Control Of A 1/4 Magnetorheological (MR) Semi-active Suspension System Using A RC Hysteresis Model For MR Damper

The semi-active suspension system based on magnetorheological(MR)damper has the characteristics of continuous adjustable damping within a certain range and low power consumption.It belongs to the field of magnetorheological technology and is expected to revolutionize the automotive suspension system.However,the insufficient accuracy of the mechanical model of magnetorheological damper will seriously affect the control performance of the magnetorheological semi-active suspension system.In this paper,we describe the nonlinear hysteresis of the MR damper based on ResistorCapacitor(RC)hysteresis model.RC hysteresis model belongs to a parametric model,which can identify the parameters in the model according to the experimental results.Identification results of the RC hysteresis model and the Bouc-Wen model show that the RC hysteresis model is more accurate in characterizing the hysteresis characteristics of MR damper and reduces the computational cost in parameter identification and model prediction.The comparison of test data and simulation data verifies the feasibility of RC hysteresis model in the control system.Based on this,a linear quadratic regulator based on state feedback is designed for the 1/4 vehicle semi-active suspension system based on MR damper.The genetic algorithm is used to optimize the weighting matrix.The selection of the weighting matrix achieves the best performance of the vehicle body acceleration,dynamic suspension deflection and dynamic tire load.The control output in the linear quadratic regulator is the optimal control force,while in the semi-active control system based on the magnetorheological damper,the effective controllable input is the excitation current.Therefore,this thesis by adjusting the excitation current in order to achieve tracking control of the required control force.In the tracking control,the influence of hysteresis characteristics of MR damper is considered,and the damping force is tracked by model-based feedforward control strategy.In order to verify the general applicability of this control strategy,the response characteristics in both time domain and frequency domain of the vehicle body acceleration,the dynamic suspension deflection and dynamic tire load under the excitation of a stationary random road surface and a pulse road surface are simulated.