基于自适应神经模糊推理系统的刀具磨损监测

Research on Method of Tool Wear Monitoring Based on ANFIS

为精确地监测高速铣床刀具在加工过程中的刀具磨损量,通过采集高速铣床加工过程中的振动信号、电流信号和噪声信号,经数据预处理与数据融合,建立基于自适应神经模糊推理系统的刀具磨损过程变化模型,实现在高速铣床不停机的前提下对铣床刀具进行状态监控。实验结果显示:针对铣床刀具磨损量的监测平均准确率为95.21%,最大监测准确率为99.74%。这表明文中所采用的方法具有较高的可行性。

In order to monitor the tool wear of a milling machine accurately,the signals of table and spindle vibration,the signals of spindle AC and DC current,and the signals of table and spindle acoustic emission were collected to establish a model of the tool wear process variations basing on ANFIS. At last,the established model will be used to monitor the milling machine tool wear without the machine stop running. According to the output of the model,the average monitoring accuracy is 95. 21%,and the highest monitoring accuracy is up to 99. 74%,which means the method of the paper proposed has a high feasibility.