New Approach to Monitor the Tool Condition in a CNC Machining Center

摘要

We propose to monitor the cutting tool condition in a CNC-machining center by using continuous Hidden Markov Models (HMM). A vibration signal database was created monitoring the vibration between the cutting tool and workpiece. We trained/tested the HMM for 18 different operating conditions. The HMM were created by preprocessing the waveforms, followed by training step using the Baum-Welch algorithm. In the decoding process, the signal waveform is also preprocessed, then the trained HMM are used for decoding. Early experimental results validate our proposal about exploiting speech recognition frameworks in monitoring tool condition. The proposed model is capable of detecting the cutting tool condition within large variations of spindle speed and feed rate. The classifier performance was of 96%.