车辆主动悬架的BP神经网络自适应PID控制  

Self-Adaptive PID Control Based on Neural Network for Active Suspension Vehicle

作　　者：王春华[1] 唐焱[1] 

机构地区：[1]桂林电子科技大学机电工程学院,广西桂林541004

出　　处：《计算机仿真》2009年第5期274-277,共4页Computer Simulation

摘　　要：提出了一种主动悬架的基于BP神经网络的自适应PID控制方法,并借助于1/4主动悬架物理模型,探讨了该控制技术在车身主动减振方面的控制问题。以白噪声模拟路面输入,对车辆主动悬架系统进行计算机仿真研究。将BP神经网络PID主动悬架、PID主动悬架和被动悬架的车身加速度、悬架动挠度及车轮动位移三项指标的均方根值进行了对比分析。仿真结果表明,具有BP神经网络PID控制器的主动悬架控制效果明显优于PID主动悬架和被动悬架,可大大减少路面对车身的振动冲击,能显著地提高车辆行驶平顺性和乘坐舒适性,且鲁棒性好。a new type of P ID control method based on BP neural network is proposed in this paper and is investigated to actively control vibrations of an automotive body based on a model of one fourth active suspension. Using the white noise as the road input, the simulation of active suspension is finished. Root mean square values of three performance indexes, body acceleration, suspension dynamic travel and tire dynamic deflection, are compared. The simulation results demonstrates that the active suspension with PID controller based on BP neural network has an advantage over the active suspension with PID controller and passive suspension, greatly reduces the vehicle body vibration, obviously improves the vehicle ride performance and comfort, and has powerful robustness.

关 键 词：主动悬架 误差反传神经网络 比例-积分-微分控制 仿真 

分 类 号：U461.4[机械工程—车辆工程]