A System Identification Method to Hammerstein Model Based on Modified Shuffled Frog Leaping Algorithm

Nonlinear system identification has always been a hot and difficult problem in the field of identification research. In this paper, a modified shuffled frog leaping algorithm is proposed for the typical nonlinear Hammerstein model. The proposed method converted the parameter identification problem in nonlinear system into a function optimization problem in the parameter space. Specifically, the learning factor is introduced into the standard shuffled frog leaping algorithm for heuristic search. This can obtain the optimal solution in the whole parameter space. Finally, the proposed method is verified by experiments in the accuracy of parameter identification and the advantage of convergence.