NANASep 9, 2011

Interior error estimate for periodic homogenization

arXiv:1109.1908119 citationsh-index: 24
Originality Synthesis-oriented
AI Analysis

Provides refined error bounds for numerical homogenization, relevant to analysts working on multiscale PDEs.

This paper improves the error estimate for periodic homogenization in bounded domains from ε^{1/2} to ε in interior subdomains, assuming C^{1,1} boundary and Dirichlet or Neumann conditions. For polygonal/polyhedral domains, global and interior error estimates are also provided.

In a previous article about the homogenization of the classical problem of diff usion in a bounded domain with su ciently smooth boundary we proved that the error is of order $ε^{1/2}$. Now, for an open set with su ciently smooth boundary $C^{1,1}$ and homogeneous Dirichlet or Neuman limits conditions we show that in any open set strongly included in the error is of order $ε$. If the open set $Ω\subset R^n$ is of polygonal (n=2) or polyhedral (n=3) boundary we also give the global and interrior error estimates.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes