Conference Contribution
Parameter Dependent Model Reduction Framework with Applications
Aivar Sootla, Anders Rantzer
Abstract
In this paper a recently proposed model reduction method for a class of linear parameterized models is investigated. A linear parameterized model is a model with coefficients of transfer function depending on a parameter. A multivariable extension is developed, for which a polynomial time algorithm is developed to obtain an approximation. The main focus of this paper is on the properties of approximations. Among those properties is stability, continuity with respect to parameter, error bounds of approximations.
Keywords
LPV identification, model reduction, convex optimization
In Proc. Reglermöte, Lund, Sweden, June 2010. Published online.