Simulation of Lightning Transients on 110 kv overhead-cable transmission line using ATP-EMTP Kresimir Fekete 1, Srete Nikolovski 2, Goran Knezević 3, Marinko Stojkov 4, Zoran Kovač 5 # Power System Department, Faculty of Electrical Engineering Osijek K. Trpimira 2B, 31000 Osijek, Croatia 1 kresimir.fekete@etfosr.hr 2 srete.nikolovski@etfos.hr * Faculty of Mechanical Engineering Slavonski Brod I. B. Mažuranić 2, 35000 Sl. Brod, Croatia 4 marinko.stojkov@gmail.com Croatian Transmission System Operator K. Franje Shepera 1A, 31000 Osijek, Croatia 5 zoran.kovac@hep.hr Abstract Due to construction of a highway in the eastern part of Croatia, a new 110/x kv substation was built with two parallel connecting underground cables. Cables were connected to already existing 110 kv overhead lines. At the place where the transition between overhead lines and cables is made, surge arresters were installed. In this paper lightning stroke at the grounding wire on the overhead line and its impact on underground cables were studied. Transient program Electromagnetic Transients Program (ATP-EMTP) is used to create a model of the system and to perform simulation of the transient process during lightning stroke. The results of the simulation are briefly presented and discussed in the paper. I. INTRODUCTION In the process of designing new facilities in the power system (substations, cables, overhead lines etc.), lightning overvoltages are important from the viewpoint of the insulation and surge arrester coordination. Due to construction of the highway in the eastern part of Croatia, a new 110/x kv substation was built with two parallel connecting underground cables as well. Cables were connected to already existing 110 kv overhead lines. A more detail presentation of the system is presented in Chapter II. At the place where the transition between overhead lines and cables is made, surge arresters were installed. In order to make surge arrester coordination it is necessary to investigate the impact of lightning overvoltages. It is very hard to observe the lightning overvoltages experimentally, and thus a numerical simulation is used to investigate it. The EMTP has been widely used for the time domain transient solution. It was first developed at Bonneville Power Administration (B.P.A.) from Dommel s basic work [1]. Nearly all system components can be represented by builtin elements in ATP-EMTP like overhead lines with line and ground wires and towers as well as underground cables [2]. In this paper ATP-EMTP is used to create a model of the power system and to simulate lightning stroke at the grounding wire on the overhead line and its impact on underground cables and surge arresters. The structure of this paper is as follows: first, a brief explanation, regarding the part of the transmission system where the new substation and cables are built, is given. The main features of the ATP-EMTP model and method are explained in Chapter III. In Chapter IV, simulation and results are presented. Based on the results, the conclusion is given in the last chapter. II. DESCRIPTION OF THE POWER SYSTEM The part of the Eastern Croatian transmission system that is studied in this paper is shown in Fig.1. The new substation TS 110/20 kv Djakovo 3 is supplied from two substations TS 220/110 kv Djakovo and TS 400/110 kv Ernestinovo. The new substation is modelled only with passive load. At the place where overhead line and cable are connected, ABB Pexlim Q surge arresters are installed. Fig. 1. Single line diagram representing the part of the transmission system that is studied Parameters of the 110 kv overhead lines which are made of Al/Fe conductors with cross sections 240/40 mm 2, and underground cables type NEXANS which has a cross section of 1000 mm 2 are shown in Table I. 978-1-4244-5794-6/10/$26.00 2010 IEEE 856
TABLE I PARAMETERS OF THE OVERHEAD LINES AND CABLES Conductor type 1 conductor/phase, 240/40 Al/St Overhead line Series resistance Series inductance (Ω/km) (mh/km) 0.1188 1.0565 Underground cable Conductor type Series resistance (Ω/km) Al 0.0251 0.346 Series inductance (mh/km) A. Tower The height of 110 kv tower used in this paper is 31.9 m. The layout of one typical 110 kv tower is shown in Fig. 3. The distances are given in meters. All necessary data about the power system have been obtained from the Croatian Electric Utility HEP Transmission System Operator. The cable is presented in Fig. 2, and its model in the ATP-EMTP model for cables and overhead lines is presented in Fig 2. Fig. 3. Layout of a typical 110 kv tower The tower is represented by four lossless Constant- Parameter Distributed Line (CPDL) models [5] as illustrated in Fig. 4, where Z t1 tower top to the upper phase = upper phase to middle phase = middle phase to lower Z t4 lower phase to tower bottom. Fig. 2. Dialog box for cable/line model in ATP-EMTP A more detailed explanation about overhead lines, tower construction, and surge arresters is presented in the following chapters. III. ATP-EMTP MODEL AND METHOD The modeling method for the back flashover analysis used in this paper is based upon various publications in this field [3], [4]. In this chapter the model of tower, transmission line, underground cable, surge arrester and lightning current will be explained. CPDL model is characterized by surge impedances Z and travelling time τ. Values of surge impedances are [3]: Z t1 = 220 Ω and Z t4 = 150 Ω. Distance between top of the tower (grounding wire GW) and the tower bottom is h = h 1 and h 2, h 3 and h 4 are distances between ground and phases starting with upper phase respectively. The propagation velocity of a travelling wave along a tower is taken to be equal to the light velocity, c 0 = 300 m/μs [3]. The tower travelling time is given by the following equation: h τ = (1) c0 Because footing impedance is represented by a linear resistance (R f ) it is recommended to take into account frequency-dependent effects for wave propagation along the tower. It is done by adding an RL parallel circuit to each part, 857
as shown in Fig. 4 in order to represent travelling wave attenuation and distortion [3]. Ri =ΔRi xi; Li= 2τ Ri (2) 2 Z t 1 1 Δ R1=Δ R2=Δ R3= ln (3) ( h x4) α1 2Z t 4 1 Δ R4 = ln (4) h α4 where: α 1 = α 4 = 0.89 attenuation along the tower. Values for resistance (R) and inductance (L) in our study are: R 1 = 15.54 Ω, R 2 = 17.66 Ω, R 3 = 15.9 Ω, R 4 = 33.48 Ω, L 1 = 3.285 mh, L 2 = 3.733 mh, L 3 = 3.36 mh and L 4 = 7.076 mh. A tower footing impedance is modelled as a simple linear resistance R f = 10 Ω. Fig. 4. A model circuit of a 110 kv tower The values of resistance (R) and inductance (L) are defined in the following equations [3]: B. Number of Towers Five towers of a part of a line route to substation TS 400/110 kv Ernestinovo and five towers of a part of line route to substation TS 220/110 kv Djakovo are represented including all line circuits. The total number of towers is ten. Direct lightning stroke at the grounding wire to tower #2 is analysed. Fig. 5 represents the model of only one part of the analysed power system. Fig. 5. Part of the ATP-EMTP model of the analysed system 858
C. Arrester In order to protect cable from lightning overvoltages zinc oxide surge arresters ABB Pexlim Q are installed at the place where overhead lines and cables are connected. In the model used in this study nonlinear branch model is used with its input V-I characteristic to represent surge arrester. Protective V-I characteristic [6] of surge arrester is illustrated in Fig. 6. Fig. 7. Lightning stroke model consisting of a current source and lightning path impedance The Heidler s function [8] is used to represent lightning current waveform: where: I ( t / τ ) it () = e η ( / ) + 1 n 0 1 n t τ1 ( t/ τ2 ) (5) Fig. 6. Protective V-I characteristic of a surge arrester D. Transmission Lines and Cables The parameters of transmission lines and cables with ground return are highly dependent on the frequency. Accurate modelling of this frequency dependence over the entire frequency range of the signals is of essential importance for the correct simulation of electromagnetic transients [7]. Models which assume constant parameters (e.g. at 50 Hz) cannot adequately simulate the response of the line over wide range of frequencies that are present during transient condition. In most cases constant-parameter representation produces a magnification of the higher harmonics of the signals and, as a consequence, a general distortion of the wave shapes and exaggerated magnitude peaks [7]. ATP-EMTP offers possibility to use various frequencydependent line models [5]. In this paper J. Marti frequencydependent line model [7] is used to represent overhead transmission lines and cables. The grounding wire is represented like a phase wire, which is connected to the top of the towers. E. Lightning Current and Impedance The lightning stroke is modelled by a current source and a parallel resistance, which represents the lightning path impedance as shown in Fig. 7. η 1/ ( 1/ 2)( 2/ 1) n = (6) e τ τ ητ τ and: I 0 = lightning current peak; τ 1 = time constant determining current rise-time; τ 2 = time constant determining current decay-time; n = current steepness factor. In this paper values for Heidler s function parameters are as follows: I 0 = 100 ka, τ 1 = 1.2 μs, τ 2 = 61.7 μs and n = 7. Fig. 8 shows the lightning current waveform used in this paper. Fig. 8. Lightning current waveform used in this paper The impedance of a lightning path is represented as a parallel resistance to a current source. The resistance value is taken to be 400 Ω, which was derived by Bewley [9]. 859
IV. SIMULATION AND RESULTS The simulation is performed in order to investigate proper work of the surge arrester when a lightning stroke at tower #2 has appeared. As mentioned before, the lightning current peak used in the simulation is 100 ka. The simulation time parameters are: - simulation time T max = 0.001 s, - simulation step T = 1E-8 s. Results of the simulation will be presented in the following. A. Voltage at the Point of the Lightning Stroke Fig. 9 presents overvoltage at the point of lightning stroke i.e. at the grounding wire of the tower #2. The peak value of the overvoltage is 6 MV. B. Voltage and Current of Surge Arrester At the point where surge arresters are installed, voltage and current are computed. Fig. 11 shows voltage at the point where surge arresters are installed. It is important to notice that only results for surge arresters that are installed between cable 1 and overhead transmission line from TS 110/20 kv Djakovo 3 to TS 400/110 kv Ernestinovo are taken into consideration in this paper. The reason for this is the location of the lightning stroke. The influence of the lightning stroke on the second surge arrester that is installed between cable 2 and overhead line to TS 220/110 kv Djakovo is disregarded in this paper. As can be observed from the Fig. 11, the maximum voltage between phases and ground is 100 kv. Fig. 9. Voltage at the point of lightning stroke on grounding wire Lightning surges that are induced on the phase conductors due to back flashover across 110 kv insulator strings are shown in Fig. 10. As it can be observed, the highest overvoltage peak (i.e. 5 MV) is induced at phase A, which is closed to the grounding wire. Induced overvoltages on phases B and C are almost the same (peak value is 2 MV). Fig. 11. Voltage at the point where surge arresters are installed Currents of surge arresters in the moment when it lead the current are shown in Fig. 12. The highest peak value has phase A because overvoltage of phase A is the highest (see Fig. 9). It can be observed that surge arresters worked according to their input V-I characteristics and in that way protecting underground cables from surges, which will be shown in the following subchapter. Fig. 10. Voltage at phase wires of the affected tower Fig. 12. Current of surge arresters 860
C. Voltage at the end of Cable 1 It can be seen in Fig. 13 that during a transient process there are no dangerous overvoltages at the end of cable 1. The maximum value of voltage is 80 kv. Fig. 13. Voltage at the end of the cable 1 It was interesting to see the value of induced voltage on the grounded shield for cables which are grounded at both ends, at the tower where the overhead line enters into the ground and at the grounding network in TS 110 kv Djakovo 3. The cable shield is grounded at four places in the buried track and is connected to copper wire with resistance R=0.0665 ohm and inductance L=0.23 mh. The induced voltage on the cable shield at TS Djakovo 3 is presented in Fig. 14. It is less than 50 V and does not present any security problems. V. CONCLUSION A flashover analysis has been performed for a 110 kv overhead line which is connected to 110 kv underground cables. Between the overhead line and cable, a surge arrester is installed. The cable shield is well grounded at both ends of the cables with a copper wire. The function of the surge arrester and influence of the lightning stroke on a cable are observed. As simulation results indicate if a surge arrestor works properly there will be no dangerous overvoltage affecting the underground cable shield. Further work on this study will be based on different lightning current waveforms, like CIGRE concave waveform [10]. Also, the influence of different stroke locations, tower structures and the cable length, as well as a detailed influence of lightning stroke on the grounding system and cable shield are interesting subjects for further analysis. ACKNOWLEDGMENT Authors wish to acknowledge Laszlo Prikler for his contribution in developing the model of overhead transmission line and tower in EMTP. The paper is supported by the Croatian Transmission System Operator HEP TSO Osijek. REFERENCES [1] H.W. Dommel, Digital Computer Solution of Electromagnetic Transients in Single-and Multiphase Networks, IEEE Trans. Power Apparaturs and Systems, vol. PAS-88, pp. 388 399, April 1969. [2] H. W. Dommel, EMTP Theory Book, Bonneville Power Administration, conversion into electronic format by Canadian/American EMTP Users Group in 1995. [3] A. Ametani, and T. Kawamura, A Method of a Lightning Surge Analysis Recommended in Japan Using EMTP, IEEE Trans. On Power Delivery, vol. 20, No 2, pp. 867 875, April 2005. [4] M. Kizilcay, and C. Neumann, Analysis of Backflashover Across 110- kv Insulator Strings of a Multi-circuit Transmission Tower, in Proc. European EMTP-ATP Meeting 2006, pp. 115-127, 2006. [5] Canadian/American EMTP User Group, ATP Rule Book, Distributed by the European EMTP-ATP Users Group Association, 2005. [6] ABB documents 1HSM 9543 13-01en Edition Protection characteristic of surge arrester PEXLIM-Q2, 2004. [7] J.R. Marti, Accurate Modeling of Frequency-dependent Transmission Lines in Electromagnetic Transient Simulations, IEEE Trans. Power Apparaturs and Systems, vol. PAS-101, No 1, pp. 147 157, January 1982. [8] F. Heidler, J.M. Cveticand B.V. Stanic, Calculation of Lightning Current Parameters, IEEE Trans. On Power Delivery, vol. 14, No 2, pp. 399 597, April 1999. [9] B. V. Bewly, Travelling Waves on Transmission Systems, New York: Dover, 1963. [10] CIGRE WG 33-01, Guide to Procedures for Estimating the Lightning Performance of Transmission Lines, Technical Brochure, October, 1991. Fig. 14. Induced voltage on the grounding shield of cable 861