Scholarlink

<p>Fast prediction of microstructural responses based on realistic material topology is vital for linking process, structure, and properties. This work presents a digital framework for metallic materials using microscale features. We explore deep learning for two primary goals: (1) segmenting experimental images to extract microstructural topology, translated into spatial property distributions; and (2) learning mappings from digital microstructures to mechanical fields using physics-informed operator learning. Loss functions are formulated using discretized weak or strong forms, and boundary conditions-Dirichlet and periodic-are embedded in the network. Input space is reduced to focus on key features of 2D and 3D materials, and generalization to varying loads and input topologies are demonstrated. Compared to FEM and FFT solvers, our models yield errors under 1–5% for averaged quantities and are over 1000× faster during 3D inference.</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

Digitalizing metallic materials from image segmentation to multiscale solutions via physics informed operator learning

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年8月13日

查看全文
  • Shahed Rezaei,
  • Kianoosh Taghikhani,
  • Alexandre Viardin,
  • Reza Najian Asl,
  • Ali Harandi,
  • Nikhil Vijay Jagtap,
  • David Bailly,
  • Hannah Naber,
  • Alexander Gramlich,
  • Tim Brepols,
  • Mustapha Abouridouane,
  • Ulrich Krupp,
  • Thomas Bergs,
  • Markus Apel

摘要

Fast prediction of microstructural responses based on realistic material topology is vital for linking process, structure, and properties. This work presents a digital framework for metallic materials using microscale features. We explore deep learning for two primary goals: (1) segmenting experimental images to extract microstructural topology, translated into spatial property distributions; and (2) learning mappings from digital microstructures to mechanical fields using physics-informed operator learning. Loss functions are formulated using discretized weak or strong forms, and boundary conditions-Dirichlet and periodic-are embedded in the network. Input space is reduced to focus on key features of 2D and 3D materials, and generalization to varying loads and input topologies are demonstrated. Compared to FEM and FFT solvers, our models yield errors under 1–5% for averaged quantities and are over 1000× faster during 3D inference.