2023•Unpublished venueRequires access

Numerical Analysis of the Finite Element Method

Patrick Ciarlet, Eric Lunéville

Open publisher page 106 citations

Abstract

In this chapter, the authors examine some aspects of finite element convergence analysis. They introduce the mesh refinement technique that improves the quality of the calculated approximate solution. One can implement mesh refinement techniques from a priori information, knowledge of the singularity of the solution, for example, from a posteriori information from the calculated solution, obtained by error estimators. In the error analysis, situations that are more complicated but which are often encountered in practice were excluded, in order to simplify things: the non-polyhedral open subset (i.e. curvilinear polyhedra) case and the data approximation case. Finally, the authors discuss approximation methods involving approximate calculations of the data and/or of the (bi)linear forms – for example, via numerical quadrature formulas – as well as external, or non-conforming, methods, for which the approximation spaces are not included within the solution and/or test-function spaces.

About this research paper

What this paper is about

In this chapter, the authors examine some aspects of finite element convergence analysis. They introduce the mesh refinement technique that improves the quality of the calculated approximate solution. One can implement mesh refinement techniques from a priori information, knowledge of the singularity of the solution, for example, from a posteriori information from the calculated solution, obtained by error estimators. In the error analysis, situations that are more complicated but which are often encountered in practice were excluded, in order to simplify things: the non-polyhedral open subset (i.e. curvilinear polyhedra) case and the data approximation case. Finally, the authors discuss approximation methods involving approximate calculations of the data and/or of the (bi)linear forms – for example, via numerical quadrature formulas – as well as external, or non-conforming, methods, for which the approximation spaces are not included within the solution and/or test-function spaces.

Why it matters

OpenAlex reports 106 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

In this chapter, the authors examine some aspects of finite element convergence analysis. They introduce the mesh refinement technique that improves the quality of the calculated approximate solution. One can implement mesh refinement techniques from a priori information, knowledge of the singularity of the solution, for example, from a posteriori information from the calculated solution, obtained by error estimators. In the error analysis, situations that are more complicated but which are often encountered in practice were excluded, in order to simplify things: the non-polyhedral open subset (i.e. curvilinear polyhedra) case and the data approximation case. Finally, the authors discuss approximation methods involving approximate calculations of the data and/or of the (bi)linear forms – for example, via numerical quadrature formulas – as well as external, or non-conforming, methods, for which the approximation spaces are not included within the solution and/or test-function spaces.

Key concepts: A priori and a posteriori, Finite element method, Polyhedron, Estimator, Curvilinear coordinates, Convergence (economics), Applied mathematics, Mathematics

Related papers

Back to paper searchBrowse research topicsOriginal source
Numerical Analysis of the Finite Element Method — Research Paper | ScholarLens