Optimal control of Sobolev type linear equations

From MaRDI portal
Publication:2388064





The paper concerns minimization of the cost functional \[ J(u)=\| Cx(\tau)-z_0(\tau)\|^2+\sum_0^{p+1}\int_0^\tau \langle N_qu^{(q)}(t),u^{(q)}(t)\rangle\,dt \] subject to the Cauchy system \(L\dot{x}(t)=Mx(t)+y(t)+Bu(t)\), \(x(0)=x_0\). Here the linear, continuous operator \(L\) and the linear, closed, densely defined operator \(M\) generate a strongly continuous semigroup for the homogeneous equation \(L\dot{x}(t)=Mx(t)\) as well as a certain nilpotent operator of degree \(p\), at most [see \textit{V.~E.~Fedorov}, ``Degenerate strongly continuous semigroup of operators, St.\ Petersbg.\ Math. J. 12, No.~3, 471--489 (2001); translation from Algebra Anal. 12, No.~3, 173--200 (2001; Zbl 0988.47027)], whereas the control \(u\) belongs to a Sobolev-Bochner space \(W_2^{p+1}\). The main results establish existence and uniqueness of a strong solution \(x\) to the Cauchy system, and existence of an optimal control \(u\) in case of certain control sets (possibly depending on \(y\) and \(x_0\)). The results are applied to a class of initial-boundary value problems for partial differential equations.




Cited in
(23)








This page was built for publication: Optimal control of Sobolev type linear equations

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2388064)