Lightning protection of transmission lines: optimal

shielding design procedure

T.T. Nguyen and R. Holt

Abstract: The paper first derives a set of shielding constraints which must be satisfied for achieving
effective shielding against lightning for all of the phase conductors of an overhead transmission line
in relation to earth-wire placement. The shielding constraints are obtained from the analysis related
to the rigorous proof of the premise that, once the earth-wire position has been determined to
achieve effective shielding for a nominated critical lightning stroke current, then shielding of the
phase conductors is still maintained for stroke currents greater than the critical value. An optimal
shielding design procedure is then developed in which the cost function relating to earth-wire
position($ is minimised subject to shielding constraints and specified clearance constraints. Results
of lightning protection designs using the procedure, to achieve effective shielding for a range of
transmission-line configurations, are presented.

List of principal symbols z

lightning stroke current
e-@
2
“0
critical lightning stroke current
nonlinear function of conductor height in attractive radius associated with a phase
Eriksson’s EGMs conductor for critical current Io
nonlinear function of lightning stroke cur- attractive radius associated with the earth
rent in Eriksson’s EGMs wire for critical current lo
attractive radius associated with a phase separation between the two earth wires in the
conductor transmission line with a horizontal con-
attractive radius associated with the earth figuration
wire S separation between two adjacent phase
phase conductor height above the earth conductors in a transmission line with a
plane horizontal configuration
earth-wire height above the earth plane kth phase conductor
horizontal distance between the earth wire height above the earth plane of phase
and the phase conductor nominated for conductor Pk
analysis gD(hk)
distance from the earth wire to nhase
gdh) conductor Pk
dHP)
shielding angle with respect to phase con-
fxr, ductor Pk
distance from the earth wire to the phase
conductor nominated in the analysis in the kf-8
general case attractive radius associated with phase con-
distance from the earth wire to the phase ductor Pk for critical current Io
conductor nominated in the analysis for location of the earth wire (also the origin of
achieving effective shielding the X-Y co-ordinates)
shielding angle in the general case location of the phase conductor nominated
shielding angle for acheving effective shield- for analysis
ing intersection between the circle having the
fao) centre at 0 and radius DsS, and the circle
having the centre at P and radius Dsc
B intersection between the horizontal line at
0 TEE. 2003 distance h above the earth plane and circle
IEE Proceeditqr online no. ZW30819 having the centre at P and radius Dsc
doi:lO.lM9/ip-ptd2~30819 Ak intersection between the circle having the
Paper fin1 received 26th June 2002 and in revised form 17th April 2033 centre at 0 and radius D!s, and the circle
T.T. Nguyen is with the Energy Systems Centre, Sfhool of Electrical, Elenronic having the centre on the phase conductor Pk
and Computer Engineering, The University of Westem Australia. 35 Stirling and radius DOScK
Highway, Crawley, Westem Australia 6W9. Australia intersection between the horizontal line at
R. Holt may be coniacted at Po Box 1192, Canlung Bndge, Applerross,
Bk
Westem Australia 6153, Australia
distance hk above the earth plane and circle

IEE Proc-Gmm T r a m Disrrlb.. Val ISO, No. 6,Now&r 2W3 659

Authorized licensed use limited to: Vardhaman College of Engineering. Downloaded on January 25, 2010 at 09:42 from IEEE Xplore. Restrictions apply.
 having the centre on the phase conductor Pk one in the case of Eriksson’s EGMs which are sigdcantly
and radius D& different from the traditional EGMs previously presented
X co-ordinates of points A and E, respec- by Brown and Whitehead.
tively The first objective of the present paper is to develop a
Y co-ordinates of points A and B, respec- formal and rigorous proof of the premise. The starting
tively point is that of the derivation based on the electrogeometric
Xco-ordinate of point A when the lightning theory of a set of effective shielding constraints for a
stroke current has the critical value nominated critical current. A formal proof is then
Y co-ordinate of point A when the lightning developed to show that, if the shielding constraints are
stroke current has the critical value satisfied at the critical current, then they will also be satisfied
phase conductors for any lightning stroke currents greater than the critical
phase conductors of the second circuit in a value. The development focuses on the Eriksson’s EGMs,
double-circuit transmission line given that these have been presented more recently.
earth wires identified by ‘I’ and ‘2, respec- The second objective is to develop a systematic and
tively optimal procedure for effective shielding design using the
intersection between the circle having the constraints derived. The previously-published procedure for
centre at 0,and radius Dss, and the circle shielding design has been based on the representation of one
having the centre at O2 and radius DSS phase conductor of a multiconductor transmission line.
D intersection between the circle having the Very often, it is the phase conductor with the greatest height
centre on the centre phase conductor and that is considered in the design. It has been assumed that the
radius D , and the vertical line passing other lower phase conductors would be protected once the
through that phase conductor highest phase conductor has been shielded [IM].
infinitesimal change in i from z, In the new procedure, all of the phase conductors of a
infinitesimal change in xA and y,, respec- transmission line are represented simultaneously. There is
tively, corresponding to infinitesimal change one set of effective shielding constraints associated with
in z by & each phase conductor. Previous assumption in relation to
the Y co-ordinates of points D and E, the use of a single phase conductor in the design is not
respectively required in the method formulated in this paper. The design
YE-YD is based on constrained optimisation, in which the earth-
earth-wire cost function wire cost is minimised subject to the shielding constraints
number of phase conductors and statutory clearance requirements. The procedure is
applicable to any transmission-lineconfiguration.
The application of the design procedure developed to
representative transmission-line configurations is reported
1 Introduction in the paper.

Fundamental to all of the methods of designing lightning 2 Eriksson’s EGMs

protection for overhead transmission lines that use earth
wires is the premise that, once effective shielding has been Based on Eriksson’s EGMs [6], the shielding analysis
established for a critical lightning stroke current, shielding diagram of Fig. 1 is constructed. The attractive radii Dsc
will be maintained for all lightning stroke currents greater and Dss in Fig. 1 associated with the phase conductor and
than the critical current [1-7]. However, no formal and
rigorous proof of this shielding premise or reference to a
proof can he found in the open literature. For the
electrogeometricmodels (EGMs) in which the earth plane
is explicitly represented [I-51, the premise is often justified in
passing, in previously-published works, by briefly making
reference to diagrams. However, this cannot be taken to be
a formal and rigorous proof.
For facilitating the subsequent discussion, the group of
EGMs that represent the earth plane as a separate structure
is denoted here as the IEEE EGMs, which include the
models presented by Brown and Whitehead [3] and those
recommended by the IEEE [I, 4, 5) More recently, EGMs
which do not explicitly represent the earth plane have been
presented [6, 71. Supported by field data, Eriksson [6] and
Rizk [qhave proposed EGMs in which an attractive radius
associated with a structure has an explicit dependence on
the height of the structure, and the earth plane represents a
default condition. For convenience, this second group of
EGMs is denoted here as Eriksson’s EGMs.
The Eriksson’s EGMs have been applied for designing
the shielding of a transmission line for a nominated critical
lightning stroke current [6].The design has also implicitly
assumed that the shielding would still be effective for any
lightning stroke currents greater than the critical current. Fig. 1 Shielding a m l y s i ~diqrom
However, there have been no published papers which give General case of incomplete shielding condition
any justification or proof that the assumption is a correct Enksson’s EGM

660 IEE Proc.-Gener. T r m Dim&.. Vol. ISO, No. 6,N o d e r 2CX73

Authorized licensed use limited to: Vardhaman College of Engineering. Downloaded on January 25, 2010 at 09:42 from IEEE Xplore. Restrictions apply.
 the earth wire are given in the following: 4 Proof

Dsc = % . f r ( I ) (‘1 4.1 Requirement

The requirement in the proof is to show that when lightning
DSS = k . f E ( I ) (2) stroke current, I, is greater than the critical current I,, the
shielding diagram becomes that shown in Fig. 3. It is,
In (I) and (2), fdf) is an increasing function of lightning therefore, required to prove the following:
stroke current I [HI;k, and k,,depend on the heights h and
H,, of the phase conductor and the earth wire, respectively: (i) Part 1: Point A is below the horizontal line at distance h
above the earth plane.
kc = g D ( h ) (3) (ii) Part 2: Point B is inside the circle having the centre at 0
(the earth wire) and radius Dss.
kz = g D ( H p ) (4)
4.2 Proof of Part 1
From previously published results derived from many
By delining z =fA& (I) and (2) become:
observed lightning flashes to a variety of structures [W],gD
in (3) and (4) is a monotonically increasing function of Dsc = k,.z (6)
conductor height. With H,,>h, (3) and (4) give
Ds = k,.z (7)
ks > kc (5) At the critical current I,, functionf, has the value off&)
which is denoted as q.For the purpose of developing the
proof, functionf(do, On, z) is defined in:
3 Effective shielding premise f ( d O , ~ O >=4 Y B - YA (8)
In (8), yA and ye are the Yco-ordinates of points A and 5 in
In Fig. 2 is shown the shielding analysis diagram of the case Fig. 3, respectively. When z= zo (or I= I,), points A and 5
where effective shielding of phase conductor P is just coincide. This leads to:
achieved at a critical current Io [q.With reference to Fig. 2,
the effective shielding premise is stated in the following: f(do,@o,zn)= 0 (9)
Once the distance & between the phase conductor and It is then required to show that:
earth wire, and shielding angle So have been determined to
achieve the effective shielding condition at the critical f(do,So,z ) >0 for zt zo (or I >Io) (10)
lightning stroke current 1, as shown in Fig. 2, then effective Point A in Fig. 3 is the intersection between the two circles
shielding of the phase conductor will he maintained for any having the centres at 0 and P with radli k 3 and k?,
lightning stroke current greater than the critical value Io. respectively. The Xand Y co-ordinates of point A , (XA, y ~ ) ,

C.
-....-.-... .
.....

-......
b

h P

27Zz arth plane m///- earth plane

Fig. 2 Shielding m l y s i r diagram Fig. 3 ShirHing annlysis diagram

Effective shielding condition at the critical current Lightning stroke c u r ” greater than the critical current
Erikssan’s EGM Eriksson’s EGM

IEE Proe-CPMr Trmm. Dlrrrh, VoL 150, No. 6, Notw!her 2003 661

Authorized licensed use limited to: Vardhaman College of Engineering. Downloaded on January 25, 2010 at 09:42 from IEEE Xplore. Restrictions apply.
 therefore satisfy the following equations of the two circles: At z= zo, Ad,, Bo, zo) = 0 and gdQ) = 0. The quadratic
form g&) has another root at a,. Assuming first that
x;+d=gz (11) a,<@, then gs(a)>O if a<al or a>@. With this
and assumption of al<a,,./&, SO, i) will be positive for z>io
because gs(a)>O for a>@. This is then consistent with the
(xa-dosin(&))’+ (yA+ d ~ c o s ( & ) )=~ k : 1 (12) property derived and given in inequality (16) thatflh, Oo. z)
The property offldo, eo, z) is investigated in the following, becomes positive when z= zo + Az, with L- being positive
when z=zo+L-, with L-being positive and infinitesimal: and infinitesimal. If it is assumed otherwise that %<al, then
With z = z o + A z , linearising (11) and (12) about x! = gda)<O for %<a<al, which means flh, BO. z)<O for
dosin(0o) + k , q and y> = ;do cos(&), which are the co- zo<z<zi (with zI = &). This is contradictory to the
ordinates of point A in Fig. 2, where z = io,and then property given in inequality (16).
solving the linearised equations gives Hence, it is concluded thatfld,, O0, i ) > O for z>zo. The
proof of part 1 is completed.

4.3 Proof of Part 2

withx~<k,zo,k,<k,, Az>Oand u & c o ~ ( ~ ~ ) AyA
> O , in (13) The equation of the circle having the centre at 0 (earth
is negative, hence, wire) with radius Dss= ksz in Fig. 3 is given by
AYA<~ (14) X 2 + Y2 -el =0 (24)
As yB= -docos(Oo) which is independent of z: The X and Y co-ordinates of point B in Fig. 3 are:
AYB = 0 (15) XB = do sin(&) + k,z (25)
From (8) and (15), and inequality (14), together withflh,
00, 20) = 0: y~ = -do COS(&) (26)
f(Q, 00,z)> 0 for z = zo + Az (16) It is required to prove that point B is inside the circle
Inequality (16) confirms that flu&,Go, z) has the property described by (24). It is sufficient to prove that, by
that it becomes positive when i increases from zo to zo+ &-. substituting X = x B and Y=yB, the left-hand side of (24)
It is now required to prove thatf(4, BO, z) remains positive is negative, i.e.
for any z>zo. (kf - e ) 2+ 2d0 sin(&)k,z + di < O for Z>ZO
(27)
From the expression for ya which is obtained by solving
(1 1) and (12) simultaneously, and yB= -&os(Qo) together The left-hand side of inequality (27) is a quadratic form in z.
with the attractive radii given in (6) and (7), eo,z) isflu&, It has two roots 22 and zo. with z2<0. Hence, the left-hand
expressed using (8): side is negative for z>zo. With the validity of inequality (27)
being established, the proof of part 2 is completed.
When z = io (or I= IO);inequality (27) becomes
(kf - <)d + 2d0 sin(Oo)kcio +do’ = 0 (28)

4.4 Discussion
In (17): The proof developed in Sections 4.2 and 4.3 has the focus
on the case in which the shielding angle is positive, i.e.
a = l (18) fl&-W. However, the proof for the case of negative
a=<-kz (19) shielding angle follows the steps similar to those which
have been derived for the case of positive angle.
It is then required to prove that, from inequality (IO) and In the proof in Sections4.2 and 4.3, it has been taken that
(17): the circle having the centre on the phase conductor and the
a.acos(00) -d~cos(H0)>2d0.~in(00) circle having the centre on the earth wire intersect, i.e. the
argument of the square root function in the right-hand side
of (17) is positive or zero. The following Section discusses
the condition in which the two circles do not intersect.
where 5 A particular case of effective shielding
a0=d (21) With Eriksson’s EGMs, the effective shielding condition
When z = io,the left-hand side of inequality (20) is equal to shown in Fig. 4 can arise where the circle having the centre
the right-hand side. Hence, the left-hand side of inequality on the phase conductor and radius &is completely inside
(20) is positive when i = z o . As the left-hand side of the circle, having the centre on the earth wire and radius
inequality (20) is an increasing function of z , it is also D:s. The condition happens when D$>&+ D”,, i.e.
positive for any z>zo (or a>%).
For positive left-hand side of inequality (20) when a>%, (L - >do
~ ) Z O (29)
the proof of inequality (20) is equivalent to proving that, If inequality (29) is satisfied at 20, then, for z>zo (i.e. I>Io),
following squaring both sides of inequality (20) and
rearrangement: (t- kc)z>do (30)
gs(a)>O for a>ao (22) Hence, shielding will he maintained for lightning stroke
In inequality (22): current greater than the critical value Io.
When the argument of the square-root function in the
gs(a) = u2a2 + 2d,,(k?cos(200) - ki)a +4 (23) right-hand side of (17) becomes negative, inequality (30),
662 IEE Pro<.-Gem“.Trmm. Dlsrrib.. Vol. 150, No. 6, Noumber 2W3

Authorized licensed use limited to: Vardhaman College of Engineering. Downloaded on January 25, 2010 at 09:42 from IEEE Xplore. Restrictions apply.
 Defining
g(z) Y E - J’D (31)
and substituting the Y co-ordinates of points D and E into
(3 I) gives
e’
g(z) = l/k:-z’ - f - kc~ + do COS(OO) (32)
In (321, eo is the distance between earth wires O1and 0’in
Fig. 5 where eo is, from Fig. 5, 2(s-&sin(OO)) and s is the
separation between two adjacent phase conductors.
At critical current Io, effective shielding is achieved hence
dzo) 2 0 (33)
For point E to exist, it is required that
HP h
k,+O 2 2 (34)
As g(z) in (32) is an increasing function ofz, it is concluded
that, from inequality (33):
I y(z)>O f o r z z z o (35)
earth plane
With the validity of inequality (35) being established, the
Fig. 4 Shielding analysir diagram proof of the effective shielding premise in relation to the
Special case where D:s>do+D!c
Enksson’s EGM
centre phase conductor in Fig. 5 is completed.

7 Effective shielding constraints for a multi-

which represents an effective shielding condition, is satisfied. conductor transmission line
This complements the proof developed in Section 4.
The results obtained from the analysis developed in Sections
6 The case of horizontal configuration with t w o 4-6 are now extended to individual phase conductors of a
earth wires multiconductor transmission line, for developing the design
procedure.
With reference to Fig. 5, shielding of the centre phase In Fig. 6 is shown the shielding analysis diagram for a
conductor PI is achieved when point E is above point D. phase conductor identilied by Pk.The position of the earth

.......... c

dh

.........
..................... ............. ......C Mh
P2 P, ’ Po

hh
h HP

+ + earth plane
7n
/+%ah
rl plane 4
%%- A....-
Fig. 6 Shielding analysis diagram
Fig. 5 Shielding anulysis diagram Effective shielding condition at the critical current
Horizontal transmission line configuration with two earth wires Minimum requirement
Eriksson’s EGM

IEE Proc.-Gener T r a m . Dbrrb.. Vol. 150, No. 6,Nvvnnber 2003 663

Authorized licensed use limited to: Vardhaman College of Engineering. Downloaded on January 25, 2010 at 09:42 from IEEE Xplore. Restrictions apply.
 wire 0 relative to this phase conductor is given by dhand Qh. Phase conductor Po is nominated for defining the earth-
The diagram in Fig. 6 represents the minimum condition in wire position in terms of & and shielding angle 0,. .From
which effective shielding is just achieved at the critical Fig. 8, d k and 0, (with kfO) are related to 4 and Qo by
current I,. As the minimum condition might not he
achieved in the final design, for each and every phase dk = [(doSin(00)+Hk)2+(vk-dnCOs00)2]''2 (42)
conductor, a more general situation of effective shielding, at
the critical current Io, is shown in Fig. 7. ln the final design,
either the diagram in Fig. 6 or that in Fig. 7 is applicable to
a phase conductor.
In (42) and (43), Hk and Vk are the co-ordinates of phase
conductor Pk with respect to phase conductor Po.

.. ... ..... 0.
..

Fig. 8 Definitwn of the earth-wire Dosition with rewecl to o h e

Z Z % / / / / m-anh plane co;;ductor Po
Po = the vhase conductor nominated for defining the shieldine ande 0,
Fig. 7 Shielding analysis diagram Pk= phase conductor P,(k#O)
I

Effective shielding condition at the critical current 0 = earth wire

Eriksson's EGM

Drawing on (9) and (28), and inequalities (10) and (27), With (42) and (43), the constraint functionsfk(dk, t?,, zo)
the effective shielding constraints applied to phase con- and gk(dk, ,e, in for )
individual phase conductors are
ductor P,, corresponding to the situation in Fig. 6 or Fig. 7, fimctions of variables & and 0.,
are given by In the case of a horizontal configuration with two earth
fk(dk,Hk,ZO)>o k = 0 , 1 , 2 , . . . . . . ,M - l (36) wires, the shielding constraints in inequalities (33) and (34)
are required in respect of the centre phase conductor.
gk(dk,ok;Zo) 5 0 k = 0 , 1 , 2 , .. . . . . : M - 1 (37) The effective shielding proof given in Section 4 has been
based on the minimum condition of Fig. 6. However, the
In inequalities (36) and (37): proof extends directly to the situation in Fig. 7 which is
discussed as follows:
Whenfk(dk, Bk, zo)>O, then it is straightfonvard to prove
thatfk(dk, e,,
z)>O for z > z , . If it is assumed thatf&fk, Oh,
z ) i O for z>zo. then there exists z, such that z n c i , i i ,
and fjdk, oh, 2,) = 0. However, based on the results in
Section 4.2, iffh(dhr Sk, zmr)=O,then, for z>z,, fk(dk, O,,
z)>O. This leads to a contradiction to the assumption.
gk(dki @ h , Z O ) = (k:h ~ k:)d + 2dh Sin(&)k<kZO+ d,' (39) Hence, if the effective shielding condition in Fig. 7 is
where achieved for z=zn, then, for z>zn (i.e. Z>Zn), fk(dh, Oh,
2
a i = k, - ek (40)
z)>O.
A similar reasoning also confirms that, if gk(dk, ok, 2,) i0,
then gh(dhr Oh, z)<O for z>zo. Hence, if the effective
kck = gD(b) (41) shielding condition of Fig. 7 is achieved for zn (i.e. for
M is the total number of phase conductors; subscript k critical current I , ) , then effective shielding is still maintained
identifies individual phase conductors. for any lightning stroke current greater than Io.
664 IEE Proc.-Gmr. T r m m . Dlsnib.. Vol. 150, No. 6,November 2W3

Authorized licensed use limited to: Vardhaman College of Engineering. Downloaded on January 25, 2010 at 09:42 from IEEE Xplore. Restrictions apply.
 When z leads to the situation where the argument (with LO1
ol = 2)in the square-root function in right-
\
cq,replaced by
hand side of (38) is negative, effective shielding will still be
maintained as proved in Section 5.

8 Optimal shielding design procedure

In addition to the effective shielding constraints, there are

other design considerations and constraints, including those
relating to clearance requirement. However, in general,
there will be a range of earth-wire positions that satisfy all
of the constraints. There are different design criteria which
can be used in selecting a particular earth-wire position
among a possible range. A particular criterion adopted here
is that based on minimising the earth-wire cost. However,
the procedure can he adopted to any other criteria.
The cost in relation to earth wire(.$ is denoted by C€&,
go) which is a function of 4 and Bo. The shielding design is
then expressed in terms of the constrained optimisation in
which C€(& 0,) is minimised with respect to 4 and Qo, and Fig. 9 Shieldinng achieved wirh earth-wire parameters obtained in
subject to: case I
Doublecircuit transmission line
the set of constraints given in inequalities (36) and (37) Vertiwl configuration with the middle phase conductors in the outer
the set of constraints given in inequalities (33) and (34), for positions
Po, PI, P2, P,, P4,Ps= phase conductors
the c a s of a horizontal configuration with two earth wires
O,, 0 2 = wrth wires
and Lo,=circle having the Centre at 0, and radius equal to the attractive
radius associated with the earth wire at the critical current
dk>dmin k = O , 1 , 2 ,...., M - I (44) Lw, LpI, Lp2= circles having the Centres at PO,P, and P2 and radii
equal to the attractiveradii associated with the phase conductors Po, PI
and P2 at lhe critical current, respectively
p<6'k59v k = O , 1 , 2 ,_...,M - l (45)

d,,,,, in inequality (44) is the minimum distance required In (47) and (48), hk is the height of the individual phase
between the earth wire and individual phase conductors. conductors, and H,, is the height of each of the earth wires.
However, any other clearance constraints can be formulated Hp is a function of 4 and So given in (from Fig. 9)
and included in the optimisation.
+
H,, = ho do COS(UO) (49)
The sequential quadratic programming (SQP) method [9] In (49), ho is the height of phase conductor Po.
is applied to solve the constrained minimisation problem. In the representative design study, a critical current IO
with a typical value of lOkA is adopted. In addition to
9 Representative designs effective shielding constraints, those relating to minimum
separation between earth wires and phase conductors are
9.1 Case 1 included. The minimum separation is set to be 7.5m.
The optimal design procedure is applied to a 330kV The results obtained from the cost nunimisation subject
transmission-line configuration shown in outline in Fig. 9. to constraints are as follows:
The transmission line is of a double-circuit construction, in
which each circuit has a vertical configuration. Phase do = 7 S m , 80 = 22.23"
conductor Po is nominated for defining the shielding angle To verify that the parameters identified give effective
So, and & is the distance between earth wire O1 and phase shielding, the shielding diagram for the critical current of
conductor Po. lOkA is constructed in Fig. 9. Owing to symmetry, the
For illustration, the earth-wire cost function is taken to shielding diagram is shown only for the circuit on the right-
comprise a h e d component Fo and a component propor- hand side of the tower. The shielding diagram for the left-
tional to the distance between each eartb wire and the hand side circuit can be constructed in the same way.
tower, and the earth-wire height above the top phase
conductors. The cost function CE(do,8,) is then 9.2 Case2
CE(d0,Bo) = Co[docos(Bo)+XDm -do sin(Bo)] In this case, the design procedure is applied to a 330kV
+ Fo (46) double-circuit transmission line shown in Fig. IO. The
conductor spacings are reduced in comparison with
In (46), COis the cost coefficient relating to the earth-wire those in case I . The other difference from case 1 is that
positions, and XDm is the perpendicular distance from all of the phase conductors in Fig. IO are at equal
phase conductor Po to the centre line of the tower. perpendicular distanczs from the tower. The minimum
Attractive radii for phase conductors and earth wires are separation between the earth wires and phase conductors is
given, respectively, in [6]: set to be 4.5 m.
D&K = 0 . 6 7 h F 6 p (47) The results obtained from the earth-wire cost minimisa-
tion subject to constraints are:
D& = 0.67Hy4.74 (48) do = 4.5m, 80 = 24.29"

IEE Proc-Cmer Trunon DiFrrLi.. Vol. ISO, h'o. 6, Nown2her 2W3 665

Authorized licensed use limited to: Vardhaman College of Engineering. Downloaded on January 25, 2010 at 09:42 from IEEE Xplore. Restrictions apply.
 Fig. 11 Shielding achieued with earth-wirepummeters obtained in
Fig. 10 Shielding achieued with earth-wire parameters obtained in cuse 3
cme 2 Singlecircuit transmission line
Doublecircuit transmission line Horizontal configuration
Vertical configuration with all phase conductors at equal perpendicular Po, P,, P, = phase conductors
distances from the tower o,,O2=earth wires
Pi,, PI, P2, P3, Pa, P5= phase conductors Lo,,Lo2=circles having the centres at 0 1 and O2and radii equal to
01,O,=earth wires the attractive radii associated with the earth wires 0 , and O2 at the
Lo,=circle having the Centre at 0 , and radius equal to the attractive critical current, respectively
radius associated with the earth wirc at the critical current Lm, Lpl, Ln = circles having the centres at Po, PI and P2 and radii
Lm, LpI. Lp2=circles having the centra at PO,P, and P2 and radii equal to the attractive radii associated with the phase wnductors Po,PI
equal to the attractive radii associated with the phare conductors PO,P, and P1 at the critical current, respectively
and P2 at the critical current, respectively Critical current adopted in the design= lOkA
Critical current adopted in the design= lOkA
Phase conductor Po is nominated for d e h n g 6 and Bo
Phase conductor Po is nominaled for defining 6 and So

The shelding diagram constructed in Fig. IO for the critical effective shielding for all of the phase conductors of a
current of I O kA confirms that effective shielding is achieved transmission line.
for all of the phase conductors, when the earth-wire With the present availability of powerful computers at
positions have the parameters do and 0,. low cost, the procedure represents an advancement in which
lightning protection design is carried out in a systematic
9.3 Case 3 way and at high speed. The design procedure removes the
transmission line with
I n this case, a 330 kV singlecircuit previous approximation, simplification and assumption
the horizontal configuration shown in Fig. 11 is considered. based on charts and graphical methods.
The minimum separation between earth wires and phase The design procedure developed has been successfully
conductors is set to be 9m. applied to many test cases in which representative results are
The results obtained from the earth-wire cost minimisa- given in the paper.
tion subject to constraints are:
11 Acknowledgments
do = 9m, Oo = 24.39"

In Fig. 1 I is shown the shielding diagram for the critical The authors gratefully acknowledge the support of the
current of I O k A . The diagram verifies that effective Energy Systems Centre at The University of Western
shielding for all of the phase conductors is achieved. Australia for the research work reported in the paper, and
the permission of Westem Power Corporation to use the
data for the design study. They express their appreciation to
10 Conclusions The University of Westem Australia for permission to
publish the paper.
A rigorous analytical proof of the effective shielding premise
in relation to the more recent EGMs referred to as
Eriksson's EGMs has been developed in the paper. The 12 References
procedure in the proof can be extended and applied to other I 'IEEE guide for improving the lightning performana of transmission
EGMs, such as IEEE EGMs, which represent the earth lines' IEEE Standards Board, Transmission and Distribution
plane explicitly. Committee of the IEEE Power Enginenng Swiety (IEEE Standard
The detailed analysis required in the derivation of the 1243-1997, 1997).
2 Young. F.S., Clayton, J.M., and Hileman, A.R.: 'Shielding of
proof has led to a set of constraints which have to be transmission lines. IEEE T r m . Power Anoar. Svrr.. 1963.. 132..
1. I

satisfied to achieve effective shielding. Based on these pp. 132-154

3 Brown, G.W., and Whitehead, ER.:'Field and analytical studies of
constraints, a systematic and practical design procedure has transmission line shielding: Part II', IEEE Power Appo. Syst.,
been developed for earth-wire placement, which achieves 1969. 88, (S), pp. 617-626

666 IEE Pro<.-Cuur T r m DVlrib.. VoL I50 No.6, Nouemher 2W3

Authorized licensed use limited to: Vardhaman College of Engineering. Downloaded on January 25, 2010 at 09:42 from IEEE Xplore. Restrictions apply.
 4 IEEE Working Group Repon: 'A simplilied method for estimating 7 k k , F.A.M.: 'Modeling of transmission line exposure to direct
lightnine oerformance af transmission tines'. IEEE T r m . Power ADW.
&r., lFi5, 104. (41,pp. 91%932
.' liehtnine strokes'. IEEE Tram Power Pelio.. 1990., 5., (
pi. 198i1997
4).
,,,

5 IEEE Working Group Report: 'Estimating lightning performance of 8 Eriksson, A.J.: 'The incidence of lightning strikes to power lines', IEEE
transmission lines T1-updates to analytical models', IEEE Trms. Power Tr-. Power Deli"., 1987, 2, (3). pp. 859470
Delia, 1993, 8, (3). pp. 1254-1267 9 Panier, E.R., and Tiw. A.L.: 'On combining feasibility. descent and
6 Eriksson, A.J.: 'An unproved eleCtrogeome1"c model for transmission superlinear convergence in inequality constrained aptimi/ation', Mnfh
line shielding analysis', IEEE T r m . Power Delir;, 1987, 2, (3), Proyrom, 1993,59, pp. 261-276
pp. 871486

IEE Proc-Gmer T r m n . Dlrr&., Vol. 150, No. 6, Nouemher 2W3

Authorized licensed use limited to: Vardhaman College of Engineering. Downloaded on January 25, 2010 at 09:42 from IEEE Xplore. Restrictions apply.