Citation: Li, Z.; Zhong, W.; Shi, Y.; Yu,
M.; Zhao, J.; Wang, G. Unsupervised
Tool Wear Monitoring in the Corner
Milling of a Titanium Alloy Based on
a Cutting Condition-Independent
Method. Machines 2022, 10, 616.
https://doi.org/10.3390/
machines10080616
Academic Editor: Kai Cheng
Received: 28 June 2022
Accepted: 19 July 2022
Published: 27 July 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Article
Unsupervised Tool Wear Monitoring in the Corner Milling
of a Titanium Alloy Based on a Cutting Condition-
Independent Method
Zhimeng Li
1
, Wen Zhong
1
, Yonggang Shi
1
, Ming Yu
2
, Jian Zhao
1
and Guofeng Wang
3,
*
1
School of Control and Mechanical Engineering, Tianjin Chengjian University, Tianjin 300384, China;
lzmcxg@tcu.edu.cn (Z.L.); zlzhongwen@163.com (W.Z.); syg_1554734524@163.com (Y.S.);
zhaojian_tju@163.com (J.Z.)
2
School of Computer and Information Engineering, Tianjin Chengjian University, Tianjin 300384, China;
maxyuming@126.com
3
School of Mechanical Engineering, Tianjin University, Tianjin 300350, China
* Correspondence: gfwangmail@tju.edu.cn
Abstract:
Real-time tool condition monitoring (TCM) for corner milling often poses significant
challenges. On one hand, corner milling requires configuring complex milling paths, leading to the
failure of conventional feature extraction methods to characterize tool conditions. On the other hand,
it is costly to obtain sufficient test data on corner milling for most of the current pattern recognition
methods, which are based on the supervised method. In this work, we propose a time-frequency
intrinsic feature extraction strategy of acoustic emission signal (AEs) to construct a cutting condition-
independent method for tool wear monitoring. The proposed new feature-extraction strategy is used
to obtain the tool wear conditions through the intrinsic information of the time-frequency image of
AEs. In addition, an unsupervised tool condition recognition framework, including the unsupervised
feature selection, the clustering based on adjacent grids searching (CAGS) and the density factor
based on CAGS, is proposed to determine the relationship between tool wear values and AE features.
To test the effectiveness of the monitoring system, the experiment is conducted through the corner
milling of a titanium alloy workpiece. Five metrics, PUR, CSM, NMI, CluCE and ClaCE, are used to
evaluate the effectiveness of the recognition results. Compared with the state-of-the-art supervised
methods, our method provides commensurate monitoring effectiveness but requires much fewer test
data to build the model, which greatly reduces the operating cost of the TCM system.
Keywords: tool wear monitoring; corner-milling; unsupervised
1. Introduction
The important position of the tool in the cutting process has caused extensive research
on the monitoring of tool wear states in the metal-cutting process [
1
,
2
], which has a history
of several decades. Among them, the complex profile of aeronautical structural parts makes
the sensor signal in the processing process very non-stationary, which greatly increases
the monitoring difficulty. Therefore, the monitoring system is required to have higher
adaptability and robustness. As an important part of signal feature extraction, its main
goal is to transform the sensor signal to achieve the dimensionality reduction and de-
redundancy of the original data, and extracting from it has high sensitivity, robustness and
reliability for monitoring targets, so as to improve the efficiency and accuracy of pattern
recognition. According to the difference of transform domain, it can be divided into feature-
extraction methods based on time domain [
3
,
4
], frequency domain [
4
,
5
] and time-frequency
domain [4,6,7].
In the time domain, the most commonly used features are root mean square, mean,
kurtosis, standard deviation, skewness, and crest factor. Yuan et al. [
8
] extracted four
Machines 2022, 10, 616. https://doi.org/10.3390/machines10080616 https://www.mdpi.com/journal/machines
