Adaptive identification of nonlinear structure uncertain perturbation system via different time scales neural networks

In this paper, an adaptive on-line identification algorithm is proposed for nonlinear structure uncertain perturbation systems via discrete different time scales dynamic neural networks. The main contributions of this paper are: (1) it is the first time to develop an identifier for nonlinear structure uncertain perturbation systems by using different time scales dynamic neural networks in discreet time domain (2) the input-to-state stability (ISS) approach is used to tune the weights of the discrete different time scales dynamic neural networks in the sense of L_∞. The commonly used robustifying techniques, such as dead-zone or σ-modification in the weight tuning, are not necessary for the proposed identification algorithm. The stability of the proposed identifier is proved by Lyapunov function and ISS theory. Simulation results are given to demonstrate the correctness of the theoretical results.