Numerical solution of singular boundary value problems by invariant imbedding (Q799080)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Numerical solution of singular boundary value problems by invariant imbedding
scientific article

    Statements

    Numerical solution of singular boundary value problems by invariant imbedding (English)
    0 references
    0 references
    0 references
    1984
    0 references
    The authors adapt the invariant embedding method to apply to approximation of the solution of a singular two-point boundary value problem for the differential equation \((i)\quad f_ 0(x)y''+f_ 1(x)y'+f_ 2(x)y=Q(x),\quad x_ 0\leq x\leq b,\quad (ii)\quad y(x_ 0)=\alpha,\quad y(b)=\beta\) where \(f_ i(x)\) are analytic and \(x_ 0\) is a regular singular point. The procedure is to approximate the solution on a small interval \([x_ 0,\delta]\) by a series expansion and then to set up and solve a regular two-point boundary value problem on the interval [\(\delta\),b] by means of an invariant embedding method described by \textit{M. R. Scott} [Numer. Solut. Bound. Value Probl. ordin. differ. Equat., Proc. Symp. Baltimore 1974, 89-146 (1975; Zbl 0335.65032)]. Results of several trial computations using a DEC-1090 are presented which show that the method compares favorably to other applicable algorithms.
    0 references
    numerical examples
    0 references
    comparison of methods
    0 references
    invariant embedding method
    0 references
    regular singular point
    0 references
    series expansion
    0 references

    Identifiers