A Method for the Identification and Modeling of Complex Component Behavior in Large-Scale Vehicle Systems Models

A novel application of a nonparametric identification technique is described for reducing the complexity of models for structural/mechanical systems components to levels where they are acceptable for inclusion in computation intensive, large-scale vehicle simulations. The technique is based on an integral-series expansion of the functional relating component response to input. Solution of the identification and modeling problem requires the determination of “kernel” functions appearing in the integral-series expressions. Various procedures for direct calculation of the kernels are demonstrated in three example problems involving: (i) the model representation of hydraulic shock absorber performance in a three degree-of-freedom vehicle model for ride analysis; (ii) simulation of the non-linear stiffness characteristics in the frequency domain of an elastomeric component; and (iii) the representation of metallic component crush including the effects of large deformation plasticity and strain rates.