Research On Model Construction And Control Strategy Of Automobile Magnetorheological Semi-active Suspension

Magneto-rheological damper(MRD),as the key force actuator of the semi-active suspension system,has the advantages of being able to maintain normal working conditions and respond quickly in harsh environments.However,due to the strong nonlinear hysteresis characteristics of the magnetorheological damper,it is difficult to directly obtain the accurate input control current of the magnetorheological damper through the state change of the suspension to control the semi-active suspension system of the automobile.In this paper,in view of the poor accuracy of the input control current of the magnetorheological damper,the difficulty in obtaining the suspension state,and the difficulty of real-time dynamic control of the car suspension performance,this paper designs an automotive magnet based on the magnetorheological damper inverse model combined with a state observer.The rheological semi-active suspension control system is validated and analyzed by simulation.First,the method of filtering white noise is used to establish four-wheel road excitation models under different working conditions.Furthermore,the suspension vibration system is established according to Newton’s second law and Lagrangian equation,and the simulation model of a quarter magnetorheological suspension and a seven-degree-of-freedom magnetorheological suspension of the vehicle is built.Secondly,considering the non-linear hysteresis characteristics of the magnetorheological damper forward dynamics model,a more versatile Spencer phenomenon model is used to establish the forward model;for the problem that the input control current of the magnetorheological damper is difficult to determine,the combination of strong The BP neural network algorithm with mapping capability establishes the inverse model of the magnetorheological damper.Considering that the BP neural network is easy to fall into the problem of local optimization and large generalization errors in the iterative process,the particles with extremely fast convergence and overall optimization capability are used.The group algorithm optimizes the BP neural network to improve the accuracy of the magnetorheological damper to control the input current.Combined with the semi-active suspension control system,the effectiveness of the optimized inverse model is verified.Then,taking the quarter magnetorheological semi-active suspension system as the research object,considering that the suspension state parameters are difficult to obtain,Kalman Filter(KF)and Unscented Kalman Filter(Unscented Kalman Filter,UKF)are used.The observer observes the suspension state,and the results show that the observation accuracy of the unscented Kalman filter observer is higher;on this basis,a magnetorheological semi-active suspension control system based on the fuzzy sliding mode algorithm is established,and the optimized magnetic current is adopted.The variable damper inverse model is combined with the control strategy to obtain the desired force to obtain the corresponding control input current,the appropriate damping force of the magnetorheological damper is output,and the semi-active suspension system is controlled in combination with the multi-condition single-wheel road excitation.From the simulation results,it can be concluded that the established fuzzy sliding mode suspension control system can obtain good performance indicators.Finally,for the vehicle’s seven-degree-of-freedom magnetorheological semi-active suspension control system,a suspension observer based on the UKF algorithm is first established to obtain the observed values of the suspension state parameters;then the seven-degree-of-freedom suspension is established by combining the fuzzy sliding mode algorithm The control system uses the expected force to reverse the input currents of the four different dampers inverse models.Under the action of the four-wheel road surface excitation model under multiple working conditions,the vehicle seven-degree-offreedom magnetorheological semi-active suspension control system is simulated.The simulation results show that the established suspension observer based on the UKF algorithm can accurately observe the suspension state parameters such as attitude,pitch and roll,and the overall performance index of the vehicle’s seven-degree-of-freedom magnetorheological suspension based on the fuzzy sliding mode algorithm is also Improved.