A Generalized Recursive Identification Algorithm Compensated by Orthogonal Weighted Kernel for Tracking Time-Variant Systems

The main purpose of this paper is to introduce a generalized recursive identification algorithm compensated by orthogonal weighted kernel. The performance index of the novel proposed algorithm considers information about both the estimation error and first time derivative of estimation error. The es...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Circuits, systems, and signal processing Jg. 39; H. 10; S. 4903 - 4929
Hauptverfasser: Tahbaz-zadeh Moghaddam, Iman, Ayati, Moosa, Taghavipour, Amir
Format: Journal Article
Sprache:Englisch
Veröffentlicht: New York Springer US 01.10.2020
Springer Nature B.V
Schlagworte:
Algorithms
Basis functions
Circuits and Systems
Computer simulation
Continuity (mathematics)
Electrical Engineering
Electronics and Microelectronics
Engineering
Fourier series
Instrumentation
Kernels
Parameters
Performance indices
Signal,Image and Speech Processing
Sliding
Tracking
Trigonometric functions
Lithium-ion battery
System identification
Time-varying parameter
Generalized recursive instrumental variable (GRIV)
Short-time sliding window
ISSN:0278-081X, 1531-5878
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The main purpose of this paper is to introduce a generalized recursive identification algorithm compensated by orthogonal weighted kernel. The performance index of the novel proposed algorithm considers information about both the estimation error and first time derivative of estimation error. The estimation process is implemented in a recursive sliding window scheme. In addition, linear combinations of sine and cosine basis functions, which lead to Fourier orthogonal series, are employed as a kernel to approximate the time-varying parameters as continuous functions in each sliding window. Properties of the proposed algorithm are extended to both least squares estimator and instrumental variable estimator; thus, it is applicable to systems with correlated and uncorrelated noise. Simulation results demonstrate the efficiency of the proposed algorithm for accurate and online tracking of unknown parameters’ trend.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0278-081X
1531-5878
DOI:10.1007/s00034-020-01394-x