Noise Source Identification In Vehicle Though Acoustic Array

The requirements for the comfort of vehicle riding environment are becoming higher and higher.Vehicle noise has become one of the conditions restricting the comfort of vehicle riding.It is imperative to optimize the driving environment of the vehicle.At the same time,vehicle noise contains rich information about vehicle equipment status.It is used for the condition monitoring and fault diagnosis of mechanical equipment,and also the basis for noise suppression.In fault diagnosis and noise reduction,it is very important to accurately find the root cause of noise,so first of all,it is necessary to accurately identify and locate the sound source.In view of this,this paper analyzes and compares various recognition methods,and combines the actuality of the project with the beamforming method for noise source identification.The summary of this paper is as follows:First of all,the research background and source of this paper are clarified.Based on the analysis of the development history of the noise source identification method,the significance and core content of this paper are proposed: numerical simulation and experimental research in the research of noise array identification technology based on acoustic array vehicles,with a view to applying it to subway trains passing noise sources Identification.Array measurement technology leads to a detailed introduction to the principle of beamforming.The conventional beamforming method based on the spherical wave hypothesis is applied to the noise source identification.At the same time,the conventional beamforming,the minimum variance distortionless response beamforming methods,and generalized inverse beamforming algorithms are specifically introduced.The advantages and disadvantages of various algorithms are compared and analyzed.Simulation analysis of the positioning performance,the number of array elements,and the measurement distance parameters of the three algorithms shows that the generalized inverse beamforming algorithm has better positioning performance and resolution.In view of the current research on modeling of directional sound source,there are few.This paper combines a more sparse complex equivalent source method and gradient descent to propose a generalized inverse beamforming method that can efficiently and accurately identify directional sound sources.Simulation and experimental analysis of coherent and incoherent sound sources show that better sound source recognition can be achieved with fewer microphones.Taking the identification of moving sound sources as the research goal,the generalized inverse beamforming method is extended to the recognition of moving sound sources.Firstly,the basic principles of acoustic radiation from moving sound sources are analyzed,and the phenomenon of Doppler effect due to motion is introduced.The method of removing the Doppler effect by cubic spline interpolation is also studied.Secondly,according to the experimental test conditions,a microphone array capable of meeting the test requirements is designed.Then the numerical analysis of dual-source and multifrequency component source identification simulation proves the effectiveness of the method and provides a basis for further practical applications.Finally,combined with the designed seven-arm ring array,the ground speed passing subway train is taken as the research object,and the noise source identification experiment is studied.