A regularized gradient-projection method in a parabolic optimal control problem
The optimal control problem for a system governed by the linear heat equation is considered. The control is included in the constant term of the equation and in the boundary condition. The permissible control set is convex. The cost function is quadratic. The convergence of the finite-difference method for the boundary value problem and cost function is proved. The gradient projection and the Tychonov regularization method are used. A convergence criterion for the considered methods is given. The proof is based on known results of F. P. Vasil'ev. Analogous results can be obtained for any linear systems with quadratic cost function and convex permissible set if the convergence of the approximation method for the considered equation is proved. The spontaneous transfer of this results to nonlinear systems is impossible.
- Method of Tikhonov regularization in a problem of optimal control of a nonlinear parabolic system
- The gradient projection method for solving an optimal control problem
- scientific article; zbMATH DE number 4040669
- scientific article; zbMATH DE number 3919166
- A smooth regularization of the projection formula for constrained parabolic optimal control problems
- Estimate of the rate of convergence of the method of lines in the problem of optimal heating of a rod
- A smooth regularization of the projection formula for constrained parabolic optimal control problems
- The gradient projection method for solving an optimal control problem
- scientific article; zbMATH DE number 1223645 (Why is no real title available?)
- scientific article; zbMATH DE number 5585710 (Why is no real title available?)
- scientific article; zbMATH DE number 5225085 (Why is no real title available?)
- Numerical Analysis and Its Applications
This page was built for publication: A regularized gradient-projection method in a parabolic optimal control problem
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1802602)