On machine 3D reconstruction of endmill tool wear

During the machining of difficult-to-machine materials, monitoring the tool wear is essential to avoid excessive wear negatively impacting the part’s surface integrity or damaging the part beyond repair. When manually monitoring the tool wear, the machine operator must physically remove the tool from the machine at regular intervals to inspect the tool wear, which can be very time-consuming. Instead, tools are often changed when the wear is significantly below the maximum allowable value, adding to the part production cost, environmental impact and machine downtime. This paper presents a method for in-situ tool wear measurement using a laser line scanner placed inside a machine tool, providing a 3-D model of the tool in its current condition. This can be performed without operator intervention and the 3-D reconstruction of the tool can then be used for further analysis. The proposed method can provide an automated system for generating tool wear database for training machine learning models. Additionally, it can be combined with machine learning-based indirect tool condition monitoring (TCM) methods, allowing the models to self-validate predicted tool wear values, further reducing operator input and increasing productivity.