Consistent immersed volumetric Nitsche methods for composite analysis

Jiarui Wang, Guohua Zhou, Michael Charles Hillman, Anna Madra, Yuri Bazilevs, Jing Du, Kangning Su

Research output: Contribution to journalArticlepeer-review

Abstract

Generating quality body-fitting meshes for complex composite microstructures is a non-trivial task. In particular, micro-CT images of composites can contain numerous irregularly-shaped inclusions. Among the methods available, immersed boundary methods that discretize bodies independently provide potential for tackling these types of problems since a matching discretization is not needed. However, these techniques still entail the explicit parameterization of the interfaces, which may be considerable in number. In this work, immersed volumetric Nitsche methods are developed in order to avoid the difficulty of generating body fitting meshes for composite materials with complicated microstructures, and overcome the issues in the surface-type methods. These approaches are developed using Nitsche's techniques to enforce volumetric continuity between the inclusion and background domains. It is shown that the proposed weak forms are fully consistent with the strong form of the composite problem. The present approach permits C0 approximations for the foreground discretization, and C1 approximations for the background. The effectiveness of these methods is demonstrated by solving homogeneous and inhomogeneous composite benchmark problems, where it is shown that the non-symmetric version of Nitsche's approach is the most robust in all settings.

Original languageEnglish (US)
Article number114042
JournalComputer Methods in Applied Mechanics and Engineering
Volume385
DOIs
StatePublished - Nov 1 2021

All Science Journal Classification (ASJC) codes

  • Computational Mechanics
  • Mechanics of Materials
  • Mechanical Engineering
  • Physics and Astronomy(all)
  • Computer Science Applications

Fingerprint

Dive into the research topics of 'Consistent immersed volumetric Nitsche methods for composite analysis'. Together they form a unique fingerprint.

Cite this