Probabilistic analysis and experimental results of sound transmission loss of composite panels. Issue 7 (16th May 2017)

• Record Type: Journal Article
• Authors: Marulo, F.; Polito, T.
• Is Part Of: Polymer engineering & science. Volume 57:Issue 7(2017)
• Page Start: 722
• Page End: 730
• Publication Date: 2017-05-16
• Subjects: Polymer engineering -- Periodicals; Polymers -- Periodicals; 668.9
• DOI: 10.1002/pen.24619
• Languages: English
• ISSNs: 0032-3888

Abstract

Abstract : In last years a growing interest has been given to composite materials to be used in an aircraft primary structure; in principle they could allow an increased noise transmission inside the fuselage, because of the composites reduced mass and damping and the increased stiffness characteristics. On the other hand, the reduction of interior noise levels in aircraft cabins is nowadays a critical aspect of maintaining the competitiveness and ensuring the commercial success of a modern aircraft. Thus, it is fundamental for the aerospace industry to be able to correctly simulate composite materials vibroacoustic behavior in order to allow a low weight–low cost design including the acoustic effects. Furthermore, composite properties suffer from high scattering, mainly related to the composites manufacturing and assembling process. Traditional deterministic approaches, such as finite element method, are not able to take into account this variability and their application can generate results which validity cannot easily be estimated. This can lead to the need to oversize structures and damping‐soundproofing treatments in order to be able to satisfy required performances, penalizing in this way weight saving issues that drive the use of composite materials. Small changes of input quantities can have strong effects on response prediction, and then reliable vibroacoustic predictive techniques have to give an assessment of the effects of parameters variability on structural and/or acoustical response. In a probabilistic finite element approach, design parameters are introduced as stochastic variables to model randomness and spatial variability of the geometrical and mechanical properties allowing optimization and sensitivity analyses to be performed once a probabilistic model has been established. In the present work, a vibroacoustic analysis has been performed interfacing a probabilistic model with the finite element one of a panel; acoustic performance in terms of achieved transmission loss has been evaluated. POLYM. ENG. SCI., 57:722–730, 2017. © 2017 Society of Plastics Engineers