Scholarlink

<p>Structural lightweight is an essential strategy for improving material efficiency, especially in thermomechanical coupling scenarios where structural failure risks are elevated. In such conditions, materials must exhibit high thermal conductivity and superior heat dissipation to enable effective energy transfer while maintaining high stiffness and minimal weight. To address these demands, our research employs a genetic algorithm for pre-optimization, seamlessly integrated with traditional topology optimization techniques, to guide the iterative refinement of macrostructure design. Advanced 3D printing technologies have highlighted the potential of porous materials, known for their ability to significantly reduce weight. Leveraging these advancements, this study focuses on applying parallel multiscale topology optimization. The optimized structure achieves an average temperature reduction of 38.5&#xa0;K compared to traditional designs. Compared to traditional designs, the proposed method integrates a binary-encoded genetic algorithm (GA) for pre-optimization, which generates high-quality initial structures with minimal computational cost. Unlike the BESO method, the GA’s efficiency allows pre-optimization to be completed rapidly and helps find more promising initial structures, ultimately improving the quality of the final optimization result, consuming negligible computational resources. The goal is to design lightweight yet robust structures with high thermal conductivity, effective heat dissipation, and rigidity.</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

Multi-objective collaborative rapid dual-scale topology optimization based on thermomechanical coupling analysis

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年5月15日

查看全文
  • Shuo Feng,
  • Lei Xia,
  • Yuhao Yang,
  • Zhen Wang,
  • Xuan Zhang,
  • Qidong Han

摘要

Structural lightweight is an essential strategy for improving material efficiency, especially in thermomechanical coupling scenarios where structural failure risks are elevated. In such conditions, materials must exhibit high thermal conductivity and superior heat dissipation to enable effective energy transfer while maintaining high stiffness and minimal weight. To address these demands, our research employs a genetic algorithm for pre-optimization, seamlessly integrated with traditional topology optimization techniques, to guide the iterative refinement of macrostructure design. Advanced 3D printing technologies have highlighted the potential of porous materials, known for their ability to significantly reduce weight. Leveraging these advancements, this study focuses on applying parallel multiscale topology optimization. The optimized structure achieves an average temperature reduction of 38.5 K compared to traditional designs. Compared to traditional designs, the proposed method integrates a binary-encoded genetic algorithm (GA) for pre-optimization, which generates high-quality initial structures with minimal computational cost. Unlike the BESO method, the GA’s efficiency allows pre-optimization to be completed rapidly and helps find more promising initial structures, ultimately improving the quality of the final optimization result, consuming negligible computational resources. The goal is to design lightweight yet robust structures with high thermal conductivity, effective heat dissipation, and rigidity.