Study of Fuzzy Generalized Predictive Control Algorithm On Nonlinear Systems

A novel control algorithm - fuzzy generalized predictive control (FGPC) algorithm for nonlinear systems is proposed. Fuzzy Gustafon-Kessel algorithm and weighted recursive least-square method are applied to identify the T-S model offline. The identified model is then combined with generalized predictive control algorithm to control non-linear systems. And the detailed steps of the algorithm have been given. The simulation results prove that the proposed algorithm has a high degree of identification accuracy, good robustness, and has obtained a good control result, and therefore is applicable for a variety of non-linear systems