Modeling And Control Of MR Damper In Vehicle Suspension System

The magnetorheological damper has low power consumption and high output damping force.Due to the inherent nonlinearity and complexity,it is utmost importance to establish accurate mathematical model of magnetorheological damping which is the key of vibration control.In this paper,the following research is conducted on MR damper in vehicle suspension systems:Firstly,the current situation and application of magnetorheological damper in various fields are introduced.The achievements of the current research in the hysteresis model of magnetorheological damper are described.Secondly,the improved Particle Swarm Optimization(PSO)is proposed,which overcomes the shortcomings of the traditional Particle Swarm Optimization(PSO),as it easy fall into local optimum and slow convergence speed.Besides,It is used to identify the parameters of the mathematical model of MR damper proposed later,and It lays the foundation for establishing accurate hysteresis model.Thirdly,for the problem of magnetorheological damper with hysteresis nonlinearity under different input current excitations,a parameterized model of magnetorheological damper based on dynamic hysteresis element is proposed.The model can adapt to different current magnetorheological characteristics,which can dynamically change the output damping force according to the input current,the verification and analysis of the model are completed with the experimental data of RD-1005.Fourthly,for the problem of magnetorheological damper with hysteresis asymmetry and damping force saturation a hybrid model based on hysteresis and two-viscosity asymmetry is proposed.The model adapts well to the hysteresisasymmetry of velocity and damping force(F-V)and the asymmetric saturation of damping force in MR damper,and the model verification and analysis are completed with the RD-1005 experimental data.Fifthly,for the problem of poor accuracy of MR damper under the influence of many external conditions,a nonparametric hysteresis model based on neural network is proposed.In this model,an improved hysteretic operator is proposed to introduce the lag operator into the input space.The multi-hysteresis can be transformed into a one-to-one mapping on the newly constructed extended input space.Therefore,the model can dynamically adapt to the external frequency and current through the experimental verification analysis,the validity of the model is proved.Sixthly,a fuzzy controller based on MR damper is designed.The simulation results show that the fuzzy controller can adjust the output damping force of the MR damper by controlling the output current,and the desired effect of damping in 1/4suspension system is achieved.Thus,the feasibility of the controller is verified.