Subscribe Now Subscribe Today
Research Article
 

Impact of Lightning Surge on Surge Arrester Placement in High Voltage Substation



M.Z.A. Ab Kadir and A.M. Azmi
 
Facebook Twitter Digg Reddit Linkedin StumbleUpon E-mail
ABSTRACT

This study generalizes the modeling details to be used in modeling the high voltage substation and performing the analysis on the impact of lightning surge on surge arrester placement in high voltage substation using the PSCAD/EMTDC software. Modeling parameters and the substation layout design are based and adapted from 132 kV substation in Johor Baharu, Malaysia, courtesy of the Tenaga Nasional Berhad (TNB). The model is based on single phase line model as it was suggested by the IEEE to be adequate to represent the substation in transient analysis simulation. The outcome of this paper would be the results of prediction of the breakdown current and effect of surge arrester placement in terms of voltage level measured at particular points in substation.

Services
Related Articles in ASCI
Similar Articles in this Journal
Search in Google Scholar
View Citation
Report Citation

 
  How to cite this article:

M.Z.A. Ab Kadir and A.M. Azmi, 2008. Impact of Lightning Surge on Surge Arrester Placement in High Voltage Substation. Journal of Applied Sciences, 8: 3298-3301.

DOI: 10.3923/jas.2008.3298.3301

URL: https://scialert.net/abstract/?doi=jas.2008.3298.3301
 

INTRODUCTION

Lightning interference occurs mainly on overhead lines and has been a problem since the earliest days of the electricity supply industry. Overvoltages which occur on the lines, travel toward the terminal or substation and can cause damage, particularly to expensive equipment such as transformers. In view of their importance, cost and the difficulty of making internal repairs, the protection of large transformers against lightning overvoltages is usually given special consideration. Lightning activity in South East Asia, especially in Malaysia, ranks as one of the highest in the world. Tenaga Nasional Berhad Research (TNBR) Malaysia has recorded as high as 320 kA lightning impulse current in Malaysia using their lightning detection network system (LDNS). Every year, million dollars worth of damage is caused by the devastating effects of lightning including to electrical power systems. The transmission line trip in Malaysia is majorly caused by lightning, which is about 70%.

Therefore, thorough knowledge on insulation coordination studies is urgently needed strategic planning and protection of the expensive assets especially in the substation section. Lightning overvoltages are fast front overvoltages with times to crest from 0.1-20 msec. For substations, shield failures, backflash and induced overvoltages generate surge voltages that impinge on the substation equipment.

Lightning induced voltages are generally below 400 kV and are important only for lower voltage systems. The incoming surges caused by the backflash are more severe than that caused by shielding failures. As these surges travel from the stroke terminating point to the station, corona decreases front steepness and the crest magnitude. The shield wire has significant impact on the wave propagation. A shield wire grounded at each tower makes the propagation velocity of the ground mode wave component very close to the conductor mode component. The magnitude of the surges caused by a backflash ranges from 70 to 120% of the positive polarity critical flashover voltage (CFO) of the line insulation. The front steepness is a function of the conductor size, the distance between the location of the backflash and the station (IEEE Power Engineering Society, 1999).

The objectives of this study are to model the high voltage substation and perform the analysis on prediction of the level of current that causes the transformer to breakdown and determine the effect of surge arrester placement at the substation. This will be done by comparing the voltage level measured close to the transformer with the suggested basic insulation level (BIL) value used by the utility.

MODELING OF THE SYSTEM

The main emphasis of this research is to model a high voltage substation for the lightning surge analysis. This modeling must include the tower, power line, tower footing resistance, lightning, substation equipments and insulation coordination.

Image for - Impact of Lightning Surge on Surge Arrester Placement in High Voltage Substation
Fig. 1: Substation model for case studies

Table 1: Key parameters used for modeling the system
Image for - Impact of Lightning Surge on Surge Arrester Placement in High Voltage Substation

Table 2: Comparison of capacitors value between TNB calculation and IEEE recommendations
Image for - Impact of Lightning Surge on Surge Arrester Placement in High Voltage Substation

Figure 1 shows the modeling arrangements at the substation, which is adopted for the case studies and based on real configurations of the TNB`s 132 kV substation. The lightning strike is placed at the tower close to the substation. The distance between the tower and the substation entrance is 50 m. Point E1 measures the entrance voltage induced by the lightning and point E2 is the point-of-connection (POC) of the surge arrester, where the voltage is expected to be clamped before passing through the capacitive voltage transformer, labeled as CCVT. Whilst points E3 and E4 are the second surge arrester, SA2 and the power transformer, labeled as CTX, respectively. Further of specific details relating to the model are described in Table 1.

Table 2 describes the comparison of capacitor value between TNB calculation approach and IEEE recommendation base on 115 kV US substation system model. For this study, TNB calculation approach of capacitor values was adapted to model the system as it more or less agreed with the value recommended by IEEE WG 3.4.11 (1992) and for the actual analysis. The distance between each substation equipments are as below:

Tsub1=3.0 mTsub5=4.0 mTsub9=14.5 m
Tsub2=3.5 mTsub6=4.5 mTsub10=3.0 m
Tsub3=3.5 mTsub7=3.0 mTsub11=3.0 m
Tsub4=3.0 mTsub8=3.0 mTsub12=5.0 m

SURGE ARRESTER MODELING

Several models of arrester had been described elsewhere in literature (IEC, 1993; Martinez and Castro-Aranda, 2004; IEEE WG 3.4.11, 1992). Most of the arrester model must include two nonlinear resistances A0 and A1 as shown in Fig. 2, with other combination of the components. However for different approach, it is basically using different type of lumped parameter arrangement. The frequency-dependent surge arrester model which was recommended by IEEE WG 3.4.11 (1992). is used in this work. This model is shown in Fig. 2 and it was reported as the most accurate representation based on single phase line model (Goudarzi and Mohseni, 2004). Adjustment procedure of parameters is described by IEEE WG 3.4.11 (1992).

Image for - Impact of Lightning Surge on Surge Arrester Placement in High Voltage Substation
Fig. 2: IEEE frequency-dependent model

RESULTS AND DISCUSSION

Surge arrester breakdown current: There are two cases are considered under the critical conditions; when the surge arrester 1 (SA1) is not operated and also when both surge arresters (SA1 and SA2) are not operated. The idea is to demonstrate the effect of floating surge arrester (missing of copper conductor connected between surge arrester and substation earthing) due to the vandalism cases as reported by the utility company in recent years.

Table 3 shows the data for the case where no SA1 is installed. As the currents increase, the voltage level also increases. The BIL used by TNB Malaysia for the 132 kV rated transformer is 550 kV. Therefore, the probability of the capacitive voltage transformer, CCVT damage can be estimated when the lightning current reaches 144 kA.

Table 4 shows the data for the case when both surge arresters are not operated. This is the worst case scenario that could possibly happen involving the case of vandalism on the surge arresters. For the case of capacitive voltage transformer, CCVT, it is estimated that the probability of the damage is at the current of 33 kA, whilst for the case of power transformer, CTX, current of 31 kA can already cause the breakdown on the equipment.

Effect of surge arrester placement: Table 5 demonstrates the effect of surge arrester placement at the substation. For the first case, SA2 is placed 8 m before power transformer, CTX, instead of the real placement which is just 5 m. For the distance of 8 m, Table 5a shows that the voltages level at point E4 is slightly higher compared to the result in Table 5c for the original placement. Whilst for the SA2 placed at 11 m away from Tx, voltages level at point E4 are also increased, as shown by the data in Table 5b. Having the differences for only few kilovolts, the results perhaps very difficult to be judged. However, this is very good analysis in determining a proper insulation coordination studies.

Table 3: Case of no SA1 is installed
Image for - Impact of Lightning Surge on Surge Arrester Placement in High Voltage Substation

Table 4: Case of both surge arresters are not operated
Image for - Impact of Lightning Surge on Surge Arrester Placement in High Voltage Substation

Table 5: Effect of surge arrester placement
Image for - Impact of Lightning Surge on Surge Arrester Placement in High Voltage Substation

In this case, having the surge arrester located at the proper location is very crucial and without having all the related knowledge, it is very difficult in making a decision.

CONCLUSION

Detail modeling guidelines and parameters for high substation are successfully presented. Results for the first part have clearly shown that the impact of lightning surge can be very dangerous even at low value of current if there is no surge arresters are in operating or used for protection. Overall results have demonstrated the importance of having a right location of surge arrester placement as it is crucially needed in order to optimize the substation performance in term of its reliability and cost effective. In other words, this surge arrester must be placed as close as possible to the equipment to be protected, as fail to do so will cause a significant damage to the equipment.

ACKNOWLEDGMENTS

The authors would like to express their sincere gratitude to the Engineering Department (Transmission and Substation) of the Tenaga Nasional Berhad for their cooperation and kind supply of various technical data.

REFERENCES
1:  IEEE Power Engineering Society, 1999. IEEE Std. 1313.2-1999: Guide for the application of the insulation coordination. ISBN: 0-7381-1761-7. URL: http://ieeexplore.ieee.org/servlet/opac?punumber=6547

2:  CIGRE., 1991. Guide to procedures for estimating the lightning performance of transmission lines. CIGRE Brochure, pp: 63. http://www.e-cigre.org/Order/select.asp?ID=67.

3:  Woodford, D., 1998. PSCAD/EMTDC: Getting Started Manual. Manitoba HVDC Research Centre Inc., Canada.

4:  IEEE WG 3.4.11, 1992. Modeling of metal oxide surge arresters. IEEE Trans. Power Delivery, 7: 302-309.
CrossRef  |  

5:  IEC, 1993. Insulation coordination Part 1: Definition, principle and rules. IEC 71-1: International Standard. ICS Codes: 29.080.30. URL: http://webstore.iec.ch/webstore/webstore.nsf/artnum/035460

6:  Martinez, L.A. and F. Castro-Aranda, 2004. Modeling overhead transmission lines for line arrester studies. Proceedings of the Power Engineering Society General Meeting, June 6-10, 2004, Colorado, USA., pp: 1125-1130.

7:  Goudarzi, A and H. Mohseni, 2004. Evaluation of mathematical models of metal oxide surge arrester for energy absorption study. Proceedings of the 39th International Universities Power Engineering Conference, September 6-8, 2004, Bristol, UK., pp: 211-214.

©  2021 Science Alert. All Rights Reserved