Adaptive nonlinear systems identification via dynamic multilayer neural networks with two-time scales

This paper presents a novel adaptive identification method for nonlinear systems including the aspects of fast and slow phenomenon via dynamic multilayer neural networks with two-time scales. The Lyapunov function and singularly perturbed techniques are used to develop the learning procedure for the hidden layers and output layers of the dynamic neural networks model. Novel correction terms are proposed in the learning algorithm to guarantee bounded tracking errors and bounded weights. The effectiveness of the algorithm is illustrated via the simulation results on an electric induction motor.