基于双响应面的地铁车辆底架稳健优化设计

Robust Optimization Design of Subway Underframe Based on Dual Response Surface

为了减少地铁车辆底架优化过程中由于不确定性因素引起的目标响应分散性问题,利用双响应面模型对其进行稳健优化设计。首先利用Monte Carlo方法筛选出对目标响应影响较大的设计变量,进行拉丁超立方抽样得到设计变量及目标响应的样本点;再运用最小二乘法对样本点进行拟合,求得目标响应的多项式响应面函数,对其进行均值估计和方差估计,得出均值和方差的响应面函数;最后以方差函数最小为优化目标,以均值函数为约束条件建立稳健优化模型,采用内点法对其进行优化求解。研究结果表明,通过双响应面模型进行稳健优化后底架结构的抗外界因素干扰能力提高,并且强度得到改善,质量减小。

In order to reduce the dispersion of the objective response due to uncertain factors in the subway underframe’s optimization process,a robust optimization design method based on dual response surface model was proposed.Firstly,Monte Carlo method was used to select the design variables which had great influence on the target response,and Latin hypercube sampling was used to get the sample points of the design variables and the target response.Then,the least square method was used to fit the sample points,and the polynomial response surface function of the target response was obtained,the mean and variance of the polynomial response surface function were estimated,and the response surface function of the mean and variance was obtained.Taking the minimum variance function as the optimization objective and the mean function as the constraints,the robust optimization model was established and solved by the interior point method.The results show that the anti-interference ability and strength of the structure is improved as well as the mass is reduced after the robust optimization of the double response surface model.