Index of content:
Volume 135, Issue 5, May 2014
- STRUCTURAL ACOUSTICS AND VIBRATION 
Dynamic and acoustic response of a clamped rectangular plate in thermal environments: Experiment and numerical simulation135(2014); http://dx.doi.org/10.1121/1.4870483View Description Hide Description
Experiments were performed to investigate the vibration and acoustic response characteristics of a clamped rectangular aluminum plate in thermal environments. Modal tests were carried out to study the influence of thermal environment on natural vibration. With the increment of structural temperature, natural frequencies of the plate decrease obviously. Mode shape interchange was observed for the modes with frequencies very close to each other. The thermally induced softening effect has unequal influences on the plate along the two in-plane directions. Numerical methods were also employed to study the experimental phenomena. Calculated results indicated that the initial deflection has a great influence on the natural vibration of the heated plate. Even a slight curvature can reduce the thermally induced softening effect obviously. Dynamic response tests were carried out under acoustic and mechanical excitations, and the measured results indicate that the variation in damping determines the response amplitudes at resonant peaks in the test.
135(2014); http://dx.doi.org/10.1121/1.4869090View Description Hide Description
Wave propagation in sandwich panels with a poroelastic core, which is modeled by Biot's theory, is investigated using the waveguide finite element method. A waveguide poroelastic element is developed based on a displacement-pressure weak form. The dispersion curves of the sandwich panel are first identified as propagating or evanescent waves by varying the damping in the panel, and wave characteristics are analyzed by examining their motions. The energy distributions are calculated to identify the dominant motions. Simplified analytical models are also devised to show the main physics of the corresponding waves. This wave propagation analysis provides insight into the vibro-acoustic behavior of sandwich panels lined with elastic porous materials.
Assessment of a hybrid finite element-transfer matrix model for flat structures with homogeneous acoustic treatments135(2014); http://dx.doi.org/10.1121/1.4871355View Description Hide Description
Modeling complex vibroacoustic systems including poroelastic materials using finite element based methods can be unfeasible for practical applications. For this reason, analytical approaches such as the transfer matrix method are often preferred to obtain a quick estimation of the vibroacoustic parameters. However, the strong assumptions inherent within the transfer matrix method lead to a lack of accuracy in the description of the geometry of the system. As a result, the transfer matrix method is inherently limited to the high frequency range. Nowadays, hybrid substructuring procedures have become quite popular. Indeed, different modeling techniques are typically sought to describe complex vibroacoustic systems over the widest possible frequency range. As a result, the flexibility and accuracy of the finite element method and the efficiency of the transfer matrix method could be coupled in a hybrid technique to obtain a reduction of the computational burden. In this work, a hybrid methodology is proposed. The performances of the method in predicting the vibroacoutic indicators of flat structures with attached homogeneous acoustic treatments are assessed. The results prove that, under certain conditions, the hybrid model allows for a reduction of the computational effort while preserving enough accuracy with respect to the full finite element solution.
135(2014); http://dx.doi.org/10.1121/1.4869086View Description Hide Description
A numerical technique for modal decomposition of the acoustic responses of structures submerged in a heavy fluid medium using fluid-loaded structural modes is presented. A Krylov subspace model order reduction approach to reduce the computational effort required for a fully coupled finite element/boundary element model is described. By applying the Krylov subspace to only the structural part of the global system of equations for the fully coupled problem, only the frequency independent finite element matrices are reduced. A fluid-loaded cylindrical shell closed at each end by hemispherical end caps is examined. The cylinder is excited by a ring of axial or transverse forces acting at one end. The individual contributions of the cylinder circumferential modes to the sound power and directivity of the radiated sound pressure are observed. The technique presented here provides a tool for greater physical insight into exterior acoustic-structure interaction problems using fully coupled numerical models, with significantly reduced computational effort.
A 2.5-dimensional method for the prediction of structure-borne low-frequency noise from concrete rail transit bridges135(2014); http://dx.doi.org/10.1121/1.4871357View Description Hide Description
Predicting structure-borne noise from bridges subjected to moving trains using the three-dimensional (3D) boundary element method (BEM) is a time consuming process. This paper presents a two-and-a-half dimensional (2.5D) BEM-based procedure for simulating bridge-borne low-frequency noise with higher efficiency, yet no loss of accuracy. The two-dimensional (2D) BEM of a bridge with a constant cross section along the track direction is adopted to calculate the spatial modal acoustic transfer vectors (MATVs) of the bridge using the space-wave number transforms of its 3D modal shapes. The MATVs calculated using the 2.5D method are then validated by those computed using the 3D BEM. The bridge-borne noise is finally obtained through the MATVs and modal coordinate responses of the bridge, considering time-varying vehicle-track-bridge dynamic interaction. The presented procedure is applied to predict the sound pressure radiating from a U-shaped concrete bridge, and the computed results are compared with those obtained from field tests on Shanghai rail transit line 8. The numerical results match well with the measured results in both time and frequency domains at near-field points. Nevertheless, the computed results are smaller than the measured ones for far-field points, mainly due to the sound radiation from adjacent spans neglected in the current model.