基于响应面方法的转向梯形优化设计

在考虑悬架跳动时前束角变化的情况下,为实现对车辆转向运动学的最优化,以某小型皮卡车为研究对象,利用ADAMS/CAR软件建立前悬架和转向总成虚拟样机模型,进行转向和轮跳仿真实验。以内、外球头的空间坐标为优化变量,分别以阿克曼误差的均方根和前束角的变化范围为目标,运用DOE方法进行2次试验设计;根据实验结果分别对影响阿克曼误差和前束角的影响因素进行灵敏度分析,运用响应面方法(RSM)拟合出回归模型。利用MATLAB优化工具箱在约束前束角的情况下,对阿克曼误差进行优化,得到硬点的最优值。研究结果表明:在前束角的变化范围减少10%的前提下,转向机构在0°~25°常用转向角度,阿克曼误差减少51.27%。优化后的结果明显地改善了转向的运动学特性,并且避免了单独优化转向,对前束角造成的影响。

Optimization of Ackerman steering linkage based on RSM

For the purpose of getting the best steering performance on the optimization of steering systems,considering the toe angle’s change,a virtual front vehicle system of the electric sightseeing vehicle was built by ADAMS/CAR.The system included double wishbone suspension and rack-pinion steer.Steer and toe angle simulations were run in this model.Experiment was designed according to DOE theory,and sensitivity analysis was given about the hard points of inner and out tie rod.Then the object functions,which was the relationship between the total root mean squared(RMS) of Ackerman error and toe angle with different variables,were estimated by using response surface methodology(RSM).Last but not least,the optimization result of design factors by MATLAB was given.The results show that steer performance of the car improves by 51.27%,and the change of toe angle reduces by 10%.