Damage detection in machining tools using acoustic emission, signal processing and feature extraction

Damage detection and fault diagnosis systems based on condition monitoring, feature extraction and sensor data guidance have achieved notoriety in the field of industrial automation for enabling the prediction of the remaining useful life of industrial assets. For that reason, this project aims to explore an alternative methodology for detecting damage in machining tools based on data from acoustic emission sensors. The study was validated from an experimental analysis carried out in the milling process. The proposed approach consists of designing condition indicators that quantify damage to the milling cutter based on the implementation of the root mean square deviation (RMSD) and correlation coefficient deviation metric (CCDM) indices. The study was carried out by testing different frequency bands of the acoustic signals collected during the process, by calculating the fast Fourier transform (FFT), seeking the most suitable to determine the wear of the material, which proved to be between 5 and 8 kHz. Finally, the results arising from the implementation of the proposed method proved to be very important for the optimization of the manufacturing process, being able to help in the automation of the exchange of the milling cutter or to alert the operators that this must be done.