变尺度特征提取在数控机床状态识别中的应用

Application of the Character Extraction with Varying Scales on State Recognition of NC Machine Tool

正确识别数控机床从正常到故障之间的演化过程,对掌握机床运行状态、保证加工精度具有重要意义。提出采用变尺度小波包特征提取方法以提高状态识别的准确性,并以数控车床主轴轴承磨损研究为例,将此方法与传统方法进行了对比分析。仿真和实验研究表明:变尺度小波包特征提取方法能有针对性地提取蕴含更多状态信息的振动信号特征用于状态识别,在192组测试样本中,变尺度特征提取方法的识别准确率达到98.44%,较传统方法有明显提高。

It is significant to recognize the evolvement from normal to fault for acquainting running-states and ensuring machining accuracy. The character extraction method with varying scales was proposed to improve the veracity of state recognition, and was com- pared with the traditional method with the example of headstock bearing＇s abrasion. The simulation and experimental results indicate that the signal characters with more state information can be obtained pertinently by using the character extraction method with varying scales. The veracity of recognition is 98. 44% in the 192 test samples, which is higher than that of the traditional method.