形态小波在轧机扭矩信号中的应用

Application of morphological wavelet in torque signal of rolling mill

轧机振动是一个复杂的耦合问题。当轧机牌坊发生垂直颤振时,采集轧机传动侧扭矩信号,在其频域图中并不能得到故障频率。利用本文中的形态小波降噪特征提取方法进行分析,降低了高频、低频以及噪声的影响,提取出了故障频率。有力地证明了轧机确实存在耦合振动现象。

Rolling mill vibration is known as a complex coupling problem. When Rolling mill frame vibrates vertically, the fault frequency cannot be distinguished in the frequency-domain spectrogram of the torque signal collected on drive side of rolling mill. The method of the feature extraction based on morphological wavelet is a-dopted to reduce the impact of the high-frequency, low frequency and noise, and extract the information of fault frequency. It is strongly proved that the coupled vibration phenomenon exists in the roiling mill indeed.