Error control and adaptivity for heterogeneous multiscale approximations of nonlinear monotone problems

Henning P, Ohlberger M

Research article (journal) | Peer reviewed

Abstract

In this work we introduce and analyse a new adaptive Petrov-Galerkin heterogeneous multiscale finite element method (HMM) for monotone elliptic operators with rapid oscillations. In a general heterogeneous setting we prove convergence of the HMM approximations to the solution of a macroscopic limit equation. The major new contribution of this work is an a-posteriori error estimate for the L2-error between the HMM approximation and the solution of the macroscopic limit equation. The a posteriori error estimate is obtained in a general heterogeneous setting with scale separation without assuming periodicity or stochastic ergodicity. The applicability of the method and the usage of the a posteriori error estimate for adaptive local mesh refinement is demonstrated in numerical experiments. The experimental results underline the applicability of the a posteriori error estimate in non-periodic homogenization settings.

Details about the publication

JournalDiscrete and Continuous Dynamical Systems - Series S
Volume8
Issue1
Page range119-150
StatusPublished
Release year2015
Language in which the publication is writtenEnglish
DOI10.3934/dcdss.2015.8.119
KeywordsA posteriori estimate; HMM; monotone operator; multiscale methods

Authors from the University of Münster

Henning, Patrick
Professorship of Applied Mathematics, especially Numerics (Prof. Ohlberger)
Ohlberger, Mario
Professorship of Applied Mathematics, especially Numerics (Prof. Ohlberger)
Center for Nonlinear Science
Center for Multiscale Theory and Computation