A parameter identification method based on time-frequency analysis for time-varying vehicle dynamics

This paper proposes a novel time-frequency analysis-based parameter identification method for time-varying vehicle dynamics under harmonic conditions. The motivation is to identify the potential time-varying system parameters from experimental data corresponding to a linear single-track model, which has the aim of model identification in order to build simulation models. Based on the single-track model and time-frequency analysis, many vehicle dynamics parameters, such as the (normalized) cornering stiffness of the axle and the instantaneous friction coefficient, are available with the assistance of this method. The method has the prominent advantage that it is capable of identifying the time-varying system parameters within the range of linear lateral dynamics. The coupling between the vertical dynamics and the horizontal dynamics is also considered with an effective axle load evaluation algorithm, by which the influence of a specific dynamic axle load on the cornering stiffness of the axle can be described. It is also integrated into the parameter identification method. Finally the method is applied to validate a complicated simulation model of a car-trailer combination obtained with IPG CarMaker (R) software using the experimental data.