Global solution curves for semilinear systems

We study semilinear elliptic systems in two different directions. In the first one we give a simple constructive proof existence of solutions for a class of sublinear systems. Our main results are in the second direction, where we use bifurcation theory to study global solution curves. Crucial to our analysis is proving positivity properties of the corresponding linearized systems. Copyright © 2002 John Wiley & Sons, Ltd.     

Nonlinear reduction for solving deficient polynomial systems by continuation methods

Summary Neither continuation methods, nor symbolic elimination methods can be directly applied to compute all finite solutions to polynomial systems, because the amount of computational time is mostly not proportional to the dimension of the system and to the number of finite solutions. The notion of S-polynomials is used to developed a reduction algorithm to lower the total degree of the deficient polynomial system, so that computing the solutions at infinity can be avoided. Applying the reduction algorithm before solving the system with continuation methods, yields a reliable solution method.     

Global solution curves for a class of elliptic systems

We study curves of positive solutions for a system of elliptic equations of Hamiltonian type on a unit ball. We give conditions for all positive solutions to lie on global solution curves, allowing us to use the analysis similar to the case of one equation, as developed in P. Korman, Y. Li and T. Ouyang [An exact multiplicity result for a class of semilinear equations, Commun. PDE 22 (1997), pp. 661–684.] (see also T. Ouyang and J. Shi [Exact multiplicity of positive solutions for a class of semilinear problems, II, J. Diff. Eqns. 158(1) (1999), pp. 94–151].). As an application, we obtain some non-degeneracy and uniqueness results. For the one-dimensional case we also prove the positivity for the linearized problem, resulting in more detailed results.     

A penalty continuation method for the ∓∞ solution of overdetermined linear systems

A new algorithm for the ∓_∞ solution of overdetermined linear systems is given. The algorithm is based on the application of quadratic penalty functions to a primal linear programming formulation of the ∓_∞ problem. The minimizers of the quadratic penalty function generate piecewise-linear non-interior paths to the set of ∓_∞ solutions. It is shown that the entire set of ∓_∞ solutions is obtained from the paths for sufficiently small values of a scalar parameter. This leads to a finite penalty/continuation algorithm for ∓_∞ problems. The algorithm is implemented and extensively tested using random and function approximation problems. Comparisons with the Barrodale-Phillips simplex based algorithm and the more recent predictor-corrector primal-dual interior point algorithm are given. The results indicate that the new algorithm shows a promising performance on random (non-function approximation) problems.     

Robust AMLI methods for parabolic Crouzeix-Raviart FEM systems

For the iterative solution of linear systems of equations arising from finite element discretization of elliptic problems there exist well-established techniques to construct numerically efficient and computationally optimal preconditioners. Among those, most often preferred choices are Multigrid methods (geometric or algebraic), Algebraic MultiLevel Iteration (AMLI) methods, Domain Decomposition techniques.In this work, the method in focus is AMLI. We extend its construction and the underlying theory over to systems arising from discretizations of parabolic problems, using non-conforming finite element methods (FEM). The AMLI method is based on an approximated block two-by-two factorization of the original system matrix. A key ingredient for the efficiency of the AMLI preconditioners is the quality of the utilized block two-by-two splitting, quantified by the so-called Cauchy-Bunyakowski-Schwarz (CBS) constant, which measures the abstract angle between the two subspaces, associated with the two-by-two block splitting of the matrix.The particular choice of space discretization for the parabolic equations, used in this paper, is Crouzeix-Raviart non-conforming elements on triangular meshes. We describe a suitable splitting of the so-arising matrices and derive estimates for the associated CBS constant. The estimates are uniform with respect to discretization parameters in space and time as well as with respect to coefficient and mesh anisotropy, thus providing robustness of the method.     

Unique continuation along curves and hypersurfaces for second order anisotropic hyperbolic systems with real analytic coefficients

In this paper we prove the following kind of unique continuation property. That is, the zero on each geodesic of the solution in a real analytic hypersurface for second order anisotropic hyperbolic systems with real analytic coefficients can be continued along this curve.

     

Algebraic perturbation methods for the solution of singular linear systems

A singular matrix A is perturbed algebraically to obtain a nonsingular matrix B. Particular solutions of Ax=b can be found as unique solutions of Bx=d, where d is an algebraic perturbation of b. More specially, null vectors and generalized null vectors of A can be found as unique solutions of linear systems. It is shown also that B^?1AB^?1 is a generalized inverse of A.     

Projected equation methods for approximate solution of large linear systems

We consider linear systems of equations and solution approximations derived by projection on a low-dimensional subspace. We propose stochastic iterative algorithms, based on simulation, which converge to the approximate solution and are suitable for very large-dimensional problems. The algorithms are extensions of recent approximate dynamic programming methods, known as temporal difference methods, which solve a projected form of Bellman’s equation by using simulation-based approximations to this equation, or by using a projected value iteration method.     

Robust fitting of parametric curves

We consider the problem of fitting a parametric curve to a given point cloud (e.g., measurement data). Least-squares approximation, i.e., minimization of the ℓ₂ norm of residuals (the Euclidean distances to the data points), is the most common approach. This is due to its computational simplicity [1]. However, in the case of data that is affected by noise or contains outliers, this is not always the best choice, and other error functions, such as general ℓ_p norms have been considered [2]. We describe an extension of the least-squares approach which leads to Gauss-Newton-type methods for minimizing other, more general functions of the residuals, without increasing the computational costs significantly. (© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)     

Non-Interior continuation methods for solving semidefinite complementarity problems

 There recently has been much interest in non-interior continuation/smoothing methods for solving linear/nonlinear complementarity problems. We describe extensions of such methods to complementarity problems defined over the cone of block-diagonal symmetric positive semidefinite real matrices. These extensions involve the Chen-Mangasarian class of smoothing functions and the smoothed Fischer-Burmeister function. Issues such as existence of Newton directions, boundedness of iterates, global convergence, and local superlinear convergence will be studied. Preliminary numerical experience on semidefinite linear programs is also reported. Received: October 1999 / Accepted: April 2002 Published online: December 19, 2002 RID="⋆" ID="⋆" This research is supported by National Science Foundation Grant CCR-9731273. Key words. semidefinite complementarity problem – smoothing function – non-interior continuation – global convergence – local superlinear convergence     

Numerical methods for analytic continuation and mesh generation

The problem of computing the analytic continuation of a holomorphic function known on a circle is considered. Several fast numerical schemes based on solving an initial-value problem for the Cauchy-Riemann equations are analyzed. To avoid instability problems, some of the schemes consist of two parts: one for integrating the Cauchy-Riemann equations, and one for smoothing the function values so obtained. We show that with appropriate integration and smoothing methods, the stability and accuracy of such schemes is sufficient for many applications. The schemes are well-suited for generating level curves and stream lines of conformal mappings. Computed examples are presented. We also indicate how the schemes can be used to generate near-orthogonal boundary-fitted grids with given mesh sizes along the boundary.     

Newton and other continuation methods for multivalued inclusions

Viability theory provides an efficient framework for looking for zeros of multivalued equations: 0 F(x). The main idea is to consider solutions of a suitable differential inclusion, viable in graph ofF. The choice of the differential inclusion is guided necessarily by the fact that any solution should converge or go through a zero of the multivalued equation. We investigate here a new understanding of the well-known Newton's method, generalizing it to set-valued equations and set up a class of algorithms which involve generalization of some homotopic path algorithms.     

An efficient step size control for continuation methods

In solving a nonlinear equation by the use of a continuation method one of the crucial problems is the choice of the step sizes. We present a model for the total computational cost of a standard numerical continuation process and solve the problem of optimal step size control for this model. Using the theoretical results as a basis, we develop an adaptive step size algorithm for Newton's method. This procedure is computationally inexpensive and it gives quite satisfactory results compared to some other numerical experiments found in the literature.     

Robust stability and stabilization methods for a class of nonlinear discrete-time delay systems

This paper establishes new robust delay-dependent stability and stabilization methods for a class of nonlinear discrete-time systems with time-varying delays. The parameter uncertainties are convex-bounded and the unknown nonlinearities are time-varying perturbations satisfying Lipschitz conditions in the state and delayed-state. An appropriate Lyapunov functional is constructed to exhibit the delay-dependent dynamics and compensate for the enlarged time-span. The developed methods for stability and stabilization eliminate the need for over bounding and utilize smaller number of LMI decision variables. New and less conservative solutions to the stability and stabilization problems of nonlinear discrete-time system are provided in terms of feasibility-testing of new parametrized linear matrix inequalities (LMIs). Robust feedback stabilization methods are provided based on state-measurements and by using observer-based output feedback so as to guarantee that the corresponding closed-loop system enjoys the delay-dependent robust stability with an L₂ gain smaller that a prescribed constant level. All the developed results are expressed in terms of convex optimization over LMIs and tested on representative examples.     

On unique continuation principles for some elliptic systems

In this paper we prove unique continuation principles for some systems of elliptic partial differential equations satisfying a suitable superlinearity condition. As an application, we obtain nonexistence of nontrivial (not necessarily positive) radial solutions for the Lane-Emden system posed in a ball, in the critical and supercritical regimes. Some of our results also apply to general fully nonlinear operators, such as Pucci's extremal operators, being new even for scalar equations.     

A continuation method for solving symmetric Toeplitz systems

A fast algorithm is proposed for solving symmetric Toeplitz systems. This algorithm continuously transforms the identity matrix into the inverse of a given Toeplitz matrix T. The memory requirements for the algorithm are O(n), and its complexity is O(log κ(T)nlogn), where (T) is the condition number of T. Numerical results are presented that confirm the efficiency of the proposed algorithm.     

The continued fraction methods for the solution of systems of linear equations

A class of iterative methods is presented for the solution of systems of linear equationsAx=b, whereA is a generalm ×n matrix. The methods are based on a development as a continued fraction of the inner product (r, r), wherer=b-Ax is the residual. The methods as defined are quite general and include some wellknown methods such as the minimal residual conjugate gradient method with one step.     

Strong unique continuation for systems of complex vector fields

In this work we study the property of strong unique continuation, at a given point, for Gevrey solutions to homogeneous systems of PDE defined by complex, real-analytic vector fields in involution. We show that when the system is minimal at the point then the strong unique continuation property holds for Gevrey solutions of order σ∈[1,2]$\sigma \in [1,2]$ and, furthermore, when the minimality property fails to hold then there are non-trivial Gevrey flat solutions of any given order σ>1$\sigma >1$. The case of Gevrey order σ>2$\sigma >2$ is also studied for some particular classes of involutive systems.