Tool Wear Monitoring Technology Based on Deep Learning of Machining State Features

Abstract

Aiming at the problem of real-time monitoring of the tool in the working state when the machine tool is producing the same part in batches, a tool wear state monitoring method based on the spindle power signal and the machining parameters of the machine tool is proposed. First, the linear model of deep learning is used to fit the linear model between machining parameters, tool wear and spindle power by using historical data, and then by setting the wear amount of the tool, the threshold value of the spindle power under different wear conditions is obtained, and the wear amount of the tool is monitored by judging the relationship between the spindle power and the threshold value. High availability compared to traditional downtime measurement methods. Finally, through the experiment, Collect the spindle power signal of the vertical machining centre to predict the wear state of the tool, and compare it with the actual wear amount. The results show that the proposed method has high accuracy and applicability, it can realize long-term online monitoring of tool wear under normal production conditions.