Black-box model identification for a continuously variable, electro-hydraulic semi-active damper

Abstract

This paper discusses the black-box identification of a continuously variable, electro-hydraulic semi-active damper for a passenger car. A neural network based output error (NNOE) model structure is selected to describe the complex nonlinear damper dynamics. This model structure is able to accurately and efficiently describe the dynamic damper behaviour, and is appropriate for full vehicle simulation. The identification procedure includes optimal experiment design, regression vector selection and model parameter estimation. The damper excitation signals are optimized multisines that yield uniform coverage of the achievable working range of the damper. A state of the art iterative procedure is used to concurrently estimate the model parameters and select an optimal set regression vector elements. Experimental validation of the proposed procedure shows that models identified from the data measured using the optimized excitations are considerably more accurate than those identified from data obtained using conventional random phase multisine excitations.