Bruno D. Welfert

A class of iterative methods for solving saddle point problems

SummaryWe consider the numerical solution of indefinite systems of linear equations arising in the calculation of saddle points. We are mainly concerned with sparse systems of this type resulting from certain discretizations of partial differential equations. We present an iterative method involving two levels of iteration, similar in some respects to the Uzawa algorithm. We relate the rates of convergence of the outer and inner iterations, proving that, under natural hypotheses, the outer iteration achieves the rate of convergence of the inner iteration. The technique is applied to finite element approximations of the Stokes equations.

A posteriori error estimates for the Stokes problem

A keyboard assembly and a keyboard switch are presented in which the keyboard switch is a layer of flexible insulating material with circuit configurations thereon and an array of flat topped protrusions which serve as key switches to effect a snap action contact with tactile feedback between a conductive element on the key switch and another conductive element. Preferably, the areas between the protrusions on the insulating layer are securely clamped relative to a backing or stiffening board, and the protruding key switches are operated by hinged key actuators which make off center contact with the protrusion.

Generation of Pseudospectral Differentiation Matrices I

We present a simple method for computing

The Newton Iteration on Lie Groups

n \times n

A posteriori error estimates for the Stokes equations: a comparison

pseudospectral differentiation matrices of order

Two-Step Runge-Kutta: Theory and Practice

p

A comparison between the mini-element and the Petrov-Galerkin formulations for the generalized Stokes problem

in

A static, physical VDMOS model based on the charge-sheet model

{\cal O}(pn^2)

About the sharpness of the stability estimates in the Kreiss matrix theorem

operations for the case of quasi-polynomial approximation. The algorithm is based on Fornbergs finite difference algorithm and is numerically stable. A Fortran implementation is included. A necessary and sufficient condition for

Numerical solution of a Fredholm integro-differential equation modelling θ˙-neural networks

D_p = D^p_1