Dichotomies and asymptotic formulas for the solutions of differential equations (Q1909982)

From MaRDI portal
Revision as of 18:43, 21 March 2024 by Openalex240321050300 (talk | contribs) (Set OpenAlex properties.)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
scientific article
Language Label Description Also known as
English
Dichotomies and asymptotic formulas for the solutions of differential equations
scientific article

    Statements

    Dichotomies and asymptotic formulas for the solutions of differential equations (English)
    0 references
    0 references
    8 October 1996
    0 references
    The paper deals with a linear system (1) \(\dot x= A(t) x\), where \(A(t)\) is an \(n\times n\) continuous matrix defined on \({\mathcal I}= [a, \infty)\). Let \(\Phi (t)\) be a fundamental matrix of (1), \(h(t)\) and \(k(t)\) be continuous functions satisfying the compensation law: \(h(t) h^{-1} (s) k(t) k^{-1} (s)\leq c_1\) \((t\geq s\geq a)\), where \(c_1\) is a positive constant. Suppose that system (1) possesses an \((h, k)\)-dichotomy, that is, there exists a projection \(P\) such that \(|\Phi (t) P\Phi^{-1} (s) |\leq c h(t) h^{-1} (s)\) \((t\geq s\geq a)\), \(|\Phi(t) (E-P) \Phi^{-1} (s)|c k^{-1} (t) k(s)\) \((s\geq 1\geq a)\) [the author, Contributions US-ACH, 1-14 (1992)], where \(c= \text{const}>0\). Let \(b(t)\) be a continuous vector, \(t\in {\mathcal I}\); \(f: {\mathcal I}\times C^n\to C^n\) be a continuous function such that \(|f(t, y_1)- f(t, y_2) |\leq \lambda (t) |y_1- y_2|\) for \((t, y_i)\in {\mathcal I}\times C^n\) \((i= 1,2)\) and for a nonnegative function \(\lambda\in L_1 ({\mathcal I})\) and \(h^{-1} (t) f(t, 0)\in L_1 ({\mathcal I})\). The author solves asymptotic equivalence problems for systems (2) \(\dot x= A(t) x+ b(t)\) and (3) \(\dot y= A(t) y+ b(t) f(t, y)\). Let by \(h_+ =h\) and \(h_-= k^{-1}\) and introduce the Banach spaces \(C_\pm= C_\pm (t_0)\) defined by \(C_\pm (t_0)= \{z\in C [t_0, \infty): |y|_\pm< \infty\}\), where \(t_0\geq a\), \(|y|_\pm= \sup_{t\geq t_0} \{h^{-1}_\pm (t)|y(t) |\}\). The main result of this paper is Theorem 1: Assume that the above-mentioned hypotheses hold. Then there exists a one-to-one continuous correspondence between the solutions \(x_+\in C_+\) (respectively, \(x_-\in C_-\)) of system (2) and the solutions \(y_+\in C_+\) (respectively, \(y_-\in C_-\)) of system (3). Moreover, we have the two formulas \(y_\pm= x_\pm+ |\Phi P|\cdot \widetilde {o} (1)+ h_\pm \cdot o(1)\), \(t\to \infty\), where \(\widetilde {o} (1)\) is a convergent function (as \(t\to \infty\)). If \(h^{-1} (t) \Phi (t) P\to 0\) as \(t\to \infty\) we obtain \(y_\pm= x_\pm+ h_\pm \cdot O(1)\), as \(t\to \infty\).
    0 references
    0 references
    0 references
    0 references
    0 references
    linear system
    0 references
    \((h, k)\)-dichotomy
    0 references
    0 references