A New Orthogonal Least Squares Identification Method for a Class of Fractional Hammerstein Models

Authors:
Open Access
Publisher website
Google Scholar
Abstract
It is known that fractional-order models can effectively represent complex high-order systems with fewer parameters. This paper focuses on the identification of a class of multiple-input single-output fractional Hammerstein models. When the commensurate order is assumed to be known, a greedy orthogonal least squares method is proposed to simultaneously identify the parameters and system orders, combined with a stopping rule based on the Bayesian information criterion. Subsequently, the commensurate order is determined by minimizing the normalized output error. The proposed method is validated by applying it to identify a CD-player arm system.
Funding Information
  • National Natural Science Foundation of China (61973137, BK20201339, 2022M711361, B23008)
  • Natural Science Foundation of Jiangsu Province (61973137, BK20201339, 2022M711361, B23008)
  • China Postdoctoral Science Foundation (61973137, BK20201339, 2022M711361, B23008)
  • Higher Education Discipline Innovation Project (61973137, BK20201339, 2022M711361, B23008)

This publication has 33 references indexed in Scilit: