Skip to main navigation Skip to search Skip to main content

A high-order three-scale reduced asymptotic approach for thermo-mechanical problems of nonlinear heterogeneous materials with multiple spatial scales

Research output: Contribution to journalArticlepeer-review

Abstract

An effective high-order three-scale reduced asymptotic (HTRA) approach is developed to simulate the coupled thermo-mechanical problems of nonlinear heterogeneous materials with multiple spatial scales. In this paper, the heterogeneous structures are constructed by periodical layout of unit cells on microscopic and mesoscopic scale. At first, the linear and nonlinear higher-order local cell functions defined at microscale and mesoscale are derived by multiscale asymptotic expansion approach. Further, two types of homogenized parameters are introduced, and related nonlinear homogenized solutions are obtained on the macroscopic domain. Then, the high-order three-scale temperature and displacement solutions are established by assembling the various multiscale local solutions and homogenization solutions. The key characteristic of the presented methods is an efficient reduced form for evaluating higher-order local cell problems, and hence greatly reducing the calculating amount in comparison to direct numerical simulations. Also, the new high-order nonlinear homogenization solutions that do not require higher-order continuity of the macroscale solutions. Finally, the effectiveness and accuracy of the multiscale algorithms are verified by some typical numerical examples.

Original languageEnglish
Article number103905
JournalEuropean Journal of Mechanics, A/Solids
Volume80
DOIs
StatePublished - 1 Mar 2020

Keywords

  • HTRA algorithms
  • Multiple spatial scales
  • Nonlinear thermo-mechanical problems
  • Reduced-order homogenization
  • Three-scale formulae

Fingerprint

Dive into the research topics of 'A high-order three-scale reduced asymptotic approach for thermo-mechanical problems of nonlinear heterogeneous materials with multiple spatial scales'. Together they form a unique fingerprint.

Cite this