摘要:

At the previous meeting of the Society, we reported that hearing‐impaired listeners required larger peak‐to‐valley differences in vowel spectra for correct identification than do normal‐hearing listeners. This outcome suggests that errors in vowel identification might be reduced by increasing the contrast between spectral peaks and valleys of natural vowels. To test this hypothesis, we created three sets of four vowels: one set with normal peak‐to‐valley amplitudes, one set with valley amplitudes decreased by 5 dB, and one set with valley amplitudes decreased by 10 dB. Identification accuracy improved minimally with 5‐dB enhancement but not at all with 10‐dB enhancement. The failure to markedly improve identification accuracy suggests that the deficiency in vowel identification does not lie in locating the spectral peaks, but rather may be due to aberrant coding of frequency in the internal auditory representation.

ISSN:0001-4966    

DOI:10.1121/1.2023067

出版商:Acoustical Society of America    

年代:1986

数据来源: AIP

摘要:

A procedure for applying the principles of statistical energy analysis (SEA) to practical noise analysis and design problems has been developed and implemented in the computer programSEAM. Ths user models a system by specifying the geometric and material properties of subsystems.SEAMcomputes the necessary SEA parameters, such as modal density, impedance, and coupling loss factor, and then solves for the system response to a particular excitation. Examples are given of the application ofSEAMto a variety of complex dynamic systems showing comparisons with measured results.

ISSN:0001-4966    

DOI:10.1121/1.2023074

出版商:Acoustical Society of America    

年代:1986

数据来源: AIP

摘要:

The response of a thin‐walled pipe section excited by an external force can be obtained using the energy accountancy method as described in the first part of this series of two papers. In order to verify this approach, an experimental study was conducted on a 2.88‐m‐long, thin‐walled steel pipe section containing two different fluids, air and water. The pipe section, 0.06 m in diameter and 0.003 m thick, was simply suspended by wires with no other outside connections. The pipe was excited externally by a calibrated hammer with the response of the pipe measured by an accelerometer attached to the outer surface of the pipe. In these experiments the internal fluid was stationary. This did not impose any limitations on the understanding obtained from the experiments since, for water systems, the flow is at a very low Mach number and therefore the effects of the flow are negligible. The results obtained show good agreement between the measured energy input and the measured dissipated and radiated energy components, the energy levels being within 1 dB of each other. This verified the application of the energy accountancy concept in describing the response of a thin‐walled pipe section. Furthermore, the results obtained can be simplified into a diagnostic tool which can be used for noise control purposes.

ISSN:0001-4966    

DOI:10.1121/1.2023076

出版商:Acoustical Society of America    

年代:1986

数据来源: AIP

摘要:

This presentation is a tutorial on the Gaussian beam method used for the asymptotic synthesis of seismic and acoustic wave fields in inhomogeneous media. The method is based on the superposition of beam solutions, each of which is an approximate solution of the wave equation along particular rays. Smoothness conditions on the medium are required for the approximate propagation of beam solutions. Within a smoothly varying medium, various choices of the beam parameters can be used. Specifying broad planar beams at the source would result in a plane‐wave decomposition of the visible spectrum. The standard ray method would result by specifying narrow planar beams at the receiver. Another choice is to specify the minimum integral beam width along each ray. There are several advantages of the Gaussian beam method, including finite amplitudes at caustics, smoothing of endpoint errors, and the reduction of amplitude variability resulting from model parameterization. Several numerical examples will be given in 2‐D and 3‐D, and comparisons will be shown with other numerical methods.

ISSN:0001-4966    

DOI:10.1121/1.2023080

出版商:Acoustical Society of America    

年代:1986

数据来源: AIP

摘要:

This paper presents examples of the evolution of ray theory of underwater acoustics at the Naval Ocean Systems Center and its predecessor laboratories since 1952. The motivations for this evolution were to bring ray‐theory results into better agreement with experiments, to distinguish true acoustic properties of the ocean from artifacts resulting from the sound‐speed model or from shortcomings in ray theory, and to develop simple controls for use in evaluating complicated propagation models. Examples of such controls are the five‐parameter Epstein profile, profiles for which various ray pencils focus at a point, and closed‐form solutions for profiles with range as well as depth dependence. The most significant topics presented are the development of continuous‐gradient sound‐speed profiles, the effect of Earth's curvature and of improvements in experimental sound speeds, the application of uniform asymptotics in the boundary layer about caustics, and the development of a method using complex parameters for the evaluation of shadow‐zone fields. Various examples of ray‐theory propagation losses are compared with experiment or with the results of normal‐mode theory.

ISSN:0001-4966    

DOI:10.1121/1.2023083

出版商:Acoustical Society of America    

年代:1986

数据来源: AIP

摘要:

An efficient range‐independent coherent shallow water propagation loss model based on eigenrays is described. The model uses logic which guarantees finding all ray paths. Travel time data are also accurately computed. Predictions at 64, 256, and 1024 Hz for a complex shallow water environment (140‐m water depth) are compared with measured data, and predictions using a detailed geoacoustic model (data and geoacoustic data reported by D. D. Ellis and D. M. F. Chapman, Defence Research Establishment Atlantic). Propagation loss predictions out to 85 km are in good agreement with measurements at 64 and 256 Hz using this relatively simple range‐independent ray tracing model with a multilayer bottom. Surprisingly, the predictions made at 1024 Hz, where ray theory may be expected to be more accurate, are not quite as good. Comparison of model predictions with other measured and theoretical data is also presented. The results generally support the validity of coherent ray‐based propagation modeling in shallow water areas over a wide frequency range.

ISSN:0001-4966    

DOI:10.1121/1.2023087

出版商:Acoustical Society of America    

年代:1986

数据来源: AIP

摘要:

Adaptive methods are being implemented increasingly to control the response of systems to meet performance objectives in the presence of unpredictable time‐varying constraints. Typical open‐ or closed‐loop control systems often fail under such conditions because their controllers (filters) have fixed coefficients. Adaptive controllers have time‐variable coefficients determined on‐line to minimize the difference at a system's output between desired and actual outputs. Adaptive methods are particularly appropriate for sound control. It is well known that active fixed‐coefficient methods for sound cancellation fail with time as they become detuned by time‐varying constraints such as changing medium temperature and motion. The purpose of this tutorial review is to provide some of the background of adaptive control with specific application to electro‐acoustic sound cancellation.

ISSN:0001-4966    

DOI:10.1121/1.2023090

出版商:Acoustical Society of America    

年代:1986

数据来源: AIP

摘要:

Boundary element methods have been used as numerical approximations of boundary integral equations to predict the sound fields in complicated three‐dimensional interior and exterior spaces. In this investigation the methods have been investigated for their capability to find and evaluate the optimal active noise controller for complex acoustic geometries. The boundary element methods are utilized to enforce the specified boundary conditions while solving for the optimal secondary source distribution necessary to minimize a giver performance equation. For the current work the performance equation is a weighted sum of mean‐square pressures at specified points. Thus the method can predict the optimal controller required to achieve local control using a single point, or to achieve a space averaged reduction by using a distribution of points. The formulation allows multiple secondary sources and either pressure, velocity, or normal specific acoustic impedance boundary conditions. The optimal secondary source's strengths and the insertion loss performance of several active controller configurations are illustrated.

ISSN:0001-4966    

DOI:10.1121/1.2023094

出版商:Acoustical Society of America    

年代:1986

数据来源: AIP

摘要:

The Obukhov quasipotential function for the acoustic field in a boundary layer of exponential profile is used to obtain a modal description of low‐frequency sound propagation. As the wind speed approaches zero, the governing equations approach the Helmholtz equation with an impedance boundary condition. The solutions for the acoustic field with a boundary layer can be given as a continuous plane‐wave spectrum with variable amplitudes given by generalized hypergeometric functions. An analysis with the hypergeometric functions gives one or more acoustic modes, depending on frequency. The acoustic modes propagate as cylindrical waves, with amplitude varying inversely with the square root of distance. An estimate of the wavenumber of the fundamental mode shows that its attenuation is proportional to the product of wind speed and boundary layer displacement thickness. The propagation theory is compared to data from a wind turbine at Medicine Bow, Wyoming. A microphone array was used to measure low‐frequency sound at ground level at distances from 200 to 20 00(3 m from the turbine. Atmospheric temperature and wind speed profiles were measured, as was ground impedance, so that the theory may be compared without ambiguity to the data.

ISSN:0001-4966    

DOI:10.1121/1.2023095

出版商:Acoustical Society of America    

年代:1986

数据来源: AIP

摘要:

Theory for predicting the diffraction of sound (creeping waves) into shadow regions, resulting from propagation in either a stratified medium or above a convex curved surface, is well known and has been used mostly in underwater sound. This theory has also been extended more recently to the case of a stratified atmosphere. However, this theory generally has not been used to predict the energy diffracted over curved ground, such as rolling hills on berms. More commonly, the diffracted energy in these cases has been predicted by adapting theory valid for thick barriers or wedges. The creeping wave theory may be more appropriate in cases such as the diffraction of sound over berms, or of sound of very low frequency over rolling ground. Preliminary experiments have been performed over convex curved (radius of curvature<100 m) ground of finite and infinite impedance in the frequency range between 250 and 8000 Hz using propagation distances up to 30 m. Controlled measurements are also planned indoors over a carefully constructed curve surface. The measurements are compared with various aspects of theory.

ISSN:0001-4966    

DOI:10.1121/1.2023104

出版商:Acoustical Society of America    

年代:1986

数据来源: AIP