Two numerical approximations for a class of weakly singular integral operators (Q1901532): Difference between revisions
From MaRDI portal
Removed claim: reviewed by (P1447): Item:Q194771 |
Changed an Item |
||
Property / reviewed by | |||
Property / reviewed by: Johannes Elschner / rank | |||
Normal rank |
Revision as of 17:56, 10 February 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Two numerical approximations for a class of weakly singular integral operators |
scientific article |
Statements
Two numerical approximations for a class of weakly singular integral operators (English)
0 references
29 April 1996
0 references
The authors study the numerical approximation of weakly singular integral operators on a finite interval by strongly stable discretizations, i.e. each nonzero eigenvalue of the integral operator is approximated by a cluster of eigenvalues of the approximate operators with preservation of algebraic multiplicities and the corresponding approximate maximal invariant subspace converges in gap to the exact one. The first method is based on truncation and numerical quadrature, and the second uses singularity subtraction. Numerical experiments on spectral computations are presented to compare both kinds of approximations, and an application to an electro-hydrodynamic model is discussed.
0 references
numerical experiments
0 references
weakly singular integral operators
0 references
strongly stable discretizations
0 references
eigenvalue
0 references
maximal invariant subspace
0 references
numerical quadrature
0 references
singularity subtraction
0 references
spectral computations
0 references