Hierarchical gradient- and least squares-based iterative algorithms for input nonlinear output-error systems using the key term separation

Feng Ding, Hao Ma, Jian Pan, Erfu Yang

Research output: Contribution to journalArticlepeer-review

Abstract

This paper considers the parameter identification problems of the input nonlinear output-error (IN-OE) systems, that is the Hammerstein output-error systems. In order to overcome the excessive calculation amount of the over-parameterization method of the IN-OE systems. Through applying the hierarchial identification principle and decomposing the IN-OE system into three subsystems with a smaller number of parameters, we present the key term separation auxiliary model hierarchical gradient-based iterative algorithm and the key term separation auxiliary model hierarchical least squares-based iterative algorithm, which are called the key term separation auxiliary model three-stage gradient-based iterative algorithm and the key term separation auxiliary model three-stage least squares-based iterative algorithm. The comparison of the calculation amount and the simulation analysis indicate that the proposed algorithms are effective. (c) 2021 The Franklin Institute. Published by Elsevier Ltd. All rights reserved.
Original languageEnglish
Pages (from-to)5113-5135
Number of pages23
JournalJournal of the Franklin Institute
Volume358
Issue number9
Early online date19 Apr 2021
DOIs
Publication statusPublished - 30 Jun 2021

Keywords

  • parameter-estimation
  • inn monolayer
  • risk model
  • identification method
  • promising candidate
  • collision-avoidance
  • bilinear-systems
  • tracking control
  • state
  • selection

Fingerprint

Dive into the research topics of 'Hierarchical gradient- and least squares-based iterative algorithms for input nonlinear output-error systems using the key term separation'. Together they form a unique fingerprint.

Cite this