汽车半主动空气悬架的神经网络控制方法-交通运输工程学报-长安大学.pdf

第6 卷　第4 期 交 通 运 输 工 程 学 报 Vol.6　No.4 2006 年12 月 Journal of T raffic and Transportation Engineering Dec. 2006 :1671-1637(2006)04-006-6 0 朱思洪,吕宝占,王　辉,张　莹, 贺　亮 ( , 　210031) 　:为了提高汽车 主动悬架的控制效果,以空气弹簧压力为控制对象,应用自适应神经网络 控制方法, 进行了不同路面激励下的 主动空气悬架的车身垂直加速度、悬架动挠度和车轮动载荷 的计算机仿真和实验研究,并与被动悬架系统的相应参数进行了对比。发现在白噪声路面和较低 频率的正弦路面激励下, 主动空气悬架采用自适应神经网络控制能够明显降低车身垂直加速度、 车轮动载荷和悬架动挠度, 降低范围为16%～8 %,提高了车辆的操纵稳定性, 改善了车辆的行驶 安全性与乘坐舒适性。 :车辆工程;空气悬架;神经网络控制方法;仿真;实验研究 :U463. 33　　　:A Neuralnetworkcontrolmethodof automotive sem-i activeair suspension Zhu S-i hong, Lu Bao-zhan, Wang Hui, Zhang Ying, He Liang (School of Engineering, Nanjing Agricultural University, Nanjing 210031, Jiangsu, China) Abstract:In order to improve the control effect of vehicle sem-i active suspension, adaptive neural netw ork control method w as developed, the air pressure of spring w as taken as controlled object, the computer simulation and experiment of body plumb acceleration, suspension dynamic deflection and w heel dynamic load with sem-i active air suspension under the excitations of different road surfaces w ere studied, the control result of neural netw ork control suspension w as compared with the control effect of passive suspension. Under the excitations of w hite noise roads and low er frequency sinusoid roads, the result show s that the sem-i active suspension with the method not only markedly reduces body plumb acceleration, w heel dynamic load, suspension dynamic deflection, the decreased range is from 16% to 8 %, but also improves automotive driving stability, riding comfortable performance and running security . 3 tabs, figs, 1 refs. Key words:vehicle engineering;air suspension;neural network control method;simulation; experimental study Author resume:Zhu S-i hong(1962-