Special two-stage input signal based neuro-fuzzy model for Hammerstein-Wiener processes
Li Jia, Aihua Yang, Min‐Sen Chiu · International Journal of System Control and Information Processing · 2012
In this paper, a special two-stage input signal based neuro-fuzzy model for Hammerstein-Wiener processes is presented. The input and output non-linear static parts of the Hammerstein-Wiener process are described by two independent neuro-fuzzy models without any prior process knowledge, thus avoiding the inevitable restrictions on static non-linear function encountered by using the polynomial approach. To construct the neuro-fuzzy-based Hammerstein-Wiener model, special two-stage input signal is carried out, and an analytical solution is developed to calculate the parameters of the linear dynamic part and two static non-linear functions. Examples are used to illustrate the applicability of the proposed method and a comparison with polynomial approach is made.