*STOCHASTIC convergence, *LEAST squares, *FINITE element method, *APPROXIMATION theory, *FIXED point theory, *BOUNDARY value problems, *NUMERICAL analysis, *ERROR analysis in mathematics
Abstract
Abstract: In this paper, a least-squares finite element method for second-order two-point boundary value problems is considered. The problem is recast as a first-order system. Standard and improved optimal error estimates in maximum-norms are established. Superconvergence estimates at interelement, Lobatto, and Gauss points are developed. Numerical experiments are given to illustrate theoretical results. [Copyright &y& Elsevier]
Abstract: This paper deals with the numerical simulation of the steady state two dimensional window Josephson junctions by finite element method. The model is represented by a sine-Gordon type composite PDE problem. Convergence and error analysis of the finite element approximation for this semilinear problem are presented. An efficient and reliable Newton-preconditioned conjugate gradient algorithm is proposed to solve the resulting nonlinear discrete system. Regular solution branches are computed using a simple continuation scheme. Numerical results associated with interesting physical phenomena are reported. Interface relaxation methods, which by taking advantage of special properties of the composite PDE, can further reduce the overall computational cost are proposed. The implementation and the associated numerical experiments of a particular interface relaxation scheme are also presented and discussed. [Copyright &y& Elsevier]
*STOCHASTIC convergence, *FINITE element method, *BOUNDARY value problems, *APPROXIMATION theory, *ASYMPTOTIC expansions, *ERROR analysis in mathematics, *NUMERICAL analysis
Abstract
Abstract: In this paper, we investigate the superconvergence properties of the - version of the finite element method (FEM) for two-point boundary value problems. A postprocessing technique for the - finite element approximation is analyzed. The analysis shows that the postprocess improves the order of convergence. Furthermore, we obtain asymptotically exact a posteriori error estimators based on the postprocessing results. Numerical examples are included to illustrate the theoretical analysis. [Copyright &y& Elsevier]