Acoustical visualization of aeroacoustic sources

Moohyung Lee, Purdue University

Abstract

The purpose of this study was to develop methods for visualizing the sound field emitted from aeroacoustic sources in order to identify their source strength distributions, radiation patterns, and to quantify the performance of noise control solutions. In this study, scan-based, multireference near-field acoustical holography was used for that purpose. First, a strategy for designing a measurement array in the context of a multireference measurement was described, and it was shown that the effect of measurement noise can be suppressed by increasing the number of references used in measurements beyond the number of significant independent sources. Secondly, the procedure for partial field decomposition was described and a new partial field decomposition formulation that takes account of measurement noise and source nonstationarity was proposed. The regularization that is applied to stabilize holographic projection solutions was described, and the applicability of using patch holography to extend the pressure measured in the finite aperture by an extrapolation process in cylindrical coordinates was demonstrated next. Finally, the various procedures developed in this study were validated by application to two types of aeroacoustic sources. First, a ducted fan was considered, and by the comparison of the visualized sound fields corresponding to various configurations, the ability of NAH to quantify noise source modification effects was shown. Secondly, results obtained from NAH measurements of a subsonic free jet were presented. When performing a measurement of a jet, microphone arrays should be designed with care: i.e., references should be positioned outside the jet plume on a surface enclosing the whole jet plume to ensure all the localized subsources are sensed. The sound field generated by the turbulent jet used in this study was decomposed into various subsources, and it was found that the dipole- and quadrupole-like components were the strongest. The sound field and the farfield radiation pattern predicted by using NAH were compared with directly measured results, and good agreement was found.

Degree

Ph.D.

Advisors

Mongeau, Purdue University.

Subject Area

Mechanical engineering

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS