基于共振稀疏分解与子带增强的滚动轴承声学诊断方法

Acoustic diagnosis of a rolling bearing based on resonance sparse decomposition and sub-band enhancement

针对轴承声信号易受环境噪声干扰,导致声学诊断结果准确率低的问题,提出一种结合共振稀疏分解与小波降噪选取核心冲击子带、对信号进行二次降噪的滚动轴承诊断方法。首先采用共振稀疏分解算法对原始声信号进行降噪处理,提取信号瞬态冲击成分;然后通过小波包变换对信号进行分解,依据各子带信号峭度值选取核心冲击子带信号进行线性叠加并重构;最终通过包络谱分析确定轴承故障。故障模拟实验结果表明,本文方法可有效增强复杂声场环境下轴承声信号的冲击特性,实现针对滚动轴承的声学诊断。

Given that bearing acoustic signals are easily disturbed by environmental noise, which leads to low accuracy of acoustic diagnosis results, a method for rolling bearing diagnosis that combines resonance sparse decomposition and wavelet noise reduction to select the core impact subband, and then performs secondary noise reduction on the signal is proposed. The resonance sparse decomposition algorithm is first used to perform noise reduction processing on the original acoustic signal, the transient impact components of the signal are extracted, and the signal is then decomposed by wavelet packet transform. The core shock subband signals are selected based on the kurtosis value of each sub-band in order to linearly superimpose the signals. The bearing structure is finally determined by envelope spectrum analysis. The results of fault simulation experiments show that the method proposed in this paper can effectively enhance the impact characteristics of bearing acoustic signals in a complex sound field environment and achieve accurate acoustic diagnosis for rolling bearings.