Mondragon Unibertsitatea face-milling dataset for smart tool condition monitoring.

This article presents a dataset of face-milling experiments for smart tool condition monitoring (TCM) performed under varying cutting conditions in the High-Perfomance Machining laboratory of Mondragon Unibertsitatea (MU). The experiments collected raw internal signals from the machine. Cutting forces, vibration signals, and acoustic emission signals were collected with external sensors. Tool wear was measured before each experiment and annotated accordingly, providing tool wear progression throughout the dataset. The dataset was technically validated using Python scripts to ensure the quality and reproducibility of the dataset. The resulting MU-TCM face-milling dataset offers a reproducible research design of experiments and associated data to carry out and advance smart TCM of milling processes. The dataset supports applications such as training machine learning and deep learning for TCM, enables sensor fusion research with diverse signal combinations, and facilitates the development of TCM solutions using only internal CNC signals for industrial environments. By supporting these applications, the dataset is expected to help reduce the gap between research and industry in smart TCM applications.