首页 | 本学科首页   官方微博 | 高级检索  
     

基于KNN方法的多约束下连续体结构拓扑优化设计
引用本文:李景奎,张义民.基于KNN方法的多约束下连续体结构拓扑优化设计[J].机械设计,2012,29(3):74-77.
作者姓名:李景奎  张义民
作者单位:东北大学机械工程与自动化学院,辽宁沈阳,110004
基金项目:长江学者和创新团队发展计划资助项目(IRT0816);科技重大专项课题资助项目(2010ZX04014-014);国家自然科学基金资助项目(50875039);“十一五”国家科技支撑计划资助项目(2009BAG12A02-A07-2)
摘    要:提出了多约束下连续体结构拓扑优化设计的K邻近(KNN)方法。首先以结构单元作为样本,将整个设计区域看作是样本空间,将连续体结构拓扑优化看作是一种对单元的模式识别;然后利用单元应力和位移灵敏度作为单元样本的特征向量,利用特征向量的欧氏距离为识别标准,引入模式识别的KNN方法进行多约束下的连续体结构拓扑优化设计;最后给出计算示例的优化结果来表明该方法是可行的。

关 键 词:结构优化  拓扑优化  K邻近  多约束  模式识别

Optimization design of continuum structure topology under multi-constraints based on K nearest neighbor
LI Jing-kui , ZHANG Yi-min.Optimization design of continuum structure topology under multi-constraints based on K nearest neighbor[J].Journal of Machine Design,2012,29(3):74-77.
Authors:LI Jing-kui  ZHANG Yi-min
Affiliation:(School of Mechanical Engineering and Automation,Northeastern University,Shenyang 110004,China)
Abstract:In this paper,the method of continuum structure topology optimization design under multi-constraints based on K Nearest Neighbor(KNN) is proposed.In which the structure unit is taken as the sample and the entire design region is considered as the sample space,and the continuum structure topology optimization design is regarded as a kind of pattern recognition to the structure unit.The element stress and element displacement sensitivity are used as the feature vectors,and the Euclidean distance of the feature vectors is utilized as the recognition criterion.The KNN method is introduced into the continuum structure topology optimization design.Several examples are given and the results show this KNN method is feasible.
Keywords:structure optimization  topology optimization  knearest neighbor  multi-constraints  pattern recognition
本文献已被 CNKI 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号