Completeness of a system of elementary solutions to a class of operator-differential equations on a finite interval (Q606566)

Let \(H\) be a separable Hilbert space and \(A\) be a positive selfadjoint operator in \(H\). Assume that the operators \(A_j\) (\(j=1,\, 2,\dots , 2m\)) are linear and possibly unbounded in \(H\), while \(\theta_\nu\) and \(\kappa_\nu\) (\(\nu=0,\, 1,\dots , m-1\)) are vectors from \(H\). This paper is devoted to the study of the system (S) \[ P\left(\frac{d}{dt}\right)u(t)=\left(-\frac{d^2}{dt^2}+A^2\right)^mu(t)+\sum_{j=1}^{2m}A_ju^{2m-j}(t)=0, \] for \(t\in (0,1)\) and \(u^{(\nu)}(0)=\theta_\nu\), \(u^{(\nu)}(1)=\kappa_\nu\), for all \(\nu=0,\, 1,\dots , m-1\). This problem is associated with the differential operator \[ P(\lambda)=(-\lambda^2E+A^2)^m+\sum_{j=1}^{2m}A_j\lambda^{2m-j}. \] The following results are communicated in the present paper: (i) conditions on the coefficients of \(P(\lambda)\) under which system (S) has a unique weak solution; (ii) conditions such that the system of derivative chains of eigen- and associated vectors corresponding to problem (S) is \(2m\)-multiply complete and minimal; (iii) conditions such that the elementary solutions are complete and minimal in the space of all weak solutions to problem (S).