Bounds for theN lowest eigenvalues of fourth-order boundary value problems

We describe a method for the calculation of theN lowest eigenvalues of fourth-order problems inH ₀ ² (Ω). In order to obtain small error bounds, we compute the defects inH ⁻²(Ω) and, to obtain a bound for the rest of the spectrum, we use a boundary homotopy method. As an example, we compute strict error bounds (using interval arithmetic to control rounding errors) for the 100 lowest eigenvalues of the clamped plate problem in the unit square. Applying symmetry properties, we prove the existence of double eigenvalues.     

Stabilised hp-Finite Element Approximation of Partial Differential Equations with Nonnegative Characteristic Form

This paper is devoted to the a priori error analysis of the hp-version of a streamline-diffusion finite element method for partial differential equations with nonnegative characteristic form. This class of equations includes second-order elliptic and parabolic problems, first-order hyperbolic problems and second-order problems of mixed elliptic-parabolic-hyperbolic type. We derive error bounds which are simultaneously optimal in both the mesh size h and the spectral order p. Numerical examples are presented to confirm the theoretical results. Received October 28, 1999; revised May 26, 2000     

A Coupled Multigrid Method for Nonconforming Finite Element Discretizations of the 2D-Stokes Equation

This paper investigates a multigrid method for the solution of the saddle point formulation of the discrete Stokes equation obtained with inf–sup stable nonconforming finite elements of lowest order. A smoother proposed by Braess and Sarazin (1997) is used and L ²-projection as well as simple averaging are considered as prolongation. The W-cycle convergence in the L ²-norm of the velocity with a rate independently of the level and linearly decreasing with increasing number of smoothing steps is proven. Numerical tests confirm the theoretically predicted results. Received January 19, 1999; revised September 13, 1999     

The Fourier-Nitsche-mortaring for elliptic problems with reentrant edges

The Fourier method is combined with the Nitsche-finite-element method (as a mortar method) and applied to the Dirichlet problem of the Poisson equation in three-dimensional axisymmetric domains with reentrant edges generating singularities. The approximating Fourier method yields a splitting of the 3D problem into a set of 2D problems on the meridian plane of the given domain. For solving the 2D problems bearing corner singularities, the Nitsche-finite-element method with non-matching meshes and mesh grading near reentrant corners is applied. Using the explicit representation of some singularity function of non-tensor product type, the rate of convergence of the Fourier-Nitsche-mortaring is estimated in some H ¹-like norm as well as in the L ₂-norm for weak regularity of the solution. Finally, some numerical results are presented.      

A Hermitian Box-scheme for One-dimensional Elliptic Equations – Application to Problems with High Contrasts in the Ellipticity

We introduce a new box-scheme, called ``hermitian box-scheme' on the model of the one-dimensional Poisson problem. The scheme combines features of the box-scheme of Keller, [20], [13], with the hermitian approximation of the gradient on a compact stencil, which is characteristic of compact schemes, [9], [21]. The resulting scheme is proved to be 4th order accurate for the primitive unknown u and its gradient p. The proved convergence rate is 1.5 for (u,p) in the discrete L ² norm. The connection with a non standard mixed finite element method is given. Finally, numerical results are displayed on pertinent 1-D elliptic problems with high contrasts in the ellipticity, showing in practice convergence rates ranging from 1 to 2.5 in the discrete H ¹ norm. This work has been performed with the support of the GDR MOMAS, (ANDRA, CEA, EDF, BRGM and CNRS): Modélisation pour le stockage des déchets radioactifs. The author thanks especially A. Bourgeat for his encouragements and his interest in this work.     

A Refined Finite Element Convergence Theory for Highly Indefinite Helmholtz Problems

It is well known that standard h-version finite element discretisations using lowest order elements for Helmholtz' equation suffer from the following stability condition: ``The mesh width h of the finite element mesh has to satisfy k <sup>2</sup> h≲1', where k denotes the wave number. This condition rules out the reliable numerical solution of Helmholtz equation in three dimensions for large wave numbers k≳50. In our paper, we will present a refined finite element theory for highly indefinite Helmholtz problems where the stability of the discretisation can be checked through an ``almost invariance' condition. As an application, we will consider a one-dimensional finite element space for the Helmholtz equation and apply our theory to prove stability under the weakened condition hk≲1 and optimal convergence estimates. Dedicated to Prof. Dr. Ivo Babuška on the occasion of his 80th birthday.     

The Streamline–Diffusion Method for Conforming and Nonconforming Finite Elements of Lowest Order Applied to Convection–Diffusion Problems

We consider the streamline-diffusion finite element method with finite elements of lowest order for solving convection-diffusion problems. Our investigations cover both conforming and nonconforming finite element approximations on triangular and quadrilateral meshes. Although the considered finite elements are of the same interpolation order their stability and approximation properties are quite different. We give a detailed overview on the stability and the convergence properties in the L ²- and in the streamline–diffusion norm. Numerical experiments show that often the theoretical predictions on the convergence properties are sharp. Received December 7, 1999; revised October 5, 2000     

Derivative Free Inclusion Methods for Polynomial Zeros

Two iterative methods for the simultaneous inclusion of complex zeros of a polynomial are presented. Both methods are realized in circular interval arithmetic and do not use polynomial derivatives. The first method of the fourth order is composed as a combination of interval methods with the order of convergence two and three. The second method is constructed using double application of the inclusion method of Weierstrass’ type in serial mode. It is shown that its R-order of convergence is bounded below by the spectral radius of the corresponding matrix. Numerical examples illustrate the convergence rate of the presented methods     

Superconvergence of a Mixed Covolume Method for Elliptic Problems

We consider a mixed covolume method for a system of first order partial differential equations resulting from the mixed formulation of a general self-adjoint elliptic problem with a variable full diffusion tensor. The system can be used to model the transport of a contaminant carried by a flow. We use the lowest order Raviart-Thomas mixed finite element space. We show the first order convergence in L ² norm and the superconvergence in certain discrete norms both for the pressure and velocity. Finally some numerical examples illustrating the error behavior of the scheme are provided. Supported by the National Natural Science Foundation of China under grant No. 10071044 and the Research Fund of Doctoral Program of High Education by State Education Ministry of China.     

h-p Spectral Element Method for Elliptic Problems on Non-smooth Domains Using Parallel Computers

We propose a new h-p spectral element method to solve elliptic boundary value problems with mixed Neumann and Dirichlet boundary conditions on non-smooth domains. The method is shown to be exponentially accurate and asymptotically faster than the standard h-p finite element method. The spectral element functions are fully non-conforming for pure Dirichlet problems and conforming only at the vertices of the elements for mixed problems, and hence, the dimension of the resulting Schur complement matrix is quite small. The method is a least-squares collocation method and the resulting normal equations are solved using preconditioned conjugate gradient method with an almost optimal preconditioner. The algorithm is suitable for a distributed memory parallel computer. The numerical results of a number of model problems are presented, which confirm the theoretical estimates.     

The SOR-k method for linear systems with p-cyclic matrices

Let the system matrix of a linear system be p-cyclic and consistently ordered. Under the assumption that the pth power of the associated Jacobi matrix has only non-positive eigenvalues, it is known that the optimal spectral radius of the SOR-k iteration matrix is strictly increasing as k increases from 2 to p. In this paper, we first show that the optimal parameter of the SOR-k method as a function of k is strictly increasing. The behaviour of the spectral radius of the SOR-k method (for fixed parameter) is then studied.     

Two-level Stabilized Finite Element Methods for the Steady Navier–Stokes Problem

In this article, the two-level stabilized finite element formulations of the two-dimensional steady Navier–Stokes problem are analyzed. A macroelement condition is introduced for constructing the local stabilized formulation of the steady Navier–Stokes problem. By satisfying this condition the stability of the Q₁–P₀ quadrilateral element and the P₁–P₀ triangular element are established. Moreover, the two-level stabilized finite element methods involve solving one small Navier–Stokes problem on a coarse mesh with mesh size H, a large Stokes problem for the simple two-level stabilized finite element method on a fine mesh with mesh size h=O(H²) or a large general Stokes problem for the Newton two-level stabilized finite element method on a fine mesh with mesh size h=O(|log h|^1/2H³). The methods we study provide an approximate solution (u^h,p^h) with the convergence rate of same order as the usual stabilized finite element solution, which involves solving one large Navier–Stokes problem on a fine mesh with mesh size h. Hence, our methods can save a large amount of computational time.     

Iterative Stokes Solvers in the Harmonic Velte Subspace

We explore the prospects of utilizing the decomposition of the function space (H ¹ ₀) ⁿ (where n=2,3) into three orthogonal subspaces (as introduced by Velte) for the iterative solution of the Stokes problem. It is shown that Uzawa and Arrow-Hurwitz iterations – after at most two initial steps – can proceed fully in the third, smallest subspace. For both methods, we also compute optimal iteration parameters. Here, for two-dimensional problems, the lower estimate of the inf-sup constant by Horgan and Payne proves useful and provides an inclusion of the spectrum of the Schur complement operator of the Stokes problem. We further consider the conjugate gradient method in the third Velte subspace and derive a corresponding convergence estimate. Computational results show the effectiveness of this approach for discretizations which admit a discrete Velte decomposition. Received June 11, 1999; revised October 27, 2000     

Two-Level Additive Schwarz Preconditioners for the h-p Version of the Galerkin Boundary Element Method for 2-d Problems

We study two-level additive Schwarz preconditioners for the h-p version of the Galerkin boundary element method when used to solve hypersingular integral equations of the first kind, which arise from the Neumann problems for the Laplacian in two dimensions. Overlapping and non-overlapping methods are considered. We prove that the non-overlapping preconditioner yields a system of equations having a condition number bounded by   where H _i is the length of the i-th subdomain, h _i is the maximum length of the elements in this subdomain, and p is the maximum polynomial degree used. For the overlapping method, we prove that the condition number is bounded by   where δ is the size of the overlap and H=max _i H _i . We also discuss the use of the non-overlapping method when the mesh is geometrically graded. The condition number in that case is bounded by clog² M, where M is the degrees of freedom. Received October 27, 2000, revised March 26, 2001     

A Domain Decomposition Preconditioner for p-FEM Discretizations of Two-dimensional Elliptic Problems

In this paper, a uniformly elliptic second order boundary value problem in 2-D discretized by the p-version of the finite element method is considered. An inexact Dirichlet-Dirichlet domain decomposition pre-conditioner for the system of linear algebraic equations is investigated. Two solvers for the problem in the sub-domains, a pre-conditioner for the Schur-complement and an extension operator operating from the edges of the elements into the interior are proposed as ingredients for the inexact DD-pre-conditioner. In the main part of the paper, several numerical experiments on a parallel computer are given.     

A method for obtaining bounds on eigenvalues and eigenfunctions by solving non-homogeneous integral equations

A new technique is presented for obtaining upper and lower bounds on eigenvalues and eigenfunctions for linear integral equations. The method is unique in that the bounds are obtained by solving non-homogeneous equations. In order to solve the non-homogeneous equations, non-linear sequence-to-sequence transformations are used to accelerate convergence of the Neumann series inside the radius of convergence and are used to “sum” the Neumann series outside the radius of convergence. Since the reciprocals of the eigenvalues appear as poles in the solution of the non-homogeneous equation, a very sensitive bounding criterion can be given. The method applies to quite general kernels, and has been successfully applied to symmetric and non-symmetric kernels. In addition, thek ^th eigenfunction may be obtained without a knowledge of the first (k?1) eigenvalues or eigenfunctions.     

The Fourier Finite-element Approximation of the Lamé Equations in Axisymmetric Domains with Edges

This paper is concerned with a priori error estimates and convergence analysis of the Fourier-finite-element solutions of the Neumann problem for the Lamé equations in axisymmetric domains with reentrant edges. The Fourier-FEM combines the approximating Fourier method with respect to the rotational angle using trigonometric polynomials of degree N (N→∞), with the finite element method on the plane meridian domain of with mesh size h (h→0) for approximating the Fourier coefficients. The asymptotic behavior of the solution near reentrant edges is described by singularity functions in non-tensor product form and treated numerically by means of finite element method on locally graded meshes. For the rate of convergence of the combined approximations in is proved to be of the order      

Two-level methods for the single layer potential in ℝ3

We consider weakly singular integral equations of the first kind on open surface pieces Γ in ℝ³. To obtain approximate solutions we use theh-version Galerkin boundary element method. Furthermore we introduce two-level additive Schwarz operators for non-overlapping domain decompositions of Γ and we estimate the conditions numbers of these operators with respect to the mesh size. Based on these operators we derive an a posteriori error estimate for the difference between the exact solution and the Galerkin solution. The estimate also involves the error which comes from an approximate solution of the Galerkin equations. For uniform meshes and under the assumption of a saturation condition we show reliability and efficiency of our estimate. Based on this estimate we introduce an adaptive multilevel algorithm with easily computable local error indicators which allows direction control of the local refinements. The theoretical results are illustrated by numerical examples for plane and curved surfaces. Supported by the German Research Foundation (DFG) under grant Ste 238/25-9.     

<Emphasis Type="Italic">H</Emphasis><Superscript>1</Superscript>-interpolation on quadrilateral and hexahedral meshes with hanging nodes

We propose a Scott-Zhang type finite element interpolation operator of first order for the approximation of H ¹-functions by means of continuous piecewise mapped bilinear or trilinear polynomials. The novelty of the proposed interpolation operator is that it is defined for general non-affine equivalent quadrilateral and hexahedral elements and so-called 1-irregular meshes with hanging nodes. We prove optimal local approximation properties of this interpolation operator for functions in H ¹. As necessary ingredients we provide a definition of a hanging node and a rigorous analysis of the issue of constrained approximation which cover both the two- and three-dimensional case in a unified fashion.      

A Numerical Method to Verify the Elliptic Eigenvalue Problems Including a Uniqueness Property

We propose a numerical method to enclose the eigenvalues and eigenfunctions of second-order elliptic operators with local uniqueness. We numerically construct a set containing eigenpairs which satisfies the hypothesis of Banach's fixed point theorem in a certain Sobolev space by using a finite element approximation and constructive error estimates. We then prove the local uniqueness separately of eigenvalues and eigenfunctions. This local uniqueness assures the simplicity of the eigenvalue. Numerical examples are presented. Received: November 2, 1998; revised June 5, 1999