A generalized multiscale finite element method for elastic wave propagation in fractured media

Eric T. Chung, Yalchin R. Efendiev, Richard L. Gibson, Maria Vasilyeva

Research output: Contribution to journalArticlepeer-review

36 Scopus citations

Abstract

In this paper, we consider elastic wave propagation in fractured media applying a linear-slip model to represent the effects of fractures on the wavefield. Fractured media, typically, are highly heterogeneous due to multiple length scales. Direct numerical simulations for wave propagation in highly heterogeneous fractured media can be computationally expensive and require some type of model reduction. We develop a multiscale model reduction technique that captures the complex nature of the media (heterogeneities and fractures) in the coarse scale system. The proposed method is based on the generalized multiscale finite element method, where the multiscale basis functions are constructed to capture the fine-scale information of the heterogeneous, fractured media and effectively reduce the degrees of freedom. These multiscale basis functions are coupled via the interior penalty discontinuous Galerkin method, which provides a block-diagonal mass matrix. The latter is needed for fast computation in an explicit time discretization, which is used in our simulations. Numerical results are presented to show the performance of the presented multiscale method for fractured media. We consider several cases where fractured media contain fractures of multiple lengths. Our numerical results show that the proposed reduced-order models can provide accurate approximations for the fine-scale solution.
Original languageEnglish (US)
Pages (from-to)163-182
Number of pages20
JournalGEM - International Journal on Geomathematics
Volume7
Issue number2
DOIs
StatePublished - Feb 26 2016

Bibliographical note

KAUST Repository Item: Exported on 2020-10-01

Fingerprint

Dive into the research topics of 'A generalized multiscale finite element method for elastic wave propagation in fractured media'. Together they form a unique fingerprint.

Cite this