Design of auxiliary model and hierarchical normalized fractional adaptive algorithms for parameter estimation of bilinear‐in‐parameter systems

Summary This study investigates the parameter identification issues of bilinear‐in‐parameter systems through fractional adaptive algorithms. An auxiliary model based ε‐normalized$$ \varepsilon \hbox{-} \mathrm{normalized} $$ modified fractional least mean square algorithm is proposed for acceleratin...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:International journal of adaptive control and signal processing Jg. 36; H. 10; S. 2562 - 2584
Hauptverfasser: Zhu, Yancheng, Wu, Huaiyu, Chen, Zhihuan, Chen, Yang, Zheng, Xiujuan
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Hoboken, USA John Wiley & Sons, Inc 01.10.2022
Wiley Subscription Services, Inc
Schlagworte:
Accuracy
Adaptive algorithms
auxiliary model
bilinear‐in‐parameter system
fractional adaptive algorithms
hierarchical identification
Mathematical models
Mean square values
Parameter estimation
Parameter identification
ISSN:0890-6327, 1099-1115
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Summary This study investigates the parameter identification issues of bilinear‐in‐parameter systems through fractional adaptive algorithms. An auxiliary model based ε‐normalized$$ \varepsilon \hbox{-} \mathrm{normalized} $$ modified fractional least mean square algorithm is proposed for accelerating the parameter estimation accuracy based on the auxiliary model identification idea and the introduced convergence index, a normalized modified hierarchical fractional least mean square algorithm is presented for improving the computational efficiency based on the hierarchical identification principle. The proposed normalized fractional adaptive strategies are effective and could provide more accurate parameter estimates comparing with conventional counterparts for bilinear‐in‐parameter identification model based on the mean square error metrics and the average predicted output error. The effectiveness and accuracy of the proposed algorithms are further verified and validated through numerical simulations for different noise variances, fractional orders and gain parameters.
Bibliographie:Funding information
National Natural Science Foundation of China, Grant/Award Numbers: 62003249; 62073250; 62173262
ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0890-6327
1099-1115
DOI:10.1002/acs.3471