Stability of a one-degree-of-freedom canonical system in the case of zero quadratic and cubic part of a Hamiltonian (Q2197282): Difference between revisions

From MaRDI portal
RedirectionBot (talk | contribs)
Removed claims
RedirectionBot (talk | contribs)
Changed an Item
Property / author
 
Property / author: Victor Lanchares / rank
 
Normal rank
Property / reviewed by
 
Property / reviewed by: William J. Satzer jun. / rank
 
Normal rank

Revision as of 13:12, 16 February 2024

scientific article
Language Label Description Also known as
English
Stability of a one-degree-of-freedom canonical system in the case of zero quadratic and cubic part of a Hamiltonian
scientific article

    Statements

    Stability of a one-degree-of-freedom canonical system in the case of zero quadratic and cubic part of a Hamiltonian (English)
    0 references
    0 references
    0 references
    31 August 2020
    0 references
    This paper studies a special case of a one-degree-of-freedom Hamiltonian system and investigates the stability of its equilibrium. The authors follow up on a question raised in the paper [\textit{A. P. Markeev}, Nelineĭn. Din. 10, No. 4, 447--464 (2014; Zbl 1372.70053)] and resolve one of the cases that that paper identifies. Their one-degree-of-freedom Hamiltonian system is assumed to be periodic in time and defined by the usual canonical equations. The authors assume that there is an equilibrium at the origin, and that the Hamiltonian \(H = H(x,y,t)\) can be expanded in a convergent power series in a small neighborhood of the origin. In particular, \(H(x,y,t) = \sum_{k=2}^{\infty} H_k (x,y,t),\) where \(H_k(x,y,t) = \sum_{\nu + \mu = k} h_{\nu \mu} x^\nu y^\mu\) and where \(\nu\) and \(\mu\) are nonnegative integers and the coefficients \(h_{\nu \mu}\) are continuous \((2 \pi)\)-periodic functions. The authors assume in addition that a resonance of first or second order occurs in this Hamiltonian system, and that the monodromy matrix is diagonal. In this case, the equilibrium at the origin is linearly stable, but the analysis of terms of \(H\) of order three and higher is necessary to determine Lyapunov stability. Under their assumptions the original system can be rewritten in canonical variables \(q\) and \(p\) so that \(H(q,p,t) = \sum_{k=3} ^ {N} H_k(q,p) + \sum_{k=N+1}^{\infty} H_k(q,p,t)\) where \(H_k = \sum_{\nu +\mu = k} h_{\nu\mu} q^\nu p^\mu\). The authors further restrict to the case where \(H_3\) is identically zero because the nonzero case had already been considered in the reference cited above. In that reference the author also notes that with a linear canonical change of variables the \(H_4\) term can be brought to one of nine simple forms. The authors here consider the ninth case where \(H_4 = q^4\). The question had been considered in [\textit{R. Gutierrez} and \textit{C. Vidal}, Regul. Chaotic Dyn. 22, No. 7, 880--892 (2017; Zbl 1403.37066)], but not completely resolved. In this paper the authors establish conditions that guarantee stability.
    0 references
    Hamiltonian systems
    0 references
    Lyapunov stability
    0 references
    normal forms
    0 references
    KAM theory
    0 references
    degeneracy
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references