Quantized tensor-structured finite elements for second-order elliptic PDEs in two dimensions (Q1692304): Difference between revisions
From MaRDI portal
Changed an Item |
Changed an Item |
||
Property / describes a project that uses | |||
Property / describes a project that uses: TT Toolbox / rank | |||
Normal rank |
Revision as of 02:17, 29 February 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Quantized tensor-structured finite elements for second-order elliptic PDEs in two dimensions |
scientific article |
Statements
Quantized tensor-structured finite elements for second-order elliptic PDEs in two dimensions (English)
0 references
26 January 2018
0 references
The authors are concerned with the tensor structured solutions to second-order linear elliptic boundary value problems in two dimensions. A dense functional background is first introduced. Namely, the authors review the tensor format for the tensor structured solution of partial differential equations, weighted Sobolev spaces and some countably normed analytic classes, uniform tensor-product partitions and the corresponding finite element (FE) spaces for functions defined on unit quadrant, etc. Then, they analyze the QTT (quantized-tensor-train) parametrization of the FE spaces in order to approximate the analytic functions with a point singularity at the origin. The main result consists in the exponential convergence of QTT-FE approximations. Two numerical experiments are carried out. In the first one the domain is an L-shaped one and in the second one the domain has a cut. In both the exponential convergence rate of QTT-FE approximation is noticed. Five appendices contain auxiliary results used along the paper.
0 references
second-order elliptic problem
0 references
curvilinear polygon
0 references
point singularity
0 references
analytic function
0 references
finite element
0 references
tensor-product mesh
0 references
tensor-structured approximation
0 references
quantized-tensor-train decomposition
0 references
exponential convergence
0 references
numerical experiment
0 references