A SVD based controller reduction method

Author

Sou, Kin Cheong ; Rantzer, Anders

Author_Institution

Dept. of Autom. Control, Lund Univ., Lund, Sweden

fYear

2010

fDate

18-21 Aug. 2010

Firstpage

2564

Lastpage

2569

Abstract

In this paper an efficient controller reduction method which guarantees closed loop stability is described. The reduced controller is obtained by solving a generalized controllability Grammian type Lyapunov inequality. This Lyapunov inequality based reduced controller search can be formulated as a generalized minimum rank matrix approximation problem to which an efficient singular value decomposition based solution procedure is available. This paper contains theoretical and numerical results on the merit of the proposed reduction method.

Keywords

Lyapunov matrix equations; approximation theory; closed loop systems; controllability; linear matrix inequalities; reduced order systems; singular value decomposition; SVD based controller reduction method; closed loop stability; controller reduction method; generalized controllability Grammian type Lyapunov inequality; generalized minimum rank matrix approximation problem; singular value decomposition; Approximation methods; Closed loop systems; Frequency control; Linear matrix inequalities; Minimization; Stability analysis; Transfer functions; controller reduction; matrix approximation; rank minimization; singular value decomposition;

fLanguage

English

Publisher

ieee

Conference_Titel

SICE Annual Conference 2010, Proceedings of

Conference_Location

Taipei

Print_ISBN

978-1-4244-7642-8

Type

conf

Filename

5602466