Journal of Computational Acoustics

The transfer matrix method (TMM) is a famous analytic method in the vibroacoustic community. It is classically considered as a high frequency approach, because of the hypothesis of acoustic plane waves impinging on a flat infinite panel. Thus, it cannot take into account directly finite-size effects or lateral boundary conditions (BCs), and it needs specific algorithms to correct its results in the low frequency range. Within the transfer matrix framework, the use of finite elements makes it possible to generalize the range of applications of transfer approaches. Thus, the study of wave propagation in poroelastic media, in presence of lateral BCs can be carried out. The links between theses waves and the acoustic response of a sample are investigated. Finally, it shows that transfer approaches are not limited in the low frequency range, as usually stated. In fact, the validity of analytic transfer approaches depends more on the material and on the geometry than on the frequency range.

The requirement by modern navies to predict sonar performance in shallow water, whether for use in research, planning or operations, led to an initiative for the validation of reverberation models in the form of two Reverberation Modeling Workshops at the University of Texas at Austin in November 2006 and May 2008 [J. S. Perkins and E. I. Thorsos, Update on the reverberation modeling workshops, J. Acoust. Soc. Am. 126 (2009) 2208]. The problem considered here (Problem XI, from the 2006 workshop) requires the computation of reverberation versus time in a Pekeris waveguide with Lambert scattering from the seabed. Results from eigenray, normal mode and (hybrid) continuum methods are presented and compared for the time window 0.05s to 1000s after pulse transmission. Approximate analytical solutions are used to provide insight into the expected behavior of the reverberation and establish regimes of validity of numerical models. In situations where the regimes of validity of different methods coincide, the solutions of models applying these methods overlap. The overlapping solutions agree with each other within ±0.3dB. Their purpose is to provide a baseline against which future model improvements can be assessed and quantified.

An efficient algorithm for long-range monostatic reverberation is developed using geometrical ray-bundle as its propagation model. In the standard approach for reverberation modeling, scattered signals are numerically integrated along the rough surface. In this work, the integration along the rough surface is converted to an integration along the delay time of source signal. It can be expressed in an analytic form under the assumption of ignoring time-dispersion effect. Repetitive eigen-ray searches for calculating scattered signals are avoided by using a split-step marching algorithm for the range based on the geometrical ray-bundle. When the bottom is range-dependent, arrival times of scattered signals from the rough bottom at same range is different according to azimuthal angles and they cannot be directly summed in time domain. The reverberation signals along the azimuthal angles are interpolated to obtain new reverberation signals at certain times. They are summed to compute a time series reverberation signal from the rough sloping bottom. The algorithm is applied to problems of reverberation modeling workshop, and its results are compared with those from the standard reverberation algorithm.

This work presents a semi-analytical technique based on the Green’s function and uniform-piston driven model to determine the end-correction length $l$ in an axially long flow-reversal end chamber muffler having an end-inlet and an end-outlet. The semi-analytical procedure is based on the 3D analytical uniform piston-driven model for obtaining the impedance [Z] matrix parameters and numerically evaluating the frequency $f_p$ at which the imaginary part of the cross-impedance parameter $Z_{E2E1}$ crosses the frequency axis at the first instance. The frequency $f_p$ corresponds to the low-frequency peak in the transmission loss (TL) spectrum of the axially long flow-reversal end-chamber muffler obtained a priori to its computation by considering the influence of higher order evanescent transverse modes. The effective chamber length (and thence, the end-correction length) in the low-frequency range are determined by using the expression for resonance frequency of a classical quarter-wave resonator. This method is employed to determine the end-correction in axially long elliptical cylindrical end chambers and circular cylindrical end chambers (with or without a rigid concentric circular pass-tube). The TL graph predicted by the 1D axial plane wave model (incorporating the end-correction length) is shown to be in an excellent agreement with that obtained by the 3D analytical approach and an experimental result (from literature) up to the low-frequency limit, thereby validating the semi-analytical technique. Parametric studies are conducted using the proposed semi-analytical method to investigate and qualitatively explain the effect of angular location and offset distance of the end ports and the pass-tube diameter on the end-correction length, thereby yielding important insights into the influence of transverse evanescent modes on dominant axial plane wave modes of the axially long end-chamber. Development of an empirical end-correction expression in a flow-reversal circular end-chamber with offset inlet and outlet ports is a practically useful contribution of this work.

We propose a pre-stack reverse time migration (RTM) seismic imaging method using the pseudospectral time-domain (PSTD) algorithm. Traditional pseudospectral method uses the fast Fourier transform (FFT) algorithm to calculate the spatial derivatives, but is limited by the wraparound effect due to the periodicity assumed in the FFT. The PSTD algorithm combines the pseudospectral method with a perfectly matched layer (PML) for acoustic waves. PML is a highly effective absorbing boundary condition that can eliminate the wraparound effect. It enables a wide application of the pseudospectral method to complex models. RTM based on the PSTD algorithm has advantages in the computational efficiency compared to traditional methods such as the second-order and high order finite difference time-domain (FDTD) methods. In this work, we implement the PSTD algorithm for acoustic wave equation based RTM. By applying the PSTD-RTM method to various seismic models and comparing it with RTM based on the eighth-order FDTD method, we find that PSTD-RTM method has better performance and saves more than 50% memory. The method is suitable for parallel computation, and has been accelerated by general purpose graphics processing unit.

Reducing the noise level in the acoustic cavities is the important problem when treating inflight conditions of commercial planes or boats. Shape optimization of the acoustic cavity that will take into account the geometrical and material uncertainties, arising during the manufacturing process, is presented in this paper. The noise level is controlled by minimizing the energy density in the cavity, obtained through an energy method called Simplified Energy Method. Such formulation is based on our previous published work where transformation function mapping 3D cavity surface on a 2D domain was proposed. The optimization process directly relies on this function and thus avoids remeshing of the geometry. Robust optimization is performed using the nondominated sorting genetic algorithm (NSGA-II) together with the Kriging surrogate model. Influence of geometrical and material characteristics on the optimal solution is identified.