Tool damage state condition monitoring in milling processes based on the mechanistic model goodness-of-fit metrics

Abstract

Monitoring of the end milling tool wear in real-time is very crucial for the quality control of a surface finish. In the model-based approaches, usually the tool wear is monitored by tracking the force model coefficients during the milling operation. In this work, we explore a monitoring approach based on two metrics derived from the goodness-of-fit of the milling mechanistic model. The cumulative sum control charts are constructed to monitor changes occurring to the tool condition during a milling operation. Compared to previously suggested methods based solely on the values of force model coefficients, our approach facilitates a preciser and less conservative identification of the point of transition from homogeneous wear to onset of breakout formation at the cutting edges. The proposed strategy is an alternative approach for real time monitoring of tool condition based on mechanistic models.