Substation Design Improvement Considering the Actual Accident On-site caused by Direct multiple lightning

Analyzing and investigating on-site incidents is important for developing countermeasures against direct lightning. Additionally, understanding the mechanisms behind lightning surge overvoltage caused by direct and multiple strikes is essential. Furthermore, the insulation between the poles of a gas circuit breaker can be weakened by the hot gas generated during the initial interruption [2]. Given the equipment cost, enlarging circuit breaker dimensions to increase dielectric strength is impractical; therefore, reviewing the substation design is a more effective solution.

Case study of an accident

This work presents a dielectric breakdown incident involving a 500kV substation equipment caused by a lightning surge, along with recommended countermeasures. The study focuses on an inter-polar flashover in a 500kV GIS (Gas-Insulated Switchgear) resulting from multiple direct lightning strikes. A breakdown occurred between the line and bus sides of the circuit breaker due to the lightning surge, leading to circuit breaker failure. The significant lightning surge phenomena observed at the substation included:

• Back flashover at a transmission tower caused by a lightning strike on the ground wire or the tower itself
• Direct lightning strikes the power line

The operational status (open or closed) of the substation line switch and circuit breaker at the time of the accident is shown in Figure 1, along with their operational history. The failed gas circuit breaker is Line A-1L. Initially, the first lightning strike (at 0ms) triggered a protection relay to operate for a grounding fault, causing the 500kV gas circuit breaker to open and isolate the ground fault. Later, the gas circuit breaker flashed over at 161ms. Next, a lightning strike was detected near transmission tower No. 11 of Line A at 0ms, with a lightning current of 18kA. Then, a second lightning strike occurred at 161ms, when an interpolar flashover happened in the 500kV Gas circuit breaker. The lightning current during this second strike was 19kA. In conclusion, the accident resulted from multiple direct lightning strikes.

Estimation of accident risk and Occurrence probability

Previous observations of lightning strikes on 500kV transmission lines show that the probability of a direct lightning strike is approximately 1.28 times per 100 km annually [3]. In comparison, the chance of multiple lightning strikes is about 85% [4]. Additionally, the likelihood that the lightning current falls between 19 kA (minimum) and 23 kA (maximum) is around 9.74%, and the arc horn on a 500 kV line would not flash over. 23 kA is empirically chosen because the surge current entering the substation, caused by the arc horn flashover due to a ground fault, is divided. Consequently, the chance of an inter-polar flashover is reduced. The total length of 500 kV transmission lines that connect to AIS circuit breakers is about 921 km in our company. The interval for accident occurrence in our company is estimated to be 10.2 years/occurrence, by the values in Table I.

Data from the Japan Meteorological Agency show changes in the yearly number of lightning strikes in our region, especially in Tokyo and the 10 surrounding prefectures [7]. The average number of lightning days per year from 1943 to 2019 was 187 days, but over the past four years, the average number of lightning strikes has been about 400 days. When considering measures against multiple direct lightning strikes, we believe the risk of accidents caused by various direct strikes should be assessed based on recent weather conditions. Since the accident happened 13 years later, in 2022, this estimate still seems quite reasonable.

Countermeasures against direct multiple lightning strikes

Our measures to prevent direct multiple lightning strikes on the 500kV power system are explained. The proposed measures include enhancing the lightning impulse withstand voltage for the 500kV circuit breaker, such as reducing the arc horn distance at the transmission line and updating the surge arrester (SA) layout. In most 500kV air-insulated substations (AIS), five components are installed in sequence from the transmission side, including the SA, potential device, line switch, station post insulator, and dead-tank breaker (DTB).

This work recommends installing an additional SA with a residual voltage of 870kV between the station post and the breaker. The electrical distance is less than 10 meters, as shown in Figure 3. All line DTBs (TEPCO PG: 98 Line) have been reviewed to assess whether a countermeasure is necessary. However, when the line DTB connects to a transmission line with more than approximately 50 meters between the DTB and the SA, there may be a risk of an accident without implementing countermeasures. Conversely, GIS systems with GIB connections are considered sufficiently protected under current designs.

A lightning surge simulation was performed to verify the effectiveness of our countermeasure, where the breaker tripped after the first lightning strike and then a second strike occurred. The results show that the additional SA, with extremely high energy performance, effectively suppressed the interpolar voltage of the DTB caused by multiple direct lightning strikes below critical levels. The analysis was conducted using XTAP [8] developed by the Central Research Institute of Electric Power Industry (CRIEPI).

Conclusion

This paper examines the failure of a 500kV AIS Gas circuit breaker caused by multiple lightning strikes, based on lightning observation and surge analysis results. It was found that the failure resulted from overvoltage between the circuit breaker poles, which was 27% higher than the JEC standard (2249kV). Additionally, the study indicates that traditional substation designs cannot handle a second lightning strike with a steep wavefront because the lightning arrester cannot protect the circuit breaker within the front time.

The probability of the accident is roughly estimated to balance the economic efficiency of the countermeasures, and their effectiveness is also confirmed by lightning surge analysis. When an additional lightning arrester is installed, the overvoltage caused by a direct lightning strike with a wavefront time of 0.2µs was reduced to 1806kV below the JEC standard withstand voltage.

Since 2022, after implementing these measures, no similar incidents have been reported despite lightning activity from 2023 to 2025. These on-site experiences confirm our countermeasure's effectiveness, leading to the decision to expand its deployment across the TEPCO grid.

Acknowledgments

I would like to thank S. Tsukao and H. Kagawa for their helpful discussions and for sharing their expertise on substation design improvements, particularly regarding direct multiple lightning.

About the Author

Keisuke MURAKITA received an MS degree in Electrical Engineering from Nagoya University, Japan, in 2017. He has been employed at TEPCO since 2017. He is currently in charge of switchgear engineering in the Substation Engineering Group, Transmission Department, and has been engaged in the development and maintenance of substation equipment.

Bibliography

1. Subcommittee for Power Stations and Substations, Study Committee for Lightning Risks. “Guide to Lightning Protection Design of Power Stations, Substations, and Underground Transmission Lines (rev. 2021)” (2022).
2. Iwao Miyachi, Tetsuo Uemura, Takashi Matsuhisa, “Countermeasures and Test Method of Circuit Braker on Multi-Lightning Strokes”,  T. IEE Japan, Vol. 112-B, No. 7, pages 563-568, (1992).
3. S. Taniguchi, T. Tsuboi, and S. Okabe, "Observation Results of Lightning Shielding for Large-Scale Transmission Lines," IEEE Trans. Dielectr. Electr. Insul., vol. 16, pp. 552-559 (2009).
4. CIGRE WG C4. 407, “Lightning parameters for engineering applications”, CIGRE TB 549 (2013).
5. Sectional Committee for Transmission Lines, Lightning Protection Design Study Committee, “Lightning Proof Design Guide-Book for Transmission Lines”, Electric Power Engineering Research Laboratory Report No. 175031, (2007).
6. Working Group on Lightning Protection for Transmission Systems, “Lightning Observations on Japan Sea Coast in Winter,” Komae Research Laboratory Report No. T10, (1989).
7. Japan Meteorological Agency [online]
8. Taku Noda, Toru Miki, Naoki Gibo and Kiyoshi Takenaka, “Development of an Electromagnetic Transient Analysis Program for Power Systems (Part I) -Fandamental Design-”, Electric Power Engineering Research Laboratory Report No. H06002, (2007).