基于BP神经网络的超声振动钻削钻头磨损状态监测试验研究

搭建了超声轴向振动钻削钻头磨损状态的钻削力和声发射信号采集系统,采集不同磨损状态下钻中区域的钻削力和声发射信号进行小波分解,得到与钻头磨损状态相关的特征量作为识别钻头磨损状态的特征参数,输入到建立的6-13-3的三层BP神经网络模型中进行融合,识别钻头磨损状态。试验结果表明,通过BP神经网络技术将钻削力和声发射信号融合识别钻头磨损的准确率约88.9%,能够有效监测钻头磨损状态。

Experimental Study on Wear Monitoring of Ultrasound Vibration Drilling Bit Based on BP Neural Network

A data acquisition system of drilling force and acoustic emission signals for ultrasonic axial vibration drilling bit wear state is built.The drilling force and acoustic emission signals in different wear states are collected and decomposed by wavelet transform.The characteristic parameters related to the wear state of drill bit are obtained as the characteristic parameters for identifying the wear state of drill bit,which are input to the established 6.-The 3-layer BP neural network model of 13-3 is fused to identify the wear state of drill bit.The experimental results show that the accuracy of recognizing bit wear by combining drilling force and acoustic emission signals is about 88.9%,which can effectively monitor the bit wear state.