Parameter Learning Algorithms of Hammerstein Nonlinear Systems

This paper deals with the modeling and parameter estimation of Hammerstein systems from samples of input and output data. Compared with the previous work which needs to identify the parameter vectors separately, this paper introduces a new approach for simplifying the complexity of identification algorithms. The proposed strategy is that the system model is transformed to the first order linear parameter identification model based on the Taylor expansion; and a novel least squares algorithm is proposed for estimating the coupled parameters simultaneously. Moreover, the simulation results are provided for demonstrating the performance of the proposed algorithms.