Shadowing orbits of ordinary differential equations (Q1342915): Difference between revisions

From MaRDI portal
Set OpenAlex properties.
ReferenceBot (talk | contribs)
Changed an Item
 
Property / cites work
 
Property / cites work: Q5566243 / rank
 
Normal rank
Property / cites work
 
Property / cites work: \(\omega\)-limit sets for axiom A diffeomorphisms / rank
 
Normal rank
Property / cites work
 
Property / cites work: On the numerical computation of orbits of dynamical systems: The one- dimensional case / rank
 
Normal rank
Property / cites work
 
Property / cites work: On the numerical computation of orbits of dynamical systems: The higher dimensional case / rank
 
Normal rank
Property / cites work
 
Property / cites work: Rigorous computational shadowing of orbits of ordinary differential equations / rank
 
Normal rank
Property / cites work
 
Property / cites work: Hyperbolicity and chain recurrence / rank
 
Normal rank
Property / cites work
 
Property / cites work: Do numerical orbits of chaotic dynamical processes represent true orbits? / rank
 
Normal rank
Property / cites work
 
Property / cites work: Numerical orbits of chaotic processes represent true orbits / rank
 
Normal rank
Property / cites work
 
Property / cites work: Q4340096 / rank
 
Normal rank
Property / cites work
 
Property / cites work: Rigorous verification of trajectories for the computer simulation of dynamical systems / rank
 
Normal rank
Property / cites work
 
Property / cites work: GIBBS MEASURES IN ERGODIC THEORY / rank
 
Normal rank

Latest revision as of 11:28, 23 May 2024

scientific article
Language Label Description Also known as
English
Shadowing orbits of ordinary differential equations
scientific article

    Statements

    Shadowing orbits of ordinary differential equations (English)
    0 references
    0 references
    0 references
    0 references
    7 June 1995
    0 references
    This note describes conditions that ensure that a numerically computed approximate orbit of a \(C^ 2\) vector field \(f\) on \(R^ n\) is shadowed by an actual orbit of the vector field. The details of the numerical method are contained in another paper by the same authors (Numer. Math., to appear). A \(\delta\) pseudo-orbit is a sequence of points \(\{y_ k\}\), \(0 \leq k \leq N\) and associated positive times \(h_ k\) with the property that the solution with initial position \(y_ k\) is located at time \(h_ k\) within \(\delta\) of \(y_{k+1}\). Their result can essentially be described as follows: Suppose that \(U\) is an open convex set in phase space and that \(\{y_ k\}\), \(0 \leq k \leq N\) is a \(\delta\) pseudo-orbit such that any orbit segment of length \(2h_ k\) beginning within some distance \(\varepsilon_ 0\) of \(y_ k\) lies inside \(U\). If \(\delta\) satisfies certain conditions then there is a constant \(\varepsilon < \varepsilon_ 0\), a point \(x \in U\), and times \(t_ k\) within \(\varepsilon\) of \(h_ k\) such that the orbit with initial position \(x\) is at time \(t_ k\) within \(\varepsilon\) of \(y_ k\). The conditions on \(\delta\) are a number of inequalities relating \(\delta\) with several other quantities, including the magnitudes of \(f\) and its first two derivatives on all of \(U\), the lengths \(h_ k\) of the orbit segments, the minimum of the magnitude of \(f\) along the pseudo-orbit, and estimates of the linearization of the flow from \(y_ k\) to \(y_{k+1}\).
    0 references
    0 references
    shadowing
    0 references
    \(\delta\) pseudo-orbit
    0 references
    numerically computed approximate orbit
    0 references
    vector field
    0 references
    0 references
    0 references