An enhanced greedy algorithm for failure resistant material design with application to composite delamination

In this work, an enhanced greedy algorithm is developed for failure resistant design of two-phase composites against delamination. Possessing the merits of a greedy algorithm, i.e. easy-to-implement and gradient-free, the computational efficiency is significantly improved. An objective functional ba...

Full description

Saved in:
Bibliographic Details
Published in:Composite structures 2021-12, Vol.278, p.114681, Article 114681
Main Authors: Weng, Huanbo, Xu, Yangjian, Chen, Junjun, Ruan, Hongshi, Nan, Chenyu, Liang, Lihua, Ju, Xiaozhe
Format: Article
Language:English
Subjects:
Composite delamination
Failure resistance
Greedy algorithm
Interface elements
Structural optimization
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this work, an enhanced greedy algorithm is developed for failure resistant design of two-phase composites against delamination. Possessing the merits of a greedy algorithm, i.e. easy-to-implement and gradient-free, the computational efficiency is significantly improved. An objective functional based on interface elements is used to represent the failure resistance and evaluated by finite element computations. The search domain for each iteration is reduced by a heuristic strategy, thus leading to a reduced number of required finite element computations. Additionally, a criterion is introduced to adaptively determine the number of switching subvolumes for each correction step, leading to a reduced number of iterations. A numerical study shows that the computational complexity of the proposed algorithm is significantly reduced on the basis of an existing algorithm. Furthermore, we develop three different generation schemes for initial solutions and conduct a comparative study on them. A maximal speed-up factor of 12.5 compared to the existing algorithm is achieved.
ISSN:0263-8223
1879-1085
DOI:10.1016/j.compstruct.2021.114681