基于自适应Autogram+OSF滤波的电机轴承故障诊断

Fault Diagnosis of Motor Bearings based on Adaptive Autogram+OSF Filtering

自相关谱峭度图Autogram方法可以对初始信号非周期脉冲与噪声进行限制的方式测定周期性脉冲,获得更精确的最优频带。设计了一种自适应Autogram方法,通过顺序统计滤波(OSF)方法完成电机轴承故障信号傅里叶谱的包络处理。研究结果表明:采用自适应Autogram方法进行处理时具备良好可行性。采用本方法测试佳解调频带的效果比滤波器快速谱峭度方法的性能更优。相对自适应Autogram方法并未形成明显的二倍与三倍频特征频率,产生了众多干扰分量。

Autocorrelation spectral kurtogram(Autogram)method can determine periodic pulses by limiting the aperiodic pulses and noise of the initial signal,so as to obtain a more accurate optimal frequency band.An adaptive Autogram method is designed to complete the envelope processing of the Fourier spectrum of motor bearing fault signals by order statistical filtering(OSF)method.The research results show that the adaptive Autogram method has good feasibility during processing.The effect of adopting this method to detect the optima frequency modulated band is better than that of adopting filter fast spectral kurtosis method.The relative adaptive Autogram method does not form obvious double-and triple-frequency characteristic frequencies,which generates many interference components.