Vehicle Noise Source Identification Based On Near-field Acoustical Holography

In near-field acoustical holography(NAH) sound field measurement, noise signal reduces the sound field reconstruction precision which affects the realization of the sound field reconstruction. A vehicle noise source identification based on NAH is put forward in this paper, aimed to identify and locate the vehicle noise source more accurately. The main content of this article include the following several parts:(1) The purpose and significance of this thesis is shown, then the Wave equation, HelmholtzKirchhoff Integral and Rayleigh integral equation are introduced to deduct the formula of continuous NAH and the discretization of NAH. Finally, simulation analysis of NAH is carried out, and the influence of acoustic field reconstruction results discuss from different reconstruction parameters such as holographic aperture, reconstruction distance and sampling interval.(2) A NAH technique based on transfer function estimation is presented in this paper. Firstly, a rational transfer function estimation(TFE) is introduced, combining with the reference microphone signal, to solve the complex sound pressure of holographic plane, then the sound field is reconstructed according to acoustic field reconstruction formula. And the acoustic experiment is carried out to make the comparison between TFE-NAH and traditional NAH to verify the feasibility and effectiveness of TFE method.(3) A double plane measurement NAH technology based on singular value energy entropy is put forward in this article. Firstly, the basic theory of entropy and singular value decomposition and properties are studied, and the acoustic field information is measured in form of double plane measuring, and the theoretical basis and the simulation scheme process are presented. Finally, in the presence of measurement noise, the simulation of double plane measurement NAH based on singular value energy entropy method is carried out and get verified.(4) The double plane measurement NAH test system based on transfer function estimation and singular value energy entropy is designed in this paper. Taking a tractor front end as the research object to complete the double plane measurement NAH experiment, the experimental research and analysis is processed using the sound source identification algorithm described earlier to identify the noise source. Comparative analysis to the experimental results between the method presented this article and traditional NAH to further validated the superiority of the method and the reliability.