基于自适应时频滤波的变转速齿轮故障特征提取

Adaptive time-frequency filtering based fault characteristics extraction method for gears under variable rotational speed

针对变转速下齿轮的故障信号分离与故障特征提取,提出了基于线调频小波路径追踪(Chirplet Path Pursuit,CPP)与S变换的自适应时频滤波方法。该方法先采用CPP算法从原始齿轮振动信号中估计出齿轮啮合频率,同时,对原始振动信号进行S变换获取其时频分布;然后根据齿轮啮合频率设计自适应时频滤波器;再采用时频滤波器对信号的时频分析进行时频滤波,并将时频滤波结果进行S逆变换,即可得到包含齿轮故障信息的滤波信号;最后对滤波信号进行阶次分析,并根据阶次谱中的调制边频带诊断齿轮故障。对变转速下齿轮的局部故障进行了算法仿真和应用实例分析,结果表明,自适应时频滤波器可根据信号的频率变化特点自适应地改变中心频率和带宽,具有较好的信号分析自适应性,且滤取的信号无相位畸变,非常适合于变转速下的非平稳信号分析。

Aiming at the fault signal separation and fault characteristics extraction of gears under variable rotational speed,an adaptive time-frequency filtering method based on the chirplet path pursuit（ CPP） and S transform was proposed. In the method,the gear mesh frequency was estimated from an original gear vibration signal by using the CPP,meanwhile,the time-frequency distribution of the original gear vibration signal was obtained by using the S transform. An adaptive time-frequency filter was designed according to the gear mesh frequency,and the time-frequency filtering was carried out on the time-frequency distribution of the original signal,which was followed by the inverse S transform so as to get the filtered signal containing gear fault informations. Then,the gear fault diagnosis was carried out according to the modulation sideband in the order spectrum,which was obtained by using order tracking to the filtered signal. The local faults of the gear were analysed both by simulations and examples. The results show that the adaptive time-frequency filter can adaptively change its centre frequency and bandwidth according to the frequency variation characteristics of the gear.It has a good adaptability for signal analysis,moreover,the filtered signal is without phase distortion. Therefore,the adaptive time-frequency filtering method is very suitable for analyzing non-stationary gear signals under variable rotational speed.