鲁棒性概率优化在乘员约束系统设计中的应用

Applications of Robust Probabilistic Optimization in Occupant Restraint System Development

利用MADYMO软件建立了国内某微型车正面碰撞数学模型,并在对模型进行试验验证的基础上结合试验设计、响应面方法和鲁棒性概率优化设计对乘员侧约束系统进行优化,优化的参数主要有安全带刚度、安全带上挂点位置、限力器撕裂力和撕裂增加长度等。采用该方法引入鲁棒性概率优化设计后能减少和控制目标函数响应的波动,降低在设计点上的敏感性,提高产品的质量。结果表明,所采用的方法既实现了对设计目标的优化,又提高了设计变量的可靠性和目标函数的鲁棒性,对工程应用具有较大指导意义。

A mathematical model of occupant restraint system of a mini-bus van in frontal impact based on software MADYMO was established.Based on that,the optimization process was conducted by conjuncting design of experiments,response surface method and robust probabilistic optimization design.The chosen design variables during this optimization were webbing stiffness,upper slipping position of the seat belt,the tearing force and the adding length.By applying the robust design,this method can reduce the sensitivity of design point and control the fluctuation of the design objects.Consequently,this will conduce to improve the quality of the products.The analysis results indicate that this method realizes the optimization of the design object,and enhances the reliability of the design parameters and the robustness of the objective function,which will make significant improvement in the engineering applications.