Vibration characteristics and damping isolation of large lightning arresters under seismic loading: experimental and machine learning analysis

This study focuses on the vibration characteristics and damping isolation technologies of large lightning arresters under seismic loading, which are critical for ensuring substation safety. A full-scale vibration table experiment was conducted on a 500kV lightning arrester to analyze its dynamic response to white noise and artificial seismic waves. Three damping and isolation measures were tested, and their effectiveness was compared. Numerical simulations were performed to validate the experimental results, showing strong agreement. Additionally, a machine learning-based prediction model was developed to assess the seismic response of different types of lightning arresters and evaluate the impact of various features on maximum displacement. The results revealed that the friction damper exhibited the most significant damping effect, effectively reducing acceleration amplitude and stress levels in the arrester. This improvement enhances the seismic performance of lightning arresters, particularly in high-strength soft soil areas. The study demonstrates that the friction damper can ensure the reliable operation of arrester equipment during earthquakes. The integrated approach, combining experimental testing, numerical simulations, and machine learning, provides a comprehensive framework for optimizing seismic protection strategies in substations. This research offers valuable insights for improving the resilience of critical electrical infrastructure in earthquake-prone regions.