基于小波包和 BP 神经网络的刀具磨损状态识别

摘要

The state of tool wear influences the metal cutting process ,so the monitoring of cutting tool wear condition is an impor-tance for improving the quality of the products .The tool wear condition monitoring system is designed .The tool vibration signals are collected with sensors and analyzed by wavelet packet .The feature value of tool wear state is extracted from the different fre-quency band energy .Using the BP neural network ,the mapping relationship between the tool wear and vibration signal feature is established.Therefore,the tool wear condition monitoring is completed .The system is realized using C ++Builder and Matlab mixed programming .The experiments show that the system identifies the tool wear state correctly .%刀具磨损状态影响金属切削过程，因此监测刀具磨损状态对提高产品质量有着重要的意义。设计刀具磨损状态监测系统，利用传感器采集刀具振动信号，通过小波包对振动信号进行数据分析，并把不同频段的能量值作为刀具磨损状态的特征值，建立BP神经网络，从而在刀具磨损状态和振动信号特征向量之间建立映射关系，完成刀具磨损状态的监测。利用C＋＋Builder和Matlab软件混合编程实现了系统的功能。试验表明，系统运行良好，能够对刀具磨损状态进行正确识别。