Study on power spectral density curves of track dynamic irregularity caused by earthquake-induced damage

A series of studies on track dynamic irregularity caused by earthquake-induced damage should be conducted in order to guarantee safe operation of high-speed trains. In this study, a nonlinear time history analysis of high-speed railway CRTS II track-bridge system under transverse random earthquakes is carried out, and the post -earthquake stiffness degradation degree of main components of track-bridge system is analyzed. A three-dimensional coupled vibration model of train-track-bridge system based on earthquake-induced damage is established, and a sample library of earthquake-induced track dynamic irregularity is constructed. Kolmogorov-Smirnov method is used to test the normal distribution of earthquake-induced irregularity samples. A calculation method of the earthquake-induced dynamic irregularity power spectral density curves based on probability guarantee rate is proposed. The power spectral density curve is fitted by piecewise polynomial, and the influence of train operation speeds and peak ground accelerations (PGAs) on the power spectrum of earthquake-induced dynamic irregularity is analyzed. The results demonstrate that the degradation degree of piers and bearings of the track-bridge system under transverse random earthquake presents the characteristics of large in the middle and small on both sides. The earthquake-induced track directional irregularity samples follow the normal dis-tribution. The shape of the power spectral density curve has obvious segmentation characteristics in the double logarithmic coordinate system. The power spectrum amplitude decreases significantly with the increase in spatial frequency and fluctuates significantly in the high frequency band. There are little influence of the train operation speeds on the power spectrum curves. In contrast, the influence of the PGAs on the power spectrum is significant. The amplitude of the power spectrum increases obviously as the PGA increases.