Tool condition monitoring and degradation estimation in rotor slot machining process

Tool wear degradation and working status of slotting cutter have a great effect on the surface quality of rotor slot; therefore, tool condition monitoring and its degradation estimation are needed for guaranteeing slot machining quality. This paper proposes a two-phase method based on acoustic emission (AE) signal classification and logistic regression model for slotting cutter condition monitoring and its degradation estimation. In the first phase, the failure reliability estimation models corresponding to different machining processes are established considering the variability of process system like tool regrinding times and material randomness of workpiece. In the second phase, the most appropriate estimation model corresponding to the optimum cluster is selected and used for failure reliability estimation and status determination of slotting cutter. This approach has been validated on a CNC rotor slot machine in a factory. Experimental results show that the proposed method can be effectively used for cutting tool degradation estimation and status determination of slotting cutter with high accuracy.