Determinação da condição de desgaste da ferramenta de corte via monitoramento de vibração e inteligência artificial

Abstract

One of the most important consequence in machining process is the tool wear. Thus, monitoring the wear of cutting tools becomes essential to ensure product success, increase productivity and avoid catastrophic damages to the equipment. Since wear is related to the vibrations of the process the vibration signal can be used to monitor it. This work presents a new approach for identification of wear condition of a tool during turning operation of VC 131 steel (AISI D6) using well known techniques with low computational cost which can provide a future industrial application to identify the ideal moment of tool change. To achieve this purpose, the vibration signals were measured during each turning step. Then, an artificial classification intelligence method (W-kNN) with features extracted from the vibration signals was used to identify the wear stage. Tests were performed with tools under different wear conditions, which were measured before and after each turning step. The results show that the combination of artificial classification intelligence method with vibratory features can successfully predict the lifespan of cutting tools which can certainly be used as an industrial tool wear-monitoring system.