A variational multiscale a posteriori error estimation method for mixed form of nearly incompressible elasticity
نویسندگان
چکیده
This paper presents an error estimation framework for a mixed displacement–pressure finite element method for nearly incompressible elasticity. The proposed method is based on Variational Multiscale (VMS) concepts, wherein the displacement field is decomposed into coarse scales that can be resolved by a given finite element mesh and fine scales that are beyond the resolution capacity of the mesh. Variational projection of fine scales onto the coarse-scale space via variational embedding of the fine-scale solution into the coarse-scale formulation leads to the stabilized method with two major attributes: first, it is free of volumetric locking and, second, it accommodates arbitrary combinations of interpolation functions for the displacement and pressure fields. This VMS-based stabilized method is equipped with naturally derived error estimators and offers various options for numerical computation of the error. Specifically, two error estimators are explored. The first method employs an element-based strategy and a representation of error via a fine-scale error equation defined over element interiors which is evaluated by a direct post-solution evaluation. This quantity when combined with the global pollution error results in a simple explicit error estimator. The second method involves solving the fine-scale error equation through localization to overlapping patches spread across the domain, thereby leading to an implicit calculation of the local error. This implicit calculation when combined with the global pollution error results in an implicit error estimator. The performance of the stabilized method and the error estimators is investigated through numerical convergence tests conducted for two model problems on uniform and distorted meshes. The sharpness and robustness of the estimators is shown to be consistent across the test cases employed. 2010 Elsevier B.V. All rights reserved.
منابع مشابه
A framework for residual-based stabilization of incompressible finite elasticity: Stabilized formulations and methods for linear triangles and tetrahedra
A new Variational Multiscale framework for finite strain incompressible elasticity is presented. Significant contributions in this work are: (i) a systematic derivation of multiscale formulations that include the classical F method as a particular subclass, (ii) an error estimation procedure for nonlinear elasticity that emanates naturally from within the present multiscale framework, and (iii)...
متن کاملA Stabilized Mixed Finite Element Method for Nearly Incompressible Elasticity
We present a new multiscale/stabilized finite element method for compressible and incompressible elasticity. The multiscale method arises from a decomposition of the displacement field into coarse (resolved) and fine (unresolved) scales. The resulting stabilizedmixed form consistently represents the fine computational scales in the solution and thus possesses higher coarse mesh accuracy. The en...
متن کاملOn finite element formulations for nearly incompressible linear elasticity
In this paper we present a mixed stabilized finite element formulation that does not lock and also does not exhibit unphysical oscillations near the incompressible limit. The new mixed formulation is based on a multiscale variational principle and is presented in two different forms. In the first form the displacement field is decomposed into two scales, coarse-scale and fine-scale, and the fin...
متن کاملA posteriori error estimators for mixed finite element methods in linear elasticity
Three a posteriori error estimators for PEERS and BDMS elements in linear elasticity are presented: one residual error estimator and two estimators based on the solution of auxiliary local problems with different boundary conditions. All of them are reliable and efficient with respect to the standard norm and furthermore robust for nearly incompressible materials.
متن کاملLocking - free adaptive mixed nite
Mixed nite element methods such as PEERS or the BDMS methods are designed to avoid locking for nearly incompressible materials in plane elasticity. In this paper, we establish a robust adaptive mesh-reening algorithm that is rigorously based on a reliable and eecient a posteriori error estimate. Numerical evidence is provided for the-independence of the constants in the a posteriori error bound...
متن کامل