Adaptive observer for on-line tool wear estimation and monitoring in turning, using a hybrid identification approach

On-line tool wear estimation in turning is essential for online cutting process optimization. In this paper, an adaptive observer based on cutting force measurement is used for a reliable on-line flank wear estimation and tool life monitoring. The design of the adaptive observer is realized using a linear observer and a Recursive Least Squares method as the adaptation algorithm. An additive white noise is considered to the process output. A continuous time hybrid identification approach is used. For model validation, the flank wear is estimated using a nonlinear model. The proposed approach improves the results previously proposed in other research works particularly concerning the number of identified parameters.