Dynamic analysis and detection of wheel polygonization on high-speed trains based on axle box vibrations

摘要

Wheel polygonization of high-speed trains has received considerable attention because it deteriorates the dynamic interaction of wheel and rail andintensifies the vibration and noise of the vehicle-track system. The dynamics of the vehicle system, the excitation of track irregularities, and the movementand deformation of wheelsets introduce extensive coupling and complexity to the vibration response of high-speed trains, especially in medium and highfrequencyranges. The paper establishes a multi-body dynamics model of a vehicle with flexible wheelsets using finite element methods; simulations areperformed to investigate the vibration characteristics of wheel polygonization. The paper measures the acceleration of axle boxes to monitor and detectwheel polygonization, adopting the complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) to decompose the signal intoseveral intrinsic mode functions (IMFs). The improved Hilbert-Huang transform (HHT) is performed, and the time-frequency spectrum is plotted tovisually indicate the characteristic frequency of polygonization. Next, the Kullback-Leibler divergency is used to select the effective IMF; the meanfrequency of the selected component is obtained to detect wheel polygonization and the corresponding order. Finally, simulations and field tests areperformed to verify the effectiveness, accuracy, and robustness of the proposed method in a complex vibration environment.