IEEE

www.ietdl.
org
Published in IET Generation, Transmission & Distribution
Received on 19th August 2009
Revised on 2nd June 2010
doi: 10.1049/iet-gtd.2009.0685
ISSN 1751-8687
Lightning attachment models and maximum

shielding failure current of overhead
transmission lines: implications in insulation
coordination of substations
P.N. Mikropoulos T.E. Tsovilis
High Voltage Laboratory, Department of Electrical Energy, School of Electrical and Computer Engineering,
Faculty of Engineering, Aristotle University of Thessaloniki, Greece
E-mail: pnm@eng.auth.gr
Abstract: The maximum shielding failure current of overhead transmission lines is an important parameter in
evaluating the shielding performance of the lines and in insulation coordination of substations. General expressions
for the estimation of the maximum shielding failure current of transmission lines, derived by employing several
lightning attachment models in shielding analysis, are presented. An application to typical 110 kV up to 1150 kV
overhead transmission lines shows that there is a great variability in maximum shielding failure current among
lightning attachment models. The importance of maximum shielding failure current in insulation coordination of
substations is demonstrated with the aid of alternative transients program-electromagnetic transients
program (ATP-EMTP) simulations. The computed overvoltages impinging on 150 and 400 kV gas insulated
system (GIS) substations because of shielding failure of the incoming overhead transmission lines, being dependent
upon shielding failure current, vary with the lightning attachment model employed in shielding analysis of the
lines. Implementation of the electrogeometric model adopted by IEEE Std 1243:1997 in shielding analysis imposes
high requirements on protection of the substations against incoming shielding failure surges.
Introduction
Shielding of overhead transmission lines against direct

lightning strokes to phase conductors is provided by shield
wires, which are metallic elements that are able to, by
physical means, launch a connecting upward discharge that
intercepts the descending lightning leader from a distance,
called striking distance, within a capture radius, commonly
called attractive radius or lateral distance. A perfect
shielding of transmission lines is achieved when lightning
strokes possessing peak current greater than the critical
current causing ashover of insulation are intercepted by
the shield wires. Apparently, some of the less intense
strokes may strike to phase conductors; however, these are
not expected to cause ashover. In practice, economical
shielding design of transmission lines is realised based on
an acceptable shielding failure ashover rate. Hence, there
is a range of currents of lightning strokes to phase
IET Gener. Transm. Distrib., 2010, Vol. 4, Iss. 12, pp. 1299 1313
doi: 10.1049/iet-gtd.2009.0685
conductors which may cause ashover. Although the lower

limit of this range can be estimated based on the
geometrical and electrical characteristics of the transmission
line, the estimation of the upper limit, that is, the
maximum shielding failure current requires extensive
geometrical analysis depending on the lightning attachment
model employed in shielding analysis.
Shielding analysis of transmission lines can be made by
implementing electrogeometric models [1]; representative
of their application is the method suggested by IEEE Std
[2], which assume the striking distance to be solely a
function of the prospective lightning peak current [3 15].
Alternatively, shielding analysis can be carried out by
employing models based on more solid physical ground of
lightning attractiveness [16 27], called hereafter in
accordance with Waters [28] generic models. A statistical
approach for shielding analysis has been introduced recently
1299
& The Institution of Engineering and Technology 2010
www.ietdl.org
[29, 30] by employing a lightning attachment model derived
from scale model experiments [31].
Table 1 Factors A, B and g to be used in (1)

Electrogeometric
model
The maximum shielding failure current of overhead

transmission lines has been formulated on the basis of
electrogeometric models in [32, 33] and of generic models
and statistical model in [34]. In the present work, the
maximum shielding failure current of typical 110 kV up to
1150 kV overhead transmission lines is estimated; results
are discussed in respect of the shielding performance of the
lines. Shielding failures of transmission lines may cause
substation outages when the incoming overvoltages exceed
the insulation level of substation equipment. The incoming
shielding failure surges depend upon shielding failure
current; thus, the maximum shielding failure current, being
the upper limit of all possible lightning stroke currents
impinging on the substation entrance owing to shielding
failure, is an important parameter in insulation
coordination of substations. This is elucidated with the aid
of ATP-EMTP simulations of 150 and 400 kV GIS
substations.
Wagner and Hileman

[3]
Young et al. [4]
2.1 Electrogeometric models

Shielding analysis according to electrogeometric models is
based on striking distance, which is dened as the distance
between the descending lightning leader and the struck
object at which the upward connecting discharge is
initiated. Striking distance to an object, S, is solely related
to the prospective lightning peak current, I, and can be
associated with striking distance to earth surface, D, by
using a factor g as
S = AI B = gD
(1)
14.2
27g
0.42
0.32
1 for h , 18 m
444/(462 2 h)
for h . 18 m
(h: shield wire
height)
Armstrong and
Whitehead [6]
6.72 0.80
1.11
Brown and
Whitehead [7]
7.1
0.75
1.11
0.65
Love [8]
2 Maximum shielding failure

current formulation
10
Whitehead [10]
9.4
0.67
Suzuki et al. [35],

derived from Golde
[36]
3.3
0.78
Anderson [13], IEEE

WG [14]
0.65
1/ba
10
0.65
1/bb
IEEE Std [2]

a
b 0.64 for UHV lines, 0.8 for extra high voltage (EHV)
lines and 1 for other lines
b
b 0.36 + 0.17 ln(43 2 h) for h , 40 m, b 0.55 for
h . 40 m where h is the phase conductor height
Despite their simplicity and widespread applicability,
electrogeometric models, with the exception of [4], do not
consider the effects of conductor height on striking
where I is in kA and S and D are in metres; factors A, B and g

are given in Table 1.
According to electrogeometric models (Fig. 1), a
descending lightning leader that reaches the arc between M
and N will strike to the phase conductor; hence a shielding
failure width, W, is dened. The latter decreases as
lightning peak current increases; the critical current
corresponding to W 0 is the maximum shielding failure
current, IMSF.
Geometrical analysis (Appendix 8.1), similar to that
detailed by Hileman [33], yields the following expression
for IMSF (kA)

IMSF =
g(hm + hp )/2
Figure 1 Shielding analysis of transmission lines
1/B
A(1 g sin a)
(2)
where factors A, B and g are given in Table 1 and hm (m), hp

(m) and a are dened in Fig. 1.
1300
hm , hp height of shield wire and phase conductor, respectively; a

shielding angle; DR horizontal separation distance between shield
wire and phase conductor; S striking distance to shield wire and
phase conductor; D striking distance to earth surface; Rm , Rp
attractive radius of shield wire and phase conductor,
respectively; and W shielding failure width
doi: 10.1049/iet-gtd.2009.0685
www.ietdl.org
distance, S. Also, most models assume a constant value for
factor g (Table 1). However, as was discussed in detail in
[31], g should depend on struck object height, lightning
peak current and interception probability, that is, the
probability for a connecting upward discharge emerging
from the air terminal.
Table 2 Factors j, E, F, z and G to be used in (5)
Rizk [18]
Petrov et al. [22]
Yuan [23]
2.2 A.J. Erikssons model

Eriksson [16], based on eld data and by using the Carrara
and Thione [37] critical radius concept for an upward
leader inception criterion, modied the electrogeometric
model by introducing in shielding analysis the attractive
radius, dened as the capture radius at which the upward
leader intercepts the descending lightning leader. The
attractive radius, R, of a conductor, is expressed as a
function of, besides lightning peak current, its height, h, as
R = 0.67h0.6 I 0.74
(3)
where I (kA) is the prospective lightning peak current and h

and R are in metres. A. J. Eriksson, performing similar
shielding analysis to that of electrogeometric models
(Fig. 1), used the attractive radius to draw arcs from
the shield wire and phase conductor up to the height of the
latter. Thus, for shielding failure width W 0 the
maximum shielding failure current IMSF (kA) is given as
(Appendix 8.2)
IMSF

1/0.74
DR + DR2 + C2 (G2 1)
=
2
0.67h0.6
p (G 1)
(4)
where G Rm/Rp (hm/hp)0.6 and C2 (hm 2 hp)2 + DR 2

and DR (m), hm (m) and hp (m) are dened in Fig. 1.
2.3 Generic models

Following Erikssons work [16], physical models for
lightning attachment, which also consider the inception of
the upward connecting discharge emerging from the
prospective struck object, were developed [17 27]. Thus,
based on different leader inception criteria, striking distance
or attractive radius expressions, taking into account
lightning parameters and prospective struck object height,
were derived. The following general expression can be used
to estimate the attractive radius of an object, R, dened as
the longest lateral distance from the object where lightning
attachment occurs
R = jh E I F + zh G
(5)
where R is in metres, I (kA) is the prospective lightning peak

current, h (m) the struck object height and factors j, E, F, z
and G are given in Table 2.
According to generic models (Fig. 1), adopting Rizks
shielding analysis [18], a shielding failure will occur when
the descending lightning leader enters the shielding failure
doi: 10.1049/iet-gtd.2009.0685
Generic model
1.57
0.45 0.69 0
0.47
0.67 0.67 0
52.47
Borghetti et al. [24],

from Dellera and
Garbagnati [19]
0.028 1
Ait-Amar and Berger

[27]
0.49 0.35 1
1
0.20 0.67 0
0.6
Using as h in (5) the object height plus 15 m

Using as h in (5) the object height minus 13 m
width W. Thus, for W 0 the maximum shielding failure

current IMSF (kA) is formulated as (Appendix 8.3)
IMSF =

1/F
DR z(hGm hGp )
j(hEm hEp )

1/F
(hm hp ) tan a z(hGm hGp )
j(hEm hEp )
(6)
where factors j, E, F, z and G are given in Table 2 and DR

(m), a, hm (m) and hp (m) are dened in Fig. 1.
It must be mentioned that the attractive radius expressions
in [22, 27] do not refer to transmission line geometry;
however, implementing these expressions in shielding
analysis of transmission lines may provide useful
information on their applicability to IMSF estimation. Also,
Yuans model [23] suffers from not yielding IMSF values
since W (Fig. 1) is not a function of lightning peak current
[E 0 in (5)]. Generic models [17 27], including
Erikssons [16], have added signicant value in knowledge
of lightning attachment. However, as was discussed in [29],
their implementation in shielding design lacks of
considering lightning interception probability.
2.4 Statistical model

Recently, investigations through scale model experiments
made possible to obtain probability distributions of striking
distance and interception radius [31, 38]; thus a statistical
approach in shielding design has been proposed [29, 39].
Interception radius of an object, dened as the lateral
distance from the object where lightning attachment
occurs, is considered as statistical quantity with a mean
value, referring to 50% interception probability, called
critical interception radius, Rci , and a standard deviation s.
It is given with reference to the striking distance to earth
1301
www.ietdl.org
Table 3 Coefcients c1 , c2 and expression of s to be used in (7)
Positive lightning
s%
c2
c1
0.235
Negative lightning
0.9
1.9(h/D)
20.75
c1
c2
s%
0.272
1.24
5.0(h/D)20.43
surface, D, as

Rci
h
+ c2
, s = c1 ln
D
D
(7)
where Rci and D are in metres and h (m) is the struck object
height. Coefcients c1 and c2 , and s in formula form are
given in Table 3 [31].
Expression (7) can be used for shielding analysis by using a
known relationship between striking distance to earth
surface, D, and lightning peak current, I, commonly of the
form D A I B. Thus, by performing shielding analysis
according to generic models (Fig. 1) and using the critical
interception radius (7), for shielding failure width W 0
the maximum shielding failure current IMSF (kA) at critical
interception is given as (Appendix 8.4)
Figure 2 Maximum shielding failure current as a function

of shield wire height with and without considering
proximity effects on lightning attachment
(kA) is well approximated by (Appendix 8.4)

IMSF
DR
=
A c1 ln(hm /hp )
1/B
=

1/B
(hm hp ) tan a
A c1 ln(hm /hp )
(8)
where DR (m), a, hm (m) and hp (m) are dened in Fig. 1 and

c1 is given in Table 3. Adopting from [8] the values of 10 and
0.65 for factors A and B, respectively, expression (8) becomes
for negative lightning

IMSF
DR
=
2.72 ln(hm /hp )
1/0.65

=
1/0.65
(hm hp ) tan a
2.72 ln(hm /hp )
IMSF
DR + 0.01h1.3
m
=
2.72 ln (hm /hp )
=
1/0.65

1/0.65
(hm hp ) tan a + 0.01h1.3
m
2.72 ln (hm /hp )
(10)
The result of (9) and (10) for hp/hm 0.75, a ratio of

conductor heights commonly found in overhead
transmission lines, is shown in Fig. 2. IMSF is always
higher when proximity effects are considered in shielding
analysis; this becomes more pronounced for higher
transmission lines and smaller shielding angles.
(9)
As was raised by Pettersson and Eriksson [40] and discussed

in detail in [30, 41], depending on transmission line
geometry, the striking distance and interception radius of
the shield wire may be affected by the presence of the
neighbouring phase conductor. In fact, the competing
upward discharge emerging from the phase conductor by
modifying the extent of development of the connecting
upward discharge that emerges from the shield wire may
cause a reduction in striking distance and interception
radius of the latter [30]. Consequently, when proximity
effects on lightning attachment are considered, W is wider
and for negative lightning assuming D 10I 0.65 [8] IMSF
1302
3 Maximum shielding failure

current of overhead transmission
lines
Fig. 3 shows the variation of maximum shielding failure
current with shielding angle, obtained by using (2), (4), (6)
and (10), for two typical 150 kV (Figs. 3a and b) and
400 kV (Figs. 3c and d) double-circuit overhead lines of the
Hellenic transmission system. It is obvious that there is a
great variability in IMSF among lightning attachment
models. However, all models agree in predicting that IMSF
increases with shielding angle; this has also been
demonstrated through simulations [42, 43].
doi: 10.1049/iet-gtd.2009.0685
www.ietdl.org
Figure 3 Maximum shielding failure current variation with shielding angle

a and b 150 kV
c and d 400 kV
Table 4 Parameters of typical overhead transmission lines

Reference Operating
voltage,
kV
Figure 4 Maximum shielding failure current variation with

shield wire height for typical overhead transmission lines,
hp/hm 0.75
a Electrogeometric models
b Generic models
doi: 10.1049/iet-gtd.2009.0685
Shield
wire
height,
m
Upper or
outer
phase
conductor
height, m
Shielding
angle,
degrees
[44]
110
25.8
20.0
27.3
[44]
110
19.4
16.3
30.1
[32]
138
17.0
13.5
29.1
[32]
138
25.0
21.3
23.4
[42]
220
19.1
11.7
20.4
[32]
230
20.0
15.6
29.6
[32]
230
35.0
29.5
17.7
[44]
330
26.1
20.0
21.5
[32]
345
20.0
14.0
29.5
[32]
345
40.0
32.4
15.4
[32]
500
25.0
17.0
14.0
[44]
500
32.0
22.2
21.7
[44]
750
38.6
24.2
17.0
[45]
750
41.1
27.9
18.1
[32]
765
30.0
18.0
18.4
[44]
765
41.8
29.9
14.6
[44]
1150
44.8
30.3
24.8
1303
www.ietdl.org
Fig. 4 shows IMSF as a function of shield wire height for
hp/hm 0.75 and by considering, according to common
practice, a reducing shielding angle with transmission line
height; the shielding angle curve in Fig. 4 is the best tting
curve depicting the reduction of shielding angle with shield
wire height for the overhead transmission lines considered
as typical in [32] (Table 4). As transmission line height
increases, IMSF augments for electrogeometric models
(Fig. 4a) but decreases for generic models (Fig. 4b).
According to the statistical model IMSF increases with
transmission
line
height,
in
agreement
with
electrogeometric models.
Fig. 5 shows the estimated IMSF values of typical

110 kV up to 1150 kV overhead transmission lines; line
parameters are listed in Table 4. All models, with the
exception of [27, 35], agree in yielding IMSF generally
less than 20 kA. The highest IMSF is obtained for the
1150 kV line, with Andersons model [13] yielding a
value of 110 kA. The latter value, determined greatly by
the factor g suggested for ultra high voltage (UHV)
lines (Table 1), is considered remarkably high when
compared to the values obtained by the rest of the
models and when considering the satisfactory shielding
performance of this line reported in [44]. It is
Figure 5 Maximum shielding failure current of typical 110 kV up to 1150 kV transmission lines as listed in Table 4
1304
doi: 10.1049/iet-gtd.2009.0685
www.ietdl.org
phase conductor (Table 5). All models, with the exception
of [24], agree in predicting shielding failures for all phase
conductors; this is in contrast to simulation results referring
to the same 275 kV line showing shielding failures only for
the upper phase conductors [46]. Also, from Table 5 it can
be deduced that IMSF varies along the length of the
transmission line as all models yield lower IMSF values at
the midspan than at the tower, as a result of reduced both
height and shielding angle in the former case; this is in
accordance with simulation results [42, 43]. It is important
to note that if the transmission line utilises negative
shielding angles, provided by shield wires or a higher phase
conductor, generic models, in contrast to electrogeometric,
yield negative IMSF values for the shielded phase
conductors, considering them as perfectly shielded (W , 0)
against all prospective lightning stroke currents. This has
also been demonstrated in [34] for typical 150 and 400 kV
overhead lines of the Hellenic transmission system.
noteworthy that according to model [24] IMSF takes

negative values; this is unrealistic since it indicates no
shielding failures of the examined transmission lines.
Summarising, from Figs. 3 5 it can be deduced that IMSF
varies signicantly among lightning attachment models used
in shielding analysis; electrogeometric models, thus also
IEEE Std [2], yield generally the highest IMSF values.
This can be attributed to the fact that the shielding
failure width is less sensitive to transmission line
geometry variations, when dened based on striking
distance (Fig. 1) and by assuming the latter independent of
conductor height.
The variability in IMSF among lightning attachment
models can also be deduced from Table 5 referring to 275
and 500 kV transmission lines; tower geometries are shown
in Fig. 6. In IMSF calculations the shielding effect provided,
besides shield wire, by the phase conductors in the doublecircuit 275 kV transmission line has been considered.
Reasonably, the lowest IMSF values are found for the lower
Fig. 7 shows the IMSF values for the upper and outer
phase of the 275 and 500 kV transmission lines,
Table 5 Maximum shielding failure current (kA) of 275 and 500 kV transmission lines
Lightning
attachment model
275 kV line
Tower
500 kV line
Midspan
upper phase
Tower
outer
phase
Midspan
outer phase
Upper
phase
Middle
phase
Lower
phase
Wagner and
Hileman [3]
69.0
5.7
3.3
34.8
10.0
4.0
Young et al. [4]
42.9
1.3
0.6
16.0
2.9
0.8
Armstrong and
Whitehead [6]
33.5
7.3
5.5
22.8
10.1
6.2
Brown and
Whitehead [7]
39.3
7.7
5.7
26.1
11.0
6.5
Love [8]
26.5
5.3
3.7
17.0
7.6
4.2
Whitehead [10]
26.3
5.5
3.9
17.1
7.9
4.4
Suzuki et al. [35]
63.5
16.6
12.4
43.9
22.5
13.7
Anderson [13]
37.3
7.4
5.2
24.0
16.8
9.0
IEEE Std [2]
90.6
6.9
4.4
31.5
10.6
4.9
Eriksson [16]
17.0
3.6
3.8
14.3
6.3
6.3
Rizk [18]
17.8
0.2
0.2
13.4
2.6
1.4
Petrov et al. [22]
25.3
0.3
0.4
21.2
3.8
2.3
Borghetti et al. [24]
10.7
7.5
Ait-Amar and
Berger [27]
94.2
0.9
1.0
64.8
12.3
5.6
Statistical model
44.3
2.7
1.1
26.0
7.2
2.5
Sag of shield wire and phase conductor: 6.7 and 7.1 m, respectively, for the 275 kV line and 7.2 and
9.5 m, respectively, for the 500 kV line
doi: 10.1049/iet-gtd.2009.0685
1305
www.ietdl.org
includes the IMSF values obtained through simulations
[42, 43]; these values, while comparing well with that
obtained from models [13, 35], are considered quite high
as the rest lightning attachment models yield IMSF values
less than 8 kA.
The maximum shielding failure current is used for the
estimation of the shielding failure rate of transmission lines,
SFR, that is, the rate of lightning strokes to phase conductors.
Based on [2], SFR (shielding failures/100 km/year)
can be calculated as
SFR = 0.2Ng
IMSF
W (I )f (I ) dI
(11)
where Ng(ashes/km2/year) is the ground ash density, f (I ) is

the probability density function of the lightning peak current
distribution and W (m) is the shielding failure width.
Following this, the shielding failure ashover rate of a
transmission line, SFFOR (ashovers/100 km/year), normally
used together with backashover rate to estimate the
expected lightning outage rate of a transmission line, is
given as
SFFOR = 0.2Ng
IMSF
W (I )f (I ) dI
(12)
Ic
Figure 6 Towers of overhead transmission lines of

a
b
c
d
275 kV [46]
500 kV [42]
150 kV
400 kV
respectively, estimated for the average heights of the shield

wire and phase conductor along the line, where average
height is the height at the tower minus two-thirds of the
midspan sag. These currents can be considered as the
average IMSF values of the transmission lines. Fig. 7b
where Ic (kA) is the minimum lightning peak current causing

ashover of insulation [2]. Both SFR and SFFOR should
reasonably vary among lightning attachment models used in
shielding analysis; this would result from the variability,
besides in IMSF, also in W.
Finally, it must be mentioned that the present analysis
refers to overhead transmission lines on at ground and to
vertically descending lightning leaders.
Figure 7 Maximum shielding failure current of overhead transmission lines of

a 275 kV
b 500 kV
1306
doi: 10.1049/iet-gtd.2009.0685
www.ietdl.org
4 Implications in insulation
coordination of substations
The lightning performance of overhead transmission lines is
considered in insulation coordination of substations [47 49].
Substation outages may be caused by impinging overvoltage
surges, owing to backashover or shielding failure of the
incoming overhead transmission lines, exceeding the
insulation level of substation equipment. According to
International Electrotechnical Commission (IEC) [47], the
substation outage rate because of incoming shielding failure
surges, Rt (outages/year), is related to the lightning peak
current, I, determining the impinging surge, through
Rt = Rp [F (I ) F (IMSF )]
(13)
Figure 8 Schematic diagrams of the evaluated systems

a 150 kV
b 400 kV
where f (I ) is the probability density function of the lightning

peak current distribution;
attachment models presented in Section 2. The

corresponding maximum current steepness, Sm (kA/ms),
was calculated as a function of lightning peak current,
I (kA), as Sm 3.9I 0.55 [51]. The front duration of the
lightning current waveshape was varied from 1 up to 5 ms,
in accordance with eld data [50, 52], to demonstrate its
effects on the overvoltages arising at the substation
equipment; the time to half value was assumed 77.5 ms [32,
50]. For both substations, the last section of 1.75 km of the
incoming overhead transmission line was represented by a
sequence of J. Marti frequency-dependent models,
considering the line span (350 m) and design. Towers were
represented as vertical lossless single-phase frequencyindependent distributed parameter lines, with surge
IMSF is the maximum shielding failure current of the

incoming line.
Table 6 Maximum shielding failure current (kA) of 150 and

400 kV double-circuit overhead transmission lines
where
Rp (shielding failures/year) is the shielding failure rate of the
incoming line within a limit distance from the substation
entrance, which can be calculated with the aid of (11);
F(I ) is the probability of lightning peak current being
greater than I given as
1
F (I ) =
f (I ) dI
(14)
Thus, as IMSF is the upper limit of all possible lightning

stroke currents impinging on the substation entrance owing
to shielding failure of the incoming transmission line
within a limit distance, from (13) for I IMSF, Rt 0
outages/year provided that the impinging overvoltage surge
is lower than the insulation level of the substation
equipment, considering also an acceptable safety factor. To
demonstrate the dependence of the overvoltages impinging
on the substation upon IMSF of the connected overhead
transmission lines, ATP-EMTP simulations have been
performed for 150 and 400 kV GIS substations. The
congurations of the evaluated substations are shown in
Fig. 8; the geometry of the 150 and 400 kV incoming
transmission lines is shown in Figs. 6c and d, respectively.
As a worst-case scenario, lightning is assumed to strike the
upper phase conductor of the double-circuit transmission
line, at the rst tower close to substation, at the time
instant of negative peak power-frequency voltage. The
lightning stroke was represented by a current source of
negative polarity producing a current waveshape with front
upwardly concave [32, 50]. The current amplitude was
assumed to be equal to IMSF (Table 6), calculated for the
upper phase conductor according to the lightning
doi: 10.1049/iet-gtd.2009.0685
Lightning attachment model
150 kV line 400 kV line
Wagner and Hileman [3]
34.0
32.0
Young et al. [4]
15.5
14.0
Armstrong and Whitehead [6]
21.7
19.2
Brown and Whitehead [7]
24.7
21.7
Love [8]
16.8
16.1
Whitehead [10]
16.9
16.2
Suzuki et al. [35]
43.4
41.9
Anderson [13]
23.6
39.1
IEEE Std [2]
33.6
44.0
Eriksson [16]
12.7
9.3
Rizk [18]
11.9
6.9
Petrov et al. [22]
17.8
8.8
5.0
Ait-Amar and Berger [27]
59.8
37.8
statistical model
27.7
23.0
Borghetti et al. [24]
1307
www.ietdl.org
impedance, Z(V), calculated as [14]

r 2 + h2T
Z = 60 ln 2
r
(15)
where r (m) is the tower base radius and hT (m) is the tower
height; the calculated surge impedance of the towers of the
150 and 400 kV lines (Figs. 6c and d) is 167 and 163 V,
respectively. Towers were terminated by a resistance of
20 V. Insulator strings, with standard lightning impulse
withstand voltage level (LIWL) of 750 and 1425 kV for
the 150 and 400 kV lines, respectively, were represented by
voltage-dependent ashover switches, modelled by transient
analysis of control systems (TACS) module. The voltage
time characteristic of the insulator strings was represented
with the aid of the following expression [53]
VFO = (400 + 710/t 0.75 )L
(16)
where VFO (kV) is the ashover voltage, t (ms) is the elapsed

time after lightning stroke and L (m) is the insulator string
length, 1.86 and 3.62 m for the 150 and 400 kV lines,
respectively. The underground single core cross linked
polyethylene (XLPE) power cables were represented by the
Bergeron model with parameters calculated at 500 kHz. Surge
arresters with specication data according to Table 7 were
modelled as non-linear current dependent resistors taking into
account their voltage-current characteristic. GIS bays were
represented as lossless stub lines with surge impedance of
75 V [54]. The step-up transformers were represented by a
capacitance pi-circuit together with BCTRAN model. Cable
connections and the surge arrester lead lengths shorter than
3 m were modelled by a lumped parameter inductance of
1 mH/m [54]. The earth resistivity was assumed 100 V m.
Fig. 9 shows an example of the computed overvoltages at the
400 kV GIS entrance and HV terminals of the transformer for
a shielding failure current with amplitude 23 kA, calculated
according to the statistical model (Table 6), and front
duration of 1 ms. The overvoltages arising in the substations,
being dependent upon IMSF, vary signicantly with the
lightning attachment model employed in shielding analysis of
the incoming lines. Actually, for a shielding failure current
with wavefront duration of 1 ms the peak overvoltage at the
GIS entrance was found to vary from 321 up to 738 kV and
Figure 9 Overvoltages at the GIS entrance and HV

terminals of the transformer; 400 kV substation
from 740 up to 1080 kV for the 150 and 400 kV substations,
respectively; such variations in arising overvoltages may affect
the selection of the standard lightning impulse withstand
voltage of the substation equipment. However, from Fig. 10
it is obvious that the dependence of the peak overvoltage
upon IMSF, thus also upon lightning attachment model,
becomes weaker as the front duration of the shielding failure
current increases; the latter results in lower arising
overvoltages in the substations, especially for relatively higher
shielding failure currents (Fig. 10).
According to common practice, surge arresters are installed
at the GIS substation entrance [55]; to evaluate their
necessity for the protection of the substation against
incoming shielding failure surges they were excluded from
the simulation circuit. Fig. 11 shows the variation of the
computed peak overvoltage arising at the GIS entrance
with the lightning attachment model employed in IMSF
calculation. In this gure the peak overvoltage is normalised
with respect to the LIWL of the GIS, 750 and 1425 kV
for the 150 and 400 kV substations, respectively; the
dashed lines correspond to safety margins of LIWL/1.15
and LIWL/1.25. From Fig. 11 it is obvious that the
overvoltages vary signicantly within the range of about
60120% of the LIWL among lightning attachment models.
Table 7 Surge arrester characteristics

System
voltage,
kV
Rated
voltage,
kV
Residual voltage, kV
10 kA,
1/2ms
10 kA,
8/
20 ms
20 kA,
8/
20 ms
1 kA,
30/
70
ms
150
144
377
346
377
294
400
342
887
821
887
698
1308
Figure 10 Peak overvoltage at the GIS entrance as a

function of front duration of the shielding failure current
doi: 10.1049/iet-gtd.2009.0685
www.ietdl.org
employed in shielding analysis of the incoming lines among
models. Besides lightning attachment model, the
transmission line and substation conguration as well as
their modelling may affect the computed overvoltages
arising at substation equipment. Nevertheless, in light of
the great variability of the computed overvoltages in the
present work, it can be stated that careful selection of
lightning attachment model is needed when in insulation
coordination of substations the impinging surges because of
shielding failure of incoming overhead transmission lines is
considered. The implementation of the electrogeometric
model adopted by IEEE Std [2] for shielding analysis,
yielding relatively high IMSF values (Table 6), imposes high
requirements on protection of the substation equipment
against incoming shielding failure surges (Fig. 11).
Figure 11 Peak overvoltage at the GIS entrance of the

substations, without surge arresters operating at the
substation entrance, of
a 150 kV
b 400 kV
Front duration of shielding failure current 1 ms
For the 150 kV substation (Fig. 11a) the overvoltages

computed by employing models [16, 18, 24] in IMSF
calculation, contrary to that obtained from the rest models,
do not indicate the need for surge arresters operating at the
substation entrance; it must be noted that the 150 kV GIS is
connected to an overhead line through an underground cable
(Fig. 8a) and for such a conguration IEC [47] generally
suggests the installation of surge arresters at the line-cable
junction. For the 400 kV GIS substation (Fig. 11b), almost
half of the models used in IMSF calculation yield overvoltages
indicating no need for surge arresters operating at the
substation entrance; it is important to note that the models
suggested in [32, 53] for shielding analysis of overhead
transmission lines yield contradictory results.
The discrepancy in the computed peak overvoltages among
lightning attachment models (Figs. 10 and 11) results from
variations in IMSF (Table 6), which, as can be deduced
from the Appendix, are directly associated with the
different striking distance and attractive radius relationships
doi: 10.1049/iet-gtd.2009.0685
Conclusions
General expressions for the estimation of the maximum

shielding failure current of overhead transmission lines have
been derived by employing several lightning attachment
models in shielding analysis. An application to typical
110 kV up to 1150 kV overhead transmission lines has
shown that the maximum shielding failure current,
depending on transmission line geometry therefore also
varying along the length of the line, varies signicantly
among lightning attachment models. The maximum
shielding failure current, being generally highest for
electrogeometric models, increases with transmission line
height for statistical and electrogeometric models, but the
opposite is true for generic models.
ATP-EMTP simulations of 150 and 400 kV GIS
substations have shown that the overvoltage surges
impinging on the entrance of the substations owing to
shielding failure of the incoming transmission lines, being
dependent upon shielding failure current, vary up to 2.4
times among the lightning attachment models employed in
shielding analysis of the lines; such signicant variation may
affect the selection of the required protection measures and
of the standard insulation level of the substation equipment.
In light of the great variability in maximum shielding failure
current among lightning attachment models employed in
shielding analysis of overhead transmission lines, conscious
selection of the lightning attachment model should be made
for the evaluation of the shielding performance of the lines
and for insulation coordination studies. In that respect, to be
in the safe side, the electrogeometric model adopted by
IEEE Std 1243:1997, yielding relatively high maximum
shielding failure current, should be implemented in shielding
analysis of overhead transmission lines.
Acknowledgment
T. E. Tsovilis wishes to thank the Research Committee of

Aristotle University of Thessaloniki for the support
provided by a merit scholarship.
1309
www.ietdl.org
7
References
[1] GOLDE R.H.: Lightning conductor, in GOLDE R.H. (ED.) :

Lightning (Academic Press, London, 1977, vol. 2), Ch. 17,
pp. 545 564
[2] IEEE 1243 1997: Guide for improving the lightning
performance of transmission lines, 1997
[3] WAGNER C.F., HILEMAN A.R.: The lightning stroke-II, AIEE
Trans. Power Appar. Syst., 1961, 80, (3), pp. 622 642
[4] YOUNG F.S., CLAYTON J.M., HILEMAN A.R. : Shielding of
transmission lines, IEEE Trans. Power Appar. Syst., 1963,
82, (4), pp. 132 154
[5] WAGNER C.F.: Lightning and transmission lines,
J. Franklin Inst., 1967, 283, (6), pp. 558 594
[6] ARMSTRONG H.R., WHITEHEAD E.R. : Field and analytical
studies of transmission line shielding, IEEE Trans. Power
Appar. Syst., 1968, 87, pp. 270 281
[7] BROWN G.W., WHITEHEAD E.R.: Field and analytical studies
of transmission line shielding-II, IEEE Trans. Power Appar.
Syst., 1969, 88, pp. 617 626
[8] LOVE E.R.: Improvements in lightning stroke modeling and
applications to design of EHV and UHV transmission lines.
MSc dissertation, University of Colorado, Denver, CO, 1973
[9] GILMAN D.W., WHITEHEAD E.R.: The mechanism of lightning
ashover on high-voltage and extra-high-voltage
transmission lines, Electra, 1973, 27, pp. 65 96
[10] WHITEHEAD E.R.: CIGRE survey of the lightning
performance of EHV transmission lines, Electra, 1974, 33,
pp. 63 89
[11] DARVENIZA L.D., POPOLANSKY F., WHITEHEAD E.R.: Lightning
protection of UHV lines, Electra, 1975, 41, pp. 39 69
[12] MOUSA A.M., SRIVASTAVA K.D.: Modelling of power lines in
lightning incidence calculations, IEEE Trans. Power Deliv.,
1981, 5, (1), pp. 303 310
ground objects. Proc. Int. Aerospace Ground Conf. on

Lightning and Static Electricity, Oklahoma City, OK, 1988,
pp. 324352
[16] ERIKSSON A.J.: An improved electrogeometric model for
transmission line shielding analysis, IEEE Trans. Power
Deliv., 1987, 2, (3), pp. 871 886
[17] CHOWDHURI P., KOTAPALLI A.K.: Signicant parameters in
estimating the striking distance of lightning strokes to
overhead lines, IEEE Trans. Power Deliv., 1989, 4, (3),
pp. 1970 1981
[18] RIZK F.A.M.: Modeling of transmission line exposure to
direct lightning strokes, IEEE Trans. Power Deliv., 1990, 5,
(4), pp. 1983 1997
[19] DELLERA L., GARBAGNATI E.: Lightning stroke simulation by
means of the leader progression model, IEEE Trans.
Power Deliv., 1990, 5, (4), pp. 2009 2029
[20] RIZK F.A.M.: Modeling of lightning incidence to tall
structures, IEEE Trans. Power Deliv., 1994, 9, (1), pp. 162193
[21] PETROV N.I., WATERS R.T. : Determination of the
striking distance of lightning to earthed structures, Proc.
R. Soc. A, Math. Phys. Eng. Sci., 1995, 450, pp. 589 601
[22] PETROV N.I. , PETROVA G., WATERS R.T. : Determination of
attractive area and collection volume of earthed
structures. Proc. 25th Int. Conf. on Lightning Protection,
Rhodes, Greece, 2000, pp. 374 379
[23] YUAN X.: Investigation on the striking distance of lightning
strokes to overhead lines. PhD dissertation, Tennessee
Technological University, Cookeville, TN, December 2001
[24] BORGHETTI A., NUCCI C.A., PAOLONE M.: Estimation of
the statistical distributions of lightning current
parameters at ground level from the data recorded by
instrumented towers, IEEE Trans. Power Deliv., 2004, 19,
(3), pp. 14001409
[25] CHOWDHURI P., TAJALI G.R., YUAN X.: Analysis of striking
distances of lightning strokes to vertical towers, IET
Gener. Transm. Distrib., 2007, 1, (6), pp. 879 886
[13] ANDERSON J.G.: Lightning performance of transmission

lines, in LAFOREST J.J. (ED.) : Transmission line reference
book 345 kV and above (EPRI, Palo Alto, CA, 1987, 2nd
edn.), Ch. 12
[26] BECERRA M. , COORAY V., ROMAN F.: Lightning striking

distance of complex structures, IET Gener. Transm.
Distrib., 2008, 2, (1), pp. 131 138
[14] IEEE Working Group: A simplied method for

estimating lightning performance of transmission lines,
IEEE Trans. Power Appar. Syst., 1985, 104, (4), pp. 919 932
[27] AIT-AMAR S., BERGER G.: A modied version of the rolling

sphere method, IEEE Trans. Dielectr. Electr. Insul., 2009,
16, (3), pp. 718 725
[15] MOUSA A.M., SRIVASTAVA K.D.: A revised electrogeometric

model for the termination of lightning strokes on
[28] WATERS
systems, in
1310
R.T.:
Lightning phenomena and protection

Advances in high
HADDAD M. , WARNE D. (EDS.) :
doi: 10.1049/iet-gtd.2009.0685
www.ietdl.org
voltage engineering (Institute of Electrical Engineering,
Power and Energy, London, 2004, vol. 40), Ch. 3, pp. 107114
[29] MIKROPOULOS P.N., TSOVILIS T.E.: Interception probability
and shielding against lightning, IEEE Trans. Power Deliv.,
2009, 24, (2), pp. 863 873
[30] MIKROPOULOS P.N., TSOVILIS T.E.: Interception probability
and proximity effects: implications in shielding design
against lightning, IEEE Trans. Power Deliv., 2010, 25, (3),
pp. 1940 1951
[31] MIKROPOULOS P.N., TSOVILIS T.E.: Striking distance and
interception probability, IEEE Trans. Power Deliv., 2008,
23, (3), pp. 1571 1580
[32] CIGRE Working Group 33.01: Guide to procedures for
estimating the lightning performance of transmission lines.
Technical Bulletin 63, 1991
[43] VAHIDI B., YAHYAABADI M. , TAVAKOLI M.R.B. , AHADI S.M. :

Leader progression analysis model for shielding failure
computation by using the charge simulation method,
IEEE Trans. Power Deliv., 2008, 23, (4), pp. 2201 2206
[44] GAIVORONSKY A.S., KARASYUK K.V., PROKOFYEVA E.N.: Numerical
investigations of lightning proofness of UHV overhead
lines. Proc. 27th Int. Conf. on Lightning Protection,
Avignon, France, 2004, pp. 1 5, paper 4p.9
[45] HOIDALEN H.K. , MORK B.A. , PRIKLER L., HALL J.L. :
Implementation of new features in ATPDraw version 3.
Proc. Int. Conf. on Power Systems Transients, New
Orleans, USA, 2003, pp. 1 5
Shielding of transmission lines, in HILEMAN

Insulation coordination for power systems
(CRC Press, Taylor & Francis Group, Boca Raton, FL, 1999),
Ch. 7, pp. 244 254
[46] BHATTARAI R., RASHEDIN R., VENKATESAN S., HADDAD A., GRIFFITHS
H., HARID N.: Lightning performance of 275 kV transmission
lines. Proc. 43rd UPEC, Padova, Italy, 2008, pp. 1 5,
paper 138
[34] MIKROPOULOS P.N., TSOVILIS T.E. : Lightning attachment

models and maximum shielding failure current:
application to transmission lines. Proc. IEEE PowerTech,
Bucharest, Romania, 2009, pp. 1 8, paper 233
[47] IEC 60071 2: Insulation coordination, 1996
[33]
HILEMAN A.R.:
[42] HE J., TU Y., ZENG R., LEE J.B., CHANG S.H., GUAN Z.: Numeral
analysis model for shielding failure of transmission line
under lightning stroke, IEEE Trans. Power Deliv., 2005, 20,
(2), pp. 815 822
A.R. (ED.) :
[48] ERIKSSON A.J., WECK K.H.: Simplied procedures for

determining representative substation impinging lightning
overvoltages. Proc. CIGRE, Paris, 1988, paper 33 16
[35] SUZUKI T., MIYAKE K., SHINDO T.: Discharge path model
in model test of lightning strokes to tall mast,
IEEE Trans. Power Appar. Syst., 1981, 100, (7),
pp. 3553 3562
[49] SHIMIZU K., MOTOYAMA H.: Substation outage rate

prediction method based on lightning characteristics,
Electr. Eng. Jpn., 2004, 148, (2), pp. 64 75
[36] GOLDE R.H.: The frequency of occurrence and the

distribution of lightning ashes to transmission lines, AIEE
Trans., 1945, 64, pp. 902 910
[50] IEEE Task Force 15.09: Parameters of lightning

strokes: a review, IEEE Trans. Power Deliv., 2005, 20, (1),
pp. 346 358
[37] CARRARA G., THIONE L.: Switching surge strength of large

air gaps: a physical approach, IEEE Trans. Power Appar.
Syst., 1976, 95, (2), pp. 512 524
[51] ANDERSON R.B., ERIKSSON A.J.: Lightning parameters for

engineering application, Electra, 1980, 69, pp. 65 102
[38] MIKROPOULOS P.N., TSOVILIS T.E.: Experimental investigation

of the Franklin rod protection zone. Proc. 15th Int. Symp.
on High Voltage Engineering, Ljubljana, Slovenia, 2007,
pp. 1 5, paper 461
[39] MIKROPOULOS P.N., TSOVILIS T.E.: Interception radius and
shielding against lightning. Proc. 29th Int. Conf. on Lightning
Protection, Uppsala, Sweden, 2008, pp. 111, paper 410
[40]
PETTERSSON P., ERIKSSON R.:
Discussion of reference 18
[41] MIKROPOULOS P.N., TSOVILIS T.E., ANANIADIS T.: The effect of an

earthed object on the interception radius of the Franklin
rod: an experimental investigation. Proc. MedPower,
Thessaloniki, Greece, 2008, pp. 1 6, paper 77
doi: 10.1049/iet-gtd.2009.0685
[52] TAKAMI J., OKABE S.: Characteristics of direct lightning

strokes to phase conductors of UHV transmission lines,
IEEE Trans. Power Deliv., 2007, 22, (1), pp. 537 546
[53] IEEE Working Group: Estimating lightning
performance of transmission lines II updates to
analytical models, IEEE Trans. Power Deliv., 1993, 8, (3),
pp. 1254 1267
[54] IEEE Task Force: Modeling guidelines for fast front
transients, IEEE Trans. Power Deliv., 1996, 11, (1),
pp. 493 506
[55] Joint Working Group 33/23.12: Insulation
co-ordination of GIS: return of experience, on site tests
and diagnostic techniques, Electra, 1998, 176, pp. 67 97
1311
www.ietdl.org
8
Appendix
From (1) S/D g, hence (18) becomes
The maximum shielding failure current, IMSF, of overhead

transmission lines is the critical lightning peak current
above which no shielding failures occur, that is, the critical
current corresponding to zero shielding failure width, the
latter as dened in Fig. 1. General expressions of IMSF can
be derived on the basis of several lightning attachment
models employed in shielding analysis according to Fig. 12.
D=
(hm + hp )/2
Also from (1) for lightning peak current I IMSF,

D (A/g)I BMSF. Combining the latter with (19), IMSF is
given as
8.1 Maximum shielding failure current

formulation according to electrogeometric
models
From Fig. 12a
sin a =
D (hm + hp )/2
(LM)
(17)
IMSF
g(hm + hp )/2
=
A(1 g sin a)
1/B

formulation according to A. J. Erikssons
model
From Fig. 12b
2
= (DR + Rp )2 + (hm hp )2
Rm
For S (KP)/2, (LM) S; thus (17) leads to

S
sin a = 1 (hm + hp )/2D
D
(19)
1 g sin a
2
= (DR + Rp )2 + (KP)2 DR2
Rm
(18)
For
G Rm/Rp (hm/hp)0.6
and
(20)
C2 (hm 2 hp)2 +
Figure 12 Estimation of maximum shielding failure current of overhead transmission lines according to
a Electrogeometric models
b A.J. Erikssons model
c Generic models
hm , hp are the heights of shield wire and phase conductor, respectively; a shielding angle; DR horizontal separation distance between
shield wire and phase conductor; S striking distance to shield wire and phase conductor; D striking distance to earth surface; Rm , Rp
attractive radius of shield wire and phase conductor, respectively
1312
doi: 10.1049/iet-gtd.2009.0685
www.ietdl.org
DR 2 (KP)2, (20) leads to
analysis according to generic models (Fig. 12c) (22) becomes
[(G2 1)Rp DR]2 = (G2 1)C2 +DR2

1
DR + DR2 +(G2 1)C2
Rp = 2
G 1
DR = Dc1 ln(hm /hp )
By using a known relationship between D and I of the form

D A I B and for I IMSF, (24) becomes
(21)
B
c1 ln(hm /hp )
DR = A IMSF
0.74
0.67 h0.6
p IMSF. Combing the latter
From (3) for I IMSF, Rp

with (21), the maximum shielding failure current is given as

1/0.74
DR + DR2 +C2 (G2 1)
IMSF =
2
0.67h0.6
p (G 1)
IMSF
From Fig. 12c

DR = Rm Rp
(22)
E
From (5) for I IMSF, Rm jhEmI FMSF + zhG
m and Rp jhp
F
G
IMSF + zhp ; thus (22) becomes
DR =
F
j(hEm
IMSF
hEp )
z(hGm
hGp )
IMSF =
DR
=
A c1 ln(hm /hp )

1/F
DR z(hGm hGp )
j(hEm hEp )

1/F
(hm hp ) tan a z(hGm hGp )
From (7) at critical interception Rm D[c1ln(hm/D) + c2]

and Rp D[c1ln(hp/D) + c2]; thus performing shielding
doi: 10.1049/iet-gtd.2009.0685

1/B
(hm hp ) tan a
DR = Dc1 ln(hm /hp )
A c1 ln(hm /hp )
0.14c1 1.3
h
D0.3 m
(26)
For negative lightning c1 0.272 (Table 3) and for

D 10I 0.65 [8] and IMSF , 50 kA (26) is approximated by
0.65
ln(hm /hp ) 0.01h1.3
DR = 2.72IMSF
m
(27)
From (27) the maximum shielding failure current is given as
j(hEm hEp )

formulation according to statistical model
1/B
When proximity effects on lightning attachment are

considered the critical interception radius of the shield wire
decreases [30]. In that case, (24) becomes DR (1 2
m)Dc1ln(hm/hp) where m 0.14 (hm/D)1.3/ln(hm/hp) [30].
Thus
(23)
(25)


formulation according to generic models
(24)
IMSF

1/0.65
DR + 0.01h1.3
m
=
2.72 ln(hm /hp )
=

1/0.65
(hm hp ) tan a + 0.01h1.3
m
2.72 ln(hm /hp )
1313

IEEE

Uploaded by

Copyright:

Available Formats

IEEE

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

IEEE

Uploaded by

Copyright:

Available Formats

www.ietdl.

Lightning attachment models and maximum

Shielding of overhead transmission lines against direct

conductors which may cause ashover. Although the lower

& The Institution of Engineering and Technology 2010

Table 1 Factors A, B and g to be used in (1)

The maximum shielding failure current of overhead

Wagner and Hileman

2.1 Electrogeometric models

2 Maximum shielding failure

Suzuki et al. [35],

Anderson [13], IEEE

IEEE Std [2]

where I is in kA and S and D are in metres; factors A, B and g

Figure 1 Shielding analysis of transmission lines

where factors A, B and g are given in Table 1 and hm (m), hp

hm , hp height of shield wire and phase conductor, respectively; a

Table 2 Factors j, E, F, z and G to be used in (5)

2.2 A.J. Erikssons model

where I (kA) is the prospective lightning peak current and h

where G Rm/Rp (hm/hp)0.6 and C2 (hm 2 hp)2 + DR 2

2.3 Generic models

where R is in metres, I (kA) is the prospective lightning peak

Borghetti et al. [24],

Ait-Amar and Berger

Using as h in (5) the object height plus 15 m

width W. Thus, for W 0 the maximum shielding failure

where factors j, E, F, z and G are given in Table 2 and DR

2.4 Statistical model

& The Institution of Engineering and Technology 2010

Figure 2 Maximum shielding failure current as a function

(kA) is well approximated by (Appendix 8.4)

where DR (m), a, hm (m) and hp (m) are dened in Fig. 1 and

The result of (9) and (10) for hp/hm 0.75, a ratio of

As was raised by Pettersson and Eriksson [40] and discussed

3 Maximum shielding failure

Figure 3 Maximum shielding failure current variation with shielding angle

Table 4 Parameters of typical overhead transmission lines

Figure 4 Maximum shielding failure current variation with

& The Institution of Engineering and Technology 2010

Fig. 5 shows the estimated IMSF values of typical

noteworthy that according to model [24] IMSF takes

Young et al. [4]

Suzuki et al. [35]

IEEE Std [2]

Petrov et al. [22]

Borghetti et al. [24]

& The Institution of Engineering and Technology 2010

where Ng(ashes/km2/year) is the ground ash density, f (I ) is

Figure 6 Towers of overhead transmission lines of

respectively, estimated for the average heights of the shield

where Ic (kA) is the minimum lightning peak current causing

Figure 7 Maximum shielding failure current of overhead transmission lines of

Figure 8 Schematic diagrams of the evaluated systems

where f (I ) is the probability density function of the lightning

attachment models presented in Section 2. The

IMSF is the maximum shielding failure current of the

Table 6 Maximum shielding failure current (kA) of 150 and

Thus, as IMSF is the upper limit of all possible lightning