随机共振降噪下的齿轮微弱故障特征提取

针对强背景噪声下的齿轮微弱故障特征提取问题，提出了一种将级联单稳随机共振与经验模式分解(EMD)-Teager能量算子解调方法相结合的特征提取方法。首先对含噪故障信号进行随机共振输出，降噪后再进行经验模式分解，分解得到具有不同特征时间尺度的固有模态函数(IMFs)，最后通过Teager能量算子解调方法求取每个有效IMF分量的幅频信息，从而提取齿轮微弱故障特征。仿真分析和实际测试结果均表明，通过随机共振降噪后，该方法能有效检测出齿轮局部损伤故障特征频率。

Weak Feature Extraction of Gear Faults Based on Stochastic Resonance Denoising

Aimed at the feature extraction problem of weak gear faults under strong background noise, an early feature extraction method was proposed based on cascaed monostable stochastic resonance(CMSR) system and EMD with Teager energy operator demodulating. Firstly CMSRS was employed as the preprocessing to remove noise, and then the denoised signals were decomposed into a series of intrinsic mode functions(IMFs) of different scales by EMD. Finally, Teager energy operator demodulating was applied to get amplitudes and frequencies of each effective IMF so as to extract the faint gear fault features. The simulation and application results show that the proposed method can detect the characteristic frequency of gear faults of local damage effectively after the noise reduction by CMSR.