Spectral Clustering Analysis of High-speed Train Seismic Events

Peking University deployed a temporary seismic array in Baoding City, Hebei Province, China. Based on a total of 10,461 seismic records induced by 951 high-speed trains recorded by 11 short-period stations under the high-speed railway viaduct, we observe how the spectral characteristics vary with changes of the speed and model of the train as well as the rail and groundsill by using the clustering algorithm K-Means. The spectrum of high-speed rail seismic event is mainly composed of equally spaced peaks and its fundamental frequency is equal to the ratio of the speed to the carriage length. We reduce the influence from the train speed by aligning the fundamental frequency to make the spectrum pattern clear and easy for clustering analysis. Clustering results show that the spectrum contains a wealth of information that can tell different train model, as well as the variation of the rail and groundsill, and has potential to be used in monitoring high-speed rail safety status.