A cascaded control strategy for magneto-rheological dampers based on Hammerstein model
Shuyou Yu, Jie Guo, Xinze Xu, Songlin Zhang, Baojun Lin
Abstract
In order to effectively attenuate the inherent hysteresis nonlinearity of magneto-rheological (MR) dampers, and achieve precise tracking control of damping force, a cascaded control strategy based on Hammerstein model is proposed in this paper. A BP neural network is utilized to construct the nonlinear module of Hammerstein model, which accurately captures the hysteresis behavior of MR dampers. The dynamic characteristics of the dampers are then described by a linear time-invariant model. For MR dampers, the cascaded control strategy enhances the robustness of the system compared to traditional open-loop control based on the inverse model schemes. The effectiveness of the proposed control algorithm has been verified through simulation experiments and hardware-in-the-loop experiments using a seat suspension testbed equipped with MR dampers.