基于代理模型的索卡减重优化设计
The Weight Reduction Optimization Design of Rope Clip Based on Surrogate Model
解昊 1王林涛 1郑首坤1
作者信息
- 1. 大连理工大学机械工程学院,辽宁 大连 116024
- 折叠
摘要
索卡作为架空缆道的核心部件具有工作环境特殊,工况复杂等特点,有关其结构优化的内容国内外少有研究.另一方面,传统的有限元优化方法对于解决高维度问题往往效率低,工程应用性不强.这里通过构建高精度Kriging代理模型对索卡结构进行减重优化,使用MIGA和NLPQL相结合的优化算法获得全局最优解.优化结果表明,在满足强度和不发生共振的情况下,结构质量减少22.05%,Kriging模型与有限元模型的最大相对误差为5.27%.验证了高精度代理模型在工程优化问题中的实用性.
Abstract
As the core part of aerial ropeway,the rope clip has the characteristics of special working environment and complicated working conditions,so there are few researches on its structure optimization at home and abroad.On the other hand,the tradi-tional finite element optimization methods for high-dimensional problems are often inefficient and inconvenient to be applied in engineering.A high-precision Kriging surrogate model is constructed for weight reduction optimization and an MIGA-NLPQL algorithm is used to obtain the global optimal solution.The results show that the structure weight is reduced by 22.05%on the premise of guaranteeing strength and non-resonance,and the maximum relative error between Kriging model and finite element model is5.27%.The practicability of the high precision surrogate model in engineering optimization is verified.
关键词
索卡/有限元/Kriging代理模型/减重优化Key words
Rope Clip/Finite Element/Kriging Surrogate Model/Weight Reduction Optimization引用本文复制引用
出版年
2024
NETL
NSTL
维普
万方数据