Tool wear monitoring in micromilling using Support Vector Machine with vibration and sound sensors

Cutting tool wear is inevitable and becomes even more critical in micromachining processes, due to the small size of the microtools, which makes it impossible to detect any damage or break in the microtool without the use of high magnification microscopy. Therefore, monitoring the wear conditions of microtools is essential to guarantee the quality of the surfaces generated by micromachining processes. Even with the use of sensors, because of the complexity and similarity of the signals, identifying changes related to variation in wear is not a simple task. To overcome these problems, this paper presents a new approach to monitor the wear of cutting tools used in the micromilling process using SVM (Support Vector Machine) artificial intelligence model, vibration and sound signals. The signals were acquired for microchannels manufactured using carbide microtools coated with (Al, Ti) N, with a cutting diameter of 400 mu m. The input features for the model were selected using the RFE method (Recursive Feature Elimination). In addition to the main objective, the behavior of the wear curve of the microtool in relation to the wear curve of the conventional machining process was studied. The results showed that the behavior of the curves were similar and the microtool with shorter cutting length had a longer life. The proposed classification methodology obtained a classification accuracy of up to 97.54%, showing that it is possible to use it to monitor the cutting tool wear.

Automated summary

This paper presents a new approach to monitor the wear of cutting tools used in micromilling processes using SVM artificial intelligence model, vibration, and sound signals. Experimental results show that microtools with different cutting lengths have different lifespans.