Derivation of Theoretical Formulas of Sequence

Currents on Underground Cable System

Teruo Ohno, Akihiro Ametani, Claus Leth Bak

Abstract-- A reliable operation of a cable system necessitates sequence impedances / currents which can consider sheath
an accurate calculation of sequence impedances of the system. It bonding and sheath grounding resistance at substations and
is recently even more important as longer cable lines are being normal joints. As a result, it has been a common practice that
constructed and planned. It has been a common practice that
those sequence impedances or currents are measured after the
these sequence impedances or currents are measured after the
installation and it is difficult to predict these values beforehand installation, and it is considered difficult to predict those
with good accuracy. This paper derives theoretical formulas of values beforehand.
the sequence currents for a cross-bonded cable and a normal- This paper derives theoretical formulas of the sequence
bonded cable. The formulas give an important advantage in currents for a cross-bonded cable and a normal-bonded cable.
setting up transient overvoltage studies as well as planning The formulas are, of course, useful for planning studies. In
studies. The accuracy of the proposed formulas is verified addition, they will help improve the initial condition setup for
through a comparison with EMTP simulations.
transient overvoltage studies. The accuracy of the proposed
formulas is verified through a comparison with EMTP
Keywords: cable, sequence current, sequence impedance.
simulations.
I. INTRODUCTION

I Nrecent years, many cable systems are under construction

and are planned for construction. For example, 400 kV
underground systems whose total length will reach 400 km
II. DERIVATION OF THEORETICAL FORMULAS OF SEQUENCE
CURRENTS

will be constructed in Denmark [1]. Since 1995, all new For underground cable systems which are longer than
transmission and distribution lines in Denmark at 100 kV or about 2 km, it is a common practice to cross-bond the metallic
below have been mandated to be undergrounded. In sheaths of three phase cables to reduce sheath currents and to
November, 2008, the Danish Parliament additionally decided suppress sheath voltages at the same time [9]. Due to the
to underground all new transmission lines up to 400 kV. Total increase of off-shore wind farms and cross-border transactions,
circuit length undergrounded at 132 kV or higher will be submarine cables, which are generally normal-bonded, are
expected to reach 3000 km. In Italy, 38 km 380 kV cable from now becoming a popular type of cable.
the main land to Sicily is under construction [2][3]. Therefore, this paper derives theoretical formulas of the
Power flow calculations, transient stability analyses, and sequence currents for a majority of underground cable systems,
relay setting studies are performed for the planning and that is, a cross-bonded cable which has more than a couple of
operation of the network. These analyses and studies require major sections. It also derives theoretical formulas for a
sequence impedances or currents as input data. The sequence normal-bonded cable, considering the increased use of
impedance / current calculation of overhead lines is well submarine cables.
known and introduced in textbooks [4]. For underground
cables, theoretical formulas are proposed for the cable itself A. Cross-bonded Cable
[5-8]. In order to derive accurate theoretical formulas, 1) 6  6 impedance matrix
however, it is necessary to consider the whole cable system, One cable system corresponds to 6 conductor system
including sheath bonding, since the return current of an composed of 3 cores and 3 metallic sheaths. The 6  6
underground cable flows through both metallic sheath and impedance matrix of the cable system is given by the
ground. Until now, there has existed no formula of the following equation [4].
Zc Zm  Zc Zm
Z    t  (1)
T. Ohno is with Tokyo Electric Power Company, Japan and the Institute of  Zm Zs  Zm Zs 
Energy Technology, Aalborg University, 9220 Aalborg, Denmark (e-mail:
[email protected], phone no. +81-3-6373-3738). c s
 Zaa Zab Zac   Zaa Zab Zac 
A. Ametani is with Doshisha University, Kyoto, Japan (e-mail:
[email protected]).
Zc   Zab Zbb Zab , Zs    Zab Zbb Zab
C. L. Bak is with the Institute of Energy Technology, Aalborg University,  Zac Zab Zcc   Zac Zab Zcc 
9220 Aalborg, Denmark (e-mail: [email protected]). m
 Zaa Zab Zac 
Paper submitted to the International Conference on Power Systems Zm   Zab Zbb Zab
Transients (IPST2011) in Delft, the Netherlands June 14-17, 2011
 Zac Zab Zcc 
 Z 
Rg Rg1 Rgn  Rg

(a) Cross-bonded cable system with m-major sections

Z 
Rg Rg

(b) Equivalent 4 conductor system

Fig. 1 Cross-bonded cable and its equivalent model.

where c: core，s: sheath，m: mutual coupling between 3) Zero sequence current

core and sheath, t: transpose The following equations are derived from Fig. 2. Here, the
In (1), cable phase a is assumed to be laid symmetrical to sheath grounding at normal joints are ignored.
phase c against phase b. The flat configuration and the trefoil V1   Z I1  (4)
configuration, which are typically adopted, satisfy this where V   E E E Vs  , I   Ia Ib Ic Is 
t t (5)
1 1
assumption.
2) 4  4 reduced impedance matrix [10][11]
Fig. 2(a) shows zero-sequence current measurement circuit
The lengths of minor sections can have imbalances due to
for a cross-bonded cable.
the constraint on the location of joints. The imbalances are
designed to be as small as possible since they increases sheath
currents and raises sheath voltages. When a cable system has
multiple major sections, the overall balance is considered to
minimize sheath currents. As a result, when a cable system has
more than a couple of major sections, sheath currents are
generally balanced among 3 conductors, which allows us to
reduce 3 metallic sheaths to one conductor.
Reducing the sheath conductors, the 6 conductor system is Z 
reduced to the 4 conductor system composed of 3 cores and 1
equivalent metallic sheath as shown in Fig. 1. The 4  4 (a) Zero-sequence current
reduced impedance matrix can be expressed as
 Zaa Zab Zac Zsa 
 Zab Zbb Zab Zsb 
Z    (2)
Zac Zab Zaa Zsa 
 
 Zsa Zsb Zsa Zss 
Here, Z (4, j )  Z ( j ,4) can be calculated from the 6  6
impedance matrix Z as
1
6 Z 
Z (4, j) 
3  Z (i, j); j  1 4 (3)
i 4 (b) Positive-sequence current
Fig. 2 Setup for measuring sequence currents for a cross-bonded cable.
 Assuming grounding resistance at substations Rg, the 1
 Ia   Z11 Z12 Z13   E 
sheath voltage Vs can be found by      2 
  Ib    Z12 Z 22 Z12   E 
(6)  Ic   Z13 Z12 Z11   E 
Vs  2 RgIs    
Since Zsa = Zsc stands in the flat configuration and the  Z11Z 22  Z122 2
Z12 ( Z13  Z11) Z12  Z13Z 22  E 
trefoil configuration, the following equations can be obtained 1  
  Z12 ( Z13  Z11) 2
Z11  Z13 2
Z12 ( Z13  Z11)  2 E  (13)
by solving (4) to (6).  2 2  
 Z12  Z13Z 22 Z12 ( Z13  Z11) Z11Z 22  Z12  E 
Ia  Ic  ( Z 22  Z12 ) E / 0 (7)
Here,
Ib  ( Z11  Z 21) E / 0
Z11  Zaa  Zsa 2 / Zss
where
 0  Z11Z 22  Z12 Z 21 Z 22  Zbb  Zsb 2 / Zss
Z11  Zaa  Zac  2 Zsa 2 / Z ss Z12  Zab  Zsa Zsb / Zss

Z 22  Zbb  Zsb 2 / Z ss Z13  Zab  Zsa Zsc / Zss

Positive sequence current is derived from (13).
Z12  Zab  Zsa Zsb / Z ss, Z 21  2 Z12
1
Z ss  Zss  2 Rg I1  ( Ia   Ib   2 Ic )
3
Zero-sequence current can be found from (7) by the (14)
following equation. 
E
( Z11  Z13 )( Z11  Z13  2 Z12 )
3 2
I 0  ( 2 Ia  Ib ) / 3
(8) 2
 Z 22 ( 2 Z11  Z13 )  3 Z12 }
 
E
 ( Z11  2 Z 22  2 Z12  Z 21 ) 2
3 0 where  2  ( Z11  Z13 ) Z 22 ( Z11  Z13 )  2 Z12
When three phase cables are laid symmetrical to each other, When three phase cables are laid symmetrical to each other,
the following equations are satisfied. (14) can be simplified using (9).
c c s s
Z aa  Z bb  Zc , Z aa  Z bb  Zs I1 
E (15)
(9) Zc  Zm
c c
Z ab  Z ac  Zm , Zsa  Zsb  Zn
Using symmetrical impedances Zc, Zm, and Zn in (9), Z11, B. Normal-bonded cable
Z22, and Z12 can be expressed as
1) 6  6 impedance matrix
Z11  Zc  Zm  2 Zn 2 / Z ss Fig. 3 shows sequence current measurement circuit for a
Z 22  Zc  Zn 2 / Z ss (10) normal-bonded cable. The following equations are given from
the 6  6 impedance matrix in (1) and Fig. 3.
Z12  Zm  Zn 2 / Z ss
E   ZcI   ZmIs  (16)
Substituting Z11, Z22, and Z12 in (7) and (8) by symmetrical
impedances, Vs  ZmI   ZsIs  2RgIs (17)
Ia  Ib  Ic  E / 1 , Is   3 ZnE / Z ss 1 (11) Here, I   Ia Ib Ia t : core current
I 0  E / 1 Is   Isa Isb Isa t : sheath current
where 1  Zc  2 Zm  3Zn / Z ss2
1 1 1
Rg   Rg 1 1 1
4) Positive sequence current 1 1 1
In Fig. 2(b), the equation Isa + Isb + Isc = 0 is satisfied at
From (17), sheath current Is  is found by
the end of the cable line. The following equations are obtained
since Vs = 0. Is  Zs  2Rg1ZmI  (18)

V1   E  Is  I 
t
 2 E E 0 Eliminating sheath current in (16), core current
(12)
can be derived as
I1   Ia Ib Ic Is t
I   Zc  ZmZs  2Rg1Zm E
1
where   exp  j 2 / 3 (19)
Solving (12) for Ia, Ib, and Ic yields (13).
 E   Z11 Z12 Z13  Ia 
 2    
  E    Z12 Z 22 Z12  Ib 
 E   Z Z11  Ic 
   13 Z12
 3) Positive-sequence current
From Fig. 3(b), E and I  are expressed as

E  E  I   Ia
Ib Iat
t
 2 E E , (26)
Core current I  is obtained from (19) and (26). Once core
current is found, positive sequence current can be calculated
Z  as I1  ( Ia  Ib   2 Ic) / 3 .
The theoretical formula of positive sequence current can
(a) Zero-sequence current also be simplified using (21).
I1 
1
3Zc  Zs 
  
E  Vs     2 E  Vs   2 E  Vs  

E (27)
Zc  Zs
Equation (27) shows that positive sequence current can be
approximated by coaxial mode current. It also shows that,
similarly to a cross-bonded cable, positive sequence current is
not affected by substation grounding resistance Rg.
Z 
III. COMPARISON WITH EMTP SIMULATIONS
(b) Positive-sequence current A comparison with EMTP simulations are conducted in
Fig. 3 Setup for measuring sequence currents for a normal-bonded cable.
order to verify the accuracy of theoretical formulas derived in
the previous chapter.
2) Zero-sequence current
Fig. 4 shows physical and electrical data of the 400 kV
From Fig. 3(a), E and I  are expressed as
cable used for the comparison. An existence of semi-
E  EE E t , I   Ia Ib Iat (20) conducting layers introduces an error in the charging capacity
Core current I  is obtained from (19) and (20), and then of the cable. Although deriving admittance is not a focus of
zero sequence current is calculated as I 0  Ia  Ib  Ic  / 3 . this paper,, relative permittivity of the insulation (XLPE) is
converted from 2.4 to 2.729 according to (28) in order to
Since the relationship Zm  Zs  stands for a cable with a
correct the error and have a reasonable cable model.
general circular shape, (16) and (17) can be simplified to (21)
using (9) when three phase cables are laid symmetrical to each
other and resistance of the cables are small.
E   Vs   Zc   Zs I  (21)
R4 R5
 Zc  Zs U I  R3
where U  : 3 × 3 unit (identity) matrix R2
Hence,
I 0  Ia  Ib  Ic Core
(22)
1
 E  Vs  Insulation
Metallic sheath
Zc  Zs
Using (22), core current I  in (17) can be eliminated, Outer cover
which yields (23).
R2 = 3.26cm, R3 = 6.14cm, R4 = 6.26cm, R5 = 6.73cm
Vs   1 Zm E   Vs   Zm Is  (23) Core inner radius: 0.0cm, Core resistivity: 1.724×10-8Ωm,
Zc  Zs
Metallic sheath resistivity: 2.840×10-8Ωm,
Adding all three rows in (23),
Relative permittivity (XLPE, PE): 2.4
Zs  2 Zm
3Vs  3 E  Vs   Zs  2 Zm Vs (24) Fig. 4 Physical and electrical data of the cable.
Zc  Zs 2 Rg
Solving (24) for Vs and eliminating Vs from (22), zero ln(R3/R2) ln(61.40/32.60) (28)
εr ' εr   2.4  2.729
sequence current is found as ln(Rso/Rsi) ln(59.50 / 34.10)
6 Rg  Zs  2 Zm (25) where Rsi: inner radius of the insulation, Rso: outer radius
I0  E
6 Rg Zc  2 Zm   Zc  Zs Zs  2 Zm  of the insulation
 From the results in Table I, it is confirmed that the
Fig. 5 shows the layout of the cables. It is assumed that the proposed formulas have satisfactory accuracy for planning
cables are directly buried at the depth of 1.3 m with the and implementation studies, compared to the results of EMTP
separation of 0.5 m between phases. simulations. An error of 7 % is observed in zero sequence
current of a cross-bonded cable. It is caused by the impedance
matrix reduction [10][11] discussed in Section II. Due to the
1.3 m matrix reduction, unbalanced sheath currents that flow into
earth at normal joints cannot be considered in proposed
formulas. Thus, the error can be increased to 14 % when the
grounding resistance at normal joints is decreased to 5 ohm. In
contrast, the error becomes negligible when the cables are laid
in a trefoil formation.
0.5 m 0.5 m
Table I shows that positive sequence impedance is smaller
Fig. 5 Layout of the cable.
for a normal-bonded cable than for a cross-bonded cable as
positive sequence current is larger for a normal-bonded cable.
The lengths of a minor section and a major section are
This is because return current flows only through metallic
respectively set to 400 m and 1200 m. The total length of the
sheath of the same cable and earth in a normal-bonded cable
cable is set as 12 km with 10 major sections.
whereas return current flows through metallic sheath of all
Table I shows zero and positive sequence currents derived
three phase cables in a cross-bonded cable (Zc – Zm > Zc –
by proposed formulas and EMTP simulations. The 6  6
Zs).
impedance matrix Z is found by CABLE CONSTANTS in
Phase angle of zero sequence current in Table I
both cases [12-14]. Calculation process in case of a cross-
demonstrates that zero sequence current is highly affected by
bonded cable using proposed formulas is shown in Appendix.
grounding resistance at substations in both cross-bonded and
In the calculations, the applied voltage is set to E = 1 kV
normal-bonded cables. As a result, there is little difference in
/ 3 (angle: 0 degree) and the source impedance is not zero sequence impedance between a cross-bonded cable and a
considered. In this paper, sequence currents are derived in normal-bonded cable. The result shows an importance of
accordance with the setups for measuring sequence currents in obtaining accurate grounding resistance at substations to
Fig. 2 and Fig. 3. The assumptions on the applied voltage and derive accurate zero sequence impedances of cable systems.
the source impedance match a condition in actual setups for
measuring sequence currents since testing sets are generally
used in the measurements. Grounding resistances at IV. CONCLUSION
substations and normal joints are set to 1Ω and 10Ω, This paper has derived theoretical formulas of sequence
respectively. currents of a cross-bonded cable and a normal-bonded cable.
These formulas are obtained by solving equations which are
TABLE I
COMPARISON OF PROPOSED FORMULAS WITH EMTP SIMULATIONS
derived from the setups for measuring sequence currents of
(a) Cross-bonded cable
cross-bonded and normal-bonded cables. For a cross-bonded
Zero Sequence Positive Sequence cable, the equations are solved utilizing the known impedance
Amplitude Angle Amplitude Angle matrix reduction technique. The derived formulas consider the
[A] [deg] [A] [deg] cable as a cable system; they can thus consider sheath bonding
EMTP and sheath grounding resistance.
133.8 -21.42 356.4 -86.35
Simulation In Section III, an accuracy of proposed formulas is verified
Proposed through a comparison with EMTP simulation results. The
formulas, eq. 124.1 -21.23 356.7 -86.79 verified accuracy of the proposed formulas shows sequence
(8)/(14) impedance / current can now be obtained before the
(b) Normal-bonded cable
installation without making measurements for a majority of
Zero Sequence Positive Sequence
cables. This gives an important advantage in setting up
Amplitude Angle Amplitude Angle
transient overvoltage studies as well as planning studies.
[A] [deg] [A] [deg]
EMTP
121.6 -21.80 694.9 -50.40
Simulation
Proposed
formulas, eq. 124.8 -22.50 722.7 -49.08
(19), (20)/(26)
 [5] IEC/TR 60909-2 ed. 2.0, “Short-circuit currents in three-phase a.c.
V. APPENDIX systems - Part 2: Data of electrical equipment for short-circuit current
calculations,” (2008)
First, the 4  4 reduced impedance matrix is found from the [6] Central Station Engineers, Electrical Transmission and Distribution
6  6 impedance matrix Z by (2) and (3). Reference Book, 4th Edition, Westinghouse Electric Corporation, 1964.
[7] J. Lewis Blackburn, Symmetrical Components for Power Systems
Engineering, CRC Press, 1993.
[Z’] (upper: R, lower: X, unit: Ω) [8] Jesus Vargas, Armando Guzman, Jorge
0.71646353 0.59136589 0.59136589 0.5913705 Robles, ”Underground/submarine cable protection using a negative-
8.44986724 6.28523815 5.76261762 6.63566685 sequence directional comparison scheme,” 26th Annual Western
Protective Relay Conference, Spokane, Washington, October 25-28,
0.59136589 0.71646353 0.59136589 0.5913705 1999
6.28523815 8.44986724 6.28523815 6.80987369 [9] CIGRE WG B1.19, “General Guidelines for the Integration of a New
Underground Cable System in the Network,” CIGRE Technical
0.59136589 0.59136589 0.71646353 0.5913705 Brochure 250, 2004.
5.76261762 6.28523815 8.44986724 6.63566685 [10] N. Nagaoka and A. Ametani, “Transient Calculations on Crossbonded
Cables,” IEEE Trans. on Power Apparatus and Systems, vol. PAS-102,
0.5913705 0.5913705 0.5913705 0.83438185 no. 4, 1983.
6.63566685 6.80987369 6.63566685 6.63485268 [11] A. Ametani, Y. Miyamoto, and N. Nagaoka, “An Investigation of a
Wave Propagation Characteristic on a Crossbonded Cable,” IEEJ Trans.
Zero Sequence Current PE, vol. 123, no. 3, pp. 395-401, 2003 (in Japanese).
[12] A. Ametani, “A GENERAL FORMULATION OF IMPEDANCE AND
 0  -3.9605900 + j12.489448 ADMITTANCE OF CABLES,” IEEE Transactions on Power Apparatus
Z11  4.0641578 + j 2.2224336 and Systems, vol. PAS-99, no. 3 May/June 1980.
[13] A. Ametani, “On the Impedance and the Admittance in the EMTP Cable
Z 22  2.1959039 + j 2.1450759 Constants / Parameters,” European EMTP-ATP Users Group Meeting,
Delft, the Netherlands, October 26-28, 2009.
Z12  2.0193420 + j 0.1372637 [14] W. Scott-Meyer, ATP Rule Book, Can / Am EMTP User Group, 1982.
Z 21  4.0386840 + j 0.2745275
I 0 (rms)  81.814700 - j31.778479

Positive Sequence Current

 2  -4.8574998 - j1.8394591
Z11  0.3670612 + j1.8221556
Z 22  0.3801952 + j 1.4707768
Z12  0.2482475 - j0.5159115
Z13  0.2419636 - j0.8650940
I1 (rms)  14.118637 - j251.86277

VI. ACKNOWLEDGMENT
The Authors acknowledge the technical support from
Energinet.dk, the Danish TSO.

VII. REFERENCES
[1] Energinet.dk, Technical report on the future expansion and
undergrounding of the electricity transmission grid, 2008, Available:
http://www.energinet.dk/NR/rdonlyres/CC966C3A-FE78-41D8-9DC7-
6B455210B502/0/TechnicalReportSummary.pdf.
[2] L. Colla, S. Lauria, and F. M. Gatta, “Temporary Overvoltages due to
Harmonic Resonance in Long EHV Cables,” IPST 2007, Available:
http://www.ipst.org/techpapers/2007/ipst_2007/papers_IPST2007/Sessio
n16/233. pdf.
[3] M. Rebolini, L. Colla, and F. Iliceto, “400 kV AC new submarine cable
links between Sicily and the Italian mainland. Outline of project and
special electrical studies,” CIGRE Session 2008, C4-116 (2008).
[4] A. Ametani, Distributed-Parameter Circuit Theory, Corona Pub. Co.,
1990 (in Japanese).