摘要
利用MADYMO软件对某车型乘员约束系统进行仿真,以加权伤害指数作为评价指标,根据正面碰撞工况下不同身材假人的损伤情况,选取相关参数为设计变量,采用拉丁方实验设计生成样本点,构造DACE-Kriging近似模型,分析变量与评价指标的相关性,利用非支配遗传优化算法,以不同身材假人伤害值最小为目标进行多目标优化,得到Pareto最优前沿,作为优选方案集合。结果表明,改进后的约束系统对不同身材乘员都能起到较好的保护效果。
MADYMO software is used to simulate the occupant restraint system of a vehicle. According to the injury situation of dummy with different statures under frontal impact condition with weighed injury criterion（ WIC） as evaluation indicator,relevant parameters are selected as design variables. Sample points are generated by the Latin-square design of experiment to construct a DACE-Kriging surrogate model. The correlation between variables and evaluation indicators are analyzed,and a multi-objective optimization is conducted by using non-dominated sorting genetic algorithm（ NSGA-Ⅱ） with minimizing the injuries of dummy with different statures as objective,so the Pareto-optimal front solutions are obtained as an optimized scheme set. The results show that the occupant restraint system optimized has better protection effects for the occupants of different statures.
出处
《汽车工程》
EI
CSCD
北大核心
2016年第11期1312-1318,共7页
Automotive Engineering
