A tensor product generalized ADI method for elliptic problems on cylindrical domains with holes

We consider solving second order linear elliptic partial differential equations together with Dirichlet boundary conditions in three dimensions on cylindrical domains (nonrectangular in x and y) with holes.We approximate the partial differential operators by standard partial difference operators. If the partial differential operator separates into two terms, one depending on x and y, and one depending on z, then the discrete elliptic problem may be written in tensor product form as (T_z⊗I + I⊗A_xy) U=F. We consider a specific implementation which uses a Method of Planes approach with unequally spaced finite differences in the xy direction and symmetric finite difference in the z direction. We establish the convergence of the Tensor Product Generalized Alternating Direction Implicit iterative method applied to such discrete problems. We show that this method gives a fast and memory efficient scheme for solving a large class of elliptic problems.     

Nonlinear Eigenvalue Problems of Schrödinger Type Admitting Eigenfunctions with Given Spectral Characteristics

The following work is an extension of our recent paper [10]. We still deal with nonlinear eigenvalue problems of the form in a real Hilbert space ℋ︁ with a semi‐bounded self‐adjoint operator A₀, while for every y from a dense subspace X of ℋ︁, B(y ) is a symmetric operator. The left‐hand side is assumed to be related to a certain auxiliary functional ψ, and the associated linear problems are supposed to have non‐empty discrete spectrum (y ∈ X). We reformulate and generalize the topological method presented by the authors in [10] to construct solutions of (∗︁) on a sphere S_R ≔ {y ∈ X | ∥y∥_ℋ︁ = R} whose ψ‐value is the n‐th Ljusternik‐Schnirelman level of ψ| and whose corresponding eigenvalue is the n‐th eigenvalue of the associated linear problem (∗︁∗︁), where R > 0 and n ∈ ℕ are given. In applications, the eigenfunctions thus found share any geometric property enjoyed by an n‐th eigenfunction of a linear problem of the form (∗︁∗︁). We discuss applications to elliptic partial differential equations with radial symmetry.     

How Competitive is the ADI for Tensor Structured Equations?

In [1] we presented an extension of the alternating direction implicit (ADI) method for the solution of Lyapunov equations (1) to higher dimensional problems. The vectorized form of the Lyapunov equation is We considered the generalization of this equation of the form (2) The tensor train structure is one possible generalization of the low rank factorization we find in the right hand side of (1). Therefor we assume B to be of tensor train structure. We showed that in analogy to the low rank ADI case the solution X can be generated in tensor train structure, too. Further we provided an algorithm that computes X using a generalization of the ADI method. Here we compare our new tensor ADI method with an density matrix renormalization group (DMRG) solver for tensor train matrix equations and with matrix equation solvers to investigate the competitiveness of our new solver. (© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim)     

Die Lösung der Prae-Maxwellschen Gleichungen mit Hilfe vollständiger Lösungssysteme

This paper deals with the Neumann problem of the pre-Maxwell partial differential equations for a vector field v defined in a region G ? R ³. We approximate its uniquely determined solution (integrability conditions assumed) uniformly on G by explicitly computable particular integrals and linear combinations of vector fields with a “fundamental” sequence of points .     

The Mortar Element Method with Locally Nonconforming Elements

We consider a discretization of linear elliptic boundary value problems in 2-D by the new version of the mortar finite element method which uses locally nonconforming Crouzeix-Raviart elements. We show that if a solution of the original differential problem belongs to the space H ²(), then an error is of the same order as in the standard nonconforming finite element method. We also propose an additive Schwarz method of solving the discrete problem and show that its rate of convergence is almost optimal.     

Radially symmetric Poisson–Boltzmann equation in a domain expanding to infinity

The electric potential u in a solution of an electrolyte around a linear polyelectrolyte of the form of a cylinder satisfies We study the problem when R → ∞.     

Numerical solution of a time-like Cauchy problem for the wave equation

Let D ? ?ⁿ be a bounded domain with piecewise-smooth boundary, and q(x,t) a smooth function on D × [0, T]. Consider the time-like Cauchy problem Given g, h for which the equation has a solution, we show how to approximate u(x,t) by solving a well posed fourth-order elliptic partial differential equation (PDE). We use the method of quasi-reversibility to construct the approximating PDE. We derive error estimates and present numerical results.     

Numerical solution to a linear equation with tensor product structure

We consider the numerical solution of a c‐stable linear equation in the tensor product space , arising from a discretized elliptic partial differential equation in . Utilizing the stability, we produce an equivalent d‐stable generalized Stein‐like equation, which can be solved iteratively. For large‐scale problems defined by sparse and structured matrices, the methods can be modified for further efficiency, producing algorithms of computational complexity, under appropriate assumptions (with n_s being the flop count for solving a linear system associated with ). Illustrative numerical examples will be presented.     

Classification of solutions for an integral equation

Let n be a positive integer and let 0 < α < n. Consider the integral equation We prove that every positive regular solution u(x) is radially symmetric and monotone about some point and therefore assumes the form with some constant c = c(n, α) and for some t > 0 and x₀ ? ?ⁿ. This solves an open problem posed by Lieb 12 . The technique we use is the method of moving planes in an integral form, which is quite different from those for differential equations. From the point of view of general methodology, this is another interesting part of the paper. Moreover, we show that the family of well‐known semilinear partial differential equations is equivalent to our integral equation (0.1), and we thus classify all the solutions of the PDEs. © 2005 Wiley Periodicals, Inc.     

Controllability of an Elliptic equation and its Finite Difference Approximation by the Shape of the Domain

In this article we study a controllability problem for an elliptic partial differential equation in which the control is the shape of the domain where the equation holds. The quantity to be controlled is the trace of the solution, with a given right hand side source term, into an open subdomain. The mapping that associates this trace to the shape of the domain is nonlinear. We first consider the linearized problem and show an approximate controllability property. We then address the same questions in the context of a finite difference discretization of the elliptic problem. We prove a local controllability result applying the Inverse Function Theorem together with a ``unique continuation' property of the underlying adjoint discrete system. Mathematics Subject Classification (1991):35J05, 93B03, 65M06     

Dynamic finite deformation viscoelasticity and energy-consistent time integration

In the present work we deal with the conserving integration of viscoelastic bodies undergoing finite deformations. Isotropic viscoelastic materials can be described by using a symmetric viscous internal variable for measuring the inelastic strains, and the right Cauchy-Green tensor as measure of the total strain (see Reese & Govindjee [1]). Then, by using the unsymmetric product tensor , the purely elastic strains enter the isotropic free energy function. Alternatively, the i application of the symmetric right stretch tensor as internal variable allows to define a symmetric elastic strain tensor which enters the free energy (see also Miehe [2]). These two different approaches lead to different evolution equations for the viscous internal variable. In this lecture, both evolution equations are discretised by an ordinary midpoint rule at each Gauss point of a standard nonlinear displacement-based finite element in space. For discretising the semidiscrete Hamilton's equations of motion in time, we use numerical time integrators which preserve the fundamental conservation laws of the underlying system. In particular, we make use of a modified midpoint rule according to the discrete gradient method, proposed in Gonzalez [3]. Numerical examples demonstrate the difference between both formulations. (© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim)     

Generalized Hopf bifurcation in Hilbert space

We consider a family of semilinear evolution equations in Hilbert space of the form with, in general, unbounded operators *A(λ), F(λ·) depending analytically on a real parameter λ. We assume that the origin is a stationary solution, i.e. F(λ,0) = 0, for all λ ε R and that the linearization (with respect to u) at the origin is given by du/dt + A(λ)u = 0. Our essential assumption is the following: A(λ) possesses one pair of simple complex conjugate eigenvalues μ(λ) = Re μ(λ) ± i Im μ(λ) such that Im μ(0) > 0 and for some m ε N or If m = 1 the curves of eigenvalues μ(λ) cross the imaginary axis transversally at ±i Im μ(0). In this case a unique branch of periodic solutions emanates from the origin at λ = 0 which is commonly called Hopf bifurcation. If μ(λ) and the imaginary axis are no longer transversal, i.e. m > 1, we call a bifurcation of periodic solutions, if it occurs, a generalized Hopf bifurcation. It is remarkable that up to m such branches may exist. Our approach gives the number of bifurcating solutions, their direction of bifurcation, and its asymptotic expansion. We regain the results of D. Flockerzi who established them in a completely different way for ordinary differential equations.     

Supercritical biharmonic elliptic problems in domains with small holes

Let D be a bounded and smooth domain in R^N, N ≥ 5, P ∈ D. We consider the following biharmonic elliptic problemin Ω = D \B_δ (P), with p supercritical, namely . We find a sequence of resonant exponents such that if is given, with p ≠ p_j for all j, then for all δ > 0 sufficiently small, this problem is solvable (© 2009 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)     

Quasilinear degenerate parabolic equations of Kirchhoff type

We investigate the evolution problem where H is a Hilbert space, A is a self‐adjoint linear non‐negative operator on H with domain D(A), and is a continuous function. We prove that if , and , then there exists at least one global solution, which is unique if either m never vanishes, or m is locally Lipschitz continuous. Moreover, we prove that if for all , then this problem is well posed in H. On the contrary, if for some it happens that for all , then this problem has no solution if with β small enough. We apply these results to degenerate parabolic PDEs with non‐local non‐linearities. Copyright © 1999 John Wiley & Sons, Ltd.     

Almost monotonicity formulas for elliptic and parabolic operators with variable coefficients

In this paper we extend the results of Caffarelli, Jerison, and Kenig [Ann. of Math. (2) 155 (2002)] and Caffarelli and Kenig [Amer. J. Math. 120 (1998)] by establishing an almost monotonicity estimate for pairs of continuous functions satisfying in an infinite strip (global version) or a finite parabolic cylinder (localized version), where ${\cal L}$ is a uniformly parabolic operator with double Dini continuous ${\cal A}$ and uniformly bounded b and c. We also prove the elliptic counterpart of this estimate. This closes the gap between the known conditions in the literature (both in the elliptic and parabolic case) imposed on u_± in order to obtain an almost monotonicity estimate. At the end of the paper, we demonstrate how to use this new almost monotonicity formula to prove the optimal C^1,1‐regularity in a fairly general class of quasi‐linear obstacle‐type free boundary problems. © 2010 Wiley Periodicals, Inc.     

Necessary and Sufficient Conditions for the Oscillation of a Second Order Linear Differential Equation

In this paper we give a necessary and sufficient condition for the oscillation of the second order linear differential equation where p is a locally integrable function and either or where We give some applications which show how these results unify and imply some classical results in oscillation theory.     

Local solvability of a nonstationary leakage problem for an ideal incompressible fluid, 3

In this paper we prove the existence and uniqueness of solutions of the leakage problem for the Euler equations in bounded domain Ω C R³ with corners π/n, n = 2, 3… We consider the case where the tangent components of the vorticity vector are given on the part S₁ of the boundary where the fluid enters the domain. We prove the existence of an unique solution in the Sobolev space W_p^l(Ω), for arbitrary natural l and p > 1. The proof is divided on three parts: (1) the existence of solutions of the elliptic problem in the domain with corners where v – velocity vector, ω – vorticity vector and n is an unit outward vector normal to the boundary, (2) the existence of solutions of the following evolution problem for given velocity vector (3) the method of successive approximations, using solvability of problems (1) and (2).     

Sparse finite elements for elliptic problems with stochastic loading

Summary. We formulate elliptic boundary value problems with stochastic loading in a bounded domain D ^d . We show well-posedness of the problem in stochastic Sobolev spaces and we derive a deterministic elliptic PDE in D×D for the spatial correlation of the random solution. We show well-posedness and regularity results for this PDE in a scale of weighted Sobolev spaces with mixed highest order derivatives. Discretization with sparse tensor products of any hierarchic finite element (FE) spaces in D yields optimal asymptotic rates of convergence for the spatial correlation even in the presence of singularities or for spatially completely uncorrelated data. Multilevel preconditioning in D×D allows iterative solution of the discrete equation for the correlation kernel in essentially the same complexity as the solution of the mean field equation. Mathematics Subject Classification (2000): 65N30Research performed under IHP network Breaking Complexity of the EC, contract number HPRN-CT-2002-00286, and supported in part by the Swiss Federal Office for Science and Education under grant number BBW 02.0418.     

A finite element method and stabilization method for a non-linear tricomi problem

We prove using the Faedo-Galerkin method the existence of a generalized solution of an initial-boundary value problem for the non-linear evolution equation 0 ? Q ? 2, in a cylinder Q_T = Ω × (0, T), where ?? u = yu_xx + u_yy is the Tricomi operator and l(u) a special differential operator of first order. We then show that the approximate generalized solution of problem (*) converges to the approximate generalized solution of the corresponding stationary boundary value problem as t → ∞.