Neural networks predictive control using an adaptive control rate

This paper deals with the predictive control of discrete time nonlinear systems based on artificial neural networks. The system behavior is described by a neural networks model and the control law is obtained by minimizing a quadratic cost function. An adaptive variable control rate which is based on Lyapunov function candidate and assumes the closed loop stability is developed. A simulation example is given in order to illustrate the performances of the proposed approach.