多项式响应面代理模型在立柱结构优化中的应用

发布时间：2018-06-15 20:31

本文选题：结构优化 + 参数化设计　；参考：《机械设计》2017年04期

【摘要】：立柱结构是某设备升降系统的重要组成部分和承力部件。由于现有结构质量过大,给装配、运输带来很大不便。文中应用多项式响应面代理模型和遗传算法结合的优化方法对立柱进行优化,选取对立柱结构力学性能影响的较大参数为设计因素,利用正交试验设计获取样本点集构建出相应的等效应力代理模型和合位移代理模型。以质量最轻为优化目标,以代理模型小于允许值为约束条件,应用遗传算法对立柱结构进行优化。优化结果表明:优化后模型满足强度刚度设计要求,其质量减轻27.4%,优化运算时间有大幅度减少,具有一定的工程应用价值。
[Abstract]:Column structure is an important part and bearing part of a lifting system. Because the quality of the existing structure is too large to assemble, transportation brings great inconvenience. In this paper, the polynomial response surface agent model and genetic algorithm are used to optimize the column, and the large parameters which affect the mechanical properties of the column are selected as the design factors. The equivalent stress agent model and displacement agent model are constructed by using the orthogonal test design to obtain the sample set. Taking the minimum mass as the optimization objective and the agent model less than the allowable value as the constraint condition, the genetic algorithm is applied to optimize the column structure. The optimization results show that the optimized model meets the requirements of strength and stiffness design, and its quality is reduced by 27.4. the optimization operation time is greatly reduced, and it has certain engineering application value.
【作者单位】：中北大学机械与动力工程学院;
【基金】：国家自然科学基金资助项目(50875247) 山西省自然科学基金资助项目(2009011026-1) 山西省研究生创新基金项目资助(2008072)
【分类号】：TN873.93;TP18

【参考文献】