Stable predictive control based on recurrent fuzzy neural networks

This paper presents a design method for stable predictive control (SPC) of nonlinear discrete-time system via recurrent fuzzy neural networks (RFNNs). An RFNN-based predictive control law is derived based on the minimization of a modified predictive performance criterion. Two theorems are presented for the stability and steady-state performance of the closed-loop systems. The results from numerical simulations show that the proposed SPC method is capable of controlling nonlinear systems with satisfactory performance under setpoint and disturbance changes.