The instability of some gradient methods for ill-posed problems

Summary For the solution of linear ill-posed problems some gradient methods like conjugate gradients and steepest descent have been examined previously in the literature. It is shown that even though these methods converge in the case of exact data their instability makes it impossible to base a-priori parameter choice regularization methods upon them.     

Pointwise rational approximation and iterative methods for III-posed problems

Summary We study the connection between the pointwise approximation of the zero function by rational functions and iterative methods for the approximate solution of ill-posed linear equations. Results are presented on convergence, stability and saturation phenomena.Dedicated to Professor Dr. G. Hämmerlin on the occasion of his 60th birthday     

A class of iterative methods for solving saddle point problems

Summary We 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.The work of this author was supported by the Office of Naval Research under contract N00014-82K-0197, by Avions Marcel Dassault, 78 Quai Marcel Dassault, 92214 St Cloud, France, and by Direction des Recherches Etudes et Techniques, 26 boulevard Victor, F-75996 Paris Armées, FranceThe work of this author was supported by Avions Marcel Daussault-Breguet Aviation, 78 quai Marcel Daussault, F-92214 St Cloud, France and by Direction des Recherches Etudes et Techniques, 26 boulevard Victor, F-75996 Paris Armées, FranceThe work of this author was supported by Konrad-Zuse-Zentrum für Informationstechnik Berlin, Federal Republic of Germany     

Numerical methods for reaction-diffusion problems with non-differentiable kinetics

Summary We consider a class of steady-state semilinear reaction-diffusion problems with non-differentiable kinetics. The analytical properties of these problems have received considerable attention in the literature. We take a first step in analyzing their numerical approximation. We present a finite element method and establish error bounds which are optimal for some of the problems. In addition, we also discuss a finite difference approach. Numerical experiments for one- and two-dimensional problems are reported.Dedicated to Ivo Babuka on his sixtieth birthdayResearch partially supported by the Air Force Office of Scientific Research, Air Force Systems Command, USAF under Grant Number AFOSR 85-0322     

A finite element lumped mass scheme for solving eigenvalue problems of circular arches

Summary This study is a continuation of a previous paper [4] in which the numerical results are given by using single precision arithmetic. In this paper, we show the numerical results which experess the sharper convergence properties than those of [4], by using double precision arithmetic.Dedicated to Prof. Masaya Yamaguti on the occasion of his 60th birthday     

A nonconforming combined method for solving Laplace's boundary value problems with singularities

Summary For solving Laplace's boundary value problems with singularities, a nonconforming combined approach of the Ritz-Galerkin method and the finite element method is presented. In this approach, singular functions are chosen to be admissible functions in the part of a solution domain where there exist singularities; and piecewise linear functions are chosen to be admissible functions in the rest of the solution domain. In addition, the admissible functions used here are constrained to be continuous only at the element nodes on the common boundary of both methods. This method is nonconforming; however, the nonconforming effect does not result in larger errors of numerical solutions as long as a suitable coupling strategy is used.In this paper, we will develop such an approach by using a new coupling strategy, which is described as follows: IfL+1=O(|lnh|), the average errors of numerical solutions and their generalized derivatives are stillO(h), whereh is the maximal boundary length of quasiuniform triangular elements in the finite element method, andL+1 is the total number of singular admissible functions in the Ritz-Galerkin method. The coupling relation,L+1=O(|lnh|), is significant because only a few singular functions are required for a good approximation of solutions.This material is from Chapter 5 in my Ph.D. thesis: Numerical Methods for Elliptic Boundary Value Problems with Singularities. Part I: Boundary Methods for Solving Elliptic Problems with Singularities. Part II: Nonconforming Combinations for Solving Elliptic Problems with Singularities, the Department of Mathematics and Applied Mathematics, University of Toronto, May 1986     

On finite element methods for the Neumann problem

Summary The Neumann problem for a second order elliptic equation with self-adjoint operator is considered, the unique solution of which is determined from projection onto unity. Two variational formulations of this problem are studied, which have a unique solution in the whole space. Discretization is done via the finite element method based on the Ritz process, and it is proved that the discrete solutions converge to one of the solutions of the continuous problem. Comparison of the two methods is done.     

Domain decomposition methods for pseudo spectral approximations

Summary Multidomain pseudo spectral approximations of second order boundary value problems in one dimension are considered. The equation is collocated at the Chebyshev nodes inside each subinterval. Different patching conditions at the interfaces are analyzed. Results of stability and convergence are given.Research supported in part by AFOSR Grant 85-0303     

A parallel projection method for solving generalized linear least-squares problems

Summary A parallel projection algorithm is proposed to solve the generalized linear least-squares problem: find a vector to minimize the 2-norm distance from its image under an affine mapping to a closed convex cone. In each iteration of the algorithm the problem is decomposed into several independent small problems of finding projections onto subspaces, which are simple and can be tackled parallelly. The algorithm can be viewed as a dual version of the algorithm proposed by Han and Lou [8]. For the special problem under consideration, stronger convergence results are established. The algorithm is also related to the block iterative methods of Elfving [6], Dennis and Steihaug [5], and the primal-dual method of Springarn [14].This material is based on work supported in part by the National Science foundation under Grant DMS-8602416 and by the Center for Supercomputing Research and Development, University of Illinois at Urbana-Champaign     

A predictor-corrector technique for constrained least-squares regularization

Summary The paper describes a numerical strategy for the approximate solution of nonlinear, discretized, inverse problems by regularization. It is assumed that the solution of the associated direct problems and the computation of Fréchet derivatives are expensive. In order to minimize the amount of work, a predictor-corrector type algorithm is proposed. From a series of solutions to problems with a coarse discretization one obtains a starting approximation for a problem with a fine discretization.     

On some boundary element methods for the heat equation

Summary The object of this paper is to study some boundary element methods for the heat equation. Two approaches are considered. The first, based on the heat potential, has been studied numerically by previous authors. Here the convergence analysis in one space dimension is presented. In the second approach, the heat equation is first descretized in time and the resulting elliptic problem is put in the boundary formulation. A straight forward implicit method and Crank-Nicolson's method are thus studied. Again convergence in one space dimension is proved.     

Numerical stability of descent methods for solving linear equations

Summary In this paper we perform a round-off error analysis of descent methods for solving a liner systemAx=b, whereA is supposed to be symmetric and positive definite. This leads to a general result on the attainable accuracy of the computed sequence {x _i } when the method is performed in floating point arithmetic. The general theory is applied to the Gauss-Southwell method and the gradient method. Both methods appear to be well-behaved which means that these methods compute an approximationx _i to the exact solutionA ^–1 b which is the exact solution of a slightly perturbed linear system, i.e. (A+A)x _i =b, A of order A, where is the relative machine precision and · denotes the spectral norm.     

Monotone iterative methods for nonlinear equations involving a noninvertible linear part

Summary We study monotone iterative methods for the numerical solution of a class of nonlinear elliptic boundary value problems by applying a theorem of Ortega and Rheinboldt. This generalizes the work of earlier authors to the case of nonlinear perturbations of linear problems with a nontrivial kernel.     

On the boundary element method for some nonlinear boundary value problems

Summary Here we analyse the boundary element Galerkin method for two-dimensional nonlinear boundary value problems governed by the Laplacian in an interior (or exterior) domain and by highly nonlinear boundary conditions. The underlying boundary integral operator here can be decomposed into the sum of a monotoneous Hammerstein operator and a compact mapping. We show stability and convergence by using Leray-Schauder fixed-point arguments due to Petryshyn and Neas.Using properties of the linearised equations, we can also prove quasioptimal convergence of the spline Galerkin approximations.This work was carried out while the first author was visiting the University of Stuttgart     

On a numerical approach to Stefan-like problems

Summary This paper is devoted to the numerical analysis of a bidimensional two-phase Stefan problem. We approximate the enthalpy formulation byC ⁰ piecewise linear finite elements in space combined with a semi-implicit scheme in time. Under some restrictions related to the finite element mesh and to the timestep, we prove positivity, stability and convergence results. Various numerial tests are presented and discussed in order to show the accuracy of our scheme.This work is supported by the Fonds National Suisse pour la Recherche Scientifique.     

On the construction of optimal mixed finite element methods for the linear elasticity problem

Summary The mixed finite element method for the linear elasticity problem is considered. We propose a systematic way of designing methods with optimal convergence rates for both the stress tensor and the displacement. The ideas are applied in some examples.     

On the numerical approximation of finite speed diffusion problems

Summary Lagrangian formulations for the Cauchy problems for the generalized-heat and porous-media equations are introduced and equivalence and existence results discussed. Efficient interface tracking finite difference and finite element discretizations of the Lagrangian formulation are discussed. Mixed Euler-Lagrange formulations for mixed problems and the one phase Stefan problem are presented. Numerical experiments are discussed.Dedicated on the occasion of Prof. Ivo Babuka's 60th birthday     

A fast iterative method for solving stationary heat conduction problems on regions partitioned into rectangles

Summary The paper deals with some finite element approximation of stationary heat conduction problems on regions which can be partitioned into rectangular subregions. By a special superelement-technique employing fast elimination of the inner nodal parameters, the original finite element problem is reduced to a smaller problem, which is only connected with the nodes on the boundary of the superelements. To solve the reduced system of finite element equations, an efficient iterative algorithm is proposed. This algorithm is based either on the conjugate gradient method or the Tshebysheff method, using a special matrix by vector multiplication procedure. The explicit form of the matrix is not used. The presented numerical method is asymptotically optimal with respect to the memory requirement as well as to the operation count.     

On the convergence of multi-grid methods with transforming smoothers

Summary In the present paper we give a convergence theory for multi-grid methods with transforming smoothers as introduced in [31] applied to a general system of partial differential equations. The theory follows Hackbusch's approach for scalar pde and allows a convergence proof for some well-known multi-grid methods for Stokes- and Navier-Stokes equations as DGS by Brandt-Dinar, [5], TILU from [31] and the SIMPLE-methods by Patankar-Spalding, [23].This work was supported in part by Deutsche Forschungsgemeinschaft     

Stability and accuracy for implicit semidiscretizations of hyperbolic problems

Summary In this paper we give bounds for the error constants of certain classes of stable implicit finite difference methods for first order hyperbolic equations in one space dimension. We consider classes of methods that user downwind ands upwind points in the explicit part andR downwind andS upwind points in the implicit part, respectively, and that are of optimal orderp=min (r+R+s+S, 2(r+R+1), 2(s+S)).In some cases the error constant of interpolatory methods [5] can be improved. The results are proved via the order star technique. They are further used to determine methods of optimal order that are stable.