汽车半主动悬架系统神经网络模型研究

Study on Neural Network Models of Vehicle Semi-active Suspension System

基于磁流变减振器的汽车悬架系统具有明显的滞后非线性,系统中的非线性阻尼和非线性刚度等对其动力学行为产生了很大的影响。本文利用系统识别理论建立了磁流变减振器的神经网络模型,该模型可根据半主动悬架系统的控制力求得控制电流,具有良好的辩识能力。

The vehicle suspension equipped with magneto-reological（MR） damper features nonlinearity.Its dynamic characteristics are related to nonlinear stiffness coefficient and nonlinear damping coefficient.To describe the nonlinear characteristics exactly,the model based on the neural network of MR damper is presented subsequently in this paper.This neural network is certified the high capacity of approximation to obtain the control current of semi-active suspension system.