Boundary contraction solution of the Neumann and mixed boundary value problems of the Laplace equation

Abstract

The boundary contraction method is generalized in such a way that it is applicable to the Neumann and mixed boundary value problems over regions of irregular shape. Various variants of mixed boundary value problems can be solved numerically in a unified way with mesh points fewer than those required of ADI and SOR. The method is not iterative and therefore does not require the positive definiteness of eigenvalues which is the necessary condition of the stability of ADI and SOR. The method is also applicable to exterior problems. Thus the applicability of the contraction method to problems of practical importance is substantially improved.