Nonautonomous bifurcation of bounded solutions. I: a Lyapunov-Schmidt approach (Q1956555)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Nonautonomous bifurcation of bounded solutions. I: a Lyapunov-Schmidt approach
scientific article

    Statements

    Nonautonomous bifurcation of bounded solutions. I: a Lyapunov-Schmidt approach (English)
    0 references
    22 September 2010
    0 references
    The author studies local bifurcations of bounded complete solutions of non-autonomous difference equations and non-autonomous ordinary differential equations in general Banach spaces. (It is interesting to stress that the bifurcating objects are not necessarily equilibria points or periodic solutions.) By bifurcation is understood a change in the number of such solutions under variation of a parameter \(\lambda\). In both cases, the main idea is to formulate the dynamic equations as operator equations in appropriate sequence spaces (in the case of difference equations) or function spaces (in the case of differential equations); then the development of a suitable Fredholm theory enables the author to use tools of branching theory, such as the Lyapunov-Schmidt reduction which yields explicit conditions in order to guarantee the bifurcation of the solutions. In Section 2, the author analyzes the non-autonomous difference equation \(x_{k+1}=f_{k}(x_k,\lambda),\) where \(f_k: \Omega\times \Lambda\rightarrow X\), \(k\in\mathbb Z\), are uniformly continuous maps and \(\Omega\subseteq X, \Lambda\subseteq Y\) are nonempty open convex sets of Banach spaces \(X,Y\). Among others, it is worth pointing out that, when the linearization admits a 1-dimensional space of bounded complete solutions, the author is able to obtain non-autonomous counterparts of the classical bifurcation patterns (fold, transcritical and pitchfork), replacing equilibria points by these complete solutions. In Section 3, the author obtains, in the same line as in Section 2, a battery of results on bifurcation for ordinary differential equations \(\dot{u}=f(t,u,\lambda).\) Here, \(f:\mathbb R\times\Omega\times \Lambda\rightarrow X\) is a continuous map satisfying certain hypotheses which ensure existence and uniqueness of solutions for the differential equation. In a final appendix, the functional-analytic tools required in the paper are briefly reviewed (Lyapunov-Schmidt reduction and abstract versions of fold, transcritical and pitchfork bifurcations). Moreover, the fine references of the paper will give the readers a good perspective of the topic of general bifurcation methods.
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    non-autonomous bifurcation
    0 references
    non-autonomous difference equation
    0 references
    non-autonomous differential equation
    0 references
    Fredholm operator
    0 references
    Lyapunov-Schmidt reduction
    0 references
    complete solution
    0 references
    dichotomy spectrum
    0 references
    0 references