Frog gaps and rail joints produces intricate wheel-rail contact conditions within subway turnouts, which results in intense vibrations. This study describes vibration tests conducted on several typical cross-sections of a turnout, which aimed at understanding the vibration transfer characteristics of typical section in turnout area. To separation the vibration signals from each carriage, wavelet analysis and peak detection algorithms were utilized. Using a 2D kernel density algorithm, representative data from the carriages were clustered, effectively isolating the influence of interference data. A coupled vehicle-turnout-tunnel-soil vibration analysis model was established, combining multiple spatial and temporal scales and the accuracy of its prediction was evaluated by comparing the calculated results to statistical test results.The test results indicate that when the train decelerated through the turnout, there was a significant variation in the vibrations of different train carriages, with a 13 dB difference in the total rail vibration levels. The rail joints and frog gap present in the turnout resulted in high vibration level in the tunnel walls. The main frequency of tunnel wall vibration in the frog gap section was 63 Hz, and the total vibration level was 84.9 dB. The accuracy of vibration response prediction was evaluated by comparing the calculated results to statistical test results using probability density functions, and the model errors were maintained within 10 % of the test data. The presented vibration testing and modelling results of vibration propagation characteristics in various directions for the subway turnout area will provide a foundation for selecting vibration mitigation measures.
Read full abstract