Loading…

Sound source identification technique using the inverse boundary element method

Numerical acoustic prediction requires the knowledge of the sound source, in most cases in terms of surface vibration data. Traditional approaches to obtain this information are by performing a structural finite-element analysis or by direct measurements. Both of these approaches are time consuming...

Full description

Saved in:
Bibliographic Details
Published in:The Journal of the Acoustical Society of America 2005-09, Vol.118 (3_Supplement), p.1918-1918
Main Authors: Shi, Jiaohua, Martinus, Ferdy
Format: Article
Language:English
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Numerical acoustic prediction requires the knowledge of the sound source, in most cases in terms of surface vibration data. Traditional approaches to obtain this information are by performing a structural finite-element analysis or by direct measurements. Both of these approaches are time consuming and may not be feasible, particularly when the source dimension is large, the excitation cannot be identified accurately or when due to safety and time concerns, direct measurements are not preferred. In this paper, an advanced source identification technique using the inverse boundary element method is presented. This approach is advantageous since only a small number of sound-pressure measurements are needed and only a boundary element model is required, therefore eliminating the necessity of building a finite-element model. As an example, the reconstruction of surface vibration on a diesel engine is presented. The reconstructed surface vibration was then used to predict the acoustic responses in the field in term of sound-pressure and radiated sound power. It will be shown that the predicted sound-pressure and radiated sound power agrees well with the measured values. The reconstructed surface vibration was also compared to several direct vibration measurements obtained using accelerometers.
ISSN:0001-4966
1520-8524
DOI:10.1121/1.4780442