One-step-ahead adaptive control for a general class of nonlinear dynamic systems based on fuzzy models

A one-step-ahead adaptive control scheme based on fuzzy models is presented for a class of nonlinear dynamic systems which have no restriction on the structure. The specific features of the fuzzy systems are taken into consideration. The proposed method has the following properties: 1) by properly designing the fuzzy system, the modeling error converges to an error band which can be made arbitrarily small; and 2) with mild assumptions, the tracking error of the closed-loop system is guaranteed within an /spl epsiv/-neighborhood of zero and the stability is ensured. Tests on the data from a real steel-heating furnace are given to show the effectiveness of the scheme.