航空学报 > 2023, Vol. 44 Issue (15): 528863-528863   doi: 10.7527/S1000-6893.2023.28863

多尺度结构拓扑优化设计方法综述

陈小前1, 赵勇2(), 霍森林2, 张泽雨2, 都柄晓2   

  1. 1.军事科学院,北京  100195
    2.国防科技大学 空天科学学院,长沙  410073
  • 收稿日期:2023-04-13 修回日期:2023-04-28 接受日期:2023-05-24 出版日期:2023-08-15 发布日期:2023-06-06
  • 通讯作者: 赵勇 E-mail:zhaoyong@nudt.edu.cn
  • 基金资助:
    国家杰出青年科学基金(11725211);空天前沿培育项目(KY0505072113);湖南省研究生科研创新项目(CX20220021)

A review of topology optimization design methods for multi-scale structures

Xiaoqian CHEN1, Yong ZHAO2(), Senlin HUO2, Zeyu ZHANG2, Bingxiao DU2   

  1. 1.Chinese Academy of Military Science,Beijing  100195,China
    2.College of Aerospace Science and Engineering,National University of Defense Technology,Changsha  410073,China
  • Received:2023-04-13 Revised:2023-04-28 Accepted:2023-05-24 Online:2023-08-15 Published:2023-06-06
  • Contact: Yong ZHAO E-mail:zhaoyong@nudt.edu.cn
  • Supported by:
    National Science Fund for Distinguished Young Scholars(11725211);Aerospace Frontier Inspiration Project(KY0505072113);Postgraduate Scientific Research Innovation Project of Hunan Province(CX20220021)

摘要:

多尺度结构是下一代结构轻量化的核心方向之一,在航空航天领域有巨大的发展潜力和应用价值。面向多尺度结构的拓扑优化技术得到了广泛研究,取得了长足的进步和诸多成果,但将其应用于实际工程仍有不少挑战和难题。首先介绍了单尺度结构拓扑优化的主流方法和特点、均匀化方法的基本思想和特殊地位、多尺度结构拓扑优化方法和单尺度结构拓扑优化方法的关系。然后以结构分布特征作为分类标准,系统地梳理了周期、功能梯度和异构3种分布形式的多尺度结构的拓扑优化设计方法。接着针对多尺度结构拓扑优化中的结构连通性、多尺度方法适用性和机器学习技术的应用等重要问题的相关研究进展进行了综述和讨论。最后对领域亟待突破的问题及重要研究方向进行了总结和展望。

关键词: 轻量化结构, 多尺度, 结构优化, 均匀化方法, 机器学习

Abstract:

Multi-scale structure is one of the important directions of next-generation structural lightweighting and has great potential and strategic value for application in the aerospace field. Topology optimization techniques for multi-scale structures have also been extensively researched, leading to great progress and achievements. However, many challenges and difficulties remain before realizing actual engineering practice. This paper firstly introduces the mainstream methods and characteristics of single-scale structural topology optimization, the basic idea and special position of homogenization methods, as well as the relationship between topology optimization methods for multi-scale structures and topology optimization methods for single-scale structures. Then, using the structural distribution characteristics as the classification criteria, this paper systematically sorted out the topology optimization design methods for three distribution forms of multi-scale structures: periodic, functional gradient and heterogeneous. For the important issues, including structural connectivity, applicability of multi-scale methods, and application of machine learning technology in multi-scale structural topology optimization, relevant research progresses are reviewed and discussed. Finally, pressing challenges and important research directions in the field are summarized and prospected.

Key words: lightweight structures, multi-scale, structural optimization, homogenization methods, machine learning

中图分类号: