Separable Recursive Gradient Algorithm for Dynamical Systems Based on the Impulse Response Signals

The identification for process control systems is considered in this paper based on the impulse response signals from the discrete measurements. By taking advantage of impulse signals and through the model parameter decomposition, two dependent identification models are constructed and two identific...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:International journal of control, automation, and systems Jg. 18; H. 12; S. 3167 - 3177
Hauptverfasser: Xu, Ling, Ding, Feng, Yang, Erfu
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Bucheon / Seoul Institute of Control, Robotics and Systems and The Korean Institute of Electrical Engineers 01.12.2020
Springer Nature B.V
제어·로봇·시스템학회
Schlagworte:
Algorithms
Control
Control Theory and Applications
Dynamical systems
Engineering
Impulse response
Mathematical models
Mechatronics
Optimization
Parameter estimation
Parameter identification
Performance tests
Robotics
제어계측공학
recursive algorithm
Gradient search
parameter estimation
system response
impulse signal
ISSN:1598-6446, 2005-4092
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The identification for process control systems is considered in this paper based on the impulse response signals from the discrete measurements. By taking advantage of impulse signals and through the model parameter decomposition, two dependent identification models are constructed and two identification sub-algorithms are presented based on the nonlinear gradient optimization. In terms of the associated items of the parameters to be estimated between two derived sub-algorithms, a separable recursive gradient parameter estimation method is proposed by designing an interactive and recursive estimation. The performance tests and the comparison experiments are carried out by the simulation examples.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
http://link.springer.com/article/10.1007/s12555-019-0940-5
ISSN:1598-6446
2005-4092
DOI:10.1007/s12555-019-0940-5