Study Of Noise Source Identification Based On NAH-BSS

Near-field Acoustical Holography(NAH)is the rise of a noise source identification method in recent years,and it has important application in noise source identification,location and acoustic field visualization.But the NAH is very strict with sound field environment and the location of sound source.For example,sound source cannot exist in a holographic measurement on both sides at the same time,and the acoustic field must be free,the sound source must be completely coherent.However,in the actual engineering applications it is often difficult to meet the near-field acoustical holography's strict requirements on acoustic field.Non-free sound field or irrelevant sound field environment widely exists in the actual situation.Blind Source Separation(BSS)is a new signal processing method.It refers to separating the source signals from the mixture of source signals under the condition that theoretical model and hybrid approach are unknown.Blind source separation which has simple principle and super calculation speed is a kind of prospective signal processing method.The traditional way to overcome the limitation of the NAH's strict requirements on acoustic field generally is by increasing the measuring of the holographic planes or measuring physical quantities,which additional increases the cost of the measurement.This paper proposes a method which combines the blind source separation with near-field acoustical holography,and is applied to the noise source identification.The method doesn't add extra measurement and overcome the limitations of near-field acoustical holography's strict requirements on sound field.It can be applied to not only the situation that sound source is not complete in holographic surface with one side but also un-coherent acoustic field.From the experiment at the end of the paper,the blind source separation can effectively separate the mixed sound field on holography surface.