基于 EMD-WVD 的车辆悬架减振器异响声品质客观评价研究

摘要

The abnormal noise from vehicle suspension shock absorber compromises the vehicle interior sound quality seriously.And the traditional psychoacoustics index is difficult to extract the sound information characteristics accurately due to that the abnormal noise signal is non-stationary and within a special bandwidth.As a kind of effective analysis method for non-stationary signal,the wavelet analysis can greatly overcome the defects of psychoacoustics mentioned above,which has been well proven in practice.However,since the classical wavelet analysis is dependent on the superposition of a certain basis function,it leads the wavelet functions to be difficult to approximate local signal signatures accurately at different scales.Therefore,on the basis of wavelet analysis,a newly developed abnormal noise objective sound quality index—SQCIMF-WVD (sound quality based on choosing IMF and proceeded WVD ) was introduced,which not only combines the advantages of empirical mode decomposition (EMD ) and Wigner-Ville distribution(WVD),but also brings in the concept of niose reduction by correlated analysis.The result shows that the correlation between the subjective evaluation and the SQCIMF-WVD is well improved,which can reflect the features of the sound quality of abnormal noise from suspension shock absorber more accurately.%车辆悬架减振器引致的车内异响问题严重削弱了车内声品质，该异响声信号为非平稳信号且带宽特殊，致使传统的心理声学客观评价指标难以准确提取其异响特征信息。而小波分析作为一种有效的非平稳信号分析方法，能够有效克服心理声学客观评价指标的上述缺陷，经实践得以验证。但是由于经典的小波分析根植于特定基函数的叠加，导致单个基函数导出的小波函数族难以在不同尺度上准确地逼近局部信号特征。为此，在小波分析的基础上，结合经验模态分解（EMD）和 Wigner-Ville 分布的优点，并引入相关分析去噪的概念，提出并建立了新的减振器异响声品质评价参数 SQCIMF-WVD（Sound Quality base on Choosing IMF and then proceed WVD），其与主观评价的相关系数进一步提高，更能准确反映减振器异响声品质的特点。