Improved eigensystem realization algorithm and its application on ocean platforms

Based on the matrix mapping theory, an improved eigensystem realization algorithm (ERA), called C/ERA, is proposed in this paper. It rebuilds a Hankel matrix by replacing all elements of each anti-subdiagonal with the arithmetic average of the elements along the anti-subdiagonal, and introduces the...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Cluster computing 2019-03, Vol.22 (Suppl 2), p.3643-3650
Hauptverfasser: Xin, Junfeng, Lei, Wang, Shixin, Li, Yongbo, Zhang
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Based on the matrix mapping theory, an improved eigensystem realization algorithm (ERA), called C/ERA, is proposed in this paper. It rebuilds a Hankel matrix by replacing all elements of each anti-subdiagonal with the arithmetic average of the elements along the anti-subdiagonal, and introduces the concept of using the Frobenius norm (L2-norm) to control the iterations number after implementing the SVD algorithm by the ERA method. With the data associated with a 5-DOF mass-spring-dashpot system and jacket-type platform under impact loading, it is demonstrated that C/ERA has a better capacity of de-noising, and a higher accuracy for low-order modes, and can identify more high-order modes than the ERA method.
ISSN:1386-7857
1573-7543
DOI:10.1007/s10586-018-2213-0