Hierarchical gradient parameter estimation algorithms for fractional order Wiener OEARMA system

In this paper, the identification of fractional order Wiener output error auto-regressive moving average (OEARMA) systems is discussed. The dynamic part of the Wiener OEARMA system is an OEARMA structure, and the static part is two-stage nonlinearity. The identification expression is obtained by key...

Full description

Saved in:
Bibliographic Details
Published in:Nonlinear dynamics Vol. 113; no. 15; pp. 19579 - 19598
Main Authors: Li, Junhong, Zhang, Hongrui, Xiao, Kang, Gu, Juping
Format: Journal Article
Language:English
Published: Dordrecht Springer Nature B.V 01.08.2025
Subjects:
Accuracy
Algorithms
Autoregressive moving average
Calculus
Engineering
Estimates
Parameter estimation
Parameter identification
ISSN:0924-090X, 1573-269X
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this paper, the identification of fractional order Wiener output error auto-regressive moving average (OEARMA) systems is discussed. The dynamic part of the Wiener OEARMA system is an OEARMA structure, and the static part is two-stage nonlinearity. The identification expression is obtained by key term separation technique and definition of the Grünwald Letnikov factional differential. Then, the hierarchical extended stochastic gradient (H-ESG) and hierarchical multi-innovation extended stochastic gradient (H-MIESG) methods are proposed for identification of the unknown parameter in the system and the the convergence is verified. Through numerical simulations, the feasibility of the derived algorithms is studied. The identification accuracy of H-MIESG is satisfactory, which reflects its excellent identification efficiency.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0924-090X
1573-269X
DOI:10.1007/s11071-025-11187-z