A TSK-type recurrent fuzzy neural network adaptive inverse modeling control for a class of nonlinear discrete-time time-delay systems

The paper presents a Takagi-Sugeno-Kang (TSK)-type recurrent fuzzy neural network (TRFNN) adaptive inverse modeling control for a class of nonlinear discrete-time time-delay systems. This type of controller uses a TRFNN as an adaptive inverse modeling controller. TRFNN is a recurrent fuzzy neural network developed from a series of TSK-type fuzzy if-then rules, and its consequent parameters learning is adopted two types of learning algorithms, the least-squared-error (off-line training) and the gradient descent learning (online training) algorithms. The adaptive inverse modeling control configuration requires no emulation of the plant and can be simple in implementation. Numerical simulations are conducted for controlling a highly nonlinear process. The results clearly indicate the excellent disturbance rejection and set-point tracking performance of the presented control method.