Analytical and numerical local sensitivity analysis of periodic spring-mass chains - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2017

Analytical and numerical local sensitivity analysis of periodic spring-mass chains

Résumé

Periodic structures can be used as mechanical filters in vibration control by creating destructive wave interferences to block the passage of propagating waves in certain frequency bands. Recently, researchers have been concentrating their effort to improve this phenomenon while creating more complex and adaptive structures. Nonetheless, there is still a lack of information about their behavior and parameters sensitivity to comprehend, for example, the effects of uncertainties. This study shows the use of analytical equations of spring-mass chains and their derivatives to compare the sensitivity of Bragg’s and resonance bandgaps. The objective is to inspect the behavior and the influence of changing stiffness and inertia properties on the attenuation zones borders. The transfer matrix method is used and a general formulation is adopted to model the unit cells. The propagation constants obtained by solving eigenvalue problems and the frequency responses are used to analyze infinite and finite chains, respectively. Analytical partial derivatives and finite differences are used to calculate the local sensitivity. The bandgap borders are found to be highly sensitive to the distance from localized modes and anti-resonances. The results from the comparison between these two kinds of attenuation zones are presented.
Fichier non déposé

Dates et versions

hal-02131284 , version 1 (16-05-2019)

Identifiants

  • HAL Id : hal-02131284 , version 1

Citer

Leandro Cunha, Morvan Ouisse, Domingos Alves Rade. Analytical and numerical local sensitivity analysis of periodic spring-mass chains. Euro-Mediterranean Conference on Structural Dynamics and Vibroacoustics, Apr 2017, Sevilla, Spain. ⟨hal-02131284⟩
73 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More