A Magnetorheological Damper Model And Its Inverse Characteristics Investigation For Control

Magnetorheological(MR)damper is a typical and most important actuator of semi-active suspension.Its dynamics modeling and control algorithm design are the hot research issues in vehicle dynamics field.MR dampers are characterized of large dynamic range,fast response speed and low power consumption and so on.To realize the fast and accurate control of MR damper’s damping characteristics,a highly accurate dynamic model is needed to describe its characteristics and design the controller.In addition,in order to track the required damping force signal in real time,the semi-active suspension controller also needs an accurate inverse model.Considering the current models can not accurately describe the strong nonlinear characteristics of the MR damper under different excitation conditions,a modified parameter model of MR damper considering the hysteresis characteristics of the shear term is proposed to represent its dynamic characteristics.Then,an investigation on inverse characteristics of MR damper based on the direct model is proposed.Moreover,the proposed models are applied to the research on semi-active suspension control algorithm.The bench test results of suspension control system for quarter vehicle show that the proposed models can describe the direct and inverse dynamic characteristics of the MR damper accurately and efficiently.The main research contents of this thesis are as follows:Firstly,a modified parameter MR model based on magic formula is proposed.The shear hysteretic width factor,friction smooth factor and viscous hysteretic width factor are introduced into the proposed model.It can more accurately represent the nonlinear dynamic characteristics of MR damper,thus providing a theoretical basis for the dynamic analysis and semi-active suspension control.Secondly,the inverse dynamics model of MR damper based on the modified parameter model is established.Adaptive neural fuzzy technique and direct inverse method are applied to build two different inverse MR models.Compared with the exact validation data generated by the direct model,the proposed inverse model can accurately predict the control current and closely track the desired damping force.Moreover,the validation of the proposed inverse models is conducted by comparison with three traditional empirical inverse approaches.The results reveal good tracking capability for the desired damping force and the weaknesses of proposed model.Thirdly,the inverse model of MR damper is applied to semi-active suspension control system.Under the conditions of sine excitation,impulse excitation and random excitation respectively,the control effects of two semi-active suspension control algorithms are compared and analyzed on the basis of the Simulink simulation model of a quarter vehicle suspension system.The results show that the proposed inverse model is effective,not only realize the ideal control effect of the semi-active suspension control algorithm,but also provides the algorithm and software basis for the bench test of the control algorithm.Finally,a test platform of suspension control system for a quarter vehicle is designed and established,which includes a modified mechanical bench,electro-hydraulic servo actuator,sensing and control system,signal transmission system,data processing system and rapid prototype controller.Validation of semi-active suspension control algorithm and the direct and inverse dynamics models are conducted by bench test.The results show that the proposed models provide a theoretical basis for the further design of more effective control algorithm,and the establishment of the test platform also provides an effective verification tool for the development of the control algorithm.