Title :
Multivariable Optimization-Based Model Reduction
Author :
Sootla, Aivar ; Rantzer, Anders ; Kotsalis, Georgios
Author_Institution :
Autom. Control Dept., Lund Univ., Lund, Sweden
Abstract :
In this technical note we introduce a multivariable extension of a recently proposed model reduction algorithm for linear time-invariant systems. Reduced models are found by solving a convex optimization problem with linear matrix inequality constraints given a state space model or frequency samples of a model. In order to illustrate the method we apply it to a large-scale model of a deformable mirror.
Keywords :
convex programming; linear matrix inequalities; linear systems; multivariable systems; reduced order systems; convex optimization problem; linear matrix inequality constraints; linear time-invariant systems; model reduction algorithm; multivariable optimization; Computational efficiency; Constraint optimization; Deformable models; Frequency; Large-scale systems; MIMO; Mirrors; Reduced order systems; State-space methods; Transfer functions; Multi-input-multi-output (MIMO); single-input-single-output (SISO);
Journal_Title :
Automatic Control, IEEE Transactions on
DOI :
10.1109/TAC.2009.2029402
