Scholarlink

<p>This paper proposes a topology optimization approach that allows human augmentation in the optimization process. The approach aims to leverage the experience and expertise of designers, enabling them to make real-time adjustments and have direct control over the topology optimization process. It allows faster topology optimization that meets design requirements, obviating the need for repeated modifications to the optimization formulation. To achieve this goal, a conversion strategy has been developed between the Solid Isotropic Material with Penalization (SIMP) method, which employs implicit geometric descriptions, and the Moving Morphable Components/Voids (MMC/MMV) method, which utilizes explicit geometric descriptions. This strategy establishes structural members for the pixel-based SIMP, thereby facilitating intuitive control over the structural topology and geometric parameters. Combined with human augmentation, the structural members can be added, adjusted, and removed. These treatments have the potential to improve structural buckling resistance, reduce stress concentration, and adjust structural complexity. Additionally, the proposed approach that combines multiple geometry descriptions can directly generate CAD models from SIMP-based topology optimization results, thereby enhancing the practicality of topology optimization. The effectiveness of this method is demonstrated through multiple examples presented in this paper.</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

Human-augmented topology optimization design with multi-framework intervention

  • 期刊论文

  • 发表日期：

    2025年1月22日

查看全文
  • Weisheng Zhang,
  • Xiaoyu Zhuang,
  • Xu Guo,
  • Sung-Kie Youn

摘要

This paper proposes a topology optimization approach that allows human augmentation in the optimization process. The approach aims to leverage the experience and expertise of designers, enabling them to make real-time adjustments and have direct control over the topology optimization process. It allows faster topology optimization that meets design requirements, obviating the need for repeated modifications to the optimization formulation. To achieve this goal, a conversion strategy has been developed between the Solid Isotropic Material with Penalization (SIMP) method, which employs implicit geometric descriptions, and the Moving Morphable Components/Voids (MMC/MMV) method, which utilizes explicit geometric descriptions. This strategy establishes structural members for the pixel-based SIMP, thereby facilitating intuitive control over the structural topology and geometric parameters. Combined with human augmentation, the structural members can be added, adjusted, and removed. These treatments have the potential to improve structural buckling resistance, reduce stress concentration, and adjust structural complexity. Additionally, the proposed approach that combines multiple geometry descriptions can directly generate CAD models from SIMP-based topology optimization results, thereby enhancing the practicality of topology optimization. The effectiveness of this method is demonstrated through multiple examples presented in this paper.