17
68
Spécial « Congrès Acoustics 2012 »
oise source identification (NSI) techniques are used to
optimise the noise emission from a wide range of products
including vehicles, household goods and wind turbines. The
goal of NSI is to identify the most important sub-sources
on an object in terms of position, frequency content and
sound power radiation. Ranking of the sub-sources can then
be used to identify where design changes will most effecti-
vely improve the overall noise radiation. With a considera-
ble number of techniques available, NVH engineers might
have need of an overview to help select the most appro-
priate solution. This paper gives some guidelines to help
the selection together with some practical applications of
NSI techniques. Emphasis is placed on the useful frequency
range and the resolution of the various methods.
NSI techniques
A number of NSI techniques use just one or a few transdu-
cers. Examples are sound mapping based on sound pres-
sure, sound intensity and selective intensity. Array based
techniques using a dozen to up to hundreds of micropho-
nes are where the greatest developments are taking place.
The main methods are :
- Near-field Acoustic Holography related techniques (STSF,
Non-Stationary STSF, SONAH)
- Beamforming (Phased Array technique)
- Refined beamforming (deconvolution methods)
- Moving source beamforming
- Spherical beamforming
The technique selected by the user will depend on the
required frequency range, the measurement distance, the
required resolution and the area covered by the measu-
rement.
Sound mapping
The most straightforward method to detect a noise source
is to map the Sound Pressure Level (SPL) at various
locations around the noisy product. The SPL is measu-
red with a single channel FFT or 1/3 octave analyser
at each point of a defined grid. The measured data is
then used to produce a contour map that gives a rough
idea of the location and characteristics of the source.
Noise source identi cation techniques :
simple to advanced applications
Applications des techniques d’identification de sources
de bruit : des plus simples aux plus complexes
Kevin-Bernard Ginn
and Karim Haddad
Brüel&Kjær Sound&Vibration A/S,
Skodsborgvej 307
2850 Nærum
Denmark
E-mail :
Abstract
The number of noise source identification (NSI) techniques available to engineers
working on noise, vibration and harshness problems has increased considerably
in recent years. The choice of the most appropriate technique depends upon the
application and the information required. This paper reviews techniques for noise
source identification and quantification ranging from simple hand-held sound intensity
systems, hand-held array systems to large ground based microphone arrays. The
methods include Beamforming, Spherical Beamforming and Acoustic Holography.
Guidelines are given to help the engineer choose a suitable technique based on the
frequency range of interest, the distance from the measurement array to the test
object and the resolution required. Practical application examples ranging from
hearing aids to wind turbines are presented to illustrate the various NSI techniques.
Résumé
Le nombre de techniques d’identification de source de bruit (NSI) disponible pour
les ingénieurs travaillant sur des problèmes de bruit, de vibrations ou de rudesse a
considérablement augmenté ces dernières années. Le choix de la technique la plus
appropriée dépend de l’application et des renseignements requis. Cet article examine
les techniques d’identification et de quantification de source de bruit allant des simples
systèmes portatifs, à des systèmes complexes de réseaux de microphones. Les
méthodes comprennent la formation de faisceaux, la formation de faisceaux sphériques
et l’holographie acoustique. Des directives sont données pour aider l’ingénieur à
choisir une technique appropriée basée sur la gamme pertinente de fréquences, sur la
distance entre la zone de mesure et l’objet testé et la résolution requise. Des exemples
d’applications pratiques allant des prothèses auditives aux éoliennes sont présentés
pour illustrer les différentes techniques de NSI.
N
