Oscillation results for linear autonomous partial delay differential equations

From MaRDI portal





A unified definition of oscillation of solutions of differential equations is given in terms of a set of truncated cones in a Banach space. Thus, if \({\mathcal K}\) is a set of truncated cones, then a continuous function \(u: [t_ 0,\infty) \to X\) is \({\mathcal K}\)-oscillating if \(u(t)\in K\) for \(t\geq t_ 0\), for some \(K\in{\mathcal K}\). A basic oscillation theorem is then proved for abstract equations of the form \[ \dot u= Au+ \int_{-r}^ 0 d\eta(s) u(t+s). \] In fact, all solutions oscillate with respect to \({\mathcal K}\), if \(\{\lambda\in \mathbb{R}\): \(\Delta^{-1} (\lambda)\in {\mathcal B}(X)\}= \mathbb{R}\) and \[ \liminf_{\lambda\to -\infty} {\textstyle {1\over {|\lambda |}}} \log\| \Delta^{-1} (\lambda)\| <\infty, \] where \(\Delta^{-1}\) is the resolvent operator. The result is applied to finite-dimensional and parabolic and hyperbolic delay equations.











This page was built for publication: Oscillation results for linear autonomous partial delay differential equations

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2367835)