varied.
Table 4. Average Current
Den
s
ity and
Electri
c
Field
for different le
vel of applied
voltage
Applied voltage (kV)
2.5
3.5
4.5
6.0
Average Cur
r
ent
Densit
y
(A/cm
2
) f
o
r
contaminant patt
e
rn
w
i
th na
rro
w
p
a
th
0.036
0.093
0.158
0.300
Average Electric Fi
eld (kV/cm) for
contaminant patt
e
rn
w
i
th na
rro
w
p
a
th
0.333
0.474
0.609
0.810
Average Cur
r
ent
Densit
y
(A/cm
2
) f
o
r
contaminant
patt
e
rn w
i
th w
i
de
pat
h
0.038
0.093
0.157
0.300
Average Electric Fi
eld (kV/cm) for
contaminant
patt
e
rn w
i
th w
i
de
pat
h
0.344
0.468
0.604
0.810
3.2. Anal
y
s
is
on Sample w
i
th Different Loading
Nanofiller
The contami
n
ant flow rate
cho
s
e
n
in thi
s
work is
0.6
mlmin
-1
with t
he appli
ed vo
ltage o
f
4.5 kV; this
voltage level
found to be
most crit
ical
on materi
al
ero
s
ion [23]
. Five different
sampl
e
s
we
re studi
ed b
y
taking int
o
co
nsid
erat
ion of dissi
m
ilar pe
rmittivity and electri
c
con
d
u
c
tivity a
s
sh
own in Table 3. It is notic
ed thatthe
unfilled LL
DPE and natural rubb
er bl
e
nds
have the
lo
west p
e
rmittivity and
high
est
co
ndu
ctivit
y. Simulation
a
nalysi
s
re
sult
s
sho
w
that t
he
polymer b
a
se
d material
without nan
ofille
r (sampl
e
P0) re
co
rde
d
hi
gher valu
e in
current d
ensity
and el
ectri
c
fi
eld compa
r
e
d
to other
sam
p
les fille
d wit
h
nan
ofiller. T
he maximum
value of curre
n
t
den
sity and maximum ele
c
tri
c
field of LLDPE-NR bl
e
nd witho
u
t filler from the g
r
aph illu
strate
d in
Figure 8
ha
d
given
a val
ue of
0.289
A/cm
2
and
1.
111 kV/cm, resp
ectively. Mean
while,
t
h
e
sampl
e
s
with
nanofille
r h
a
ve the sm
a
llest value
o
f
current de
nsity and el
ectri
c
field.
This
demonstrates that the addition of
nanofiller i
n
the blends
could in
crease the tracking and erosion
resi
stan
ce. A
n
expe
rime
ntal test
on th
e same
co
m
positio
n
of sample
A
3
was co
ndu
cte
d
by
other rese
archers and the
y
found
that this compo
s
iti
on sho
w
s
lo
we
st LC and
no carbon track
formation. [5]. However, simulation re
sults also sho
w
that sampl
e
A1, A3 and A5 have a goo
d
resi
stan
ce
in
trackin
g
an
d erosio
n te
st due to
th
e lowest
cu
rrent d
e
n
s
ity and el
ectri
c
field
distrib
u
tion of
range of 0.2
85 to 0.286 A/cm
2
and 1.0
9
to 1.10 kV/cm, re
spe
c
tively. Fig. 8
shows
the current d
ensity along t
he insul
a
tor
surfa
c
e fr
om HV elect
r
ode
to the groun
d electrode
with
sampl
e
A7
shows th
e hi
g
h
average
value
of curre
n
t den
sity. T
he n
anofille
r more tha
n
f
i
ve
percent a
r
e
reporte
d a
s
n
o
t benefi
c
ial to the bl
en
d [24]. Highe
r n
anofiller l
oadi
ng in the bl
e
nds
tend to
ag
gl
omerate in
t
he
com
p
o
s
ition.
When
th
e mixture
i
s
not unifo
rmly
dispe
r
si
ng,
the
comp
oun
ds b
e
com
e
ro
ugh
er and thi
s
co
uld lead to hi
gh LC d
e
n
s
ity.
The ave
r
a
ge
value of
cu
rre
n
t den
sitie
s
meas
ured i
n
t
he
sampl
e
s is in the
ra
nge
of 0.156
to 0.157 A/
cm
2
and
average el
ect
r
ic fi
eld of the
sa
mples are 0.
60 to 0.6
1
kV/cm re
sp
ect
i
vely.
Although the
differen
c
e f
r
om the
ave
r
age val
ue
wa
s mino
r, the re
sulte
d
of the tracki
ng
resi
stan
ce i
s
signifi
cant. T
h
is was
hap
p
ened p
o
ssibl
y
becau
se th
e field simul
a
tion itself ha
s a
limitation. Table 5 depi
cte
d
the SD of
the curre
n
t den
sity and ele
c
tric field.
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
Modellin
g on
Tra
cki
ng Te
st
Conditio
n
of Polym
e
r Na
n
o
com
p
o
s
ite u
s
ing Finite …
(Fatin L.M.)
1202
Table 5. Stan
dard
Deviatio
n (SD)
Sample
P0 A1
A3 A5 A7
SD of Cur
r
ent D
e
nsity
0.188
0.069
0.185
0.184
0.072
SD of Electric Field
0.723
0.265
0.711
0.709
0.278
Dis
t
ance
from hig
h
vol
t
age
electr
ode to gro
und e
l
ectro
de (c
m)
1.
5
2
.
0
2.
5
3
.
0
3
.
5
Current
D
ensi
t
y
(A/
c
m
2
)
0.
16
0.
18
0.
20
0.
22
0.
24
0.
26
0.
28
P0
A1
A3
A5
A7
Figure 8. Current den
sity distributio
n
4. Conclusio
n
The mo
del
geomet
ry of plane
-p
arall
e
l two
dime
nsio
nal h
a
s been
devel
oped
i
n
QuickFiel
d
Fi
nite Eleme
n
t
Analysis software to
st
u
d
y the effe
ct of
ele
c
tri
c
field
distrib
u
tion
a
nd
curre
n
t den
sity when
co
ntrolled
pa
ra
meters pa
rti
c
ula
r
ly ap
plied voltage,
permittivity and
con
d
u
c
tivity are va
ried.
The
simulati
on an
alysi
s
t
hat is
co
ndu
cted
on the
test conditio
n
of
inclin
ed pla
n
e
trackin
g
(I
PT) set
-
up
sho
w
some
correl
ation b
e
twee
n physical pa
ram
e
ters
involved in th
e IPT test co
ndition. The
physi
cal
shap
ed of co
ntami
nant sol
u
tion
path affect
s the
curre
n
t den
sity and the electri
c
field di
stributio
n
on
the insulatin
g
sample
surf
ace. Analysi
s
on
the sam
p
le
with differe
nt loading of n
anofiller d
e
m
onstrates th
a
t
sample
with 1-5 % sili
cone
oxide nanofiller gives
a good resi
stance
to tracking and erosi
on due to
the lower value of current
den
sity and e
l
ectri
c
field o
b
tained from
simulatio
n
an
alysis. Th
us,
it can be
co
n
c
lud
ed that the
results
obtain
ed fro
m
simu
lation an
alysi
s
i
s
ing
ood
a
g
ree
m
ent
wit
h
the
previo
u
s
exp
e
rim
ent
al
results. It i
s
b
e
lieved th
at the results f
r
o
m
the
simul
a
tion a
nalysi
s
can b
e
imp
r
ov
ed in
the futu
re
by con
s
ide
r
in
g other fa
ctors to be incl
ud
ed in simul
a
tion wo
rks.
Ackn
o
w
l
e
dg
ements
The a
u
thors
expre
s
s their sin
c
e
r
e g
r
ati
t
ude to
Unive
r
siti Te
kn
olog
i Malaysia
(UTM), for
the use of facilities an
d b
y
awardi
ng a
rese
arch uni
versity gra
n
ts under vote0
3
H8
6 and 4L
133
from
the Re
sea
r
ch
Ma
na
gement Ce
ntre UTM and
Malaysi
a
M
i
nistry of
Hi
gher Edu
c
ati
o
n
(MO
H
E).
Referen
ces
[1]
Vasud
e
v N, S
Ganga, RS S
h
iv
akum
ara Ara
d
h
y
a, B L
a
lith
a
Pai
. Effect o
f
ATH filler co
ntent on th
e
perfor
m
a
n
ce o
f
silicon
e rub
b
e
r by inc
lin
ed
pla
ne tracki
ng
and
erosi
on te
st metho
d
.
2012 IEEE 10
th
Internatio
na
l C
onfere
n
ce o
n
the Prop
erties
a
nd App
licati
on
of Diel
e
ctric Materia
l
s
.
2012:
1-4.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 13, No
. 4, Decem
b
e
r
2015 : 119
4 – 1203
1203
[2]
T
u
miran T
et.al. Accelerate
d
Ag
i
n
g
Effe
ct on
Epoxy
-
p
o
ly
si
lo
xa
ne-Ric
e Husk Ash Poly
meric Insulator
Material
. T
E
L
K
OMNIKA (T
ele
c
ommunic
a
tio
n
Co
mp
utin
g El
ectronics
an
d
Contro
l)
. 20
01;
10(4)
: 65
5-
666.
[3]
Venkatar
ama
n
S et.al.
Arc tr
acking resist
ance of poly
m
e
r
i
c ma
te
ri
al
s in
o
xyg
en
de
fi
ci
en
t co
nd
i
t
i
o
n
s
.
200
3 IEE Conf
erenc
e in Elec
trical Insul
a
tion
and Di
electric
Phen
ome
na. 2
003: 10
4-1
07.
[4]
Piah
M et.al.
Effect of AT
H filler
on
the
el
ectric
al track
i
n
g
a
nd
erosi
o
n
prop
erties
of n
a
tural
rub
ber-
LLDPE ble
nds und
er
w
e
t
co
ntamin
ated
co
nd
i
t
ions.
Jour
nal
o
f
Industrial T
e
c
hno
logy
. 2
0
0
4
; 13(1): 27-
40.
[5]
Jamail NAM et.al.
Electrical tracking ch
ara
c
teri
z
a
t
i
o
n
of LLDPE-
Natura
l
Rubb
er ble
n
d
s
filled w
i
th
nan
ofill
ers
.20
1
3
IEEE Co
nfer
ence
in E
l
ectri
c
al Insu
la
ti
on and Die
lectric Phen
ome
na
(C
EIDP).
201
3:
695-
698.
[6]
Cresp
o
-San
do
va J et.al. Rate of energ
y
absor
pt
ion
as
indic
a
tor for the tracking/er
osio
n test of
silico
ne ru
bb
er.
IEEE Transactions
on
Die
lect
rics and El
ectri
c
al Insul
a
tio
n
. 201
0; 17(6): 17
72-1
780.
[7]
Vas JV, B Ve
nkatesu
l
u
and
MJ T
homas.
T
r
ackin
g
an
d er
osio
n of si
lico
ne ru
bb
er na
n
o
comp
osite
s
und
er D
C
v
o
lt
ages
of
both
pol
arities.
IEE
E
T
r
ansacti
on
s on
Di
electric
s an
d El
ectric
al Ins
u
lati
on.
201
2; 19(1): 9
1
-98.
[8]
Kumag
a
i S.
L
eaka
ge c
u
rren
t
suppr
essio
n
and r
e
sista
n
ce
to trackin
g
a
nd er
osi
on
of
HT
V silicon
e
rubb
er
w
i
t
h
ad
ded sil
i
co
ne pl
asticizer
. IEEE Transactions
on Die
lectrics
and El
ectrical
Insulati
on
.
200
7; 14(2): 38
4-39
2.
[9]
Ro
w
l
an
d SM et.al. Use of i
m
age a
n
a
l
ysis
in
DC i
n
cli
n
e
d
pla
ne tracki
ng tests of na
no a
nd mic
r
o
compos
ites
IEEE Transactio
n
s on Di
electri
cs and Electric
al Insul
a
tio
n
. 2011; 18(
2): 365
-374.
[10]
Piah
MAM, A
Darus, A
Hass
an. El
ectrical
tr
a
ckin
g
p
e
rform
ance
of
LLDP
E
-natura
l
ru
bb
er b
l
en
ds
b
y
empl
o
y
in
g co
mbin
ation
of l
eaka
ge c
u
rren
t
leve
l
an
d r
a
te of car
bon
track prop
ag
ation.
IE
EE
T
r
ansactio
n
s o
n
Diel
e
ctrics a
nd Electric
al In
sulati
on
. 20
05; 12(6): 12
59-
12
65.
[11]
S
y
ak
ur A et.al.
Electrical T
r
acking F
o
rmati
o
n
on Si
lan
e
E
p
o
x
y Res
i
n u
n
der Vari
ous C
ontami
nants
.
T
E
LKOMNIKA (T
eleco
m
mu
ni
cation C
o
mputi
ng Electro
n
ics
and C
ontrol)
. 2
013; 11(
1): 17-
28.
[12]
S
y
ak
ur A et.al.
Leaka
ge curr
ent measur
em
ent
of epo
xy r
e
sin com
p
o
u
n
d
w
i
t
h
sil
i
con r
ubb
er.
201
1
Internatio
na
l C
onfere
n
ce o
n
Electric
al E
ngi
ne
erin
g and Infor
m
atics (ICEEI)
. 2011: 1-5.
[13]
Yong Z
et.al., Mecha
n
ism for
chan
ge
in
leak
age c
u
rre
nt
w
a
veform on
a
w
e
t silic
one
rub
b
e
r surface-
a
stud
y
us
in
g a d
y
nam
ic 3-D mode
l
. IEEE Transactions on Dielectri
cs and Electrical Ins
u
lation
. 200
5;
12(3): 55
6-5
6
5
.
[14]
Da Silv
a E a
n
d
SM Ro
w
l
a
n
d
.
T
he de
pe
n
dency
of w
a
ter dropl
et be
ha
viour a
nd
leak
age c
u
rrent
pattern on el
ectrode
confi
g
uratio
n
. IEEE Conf
erenc
e
on E
l
ectrical
Insulation and
Dielectric
Phen
ome
na (C
EIDP '
09). 200
9: 242-2
45.
[15]
Basap
pa
P
et.al.
Si
mu
latio
n
of Electric F
i
e
l
d distri
butio
n a
r
oun
d w
a
ter dr
opl
ets on
outd
oor i
n
sul
a
tor
surfaces
. Conf
erenc
e Recor
d
of the 2010 I
EEE Internatio
nal S
y
mp
osiu
m on in Electri
c
al Insul
a
tio
n
(ISEI)
).
2008: 50-54.
[16]
Haife
ng G et.al
.
Effect of H
y
dr
oph
obic
i
t
y
on
El
ectric
F
i
el
d Distributi
on an
d
Disch
arg
e
s Alon
g
Vari
ou
s
W
e
tted H
y
dro
pho
bic Surfac
es.
IEEE Transactions
on Dielectri
cs and
Electrical Ins
u
lation
. 20
08;
15(2): 43
5-4
4
3
.
[17]
Sarang B, P
Basappa, V
Lakda
w
ala.
Effect of w
a
ter dropl
ets an
d w
a
ter films
on t
h
e el
ectrica
l
perfor
m
a
n
ce o
f
practical out
door insu
lator
s
. Conference Recor
d
of
the 2010 IEEE
International
S
y
mp
osi
u
m on
in Electrica
l
Insulati
on (ISEI). 2010: 1-6.
[18]
Haor
an W
et.
a
l.
Si
mu
latio
n
study on
E-fiel
d distri
butio
n
and c
o
ro
na c
h
aracteristics
of compos
ite
i
n
su
la
to
r wi
th
w
a
ter drop
lets
. IEEE Confer
ence on Electrical
Insulation and Dielectric Phenomena
(CEIDP). 2013:
422-4
25.
[19]
Illias H
A
, G Chen
an
d PL
L
e
w
i
n. M
easur
e
m
ent
an
d mo
d
e
lli
ng
of parti
al
disch
arge
be
havi
our i
n
a
spher
ical cav
i
ty
w
i
th
in a sol
i
d
diel
ectric
ma
terial as a fun
c
tion of cavit
y
diam
eter.
201
0 10th IEEE
Internatio
na
l C
onfere
n
ce o
n
Solid D
i
e
l
ectrics
(ICSD).
2010: 1-4.
[20]
Du B et.al. D
y
namic b
ehav
io
r of
w
a
ter dro
p
l
et for evalu
a
ti
ng outd
oor i
n
s
u
lator.
IEEE 9th International
Confer
ence
on
the Properties
and Ap
plic
atio
ns of Diel
e
ctric
Materials (ICP
ADM 200).
20
0
9
: 272-2
75.
[21]
Standar
d
B.
IEC 60
587 E
l
e
c
trical ins
u
l
a
tin
g
materi
als us
ed u
nder s
e
ve
re a
m
bi
ent co
nditi
ons-T
es
t
meth
ods for ev
alu
a
ting res
i
sta
n
ce to tracking
and er
osio
n
, British Stand
ard.
200
7: 13.
[22]
Jamail
N
et.al.
Effect of Nan
o
fillers
on
the
Po
larizati
on
an
d Dep
o
lar
i
zatio
n
Current Ch
ara
c
teristics
o
f
LLDPE-
NR Co
mpou
nd for Hi
gh Volta
ge Ap
plicati
o
n
. Adv
a
nces in Mater
i
al Scie
nce a
n
d
Engin
eeri
n
g
.
201
4: 1-7.
[23]
Heg
e
r G et.al. A comparativ
e
stud
y
of insu
lat
o
r materia
l
s e
x
pose
d
to hig
h
voltag
e AC an
d DC surfac
e
disch
arges.
IEEE Transactio
n
s on Di
electri
cs and Electric
al Insul
a
tio
n
.
2010; 17(
2): 513
-520.
[24]
Me
yer
L
H
,
EA Cher
ne
y an
d SH
Ja
yar
a
m. T
he
role
of in
o
r
gan
ic fill
ers i
n
silic
on
e ru
bbe
r for out
doo
r
insul
a
tio
n
al
um
ina tri-h
y
d
r
ate
or silica.
IEEE Electrical Ins
u
lation Maga
z
i
ne
. 2004; 20(
4): 13-21.
Evaluation Warning : The document was created with Spire.PDF for Python.