Research On Robust H_∞ Control Of Semi-active Suspension On In-wheel Motor Drive Electric Vehicle

The hub motor drive system for electric vehicles has advantages such as compact structure,short transmission chain,and high transmission efficiency,which has been favored by researchers at home and abroad.However,since the motor is directly installed in the driving wheel,it increases the non-sprung mass and introduces electromagnetic interference,affecting the smoothness and ride comfort of the vehicle.The semi-active suspension system based on magnetorheological dampers has advantages such as fast response,continuous damping adjustment,and low energy consumption.With corresponding control algorithms,it can effectively improve the smoothness and handling stability of the vehicle.An improved hyperbolic tangent model is established through dynamic characteristic tests on the selected magnetorheological damper,and genetic algorithms and nonlinear least squares methods are used for parameter identification.The inverse model of the improved hyperbolic tangent model is derived.The model prediction results match well with the test data,indicating that the improved hyperbolic tangent model can describe the hysteresis characteristics of the magnetorheological damper.To analyze the effect of electromagnetic interference on semi-active suspension control,an accurate motor model must be established,and the electromagnetic excitation must be solved.Based on the physical parameters of an external rotor brushless DC hub motor,a finite element model is established,and the frequency characteristics and spatial orders of the magnetic field are analyzed under no eccentricity,static eccentricity,and dynamic eccentricity.It is found that the finite element simulation results are consistent with the analytical results under the corresponding eccentricity state,indicating that the motor model meets the requirements.Based on this,the electromagnetic force characteristics under different eccentricity states are analyzed,and it is found that the electromagnetic force slightly fluctuates when there is no eccentricity,the amplitude increases significantly under static eccentricity,and the balance position moves up compared to no eccentricity,while the balance position under dynamic eccentricity is close to no eccentricity and exhibits quasi-harmonic vibration.Considering the combined effects of road surface and motor excitation on the suspension,a 1/4-car semi-active suspension model based on magnetorheological dampers is established,and an H-state feedback controller with strong robustness to parameter uncertainty is designed and compared with passive control and skyhook control.The results show that compared with passive suspension and skyhook control,the root means the square value of the center of gravity acceleration under random road surfaces is reduced by 27.4% with robust H_∞ control,and the control effect is even better under changing road surfaces,with the root mean square values of the three smoothness indicators reduced by 33.4%,28.7%,and 18.1%,respectively.Therefore,robust H_∞ control can more effectively suppress vehicle body acceleration and has good operating condition adaptability.To more comprehensively reflect the control effect of the semi-active suspension,a half-car model of the hub motor-driven electric vehicle is established,and an augmented matrix containing road preview information is established using axle preview.Based on this,a robust preview control strategy considering parameter uncertainty is designed,and its superiority is verified.The research shows that the control strategy considering parameter uncertainty has strong robustness to parameters,and the control strategy fully considering road preview information can further improve the performance of the suspension.