摘要
针对某电动三厢轿车的麦弗逊式前悬架,在悬架K&C仿真试验中出现的车轮跳动转向梯度偏大易造成不足转向度过大、车轮跳动外倾梯度过大易造成轮胎磨损的悬架运动学问题进行优化。优化策略为:首先采用灵敏度分析确定优化变量,即通过试验设计方法对设计变量及响应函数进行迭代计算,通过响应面法对数据进行数学回归分析,从而确定设计变量的灵敏度,进而确定优化变量;然后基于改进的遗传算法(NSGA-II),即在传统NSGA-II算法中加入了精英保持策略,对确定出来的优化变量进行多目标优化。优化结果表明,该方法去除了相关度较小的设计变量,提高了算法精度,缩短了计算时间。
The research object of article is a electromobile＇s Macpherson front suspension. There are two problems in suspension kinematics test. The first problem is that bump steer gradient is bigger than normal value which will cause too much under-steer. The second problem is that bump camber gradient is bigger than normal value which will cause tire ＇s abrasion. The article will proposeoptimization for above two problems. The optimization strategy is that： firstly,the sensitivity analysis is used to determinate optimization variable. During the sensitivity analysis, DOE（design of experiment） method is used to design simulation experiment,and response surface method is used to process stimulation experiment data results,and finally it determines the sensitivity of the design variables. After that,we use improved NSGA-II method to optimize multi-objectives which are bump steer gradient and bump camber gradient. The optimized results indicate that the above optimizing strategy reduced irrelevant design variables, improved accuracy of the algorithm, and reduced computation time.
作者
郝志宽
黄江
张乃文
吴伟
HAO Zhikuan;HUANG Jiang;ZHANG Naiwen;WU Wei(Chongqing University of Technology, Chongqing 401320, China;Changan Automotive Engineering Research Institute, Chongqing 401120, China;Chongqing FuBei Information Technology Co. , Ltd. , Chongqing 401120, China)
出处
《重庆理工大学学报（自然科学）》
CAS
北大核心
2018年第4期27-34,共8页
Journal of Chongqing University of Technology：Natural Science
基金
国家自然科学基金资助项目(51105136)
