针对现有副车架轻量化设计要么优化效果不理想、要么对模型精度要求较高的问题,采用Kriging近似模型替代物理模型对副车架进行了轻量化设计:利用拉丁超立方试验设计方法对副车架的重要特征点进行采样,建立Kriging近似模型,并将副车架参数化;以副车架质量为目标函数,选取11个厚度尺寸作为设计变量对其进行优化,通过约400次的迭代得到最优解.优化后的副车架质量为11.3 kg,较优化前减轻10.3%,跳动工况下最大应力值仍满足屈服条件,且自由模态频率也满足模态要求,较好地实现了轻量化设计.该方法的优点在于既实现了较好的优化效果,又无需在不同软件间转换,有效避免了模型的失真.%The light weight design of the subframe is either not ideal,or the accuracy of the model should be higher. To deal with the problems,the Kriging approximate model was used to make a lightweight design for the subframe.Firstly,the Latin hypercube sampling test design method was used to sample the important fea-ture points of the subframe,and the Kriging approximate model was combined to parameterize the subframe. Then,the fatigue of subframe was taken as the target parameter,and 11 thickness dimensions were selected as design variables and optimized. The optimal solution was obtained through about 400 iterations,and the opti-mized subframe quality was 11.3 kg,which was reduced by 10.3% compared with that before optimization. The maximum stress value under beating condition still satisfied the yield condition,and the free model fre-quency also met the model requirement,which realized the lightweight design better. The advantage of this method was that it not only achieved better optimization results,but also did not need to transform among dif-ferent softwares and effectively avoided the distortion of the model.
展开▼
