The performance of robust controllers depends on the set of candidate plants, but at present this intimate connection is untransparent. The aim of this paper is to construct a model set to improve the performance in a subsequent robust control design. Analysis of uncertainty structures reveals that there is an unexploited freedom in the realization of coprimefactorizations in the dual-Youla uncertainty structure. The main result of this paper is aspecific coprime factorization that results in model sets that are tuned for robust control. Thepresented coprime factorization can be identified directly from data. Application of the proposedmethodology to an industrial wafer stage reveals improved model validation results.
|Title of host publication||Proceedings of the 15th IFAC Symposium on System Identification (SYSID 2009) 6-8 July 2009, Saint-Malo, France|
|Place of Publication||Saint Malo, France|
|Publication status||Published - 2009|