Balancing truncation and round-off errors in FEM: one-dimensional analysis

DOI10.1016/J.CAM.2020.113219zbMATH Open1461.65220arXiv1912.08004OpenAlexW3092182011MaRDI QIDQ2222170FDOQ2222170

Authors: Jie Liu, Matthias Möller, Henk M. Schuttelaars 

Publication date: 3 February 2021

Published in: Journal of Computational and Applied Mathematics (Search for Journal in Brave)

Abstract: In finite element methods (FEMs), the accuracy of the solution cannot increase indefinitely because the round-off error increases when the number of degrees of freedom (DoFs) is large enough. This means that the accuracy that can be reached is limited. A priori information of the highest attainable accuracy is therefore of great interest. In this paper, we devise an innovative method to obtain the highest attainable accuracy. In this method, the truncation error is extrapolated when it converges at the analytical rate, for which only a few primary $h$-refinements are required, and the bound of the round-off error is provided through extensive numerical experiments. The highest attainable accuracy is obtained by minimizing the sum of these two types of errors. We validate this method using a one-dimensional Helmholtz equation in space. It shows that the highest attainable accuracy can be accurately predicted, and the CPU time required is much less compared with that using the successive $h$-refinement.

Full work available at URL: https://arxiv.org/abs/1912.08004

Recommendations

• The use of the local truncation error to improve arbitrary-order finite elements for the linear wave and heat equations
• scientific article; zbMATH DE number 3881795
• scientific article; zbMATH DE number 4153824
• Round-off errors in the stiffness equation
• Estimating manipulation errors in finite element analysis. II

zbMATH Keywords

finite element methoda posteriori error estimationround-off error\(hp\)-refinement strategyoptimal number of degrees of freedom

Mathematics Subject Classification ID

Roundoff error (65G50) Finite element, Rayleigh-Ritz, Galerkin and collocation methods for ordinary differential equations (65L60)

Cites Work

• Algorithm 832
• The h-p version of the finite element method. I. The basic approximation results
• Title not available (Why is that?)
• Mixed finite element methods and applications
• Understanding and implementing the finite element method.
• Title not available (Why is that?)
• A survey on direct solvers for Galerkin methods
• Title not available (Why is that?)
• Numerical Methods for Ordinary Differential Equations
• Derivative calculation from finite element solutions
• The conjugate gradient method
• The deal.II library, version 9.0
• On roundoff error growth in elliptic problems

Cited In (1)

• Textbook efficiency: massively parallel matrix-free multigrid for the Stokes system

Uses Software

• UMFPACK

This page was built for publication: Balancing truncation and round-off errors in FEM: one-dimensional analysis

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2222170)