No label defined (Q3750043)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 3986547
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | No label defined |
scientific article; zbMATH DE number 3986547 |
Statements
1985
0 references
block structure
0 references
elimination methods
0 references
eigenvalues
0 references
parallel computation
0 references
convergence
0 references
singular perturbation problems
0 references
0.89163864
0 references
0.87763226
0 references
0.87298596
0 references
0.8726569
0 references
0.8720786
0 references
0.86859286
0 references
0.8685869
0 references
The author deals with the equation (1) \(Au=f\), where A is a linear operator, which allows a block representation. The block structure of the operator A make easy to use elimination methods for the solution of (1) and to localizate the eigenvalues of A. One suggests the algorithm for the solution of equations in infinite dimensional spaces, using parallel computation on a multiprocessor computer. The important property of the algorithm is, that the convergence of the approximation by the fixed number of blocks allows to reach the solution of (1) with arbitrary accuracy on a fixed bounded number of processors. Analytical aspects of the block representation of operators and the solvability for singular perturbation problems with different type of perturbations are treated. In {\S} 1 the equation \(A(\epsilon)u=f\) is solved with \(A(\epsilon)=A_ 0+A_ 1\epsilon +\epsilon^ 2A_ 2+...\), where \(A_ i\) are matrices, u, f vectors, \(\epsilon\) a small parameter, \(A_ 0\) has a nontrivial kernel, and the series \(\sum \epsilon^ iA_ i\) converges. In {\S} 2 one deals with the equation \((A_ 0+\epsilon A_ 1)u=f\), where \(A_ 0\) is a bounded and \(A_ 1\) an unbounded linear operator, \(\epsilon\) is a perturbation parameter. In {\S} 3 the iterative procedure and the parallelism are applied to a boundary value problem for a second order ordinary differential equation. Three existence theorems are presented.
0 references