Accurate frequency response function identification of LPV systems: A 2D local parametric modeling approach

Abstract

In the last decades, LPV control has emerged as a systematic approach to design gain-scheduled controllers. This development has led to the emergence of LPV identification techniques. The aim of this paper is to present a new approach that enables fast and accurate identification of nonparametric frequency response functions of LPV systems. Main applications of the proposed approach include pre-testing for global LPV modeling and the use of nonparametric models as a basis for local LPV modeling. The proposed method extents recent developments of local parametric modeling techniques to a 2D situation. Pre-existing LPM/LRM methods are recovered as a special case. The advantages of the proposed approach are illustrated on a simulation model.