Feature Selection for Tool Condition Monitoring in Turning Processes
The aim of the present work is to develop a tool condition monitoring system (TCMS) using sensor fusion and artificial neural networks. Particular attention is paid to the manner in which the most correlated features with tool wear are selected. Experimental results show that the proposed system can reliably detect tool condition in turning operations and is viable for industrial applications. This study leads to the conclusion that the vibration in the feed direction and the motor current signals are best suited for the development of a TCMS than the sound signal, which should be used as an additional signal.
M. Marcos and L. Sevilla
D. R. Salgado et al., "Feature Selection for Tool Condition Monitoring in Turning Processes", Materials Science Forum, Vol. 526, pp. 97-102, 2006