论文部分内容阅读
为了直接利用连续小波变换进行内燃机噪声源识别,提出了一种连续小波变换的改进算法.研究了信号组成成分的幅值和频率变化对小波系数的影响,根据连续小波变换的基本性质,对连续小波变换后产生的与信号频率有关的衰减系数进行修正,通过对比分析从Matlab6.5的小波工具箱中找出了适合振动噪声信号分析的小波函数.工程信号的分析结果表明,经过对连续小波变换后小波系数的修正,使小波系数的大小能正确反映信号各组成成分的幅值大小,从而可以利用小波系数形成的云图直接进行内燃机噪声源识别.
In order to directly identify the noise sources of continuous combustion engine using continuous wavelet transform, an improved algorithm of continuous wavelet transform is proposed. The influence of amplitude and frequency of signal components on wavelet coefficients is studied. According to the basic properties of continuous wavelet transform, After wavelet transform, the attenuation coefficient related to the signal frequency is corrected, and the wavelet function suitable for the analysis of vibration and noise signals is found out from the Matlab6.5 wavelet toolbox through the comparative analysis.The analysis results of the engineering signal show that after the continuous wavelet After the wavelet coefficients are transformed, the size of the wavelet coefficients can correctly reflect the amplitude of each component of the signal, so that the cloud image formed by the wavelet coefficients can be directly used to identify the noise source of the internal combustion engine.