High-speed Vehicle Sound Source Quantitative Identification

The noise of high-speed vehicles such as high speed cars or trains will cause severe noise pollution. It is a kind of complex noise source, which is the combination of multiple noise sources that move at high speed and contain broad bands of waves in a large open space. In order to comprehensively understand the mechanism of the noise generation and therefore control it in the end, quantitative measurement and sound visualization of these noise sources are very important prerequisite. Existing measurement methods and techniques are not able to achieve these goals because of their limitations with theories and properties.This thesis proposes the de-dopplerization method for high-speed sound sources and establishes acoustical holography method for high-speed sound sources. Taking high-speed vehicles as the research objective, dynamic sound source radiation model is established, non-simplified sound radiation model with lossless accuracy is applied in the high-speed sound source de-dopplerization. Sound source characteristic function and single point integration method are introduced to achieve inverse calculation of complex model. Based on single point source de-dopplerization method, we propose the independence judging condition of multiple coupling sources and solve the multiple sound source de-dopplerization with inverse decoupling method. Acoustical holography method which is applicable for high-speed sound source identification is established and the effects of this de-dopplerization method in correcting the amplitude and frequency shift of high-speed sound source signal are validated with experiments. Identification and signal correction lay the foundation for high-speed sound sources quantitative reconstruction.The aperture error mechanism of acoustical holography based on the Kirchhoff diffraction integral theory is proposed and the improved acoustical holography method with aperture error corrected is established. The classic Kirchhoff diffraction integral theory is extended from closed surface to finite area, Fresnel zone approximation method is applied to derivate the relationship between integral domain and the quantitative accuracy of integral result, which shows the relationship between acoustical aperture angle and acoustical reconstruction quantitative accuracy. Spatial window function is applied to restrain the signal leakage of spatial truncation and increase the sound source identification resolution of acoustical holography. Quantitative acoustical holography method, which is combined with equivalent sources and dynamic wave superposition, is established based on aperture error correction, this method can reconstruct holographic data, enlarge the holographic plane and increase the holographic aperture angle. Quantitative acoustical holography method optimizes the identification quantitative accuracy of sound sources and can be validated with experiments.Approximation sound source model method for analyzing the near-field evanescent wave component of actual sound source is proposed, which solves the problem of sound source model error in acoustical holography quantitative reconstruction validation and reveals the physical meaning of the reconstruction results of acoustical holography under the far-field condition. The practicability of simulating line source and surface source with simple source model is proved. The evanescent wave character of general simple source such as point source, dipole source and quadrapole source is analyzed and approximation sound source model method of actual sound source is proposed. With this method, evanescent wave component at arbitrary position in front of the sound source and be calculated. Experiments with actual sound sources prove that this method can solve the problem of correcting the sound source model error in acoustical holography effectively and achieves strict validation of high-speed sound source quantitative identification.Based on the high-speed vehicle sound sources quantitative identification methods, we develop and establish high-speed vehicle sound sources quantitative identification measurement system, using microphone array and binocular vision cameras to collect sound signal and images. With this system, measurement experiments of the operating Beijing-Tianjin high-speed train are carried out, the sound sources character analysis results and quantitative identification visualization results are obtained and the methods in this thesis are validated effective. These are the first quantitative identification results of operating high-speed vehicle with the acoustical holography method.With the innovative methodologies and technologies developed in this thesis, we expand the quantitatively measure and visualize the complex moving sound sources to the high-speed field. They have great potential in exploring unknown acoustic phenomenon, revealing the sound sources distribution of high-speed vehicles and contributing to noise pollution prevention and protection. They also provided technique foundations for the design and development of low-noise high-speed vehicles.