Non-linear optimization of the shape functions when applying the finite element method to vibration problems

The finite element method constitutes, undoubtedly, one of the most powerful tools of computational mechanics. Minimizing the discretization error, on the other hand, is a question of basic concern in order to insure maximum reliability without increase in computer memory and/or CPU time. The presen...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Journal of sound and vibration 1986-12, Vol.111 (2), p.219-228
Hauptverfasser: Laura, P.A.A., Utjes, J.C., Sarmiento, G. Sanchez
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The finite element method constitutes, undoubtedly, one of the most powerful tools of computational mechanics. Minimizing the discretization error, on the other hand, is a question of basic concern in order to insure maximum reliability without increase in computer memory and/or CPU time. The present paper introduces the concept of the “ k-optimization parameter” contained in the shape functions. Once the Rayleigh-Ritz formulation of the finite element algorithm is performed and the eigenvalues are determined in the case of vibrations or elastic stability problems one is able to minimize them by optimizing with respect to k. The proposed method is applied to a classical vibrations problem—the vibrating string—and it is concluded that the main advantage of the method is the fact that excellent accuracy can be achieved with a small number of elements. In other words, the requirement of computer memory is considerably lower for a pre-assigned relative error.
ISSN:0022-460X
1095-8568
DOI:10.1016/S0022-460X(86)80158-4