Nonlinear Normal Modes of a Rotating Shaft Based on the Invariant Manifold Method - Archive ouverte HAL
Article Dans Une Revue International Journal of Rotating Machinery Année : 2002

Nonlinear Normal Modes of a Rotating Shaft Based on the Invariant Manifold Method

Résumé

The nonlinear normal mode methodology is generalized to the study of a rotating shaft supported by two short journal bearings. For rotating shafts, nonlinearities are generated by forces arising from the supporting hydraulic bearings. In this study, the rotating shaft is represented by a linear beam, while a simplified bearing model is employed so that the nonlinear supporting forces can be expressed analytically. The equations of motion of the coupled shaft-bearings system are constructed using the Craig-Bampton method of component mode synthesis, producing a model with as few as six degrees of freedom (d.o.f.). Using an invariant manifold approach, the individual nonlinear normal modes of the shaft-bearings system are then constructed, yielding a single d.o.f. reduced-order model for each nonlinear mode. This requires a generalized formulation for the manifolds, since the system features damping as well as gyroscopic and non-conservative circulatory terms. The nonlinear modes are calculated numerically using a nonlinear Galerkin method that is able to capture large amplitude motions. The shaft response from the nonlinear mode model is shown to match extremely well the simulations from the reference Craig-Bampton model.
Fichier principal
Vignette du fichier
419761bbb.pdf (474.79 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01315570 , version 1 (13-05-2016)

Licence

Identifiants

Citer

Mathias Legrand, Dongying Jiang, Christophe Pierre, Steven Shaw. Nonlinear Normal Modes of a Rotating Shaft Based on the Invariant Manifold Method. International Journal of Rotating Machinery, 2002, ⟨10.1155/S1023621X04000338⟩. ⟨hal-01315570⟩

Relations

486 Consultations
326 Téléchargements

Altmetric

Partager

More