Stoll, Martin and Wathen, A. J.
Preconditioning for active set and projected gradient methods as
semi-smooth Newton methods for PDE-constrained optimization
with control constraints. Not specified . (Submitted)
Optimal control problems with partial differential equations play an important role in many applications. The inclusion of bound constraints for the control poses a significant additional challenge for optimization methods. In this paper we propose preconditioners for the saddle point problems that arise when a primal-dual active set method is used. We also show for this method that the same saddle point system can be derived when the method is considered as a semi-smooth Newton method. In addition, the projected gradient method can be employed to solve optimization problems with simple bounds and we discuss the efficient solution of the linear systems in question. In the case when an acceleration technique is employed for the projected gradient method, this again yields a semi-smooth Newton method that is equivalent to the primal-dual active set method. Numerical results illustrate the competitiveness of this approach.
|Subjects:||H - N > Numerical analysis|
|Research Groups:||Oxford Centre for Collaborative Applied Mathematics|
|Deposited By:||Dr M. Stoll|
|Deposited On:||24 Sep 2009 12:34|
|Last Modified:||29 May 2015 18:31|
Repository Staff Only: item control page