DIgSILENT Technical Documentation

Conductor and Overhead Line

Modelling
 DIgSILENT GmbH
Heinrich-Hertz-Strasse 9
D-72810 Gomaringen
Tel.: +49 7072 9168 - 0
Fax: +49 7072 9168- 88
http://www.digsilent.de
e-mail: [email protected]

Overhead Lines Modelling

Published by
DIgSILENT GmbH, Germany

Copyright 2007. All rights

reserved. Unauthorised copying
or publishing of this or any part
of this document is prohibited.

Document No

Build 246 30.03.2007

TechRef Overhead Line Parameters (DRAFT) 2

 Table of Contents

Table of Contents

1 CONDUCTOR TYPES .............................................................................................................................................. 4

1.1 GEOMETRICAL MEAN RADIUS (GMR) OF A CONDUCTOR ................................................................................................. 4
1.1.1 GMR of Solid Conductors ............................................................................................................................................. 5
1.1.2 GMR of Tubular conductors ......................................................................................................................................... 5
1.1.3 Numerical example...................................................................................................................................................... 6
1.2 BUNDLE CONDUCTOR.................................................................................................................................................... 6

2 TOWER TYPES: OVERHEAD LINE PARAMETER CALCULATION IN DIGSILENT POWERFACTORY. ......................... 8

2.1 OVERHEAD LINE IMPEDANCE CALCULATION IN DIGSILENT POWERFACTORY. ................................................................. 8
2.1.1 Geometrical Impedance............................................................................................................................................... 9
2.1.2 Earth Correction. Carson´series. .................................................................................................................................. 9
2.1.3 Impedance Matrix ..................................................................................................................................................... 11
2.2 OVERHEAD LINE CAPACITANCE CALCULATION IN DIGSILENT POWERFACTORY. ............................................................ 11

TechRef Overhead Line Parameters (DRAFT) 3

1 Conductor Types
Conductor Type (TypCon) allows the definition of geometrical and electrical characteristics for a conductor or a bundle of
subconductors, which are used later by the Tower Type (TypTow) routine for the calculation of electrical parameters of lines
based on tower types. Thus, this is the very first step to be carried out when deriving a transmission line model from its
geometrical characteristics.

Figure 1 shows the conductor type dialog. A conductor may be constituted by more than one subconductor, i.e. a bundle
conductor. In this case, the number of subconductors per bundle is to be specified. The nominal current and voltage refer to
the bundle conductor.

Figure 1: Conductor type dialog.

Subconductors parameters are used to calculate its internal impedance: DC resistance, diameter (or radius) and internal
inductance, which can also be specified in terms of relative permeability or geometrical mean radius GMR (for relationships
between these parameters see below).

The internal impedance is adjusted according to skin effect if the flag is selected.

More details on subconductor specifications and skin effect in the following sections.

1.1 Geometrical Mean Radius (GMR) of a Conductor

Geometrical Mean Radius (GMR) is usually provided in manufactures´ datasheets. This is the data the user is encouraged to
give as input for the conductor type definition.

TechRef Overhead Line Parameters (DRAFT) Page 4 of 12

If not available, Power Factory can also calculate the GMR at power frequency starting from the conductor diameter (or radius)
or the relative permeability; however, under the assumption of an uniform current distribution over its cross section. In other
words, the influence of elementary wires is not taken into account.

1.1.1 GMR of Solid Conductors

Consider a solid conductor of nonmagnetic material and radius r as depicted in Figure 1.

Figure 2: Solid conductor

From the magnetic field theory, the auto inductance of the conductor can be calculated as in (1),

µ0 µ0 ⎛ 1 ⎞
Lself = + ⋅ ln⎜ ⎟ (1)
8π 2π ⎝ 2⋅r ⎠

where the first term of the sum is the internal inductance associated with the magnetic flux inside the conductor (2) and the
second one represents the external inductance associated to the external flux (3).

µ0
Lint ernal =
8π (2)
µ ⎛ 1 ⎞
Lexternal = 0 ⋅ ln⎜ ⎟
2π ⎝ 2⋅r ⎠ (3)

In means of the Geometrical Mean Radius (GMR) of the conductor, (1) can be rewritten as (4)
µ0 ⎛ 1 ⎞
Lself = ⋅ ln⎜ ⎟ (4)
2π ⎝ 2 ⋅ GMR ⎠

Therefore, between (1) and (4) the following expression can be deduced for the GMR of a solid conductor under the above
mentioned assumptions:
1
−
GMR = r ⋅ e 4
(5)

1.1.2 GMR of Tubular conductors

A similar procedure can be used to calculate the GMR of a tubular conductor. A uniform magnetic distribution of the current
over the conductor section is here to be assumed as in the preceding case as well.

For the tubular conductor depicted in Figure 3, the auto inductance is calculated as in (6):
.

TechRef Overhead Line Parameters (DRAFT) Page 5 of 12

Figure 3: Tubular conductor.

µ0 ⎡ q4 r 3q 2 − r 2 ⎤ µ 0 ⎛ 1 ⎞
Lself = ⋅⎢ ⋅ − 2 ⎥
+ ⋅ ln⎜ ⎟
q 4 ⋅ (r − q )⎦⎥ 2π
ln
2π ⎣⎢ (r − q )
(6)
⎝ 2⋅r ⎠
2 2 2 2

Again, the first term represents the internal inductance, and thus for the tubular conductor:

µ0 ⎡ q4 r 3q 2 − r 2 ⎤
Lint = ⋅⎢ ⋅ ln − ⎥
2π (
⎢⎣ r 2 − q 2 )
2
(
q 4 ⋅ r2 − q2 )
⎥⎦
(7)

Between (4) and (6) results (8) for the GMR of a tubular conductor:

⎡ 3q 2 − r 2 q4 r⎤
GMR = r ⋅ exp ⎢ 2 − ⋅ ln ⎥
⎣⎢ 4 r − q(2
r 2 − q2 ) ( )
2
q ⎦⎥
(8)

It should be emphasized here again that (5) and (8) assume a uniform distribution of the current over the cross section of the
conductor and therefore the elementary wires are not taken into account.

1.1.3 Numerical example

For an Al/St Conductor, 120/20 mm2, Radius 7,75 mm and q/r= 0,226, the following values can be verified in Power factory:

From (7), Linternal=0,045479 mH/km

From Eq. (6), Lself= 0,878862 mH/km
From Eq. (8), GMR = 6,17369 mm

1.2 Bundle Conductor

Bundle conductors are frequently use in high voltage transmission lines to reduce corona losses and fulfill electromagnetic
radiation requirements. Two or more subconductors per phase are held together by spacers conforming a symmetrical bundle
conductor.

For the calculation of electrical parameters, Power Factory replaces the bundle subconductors with a conductor of equivalent
radius. Subsequent calculations of line parameters (internal impedance and geometrical impedance due to external flux) are
then carried out considering this equivalent single conductor as if it were located in the middle of the bundle.

This approach is based on the following two assumptions:

− The bundle is symmetrical
− The current distribution among the subconductors within a bundle is uniform

TechRef Overhead Line Parameters (DRAFT) Page 6 of 12

Under these assumptions, the radius of the equivalent conductor will be:

rB = n n ⋅ r ⋅ R n −1 (9)

where r is the radius of an individual subconductor, n the number of subconductors and R the radius of the bundle (calculated
from the bundle spacing a as depicted in Figure 4 ).

2r

Figure 4: Symmetrical bundle of radius R with n subconductors. Bundle spacing a

The relationship between the bundle spacing a and angle α is given by:

⎛π ⎞
a = 2 R ⋅ sin ⎜ ⎟
⎝n⎠

The equivalent GMR results:

GMRB = n n ⋅ GMR ⋅ R n −1 (10)

where GMR is the geometrical mean radius of individual subconductor in bundle (see (5) and (8)).

A different method for calculating line parameters of bundle conductors consists in computing the parameters for each
subconductor as if it were represented as an individual conductor. Since all subconductors within a bundle have the same
voltage, the order of the geometrical matrices is then reduce to matrices of equivalent phase conductors. Even though
nonsymmetrical bundle conductor can also be considered in this case, a uniform current distribution among subconductors
within the bundle is still to be assumed. Differences between both methods seem however not to be relevant. Power factory
supports the first approach described above.

For calculating the internal impedance of the bundle, the internal impedance of one sub-conductor must be divided by the
number of sub-condcutors n

TechRef Overhead Line Parameters (DRAFT) Page 7 of 12

2 Tower Types: Overhead Line Parameter Calculation in
DIgSILENT PowerFactory.

2.1 Overhead Line Impedance Calculation in DIgSILENT PowerFactory.

The line impedance consists of three components:

1. The internal impedance Z ' L = R' (ω ) + jω ⋅ L´int (ω ) , considering the internal field of the conductor (only self
impedance) and the conductor resistance. Due to skin effect, the internal reactance as well as the internal resistance is
frequency dependent.

2. Geometrical impedance Z ' G , being the impedance of an ideal conductor without any magnetic field inside and an ideally
conducting ground. The geometrical inductance is not frequency dependent.

3. Earth correction term Z ' E = R't (ω ) + jω ⋅ L´t (ω ) , a frequency dependent term considering finite earth
conductivity and proximity effects. It depends on the earth resistivity and the line geometry (different coefficients for self
and mutual impedances).

Figure 5: Definition of the self (left) and mutual (right) impedance of a line.

According to Figure 5, self and mutual impedance can hence be expressed as follows:

Z 'ii = Z 'Gii + Z ' Eii + Z ' Lii ( 11 )

Z 'ij = Z 'Gij + Z ' Eij ( 12 )

Remarks: even though Z 'ij defines the mutual impedance, it has also a real component due to the resistance of the earth
return path.

TechRef Overhead Line Parameters (DRAFT) Page 8 of 12

2.1.1 Geometrical Impedance
µ0
Z 'Gii = jω N ii ( 13 )
2π

µ0
Z 'Gik = jω N ik ( 14 )
2π

The frequency independent coefficients N can be calculated as follows:

2hi
N ii = ln ( 15 )
ri
d ik
N ik = ln ( 16 )
d ' ik

with hi , d ij and ri as defined in Figure 6.

ri
i rk
d ik
k
hi

hk
d ik'

'
k
'
i

Figure 6: Calculation of the geometrical coefficients. Conductor profiles between towers.

In case of bundle conductors, the radius ri in (15) is to be replaced by the equivalent radius as calculated in (9).

hi is the average height above ground of conductor i. If the conductor profile can be described as a parabola (which is quite
accurate for spans below 500 meters), then the average height above ground is:
2 1
hi = haverage = ⋅ hmidspan + ⋅ htower
3 3
where hmidspan is the conductor height at midspan and htower at tower as shown in (6).

2.1.2 Earth Correction. Carson´series.

The earth correction term and hence the impedance of the earth return path is calculated according to the Carson’ series.

TechRef Overhead Line Parameters (DRAFT) Page 9 of 12

 µ0
Z 'E = ω (P + jQ ) ( 17 )
π
The coefficients P and Q are highly frequency dependent. They can be calculated as follows:

π
P= −
8
[ ]
− b1 x cos ϑ + b2 (c 2 − ln x )x 2 cos 2ϑ + x 2ϑ sin 2ϑ + b3 x 3 cos 3ϑ − d 4 x 4 cos 4ϑ
5
[ 6 6
]
− b5 x cos 5ϑ + b6 (c6 − ln x )x cos 6ϑ + x ϑ sin 6ϑ + b7 x 7 cos 7ϑ − d 8 x 8 cos 8ϑ K ( 18 )

Q = (k − ln x )x 6 +
1
2
[
+ b1 x cos ϑ − d 2 x 2 cos 2ϑ + b3 x 3 cos 3ϑ − b4 (c 4 − ln x )x 4 cos 4ϑ + x 4ϑ sin 4ϑ ]
6 7
[
+ b5 x cos 5ϑ − d 6 x cos 6ϑ + b7 x cos 7ϑ − b8 (c8 − ln x )x cos 8ϑ + x ϑ sin 8ϑ K
5 8 8
] ( 19 )

where

x = 2hi µ ⋅ κ ⋅ ω and ϑ = 0 for the self impedance

and x = d ij µ ⋅ κ ⋅ ω for the mutual impedance

The coefficients are calculated as follows:

sign
bi = bi − 2
i ⋅ (i + 2)
sign = +1 for I = 1,2,3,4….
sign = +1 for I = 5,6,7,8 … and alternating after 4 terms

1 1
ci = ci − 2 + +
i i+2

π
di = ⋅ bi
4

1
k= + ln 2 − C = 0,61593
2

⎛ n 1 ⎞
C = lim⎜ ∑ − ln n ⎟ = 0,577215
n →∞
⎝ ν =1 ν ⎠

with the following initial values:

2 1
b1 = b2 =
6 16

5
c2 = − C + ln 2 = 1,3659315
4

At fundamental frequency, Carson´ serie (eq. (18) and (19)) can be approximated as follows:

TechRef Overhead Line Parameters (DRAFT) Page 10 of 12

 π
Pik ≈
8
1
Qik ≈ (k − ln x ik )
2
2.1.3 Impedance Matrix

The impedance matrix relates the voltage across the line to the line currents:

∆U = Z × I

The dimension of the “natural” impedance matrix corresponds to the number of phase conductors + the number of earth wires.

⎡∆U E ⎤ ⎡Z EE Z EP ⎤ ⎡I E ⎤
⎢ ∆U ⎥ = ⎢ Z ×
Z PP ⎥⎦ ⎢⎣I P ⎥⎦
⎣ P⎦ ⎣ PE

For using the matrix in a three phase model, it must be reduced to the number of phase conductors. Therefore, an ideal
grounding is considered leading to the assumption that there is no voltage drop across the earth conductors. The reduced
matrix is hence:

[ −1
]
∆U P = Z PP − Z PE Z EE Z EP I P = Z red I P

If a detailed modeling of earth wires is required, earth wires must be entered as phase conductors.

The symmetrical impedance matrix is calculated as follows:

Z 012 = S ⋅ Z red ⋅ T

⎡1 1 1⎤
⎢
where T = 1 a
2
a ⎥⎥ , S = T −1 and a = e j120°
⎢
⎢⎣1 a a 2 ⎥⎦

Only for perfectly transposed lines the resulting symmetrical matrix is diagonal, with all elements outside the main diagonal
equal to zero, and hence there is no coupling between sequence modes. In this case, the resulting matrix looks like:

⎡Z s + 2Z m 0 0 ⎤
Z 012 = ⎢⎢ 0 Zs − Zm 0 ⎥⎥
⎢⎣ 0 0 Z s − Z m ⎥⎦

where Z s is the self impedance and Z m the mutual impedance of the perfectly transposed line.

Power Factory prints the natural, reduced (or phase equivalent) and symmetrical matrix into the output window.

2.2 Overhead Line Capacitance Calculation in DIgSILENT PowerFactory.

The natural potential coefficient matrix P relates the voltage to the charge of each conductor. The dimension of this “natural”
matrix corresponds to the number of phase conductors + the number of earth wires.

U = P×Q
TechRef Overhead Line Parameters (DRAFT) Page 11 of 12
Analogous to the impedance matrix, also the matrix of potential coefficients is reduced to the number of phase conductors,
under the assumption that the voltage at the earth wires is zero. It follows:

⎡U E ⎤ ⎡PEE PEP ⎤ ⎡Q E ⎤
⎢U ⎥ = ⎢P ×
PPP ⎥⎦ ⎢⎣Q P ⎥⎦
⎣ P ⎦ ⎣ PE

UE = 0

[ −1
U P = PPP − PPE PEE PEP ⋅ Q P ]
and hence the reduced coefficient matrix is

[
Pred = PPP − PPE PEE PEP
−1
]
The matrix of capacitance coefficients can be obtained now by inverting the potential coefficient matrix:

−1
C red = Pred

Using the same transformation matrix T and S as for the impedance, we can now calculate the symmetrical admittance
matrix as follows:

C 012 = S ⋅ C RST ⋅ T

As in the impedance case, the resulting symmetrical matrix Z 012 is diagonal only for perfectly transposed lines, with all
elements outside the main diagonal equal to zero. Hence there is no coupling between sequence modes. In this case, the
resulting admittance matrix looks like:

⎡C s + 2 ⋅ C m 0 0 ⎤
C 012 = ⎢⎢ 0 Cs − Cm 0 ⎥⎥
⎢⎣ 0 0 C s − C m ⎥⎦

where C s is the self capacitance and C m the coupling capacitance of the perfectly transposed line.

TechRef Overhead Line Parameters (DRAFT) Page 12 of 12