Identification of Wiener Time Delay Systems Based on Hierarchical Gradient Approach

In this paper, we propose a hierarchical gradient based algorithm to identify simultaneously the parameters of the linear and nonlinear blocks and the time delay of Wiener time delay systems. We separate the nonlinear cost function into two cost functions and present the gradient algorithm for estimating directly the time delay and the parameters. Furthermore, we give an appropriate choice of convergence factor. Numerical simulation results are presented to demonstrate the performance of the proposed method.