基于ATF与EEMD变转速齿轮故障特征降噪方法

Denoising Method for Gear Fault Characteristic under Variable Rotation Speeds Based on ATF and EEMD

针对变转速下的齿轮故障特征的降噪问题,提出一种基于自适应时变滤波(Adaptive time-varying filtering,ATF)与集合经验模态分解(Ensemble empirical mode decomposition,EEMD)的齿轮故障特征降噪方法。该方法首先用线调频小波路径追踪(Chirplet path pursuit,CPP)算法从变转速下的齿轮故障振动信号中估计出齿轮啮合频率,并依据该啮合频率设计时变滤波器;再利用该时变滤波器对齿轮故障振动信号进行滤波,将滤波器阻带内的噪声予以去除;然后采用EEMD方法对滤波后的信号进一步降噪,减少滤波器通带内的噪声干扰;接着利用时变滤波器对降噪后的信号再次进行滤波,消除EEMD降噪时在阻带带来的噪声干扰;最后对降噪后的信号进行阶次分析,提取齿轮故障特征。对齿轮局部故障的算法仿真和应用实例分析表明,该方法不仅可以消除阻带的噪声干扰,而且对通带内的噪声也有较好的抑制作用,可有效凸显齿轮的故障特征。

Aiming at the noise reduction for gear fault characteristic under variable rotation speeds,a denoising method for gear fault characteristics based on adaptive time-varying filtering(ATF)and ensemble empirical mode decomposition(EEMD)is proposed.In this method,the gear meshing frequency is estimated from the gear fault vibration signal under variable speed by using chirplet path pursuit(CPP),and a time-varying filter is designed according to the meshing frequency.Then,the time-varying filtering is used to filter the gear fault vibration signal so as to remove the noise in the filter stopband,and the EEMD method is used to further reduce the noise interference in the filter passband.Afterwards the time-varying filter is used to filter the denoised signal again to eliminate the noise interference caused by EEMD denoising in the stopband.Finally,the order analysis of the denoised signal is carried out to extract the gear fault characteristic.The simulation and application example analysis of gear local faults show that the method can not only eliminate the noise interference of stopband,but also has a good noise suppression effect in the passband,which can effectively highlight the gear fault characteristics.