Abstract: As diesel engine vibration signal characteristic frequency band and noise frequency band exist overlap phenomenon, so it was diffcult to choose suitable wavelet threshold for wavelet threshold denoising in vibration signal processing. In order to extract fault features from engine vibration signal, a new method using LMS Adaptive Filtering Denoising based on wavelet transform for diesel engine vibration signals was proposed. Reconstruct signal that removed wavelet detail parts including noise of the multi-scale decomposition of wavelet transforms was used as input signal for LMS adaptive filter firstly. the first layer wavelet detail parts of the multi-scale decomposition of wavelet transform was used as reference input signal secondly, and Then the optimal filter for signal-noise decomposition was realized. Using the adaptive filter based on the wavelet transform to remove noise from diesel engine vibration signals, the filter has excellent filtering capability. Application example demonstrates that this method is excellent to realize the optimal estimate for the valuable signal and noise of the diesel engine vibration signal, it don’t need to choose suitable wavelet threshold for wavelet threshold denoising in vibration signal processing, its set parameter is few and easy to control. using this new method, the engine vibration fault discrimination is increased.

Key words: vibration and wave, wavelet transform, adaptive filtering, diesel engine, vibration signal, de-noising p, fault diagnosis

摘要： 由于柴油机振动信号的特征频带和噪声频带存在重叠现象，利用小波阈值消噪时难以选取合适的小波阈值，针对该问题提出一种基于小波包的LMS自适应滤波降噪方法。该方法将小波包与LMS自适应滤波相结合，首先利用小波包变换对信号进行多层分解，然后以噪声干扰对应尺度上的第一层“细节”分量及最大分解尺度上的逼近分量重构信号，将重构后的信号作为LMS自适应滤波器原始输入信号，再以小波包最大分解尺度上的高频细节信号作为自适应抵消器的参考输入信号，进行LMS自适应滤波降噪处理。仿真计算和工程应用表明，该方法参数设置较少，易于控制，不涉及小波阈值降噪中阈值的选取问题，对比试验信号的分析验证了方法的有效性，将该法应用在柴油机振动诊断中提高了故障识别率。

关键词: 振动与波, 小波包变换, 自适应滤波, 柴油机, 振动信号, 降噪, 故障诊断