On the use of viscoelastic materials characterized by Bayesian inference in vibration control

Abstract

Viscoelastic materials are used to reduce vibrations in mechanical systems due to their control efficacy. Considering that the dynamic behavior of those materials may be described by means of complex moduli, and experimental data may present ucertainties, an alternative is to use probabilistic methods, especially the Bayesian inference approach. By that approach, probability distribution functions are obtained for parameters of a model which describes the behavior of a given material. The present work employs a viscoelastic material modeled by the Bayesian approach in two vibration control actions, namely: a) use of vibration isolators; b) use of dynamic neutralizers. Transmissibility and receptance curves are displayed as well as dimensions of the control devices. Performance predictions are carried out in both cases. It is shown that the Bayesian approach can favourably reflect the presence of the uncertainties and advance their effects. Thus, more information can be provided for the designer of viscoelastic vibration control devices to anticipate eventual corrective measures.