Adaptive fuzzy model-based control
Martin Fischer, Oliver Nelles, Alexander Fink · TUbilio (Technical University of Darmstadt) · 1999
A novel approach for adaptive fuzzy model-based control is proposed. A nonlinear predictive controller is designed on the basis of a Takagi-Sugeno fuzzy model. By on-line adaptation of the fuzzy model, high control performance can be achieved even with time-variant process behaviour or changing unmodelled disturbances. For this purpose, a local recursive weighted least-squares algorithm is utilised which exploits the local linearity of Takagi-Sugeno fuzzy models. In order to cope with problems resulting from insufficient excitation a variable forgetting factor is introduced. The effectiveness and real-world applicability of the proposed approach are demonstrated by application to temperature control of a heat exchanger.