Convergence results for linear multistep methods for quasi-singular perturbed problem systems

ثبت نشده
چکیده

Stiff behavior occurs in a variety of ODE systems relevant in applications. The notion of stiffness is a phenomenological one, and a stability and error analysis of numerical methods has been based either on simple models or particular problem structures. In particular, stiff initial value problems in standard singular perturbation form are well understood. However, problems of this type exhibit a very simple phase space geometry, namely the stiff eigendirections also behave stiff in another sense, i.e., they are almost parallel. This motivates us to consider a more general nonlinear class of stiff ODE systems depending on a small parameter. In particular, we investigate the convergence properties of the implicit Euler scheme applied to problems of this type, and linear multistep schemes will further be investigated. 1 The class of quasi-singular perturbed problems The idea of extending stability theory by considering a more general class of problems (not only of singular perturbed form) is not new, but has already been done by Auzinger, Schranz-Kirlinger and Frank (see [1]). In this paper, a new ansatz, basing on a generalised form of singular pertubation, is introduced. Definition 1 We call an ODE sytem a singular perturbed (SPP) problem iff it takes the form ż1 = F (z1, z2), (1) ż2 = 1 ǫ F (z1, z2), (2) where the real part of the spectrum of the Jacobian matrix Dz2F ♯ is smaller than some −b0 for b0 > 0. There is a well developped theory about the dynamics of SPP, basing on the works of Fenichel [2]. In [3] Nipp has shown that under some reasonable assumptions, SPP have a strongly attractive smooth invariant manifold with moderate flow. Definition 2 A system ẏ = G(y) is called quasi-singular perturbed (QSPP) iff there exists a transformation y = Φ(z) with ‖DΦ‖ = O(1) and ‖D(Φ)‖ = O(1) which transforms the system to singular perturbed form.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Convergence of product integration method applied for numerical solution of linear weakly singular Volterra systems

We develop and apply the product integration method to a large class of linear weakly singular Volterra systems. We show that under certain sufficient conditions this method converges. Numerical implementation of the method is illustrated by a benchmark problem originated from heat conduction.

متن کامل

On the convergence speed of artificial neural networks in‎ ‎the solving of linear ‎systems

‎Artificial neural networks have the advantages such as learning, ‎adaptation‎, ‎fault-tolerance‎, ‎parallelism and generalization‎. ‎This ‎paper is a scrutiny on the application of diverse learning methods‎ ‎in speed of convergence in neural networks‎. ‎For this aim‎, ‎first we ‎introduce a perceptron method based on artificial neural networks‎ ‎which has been applied for solving a non-singula...

متن کامل

Reproducing Kernel Hilbert Space(RKHS) method for solving singular perturbed initial value problem

In this paper, a numerical scheme for solving singular initial/boundary value problems presented.By applying the reproducing kernel Hilbert space method (RKHSM) for solving these problems,this method obtained to approximated solution. Numerical examples are given to demonstrate theaccuracy of the present method. The result obtained by the method and the exact solution are foundto be in good agr...

متن کامل

Optimal Control for Descriptor Systems: Tracking Problem (RESEARCH NOTE)

Singular systems have been studied extensively during the last two decades due Abstract to their many practical applications. Such systems possess numerous properties not shared by the well-known state variable systems. This paper considers the linear tracking problem for the continuous-time singular systems. The Hamilton-Jacobi theory is used in order to compute the optimal control and associa...

متن کامل

The inexact, inexact perturbed, and quasi-Newton methods are equivalent models

A classical model of Newton iterations which takes into account some error terms is given by the quasi-Newton method, which assumes perturbed Jacobians at each step. Its high convergence orders were characterized by Dennis and Moré [Math. Comp. 28 (1974), 549–560]. The inexact Newton method constitutes another such model, since it assumes that at each step the linear systems are only approximat...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2006