Author : Adrian Ziessler
Release : 2018
Genre :
Kind : eBook
Book Rating : /5 ( reviews)
Book Synopsis Analysis of Infinite Dimensional Dynamical Systems by Set-oriented Numerics by : Adrian Ziessler
Download or read book Analysis of Infinite Dimensional Dynamical Systems by Set-oriented Numerics written by Adrian Ziessler. This book was released on 2018. Available in PDF, EPUB and Kindle. Book excerpt: One central goal in the analysis of dynamical systems is the characterization of long term behavior of the system state. To this end, the so-called global attractor is of interest, that is, an invariant set that attracts all the trajectories of the underlying dynamical system. Over the last 20 years so-called set-oriented numerical methods have been developed that allow to compute approximations of invariant sets. The basic idea is to cover the objects of interest, for instance attractors or unstable manifolds, by outer approximations which are created via subdivision techniques. However, the applicability of those techniques is restricted to finite dimensional dynamical systems, i.e., ordinary differential equations or discrete dynamical systems. In this thesis, we extend the set-oriented numerical methods to the infinite dimensional context. With those extensions we will be able to compute finite dimensional invariant sets for infinite dimensional dynamical systems, e.g., for delay and partial differential equations. The idea is to utilize infinite dimensional embedding techniques in our numerical treatment. This will allow us to construct a finite dimensional dynamical system, the core dynamical system (CDS), on an appropriately chosen observation space. Using the CDS, we then can approximate finite dimensional embedded attractors or embedded unstable manifolds. Furthermore, we will be able to compute approximations of the embedded invariant measure in the observation space which gives a statistical description of the dynamical behavior of the infinite dimensional dynamical system. We present numerical realizations of the CDS for delay and partial differential equations and illustrate the efficiency of our approach in several examples. Furthermore, we present modifications for the set-oriented subdivision and continuation method. ... ; eng