Simplified and Rapid Prediction of Earthquake-Induced Track Dynamic Irregularity of High-Speed Railway Bridges under Different Site Conditions

Rail transportation, as a critical component of social development, plays a pivotal role during earthquake relief efforts. To ensure the safety and comfort of post-earthquake travel, it is essential to accurately predict the earthquake-induced dynamic irregularities of the track. In this study, 120 seismic waves are chosen from the PEER strong earthquake database based on various site classifications. A three-dimensional coupled vibration model of the train-track bridge, considering seismic damage, is developed. This research also involves the creation of a database for earthquake-induced track dynamic irregularities, and these irregularities are then simplified using variational mode decomposition. The Kolmogorov-Smirnov method is used to test the distribution of these simplified samples. Furthermore, a calculation method based on the probability guarantee rate is introduced for the earthquake-induced simplified dynamic irregularities. A sine function is employed to fit the dynamic irregularity curve. The study also examines the impact of site classification on earthquake-induced irregularities and provides a parameter table for the simplified fitting curve of earthquake-induced irregularities, using a 95% guarantee rate for various site types. This table aids in the rapid prediction of earthquake-induced track dynamic irregularities. The findings indicate that earthquake-induced dynamic irregularities are primarily characterized by a single peak component. This single peak component can serve as a simplified dynamic irregularity. This single peak component also exhibits pronounced sinusoidal characteristics suitable for sinusoidal function fitting. The peak value of the simplified fitting curve for dynamic irregularity rises with the increasing characteristic period of ground motion. Notably, the peak value for the class IV site's simplified fitting curve is significantly greater than that for the class I site.