Solving singularity problems in unbounded domains by coupling of natural BEM and composite grid FEM (Q5932820)

From MaRDI portal
scientific article; zbMATH DE number 1607452
Language Label Description Also known as
English
Solving singularity problems in unbounded domains by coupling of natural BEM and composite grid FEM
scientific article; zbMATH DE number 1607452

    Statements

    Solving singularity problems in unbounded domains by coupling of natural BEM and composite grid FEM (English)
    0 references
    0 references
    0 references
    7 February 2002
    0 references
    The authors discuss the fast adaptive composite grid (FAC) method combined with a coupled finite element-boundary element procedure for approximating the solution of the Laplace equation (with a Dirichlet boundary condition) in an exterior domain. The coupling uses a circular auxiliary boundary to derive a variational problem in a bounded subdomain of the exterior domain. A composite grid is then used to discretize the finite element space and the discrete problem is solved by a FAC-steepest descent algorithm. The method handles effectively the singularities caused by the boundary of the exterior domain and has the advantage that no boundary integral equation associated with the unbounded subdomain is solved, but instead only certain singular integrals are computed. Error estimates and convergence for the method are proved. Some numerical results are presented.
    0 references
    0 references
    natural boundary reduction
    0 references
    composite grid
    0 references
    iteration algorithm
    0 references
    fast adaptive composite grid method
    0 references
    steepest descent algorithm
    0 references
    error estimates
    0 references
    Laplace equation
    0 references
    exterior domain
    0 references
    convergence
    0 references
    numerical results
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references