An Improved Interpolation Error Estimate from a New Taylor-Like Formula: Application to Finite Element Method

Joël Chaskalovic, Franck Assous

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In this paper, we propose an improved interpolation error estimate based on a new Taylor-like formula, which we apply to the finite element method. We first present a new first-order and second-order expansion formula with a reduced remainder. Then, we derive a new interpolation error estimate in W1,p. We compare this with the classical error estimates based on the standard Taylor formula and the corresponding interpolation error estimate derived from the mean value theorem. We illustrate, with examples, the significant reduction this yields in finite element computation costs.

Original languageEnglish
Title of host publicationNew Trends in the Applications of Differential Equations in Sciences - NTADES 2024
EditorsAngela Slavova
PublisherSpringer
Pages277-290
Number of pages14
ISBN (Print)9783031833977
DOIs
StatePublished - 2025
Externally publishedYes
Event11th International Conference on New Trends in the Applications of Differential Equations in Sciences, NTADES 2024 - Saints Constantine and Helena, Bulgaria
Duration: 7 Jul 202410 Jul 2024

Publication series

NameSpringer Proceedings in Mathematics and Statistics
Volume488
ISSN (Print)2194-1009
ISSN (Electronic)2194-1017

Conference

Conference11th International Conference on New Trends in the Applications of Differential Equations in Sciences, NTADES 2024
Country/TerritoryBulgaria
CitySaints Constantine and Helena
Period7/07/2410/07/24

Bibliographical note

Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

Keywords

  • Approximation error
  • Finite elements
  • Interpolation error
  • Taylor-like formula
  • Taylor’s theorem

Fingerprint

Dive into the research topics of 'An Improved Interpolation Error Estimate from a New Taylor-Like Formula: Application to Finite Element Method'. Together they form a unique fingerprint.

Cite this