Finite element methods are not always optimal

DOI10.1016/0196-8858(87)90028-5MaRDI QIDQ1096364zbMATH OpenOpenAlexFDO

Authors A. G. Werschulz

Publication date 1987

Published in Advances in Applied Mathematics (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/0196-8858(87)90028-5

zbMATH Keywords

finite element methodGalerkin methodN-widths of compactsoptimal algorithmsSobolev spacesspline algorithm

Mathematics Subject Classification ID

Boundary value problems for higher-order elliptic equations (35J40) Finite element, Rayleigh-Ritz and Galerkin methods for boundary value problems involving PDEs (65N30) Analysis of algorithms and problem complexity (68Q25)

Review text

It is a well known fact that the theory of optimal algorithms for elliptic problems is connected with the theory of N-widths of compacts in Sobolev spaces. The author considers a regular elliptic 2n-th order problem \(Lu=f\) with \(f\in H^ r\), \(r>-m\). He shows that the finite element method with piecewise polynomials of degree k is not asymptotically optimal if \(k<2m-1+r\). A special spline algorithm (a generalized Galerkin method) is recommended in this case. In the complexity analysis the author considers the number of unknowns only as a cast of the algorithm. Usually the computational work to solve grid systems with desired accuracy is considered as the key problem.

Recommendations

• scientific article; zbMATH DE number 495827
• Finite element approximation of an optimal design problem
• A finite element alternative to infinite elements
• scientific article; zbMATH DE number 868038
• Optimal finite element methods for interface problems
• The finite element method. Extensions and alternatives
• Publication:3026464
• scientific article; zbMATH DE number 3898167
• scientific article; zbMATH DE number 3878292

Cites work

• scientific article; zbMATH DE number 3423512 (Why is no real title available?)
• scientific article; zbMATH DE number 3688714 (Why is no real title available?)
• scientific article; zbMATH DE number 3696612 (Why is no real title available?)
• scientific article; zbMATH DE number 3520921 (Why is no real title available?)
• scientific article; zbMATH DE number 3525575 (Why is no real title available?)
• scientific article; zbMATH DE number 3561771 (Why is no real title available?)
• scientific article; zbMATH DE number 3601500 (Why is no real title available?)
• scientific article; zbMATH DE number 3438337 (Why is no real title available?)
• Interpolation theory over curved elements, with applications to finite element methods
• Optimal Error Properties of Finite Element Methods for Second Order Elliptic Dirichlet Problems

Cited in

(12)

• Complexity of differential and integral equations
• Galerkin method and optimal error algorithms in Hilbert spaces
• Optimal algorithms for a problem of optimal control
• Optimal approximation of elliptic problems by linear and nonlinear mappings. I
• On the utilisation of the theory of elements finish in problems of optimal order
• scientific article; zbMATH DE number 3898167 (Why is no real title available?)
• An Optimal Adaptive Finite Element Method
• scientific article; zbMATH DE number 4186297 (Why is no real title available?)
• On optimal recovery and information complexity in numerical differentiation and summation
• Optimization of methods for mixed derivatives of bivariate functions
• Regularized summation of Fourier-Laguerre series in the weighted sup-norm
• Numerical differentiation of functions on the half-line in a weighted uniform metric

This page was built for publication: Finite element methods are not always optimal

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1096364)