A multiscale method for porous microstructures

Abstract

In this paper we develop a multiscale method to solve problems in complicated porous microstructures with Neumann boundary conditions. By using a coarse-grid quasi-interpolation operator to define a fine detail space and local orthogonal decomposition, we construct multiscale corrections to coarse-grid basis functions with microstructure. By truncating the corrector functions we are able to make a computationally efficient scheme. Error results and analysis are presented. A key component of this analysis is the investigation of the Poincaré and inverse inequality constants in perforated domains as they may contain micro-structural information. Using first a theoretical method based on extensions of functions and then constructive method originally developed for weighted Poincaré inequalities, we are able to obtain estimates on Poincaré constants with respect to scale and separation length of the pores. Finally, two numerical examples are presented to verify our estimates.