A stochastic inversion spectrum analysis method for rail vehicle structural load boundaries

Abstract To improve the efficiency and accuracy of complex dynamic load data acquisition under actual rail vehicle operating conditions (e.g. track excitation and aerodynamic effects), a stochastic spectrum compilation method for dynamic load is proposed based on rail vehicle standards and specifications. To this end, a multi-body dynamics model of trains considering track excitation and aerodynamic effects is constructed. Then, the dynamic load time histories of vehicle components under different working conditions are obtained by using the multi-body dynamics model. Subsequently, a statistical model containing dynamic load features (e.g. mean value and amplitude) is established based on the calculated load data. Finally, the complex dynamic load spectra (including the effects of track excitation and aerodynamics) are generated using a statistical inverse transformation approach, i.e. a Monte Carlo random generation method. The results show that the generated load spectrum is consistent with the measured data and can accurately simulate the dynamic load variations experienced by rail vehicles under different operating conditions. The load spectrum obtained through statistical envelope can reflect the most extreme dynamic loads that vehicle components may experience during service, providing an basis for the design and safety assessment of vehicle structures.