Scholarlink

<p>This work addresses a level set-based framework for connectivity constraints in shape and topology optimization, using the concept of topological derivative. By controlling the presence of disconnected material components in optimal designs, the goal is to ensure that optimal solutions are also physically realizable in terms of additive manufacturing and of its final application. The proposed strategy is an extension of the continuous spectral graph approach, originally introduced in Donoso et al. (Struct Multidisc Optim 66(4):71, 2023. https://doi.org/10.1007/s00158-023-03526-8) in the density-based framework, which identifies and enforces connectivity based on the spectrum of a two-phase differential operator. The effectiveness of the strategy is demonstrated through several benchmark problems in structural optimization, in two and three dimensions. Numerical challenges and limitations such as filtering strategies and parameter dependence are illustrated and successful simulations are obtained with either the addition of perimeter penalization or a dilation operator.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

On a level set and topological derivative-based strategy for connectivity constraints in topology optimization

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年5月21日

查看全文
  • Giovanna C. Andrade,
  • Alberto Donoso,
  • David Ruiz,
  • Alex Ferrer

摘要

This work addresses a level set-based framework for connectivity constraints in shape and topology optimization, using the concept of topological derivative. By controlling the presence of disconnected material components in optimal designs, the goal is to ensure that optimal solutions are also physically realizable in terms of additive manufacturing and of its final application. The proposed strategy is an extension of the continuous spectral graph approach, originally introduced in Donoso et al. (Struct Multidisc Optim 66(4):71, 2023. https://doi.org/10.1007/s00158-023-03526-8) in the density-based framework, which identifies and enforces connectivity based on the spectrum of a two-phase differential operator. The effectiveness of the strategy is demonstrated through several benchmark problems in structural optimization, in two and three dimensions. Numerical challenges and limitations such as filtering strategies and parameter dependence are illustrated and successful simulations are obtained with either the addition of perimeter penalization or a dilation operator.