Research on load inversion of bogie frame based on vibration acceleration

It is critical in the implementation of structural health monitoring for rolling stock to acquire load characteristics of vital interfaces such as air suspensions, auxiliary components, axle box springs, and rubber joints during the service life. This paper explored an estimation method for loads on bogie frames. Based on an initially established vehicle system dynamics model, lateral and vertical dynamics models were derived, incorporating localized refinements specific to rocker type axle boxes. Based on these models, Kalman filtering was applied for inversion estimation of system state quantities through the system state equation. The study results, which considered excitations from track irregularities, show a correlation surpassing 0.8 between the estimated and simulated values for vertical loads on the bogie frame, while a correlation exceeding 0.6 for lateral loads. These findings indicate consistent change trends in the time-frequency domain between the estimated and simulated values. Moreover, as the vehicle speed increases, the accuracy of the inversion results based on Kalman filtering progressively improve in correlation with the simulated values. Based on the above outcomes, conclusions can be drawn that the proposed load estimation method is sufficiently accurate and stable in the inversion calculations for loads on the critical interfaces of bogie frames. This method can serve as a data foundation for subsequent damage monitoring and evaluation, remaining life calculations, and optimizing the structural design for key bogie components.