Fuzzy modelling and control using parameterized linear filters
conference paper
The paper presents a nonlinear identification scheme, which consists of a linear dynamical section (filter) and a nonlinear zero-memory section (implemented by a fuzzy mapping). Only the filter section is on the primary signal path. The nonlinear mapping (depending on the system input and state) delivers the filter parameters. The identification assumes structural knowledge about the process with proper parameterization. An adaption procedure is introduced, which tunes the nonlinear mapping (e.g. membership function parameters) to minimize identification error. The adaption procedure is driven by the approximate dynamical sensitivity model of the system thus the method is very effective with respect to the number of training steps necessary to reach the accuracy required. The scheme proposed can incorporate a priory knowledge on two levels (structure and fuzzy rule set). One of the most distinctive features of the scheme is that it directly supports controller design and/or (on-line) tuning.
Topics
TNO Identifier
233754
Publisher
IEEE
Source title
Proceedings of the 1997 IEEE Instrumentation & Measurement Technology Conference, IMTC, 19-21 May 1997, Ottawa, Canada
Place of publication
Piscataway, NJ, USA
Pages
778-783
Files
To receive the publication files, please send an e-mail request to TNO Repository.