基于ARMA-MOMEDA的电机轴承复合故障诊断研究

Research on composite fault diagnosis of motor bearings based on ARMA-MOMEDA

由于电机工作时背景噪声过大,当其轴承发生微弱故障时不能被准确检测,而电机在持续运转过程中极可能对其余轴承造成应力冲击从而造成更大的损失,为此提出了基于最优最小熵反褶积方法(MOMEDA)和自回归滑动平均模型(ARMA)的轴承复合故障诊断方法。首先通过ARMA对原始振动信号进行平滑处理,消除背景噪声对振动信号的干扰,并通过不同信噪比情况下的仿真信号来验证其降噪性能,然后通过MOMEDA对平滑后信号中的复合故障特征信息进行提取,最后通过实验验证了该方法的可行性。

In order to diagnose the composite faults of motor bearings,a composite fault diagnosis method based on the optimal minimum entropy deconvolution method(MOMEDA)and the autoregressive moving average model(ARMA)was proposed in order to diagnose the compound faults of the motor bearings.Firstly,the original vibration signal is smoothed by ARMA to eliminate the interference of background noise on the vibration signal,and the noise reduction performance is verified by the simulated signal under different signal-to-noise ratios,and then the composite fault feature information in the smoothed signal is extracted by MOMEDA,and finally the feasibility of the proposed method is verified by experiments.