Premium
A DIFFUSE ELEMENT‐FINITE ELEMENT TECHNIQUE FOR TRANSIENT COUPLED ANALYSIS
Author(s) -
MODARESSI H.,
AUBERT P.
Publication year - 1996
Publication title -
international journal for numerical methods in engineering
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.421
H-Index - 168
eISSN - 1097-0207
pISSN - 0029-5981
DOI - 10.1002/(sici)1097-0207(19961130)39:22<3809::aid-nme26>3.0.co;2-2
Subject(s) - finite element method , mixed finite element method , extended finite element method , galerkin method , smoothed finite element method , finite element limit analysis , transient (computer programming) , weighting , hp fem , stability (learning theory) , mathematics , mathematical analysis , porous medium , computer science , boundary knot method , physics , materials science , structural engineering , engineering , porosity , boundary element method , operating system , machine learning , acoustics , composite material
A coupled Diffuse Element–Finite Element technique (DEM–FEM) for transient coupled‐field analysis is proposed. The method combines the standard finite element method and a diffuse element technique (also called Element–Free Galerkin Method). The resulting approach takes advantage of the capabilities of both of these numerical techniques. For instance, different approximations can very easily be introduced for different fields. Moreover, the finite element mesh provides the cells for numerical integration of the DEM equations. The matrix formulation is derived from the variational formulation of the coupled equation system. The case of the coupled hydromechnical formulation of two‐phase porous media is studied as an applied example. It is shown that, as only few pore‐pressure nodes are necessary, the stability requirements for almost impervious porous media are satisfied. The precision and stability of the obtained formulation was studied, and using several examples, we investigated the role of different parameters on the accuracy of results, such as the influence domain, the weighting function type, or the number of nodes.