On The Approach Of Identifiing Sound Source By Utilizing Sound Pressure Effective Value To Conduct Inner Product Operation

Sound source recognition attaches great importance to the applications of noise control engineering and acoustic fault diagnosis.The current methods of sound source identification based on microphone array measurement mainly include beam-forming and near-field acoustic holography,both of which require the phase of the acoustic signals as the information basis to identify the sound source.In the project,there is also the case that the sound source radiates white noise,and the sound signals at each measurement point are not coherent,so the sound source cannot be identified through the phase information which is regarded as the core signal.The paper proposes the approach of using the effective value information at each measurement point to conduct the inner product operation.The basic principle of this is using the effective value of the sound pressure signal measured by the microphone array to conduct inner product operation,searching for the maximum value of the inner product mode,and then determine the sound source position according to Cauchy Schwartz’s inequality.By establishing the corresponding identification steps,the simulated experiment respectively identified the sound sources of radiated harmonics and white noise,and also accurately identified the spatial position and intensity of the sound source and other information.Thus,it is proved that the approach of identifing sound source by utilizing sound pressure effective value to conduct inner product operation is feasible in theory.The paper explores the relevant characteristics of utilizing the effective value to conduct the inner product operation to identify the sound source,analyzing the influences on the identifying efficiency which are caused by the microphone array,the number of microphones,the array aperture,the measurement distance,the frequency of the sound source and the position error of the microphone.The research results are: First,the inner product mode value distribution of the circular array has a better dynamic range.Second,more microphones and larger array aperture can make this method have stronger ability to resist noise interference.Third,the recognition sensitivity is reduced when the measurement distance increases.Fourth,the beam forming method is prone to "ghosting" phenomenon when the wave length of the sound source radiated is smaller than the microphone interval.At the same time,it is worthy of noticing that the recognition effect of the approach put forward in this paper is not limited by the frequency of the sound source,and it has a stable recognition effect for high and low frequency sound sources.What’s more,the use of sound pressure amplitude information is less susceptible to interference than the use of phase information and is more accurate for the recognition under the situation of deviations in the microphone position.On the other side,based on the accurate recognition of single-point sound sources,the identification of multi-point ones has been conducted.In order to reduce the inaccurate recognition caused by the interference among sound sources when their distance is short,the author proposes to identify multi-point sound sources by combining the method explored in this paper with Schmidt’s orthogonality.To testify the approach,by establishing the recognition steps and performing simulation calculations,the results show that the method accurately recognizes the location and intensity of multi-point sound sources.Sound source identification experiments were carried out in a semi-anechoic room.By controlling the non-directional spherical sound source to radiate the simple harmonics of different frequencies,and using the microphone array and digital acquisition system to collect the sound pressure signal.The time-delay sum beamforming method and the proposed method are used to identify the spherical sound source respectively,and the distribution differences of the two methods in acoustic imaging are compared.Control the spherical sound source to radiate white noise with no certain phase information,while changing the depth of the sound source,using the method in this paper to accurately identify the white noise source under different sound source depths.