Analysis of the structure of vibration signals for tool wear detection

Abstract

The objective of this work is to develop a reliable tool condition monitoring system (TCMS) for industrial application. The proposed TCMS is based on the analysis of the structure of the tool vibration signals using singular spectrum analysis (SSA) and cluster analysis. SSA is a novel non-parametric technique of time series analysis that decomposes the acquired tool vibration signals into an additive set of time series. Cluster analysis is used to group the SSA decomposition in order to obtain several independent components in the frequency domain that are presented to a feedforward back-propagation (FFBP) neural network to determine the tool flank wear. The results show that this use of SSA and cluster analysis provides an efficient automatic signal processing method, and that the proposed TCMS based on this procedure, is fast and reliable for tool wear monitoring.