利用模糊神经网络的汽车主动悬架与EPS集成控制

摘要

The dynamic model of vehicle's active suspension for steering analysis is established. The EPS model, tire model and road-condition input model are introduced. Fuzzy neural network control method is used to design the vehicle's active suspension and the integrated EPS control system. According to the changes of the body posture of the vehicle, the active force and the EPS assistant torque of the actuator of the active suspension system can be adjusted dynamically. The simulation and analysis are carried out for different steering-wheel angle input. The results show that compared to the uncontrolled suspension and steering system, the vertical acceleration peak and standard deviation of the active suspension and the EPS integrated control system are decreased by 40.94% and 26.06%, and the peak and standard deviation of transverse angular velocity of the vehicle are decreased by 6.24% and 48.15%. Thus, the vehicle body' s vibration is suppressed effectively, and the ride-comfort and handling stability are greatly improved.%在基于转向的主动悬架整车动力学模型基础上,引入EPS模型、轮胎模型和路面输入模型,应用模糊神经网络控制方法,设计汽车主动悬架与EPS集成控制系统及其控制策略,集成控制器根据车身姿态的变化,动态调节主动悬架系统的作动器作用力和EPS的助力转矩,进行转向盘转角角阶越输入的仿真计算和分析,结果表明,相对于不加控制的悬架和转向系统,基于模糊神经网络的主动悬架与EPS集成控制系统的质心垂直加速度峰值和标准差分别下降40.94％和26.06％,整车横摆角速度峰值和标准差分别下降6.24％和48.15％,有效抑制车身的振动,提高汽车的行驶平顺性和操纵稳定性.