Digsilent Powerfactory: Technical Reference Documentation
Digsilent Powerfactory: Technical Reference Documentation
Cable Systems
TypCabsys
DIgSILENT GmbH
Heinrich-Hertz-Str. 9
72810 - Gomaringen
Germany
http://www.digsilent.de
info@digsilent.de
r994
Copyright ©2013, DIgSILENT GmbH. Copyright of this document belongs to DIgSILENT GmbH.
No part of this document may be reproduced, copied, or transmitted in any form, by any means
electronic or mechanical, without the prior written permission of DIgSILENT GmbH.
Contents
1 Introduction 3
3.4.1 Impedance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
3.4.2 Admittance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3.5.1 Impedance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3.5.2 Admittance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
A Parameter Definitions 14
B Calculation Results 16
C References 19
List of Figures 20
List of Tables 21
1 Introduction
This document describes the definition of a cable system in terms of its geometry, the properties
of the conducting, semi-conducting and insulating layers, installation characteristics (buried di-
rectly underground, in a pipe). Besides it details calculation of its frequency-dependent electrical
parameters.
The definition of a frequency-dependent cable system success in PowerFactory with help of two
type objects: a single core cable type TypCab which describes the constructive characteristics
of the cable and a cable system type TypCabsys, which defines the coupling between phases,
i.e. the coupling between the single core cables in a multiphase/multi-circuit cable system.
A built-in cable constants function in the cable system type calculates then the frequency-
dependent electrical parameters (impedance and admittance matrices). The function can han-
dle coaxial cables consisting of a core, sheath and armour directly undergrounded or installed
in pipes (pipe-type cables). This function can be started in a stand-alone case from the Cal-
culate button on the edit dialog of the cable system, in which case the results are printed to
the output windows, or be automatically called by any simulation function in PowerFactory , eg.
when running a frequency scan or when adjusting the model for an electromagnetic transient
simulation.
Finally, the reader should notice that this cable system type supports the definition of the cable
in terms of geometrical data; is the cable to be defined in terms of electrical data, the reader is
referred to [3] where the general line/cable element (ElmLne) is used instead.
The single core cable type TypCab supports up to three tubular conducting layers in coaxial
arrangement, i.e. core, sheath and armour, separated by three insulating layers. Figure 2.1
shows the typical layout of a HV AC single core cable. The model also supports the definition
of a core-outer and insulation-outer semiconducting layer.
Figure 2.1: Cross section of a single core cable including the core, sheath and armour.
Section A shows the complete list of input parameters including units, range and the symbol
used in this document. Hover the mouse pointer over the input parameters in the edit dialog of
TypCab to display the name of the input parameter. This is the name listed in the first column of
the table.
The input data in the edit dialog of TypCab is organized according to layers, i.e. the conducting,
insulation and semiconducting layers, if available. Use TypCab to enter all the geometrical data
defining the cross section of the single core cable and the properties of all constitutive materials.
To account for the compacting ratio of the cross section of the conducting layers (stranded
conductors, shaped compact, etc.) the users can enter a filling factor Cf.. This filling factor is
related with the dc resistance of the cable by the following equation:
1
RDC [Ω/km] = ρ[µΩ · cm] · × 10
π · (r2 − q 2 ) · Cf
where r and q are the outer and inner radius of the conducting layer respectively.
The user chooses the input parameter between the filling factor in % or the DC resistance in
ohm/km by clicking the selection arrow note that one of them is always greyed out indicating
its dependency on the other one.
If the temperature dependency of line/cables option is enabled in the load flow calculation, the
resistivity of the conducting layers is adjusted by the following equation
ρT = ρ20◦ C · [1 + α(20 − T )]
where α is the temperature coefficient of resistance. The resistivities and temperature coeffi-
cient of common metals are given in Table 2.1 for reference.
The cable system type TypCabsys is used to complete the definition of a cable system. It
defines the coupling between phases, i.e. the coupling between the single core cables in a
multiphase/multi-circuit cable system. As in general the cables are laid close together this cou-
pling has to be taken into account.
Among other factors, this coupling depends on how the cables are laid. The PowerFactory
model supports following two options:
• Parallel single-core cables: the cables are grounded direct into ground. This is normally
the case of underground HV AC cables.
• Pipe-Type cables: the cables are drawn into a pipe, usually made of steel, and the pipe
laid into ground. This is in widespread use in submarine cables.
The input parameter ”Buried: Direct in ground/ in Pipe” lets the user choose between both
models. In case of pipe-type cables additionally data is required for the pipe. The complete list
of input parameters is shown in Table A.2 in section A.
The cable system type also defines the bonding conditions of the sheath and armours when
available.
The calculation of the impedance and admittance of the cable is based on the cable constants
equations formulated by [2] and underlies the following assumptions:
• Coaxial arrangement of the conducting and insulating layers inside the single core cable
• Single core cables inside the pipe are concentric with respect to the pipe
• Each conducting layer of the cable has constant permeability. Furthermore, conducting
layers are non-magnetic so that the cable model does not account for current-dependent
saturation effects
• Displacement currents and dielectric losses of the insulating layers are negligible.
A general formulation of the series impedance and shunt admittance of the cable is given by:
∂
[U ] = −[Z] · [I] (1)
∂x
∂
[I] = −[Y ] · [U ] (2)
∂x
where [U ] and [I] are the voltage and currents vectors at a distance x along the cable.
The dimension of [Z] and [Y ] depends on the total number of cables in the system and the total
number of layers per single core cable. For instance in a three phase cable system with three
conducting layers per single core cable (core, sheath and armour) the dimension of the [Z] (i.e.
[Y ]) results 9 (=3 phases x 1 single core cable/phase x 3 conducting layers/cable).
[Z] and [Y ] are symmetric square matrices that can be expressed in the following terms:
[Z] = [Z I ] + [Z P ] + [Z C ] + [Z 0 ] (3)
[Y ] = s · [P ]−1
[P ] = [PI ] + [PP ] + [PC ] + [P0 ] (4)
where [P ] is a potential coefficient matrix and the Laplace’s operator (complex frequency).
The matrices with subscript I account for the internal impedance and admittance respectively
and matrices with subscript O for the earth or air return path. In case if a pipe enclosure cable
the matrices with subscript C and P define the impedance and admittance of the pipe; these
matrices becoming zero if the cable is laid directly underground. In the next sub-chapters we will
discuss the physical meaning of these sub-matrices and the formulas used to calculate them.
• Subscript I accounts for the internal impedance, subscript O for the earth or air return path
and subscripts C and P for the pipe enclosure if available.
• Subscripts c,s and a (lower case) are used for core, sheath and armour in cable layer
equations.
• Subscripts i,j and k refer to the cables in the system (typically three cables in a three
phase cable system).
The internal impedance is associated to the longitudinal voltage drop due to the magnetic field
inside the single core cable and it is given by the following equation:
U Z Z cs Z ca Ic Ic
∂ c cc
U s = Z sc Z ss Z sa · I s = −[ZI ] · I s (5)
∂x
Ua Z ac Z as Z aa Ia Ia
where the layer internal impedances in 5 are defined in terms of coaxial loop impedances as
follow:
Z cc = Z 11 + 2 · Z 12 + Z 22 + 2 · Z 23 + Z 33
Z cs = Z sc = Z 12 + Z 22 + 2 · Z 23 + Z 33
Z ca = Z ac = Z sa = Z as = Z 23 + Z 33
Z ss = Z 22 + 2 · Z 23 + Z 33
Z aa = Z 33 (6)
The impedances with subscript 1,2 and 3 are referred as loop impedances. For instance Z 11 is
the impedance of the inner most loop of concentric tubular conductors and therefore that of the
loop core-sheath.
The impedance of the tubular conductors are found with the modified Bessel functions with tube
= c, s and a respectively:
ρm
Z tube,IN = {I0 (mq) · K1 (mr) + K0 (mq) · I1 (mr)}
2πqrD
ρm
Z tube,OU T = {I0 (mr) · K1 (mq) + K0 (mr) · I1 (mq)}
2πqrD
ρ
Z tube,M U T U AL = (8)
2πqrD
where
s
jωµ 1
m= = (10)
ρ p
The parameter m is the reciprocal of the depth of penetration p and are both frequency-dependent
complex values.
Z IN S accounts for the longitudinal voltage drop due to the magnetic filed in the insulating layers.
For the general case of non-concentric tubular conductors it results:
( " 2 #)
µ0 qk di
Z IN S = jω · ln 1− (11)
2π ri qk
µ0 qk
Z IN S = jω · ln
2π ri
The internal admittance matrix is associated to the capacitive coupling and dielectric losses due
to the insulating layers within the single core cable. The capacitance and dielectric losses of
each insulating layers is given by:
2πε0 εr 1
Ci = =
ln(r/q) Pi
Gi = ωCi · tg(δ) (12)
Assuming that the single core cable consist of three layers, hence the insulation between core
and sheath, sheath and armour and outermost insulating layer of the single core cable, it follows:
Pc + Ps + Pa Ps + Pa Pa
[PI ] = Ps + Pa Ps + Pa Pa (13)
Pa Pa Pa
1
[CI ] = (14)
[PI ]
The model supports the definition of a semiconducting layer on the conductor’s outer surface
and the insulation’s outer surface. These semiconducting layers mainly influence the admittance
of the insulation. Their effect on the impedance of the conductor is rather minor and therefore
not considered by the moment in the model.
The capacitance and conductance of the tubular semiconducting layer is given by the following
equations:
1
CSC = 2πε0 · εrSC
ln(rSC /qSC )
2π 1
GSC = ·
ρSC ln(rSC /qSC )
where rSC and qSC are the outer and inner radius of the tubular semiconducting layer respec-
tively, εrSC the relative permittivity and ρSC the resistivity.
Hence the equivalent admittance of the insulation under consideration of the semiconducting
layers is calculated in the following form:
1 1 1
= +
GiIns GIL GSC
1 1 1
= +
CiIns CIL CSC
3.4.1 Impedance
Let’s assume being i, j, k three parallel single core cables each of them consisting of core,
sheath and armour. Equation (1) can be expanded as:
[Z I,ii ] [0] [0] [Z 0,ii ] [Z 0,ij ] [Z 0,ik ]
[U ] [I i ]
.. ..
∂ i
[U j ] = . [Z I,jj ] [0] + . × [I j ]
[Z 0,jj ] [Z 0,jk ] (16)
∂x . .
[U k ] [I k ]
.. ..
··· [Z I,kk ] ··· [Z 0,kk ]
where [Z 0,s ] and [Z 0,m ] are the self and mutual earth-return impedance matrices of the cable
system given as:
Z e,s Z e,s Z e,s
[Z 0,s ] = Z e,s Z e,s Z e,s ...s = jj, kk, ll (17)
Z e,s Z e,s Z e,s
Z e,m Z e,m Z e,m
[Z 0,m ] = Z e,m Z e,m Z e,m ...m = jk, kl, lj (18)
Z e,m Z e,m Z e,m
Z e,m is the mutual earth-return impedance between two parallel cables i, j given by:
ωµ0
Z e,jk = j [K0 (m · dik ) − K0 (m · Dik )] + (Pik + jQik ) (19)
π
and Pik + jQik the terms of the Carson’s serie (see [1] for further reference).
Z e,s is the self earth-return impedance of the single core cable. Its value is obtained from (19)
by replacing d with R, D with 2h and h+y with 2h.
3.4.2 Admittance
As the cable is directly laid underground and the earth surrounding the cable being assumed
an equipotential surface, there is no capacitive coupling effect among the single core cables. It
follows then that [P 0 ] = Θ in equation 4 and therefore the admittance matrix of the cable results:
[PI,i ] 0 0
[P ] = [PI ] = 0 [PI,j ] 0 (20)
0 0 [PI,k ]
[Y ] = [P ]−1
[I i ] [U i ]
∂
[I j ] = −[Y ] · [U j ] (21)
∂x
[I k ] [U k ]
3.5.1 Impedance
Assuming again a system of three single core cables i, j, k each of them consisting of core,
sheath and armour, equation (3) can be expanded as follows for the case of a pipe type cable:
[Z I,ii ] [0] [0] 0
[U i ]
.
..
∂ [U j ] = [Z I,jj ] [0] 0
∂x [U k ] ..
. ··· [Z I,kk ] 0
[U p ]
0 0 0 0
[Z P,ii ] [Z P,ij ] [Z P,ik ] 0
..
. [Z P,jj ] [Z P,jk ] 0
+
...
··· [Z P,kk ] 0
0 0 0 0
[Z C1 ] [Z C1 ] [Z C1 ] Z C2
..
. [Z C1 ]
[Z C1 ] Z C2
+
...
··· [Z C1 ] Z C2
Z C2 Z C2 Z C2 Z C3
[Z 0 ] [Z 0 ] [Z 0 ] Z 0
..
[I i ]
× [I j ]
. [Z 0 ] [Z 0 ] Z 0
+
. [I k ] (22)
.. · · · [Z 0 ] Z 0
Ip
Z0 Z0 Z0 Z0
[Z P ] defines the self and mutual impedances of the pipe-return path of the single core cables.
A submatrix is given by:
Z P,ij Z P,ij Z P,ij
[Z P,ij ] = Z P,ij Z P,ij Z P,ij (23)
Z P,ij Z P,ij Z P,ij
Self impedance of the with pipe-return path for the i-th cable (i=j):
∞
" 2n #
µ0 µr K0 (mq) X di 2µr Kn (mq)
Z P,ii = jω + (24)
2π mqK1 (mq) n=1 q nµr Kn (mq) − mqKn0 (mq)
Mutual impedance between the i-th and the j-th cables with common pipe-return path (16=j):
µ0 q K0 (mq)
Z P,ij = jω ln √ 2 2 + µr mqK 1 (mq)
2π di +dj −2di dj cos ϑij
∞
X di dj n )
2µr Kn (mq) 1
+ cos (nϑij ) − (25)
n=1
q2 nµr Kn (mq) − mqKn0 (mq) n
[Z C ] is the connection impedance matrix between the pipe inner and outer surfaces. The sub-
matrix [Z C1 ], [Z C2 ] and [Z C3 ] are given by:
Z C1 Z C1 Z C1
[Z C1 ] = Z C1 Z C1 Z C1 (26)
Z C1 Z C1 Z C1
where Z C1 , Z C2 and Z C3 are calculated using equations (8) to (11) for the impedance of tubular
conductors and tube being the pipe as follows:
Finally, [Z 0 ] represent the impedance of the earth return-path of the pipe. The diagonal subma-
trix [Z 0 ] is given by:
Z0 Z0 Z0
[Z 0 ] = Z 0 Z0 Z 0 (28)
Z0 Z0 Z0
where Z 0 is the self earth return impedance of the pipe according to equation (19).
3.5.2 Admittance
The admittance follows the general definition in terms of the potential coefficient matrix as fol-
lows:
[PI,ii ] [0] [0] 0
.
..
[PI,jj ] [0] 0
[P ] =
..
. ··· [PI,kk ] 0
0 0 0 0
[PP,ii ] [PP,ij ] [PP,ik ] 0
..
. [PP,jj ] [PP,jk ] 0
+
...
··· [PP,kk ] 0
0 0 0 0
[PC ] [PC ] [PC ] PC
..
. [PC ] [PC ] PC
+
. (29)
..
· · · [PC ] PC
PC PC PC PC
where
[GI,i ] 0 0 0
0 [GI,j ] 0 0
[Y ] =
+ jω ([P ])−1 (30)
0 0 [GI,k ] 0
0 0 0 0
[I i ] [U i ]
∂ [I j ] = −[Y ] · [U j ]
(31)
∂x [I k ] [U k ]
Ip Up
Note in equation (30) that dielectric losses of the pipe are not being considered.
Each of the [PI,ii ] submatrices of [PI ] is the internal potential coefficient matrix of the single
core cable according to (20).
[PP ] is the pipe internal potential coefficient matrix and defines the capacitive coupling between
the outermost layer of the single core cables and the pipe and hence the dielectric medium
between the cables and the pipe. Each of the submatrices [PP,ij ] of [PP ] is a matrix with equal
elements given in the following form:
PP,ij PP,ij PP,ij
[PP,ij ] = PP,ij PP,ij PP,ij (32)
PP,ij PP,ij PP,ij
with
( " 2 #)
1 q di
Pii = ln 1− (33)
2πε0 εr Ri q
∞ n
1 q −
X 1 d d
i j
Pij = ln q · · cos ϑ ij (34)
2πε0 εr d2i + d2j − 2di dj cos ϑij n=1
n q2
[PC ] is the potential coefficient matrix between the pipe inner and outer surfaces and hence the
capacitance due to the dielectric layer surrounding the pipe. A submatrix and the last column
and row elements are given by:
PC PC PC
[PC ] = PC PC PC (35)
PC PC PC
1 r
PC = · ln (36)
2πε0 εr q
It is assumed in the model that the pipe is underground. Therefore the outer surface of the
insulating layer surrounding the pipe is in direct contact with the earth (equipotential surface
with U=0). Hence no additional capacitive effect exists between the insulating layer of the pipe
and ground.
A Parameter Definitions
Table A.1: Input parameter of the single core cable type (TypCab)
B Calculation Results
The cable constants function in stand-alone mode can be started from the ”Calculate” button
on the edit dialog of the cable system type TypCabsys. Then PowerFactory prints the resulting
impedance and admittance matrices to the output windows.
It follows an extract of the output window for a 132 kV, 3-phase cable system, 630 mm2 , directly
underground. The first two matrices correspond to the unreduced layer impedances and admit-
tances in phase components; cores first, followed by sheaths. Cables are in the same order as
the input. Rows follow real and imaginary part.
The next two matrices are the impedances and admittances in symmetrical components in 0-
1-2 sequence. Idem before, cores come first followed by the sheaths. Cables are in the same
order as the input. Rows follow real and imaginary part.
C References
[1] Technical Reference Overhead Lines Constants, 2009.
[2] A. Ametani. A General Formulation of Impedance and Admittance of Cables. IEEE Trans-
actions on Power Apparatus and Systems, PAS-99(3), May 1980.
[3] B. K. Bose. Modern Power Electronics and AC Drivers. Prentice Hall of India, 1 edition,
2008.
List of Figures
2.1 Cross section of a single core cable including the core, sheath and armour. . . . 4
List of Tables