Experimental and theoretical investigation of tool wear in milling processes, towards a digital twin
Loading...
Date
2024-12-10
Authors
Τριαντοπούλου, Ελευθερία
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Tool condition monitoring is an essential component of any modern machining environment, contributing significantly to cost-effectiveness, product quality and operational efficiency. Most existing approaches rely on the installation of commercially available sensors to the milling machine and/or workpiece. However, this setup has limitations regarding workpiece size, maintaining process repeatability and there are potential measurement variations due to installation variations. Sensor integrated cutting tools offer improved measurement accuracy and precision, as they are installed closer to the cutting zone and eliminate errors from repetitive installations, ensuring consistent process repeatability. This leads to higher-quality parts, enhanced production efficiency and cost reduction. This study presents an approach for tool wear monitoring based on the fusion of cutting force and temperature signals, collected from cutting tool integrated sensors. The digital model of the cutting tool utilizes signal features that are particularly sensitive to wear, selected through correlation analysis. Two Artificial Intelligence algorithms, Decision Tree and Random Forest are trained and employed for the digital model construction, achieving high accuracy and speed that enables the real time monitoring and prediction of tool wear.
Description
Keywords
Tool wear, Tool wear monitoring, Integrated sensors, Milling