4.7 Article

Isogeometric sizing and shape optimisation of beam structures

期刊

COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING
卷 199, 期 17-20, 页码 1216-1230

出版社

ELSEVIER SCIENCE SA
DOI: 10.1016/j.cma.2009.12.010

关键词

Isogeometric analysis; NURBS; Continuous sizing; Multilevel design; Method of projected sensitivities

向作者/读者索取更多资源

Based on the recently proposed framework of isogeometric analysis, a study of structural sizing and shape optimisation of curved beam structures is presented. The idea of the isogeometric approach is to analyse the structure using its intrinsic and hence exact geometric definition. The structural response is approximated using the same basis functions describing the geometry. The presented isogeometric framework employs non-uniform rational B-splines. Shape changes can be represented by altering both spatial location of control points and the corresponding weights towards the optimal design. Adaptation of identical geometric representations for both design and analysis resolves the problem of repetitive conversion between the computer-aided and facet based geometries used in conventional finite element analysis. Sizing variables defined at the control points are interpolated using the same basis functions. Due to the fact that both analysis, sizing and shape design share the same underlying description, direct sensitivities can be easily evaluated analytically which improves the performance of the optimisation process. While sizing should reflect local stress states, shape design is preferably performed on a global level. Thus, a multilevel design approach is proposed, where shape design is carried on a coarser level. Projecting the shape design sensitivities bridges the gap between the different levels. The proposed design formulation is applied to compliance minimisation problems. Numerical results indicate the applicability of the method. (c) 2009 Elsevier B.V. All rights reserved.

作者

我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.7
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

暂无数据
暂无数据