Error estimates for two-phase Stefan problems in several space variables. II: Nonlinear flux conditions (Q1084846)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Error estimates for two-phase Stefan problems in several space variables. II: Nonlinear flux conditions
scientific article

    Statements

    Error estimates for two-phase Stefan problems in several space variables. II: Nonlinear flux conditions (English)
    0 references
    0 references
    0 references
    1985
    0 references
    [For part I see ibid. 22, 457-499 (1985; reviewed above)] We derive \(L^ 2\)-error-estimates for multidimensional two-phase Stefan problems involving further nonlinearities such as nonlinear flux conditions. We approximate the enthalpy formulation by a regularization procedure combined with a \(C^ 0\)-piecewise linear finite element scheme in space, and the implicit Euler scheme in time. Under some restrictions on the initial datum and on the finite element mesh, we obtain an \(L^ 2\)-rate of convergence of order \(\epsilon^{1/2}\) for regularized problems, and an \(L^ 2\)-rate of convergence of order (h\(| \log h|^ k+\tau +\epsilon)^{1/2}\) for the discrete problems \((k=0,1)\). These estimates lead to the choice \(h\sim \epsilon \sim \tau\) and yield a global \(L^ 2\)-rate essentially of order \(h^{1/2}.\) The a priori relationship between the approximation parameters allows the discrete problem to satisfy a maximum principle (without a lumping procedure), from which a priori bounds and monotonicity properties of the discrete solutions are shown.
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    error-estimates
    0 references
    two-phase Stefan problems
    0 references
    nonlinear flux conditions
    0 references
    regularization
    0 references
    linear finite element
    0 references
    implicit Euler
    0 references
    rate of convergence
    0 references
    maximum principle
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references