scientific article; zbMATH DE number 431575
This paper is devoted to the study of an optimal control problem where the state equation is a linear, elliptic, partial derivative equation \((- \Delta y=v\) on \(\Omega\subset\mathbb{R}^ n)\) and the cost functional is quadratic. The main difficulty comes from the fact that the state is constrained: \(y\in K\) (where \(K\) is closed and convex). It is well-known that a qualification condition is needed to get decoupled optimality relations characterizing the solution of such a problem. The most standard one is the Slater condition, which requires (in particular) the \(\mathbb{L}^ \infty\)-interior of \(K\) to be non-empty. This condition is rather strong and it is important to weaken it. For this purpose, the state-equation is penalized so that the state and the control are ``independent one of each other, and a penalized optimality system is obtained related to the penalized optimal control problem. At this step, a qualification hypothesis \((H)\) is done to be allowed to pass to the limit in the penalized system. Thus, optimality conditions may be set where the state and the control are completely ``decoupled. The main novelty of the paper is that assumption \((H)\) is much weaker than the Slater condition, so that a quite large class of problems are concerned. For the entire collection see [Zbl 0771.00019].
- A Penalization Method for Optimal Control of Elliptic Problems with State Constraints
- Elliptic control by penalty techniques with control reduction
- Optimality Conditions for Some Nonqualified Problems of Distributed Control
- Method of penalization for the state equation for an elliptical optimal control problem
- Optimal control of parabolic problems with state constraints: A penalization method for optimality conditions
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3138347)