基于MOMEDA和包络谱的齿轮微弱故障特征提取

Feature Extraction of Weak Fault for Gear based on MOMEDA and Envelope Spectrum

复合故障下的齿轮微弱故障易被强故障掩盖而出现漏诊现象,对齿轮复合故障下的微弱故障特征提取进行研究。首先采用多点优化最小熵解卷积调整(Multipoint Optimal Minimum Entropy Deconvolution Adjusted,MOMEDA)作为前置滤波器对原信号进行降噪,增强信号中的周期性冲击成分,然后进行Hilbert变换得到包络谱;通过分析其中明显的频率成分识别故障,实现微弱故障特征的提取。仿真信号和变速器故障诊断实例表明,该方法能有效实现齿轮微弱故障特征提取。

The gear weak fault signal is easy to be covered by the strong fault signal in a complex fault,and there will be missed diagnosis phenomenon. The the feature extraction of weak fault under gear complex failure is studied. Firstly,Multipoint Optimal Minimum Entropy Deconvolution Adjusted( MOMEDA) is used as the pre-filter to denoise the original signal and enhance the periodic impact components in the signal. Then,the Hilbert transform is used to obtain the envelope spectrum. By analyzing the obvious frequency components to identify the fault,the extraction of weak fault characteristics is achieved. Simulation example of fault signal and transmission shows that this method can effectively realize the feature extraction of gear weak fault.