In general, tool failure contributes about 7% to the down time of machine centers. And more severely, tool failure will reduce the machining quality of parts, and even damage the machine. Therefore, Tool wear condition monitoring (TWCM) is an important part of intelligent manufacturing, and has been a research hotspot since 1968.

TWCM can be divided into two methods according to the sensor types: direct method and indirect method. The direct method employs vision or laser sensors. Despite its superior precision and reliability, achieving online real-time monitoring is challenging, as the sensors are easily interfered with by cutting fluid, chips and other elements inherent to the processing environment. Conversely, the indirect method holds promise for online real-time monitoring capabilities, yet its practical application remains limited despite an abundance of academic publications.

A key obstacle is the selection of suitable monitoring sensors, such as those that measure force, vibration, temperature, and acoustic emissions, which can inevitably disrupt the natural processing conditions. The microphone sensor, which captures audible sound signals, has emerged as a promising strategy. Its potential is contingent upon enhancing its accuracy through initiatives denoising system and the development of interpretable decision-making algorithms.

Recently, a team of intelligent manufacturing scientists led by Guochao Li from Jiangsu University of Science and Technology in China conducted the first systematic review of the status and trend of audible sound-based tool wear monitoring. This comprehensive work not only serves as a valuable resource for researchers and manufacturers by summarizing recent trends, but also highlights four promising research directions: the development of datasets, initiative denoising system, specialized feature extraction techniques, and the creation of interpretable decision-making algorithms.

The team published their work in Journal of Advanced Manufacturing Science and Technology.

"In this report, we've conducted a review of the current state and future direction of tool wear monitoring focused on audible sound, spanning a decade of scholarly publications," said Guochao Li, Associate Professor at the School of Mechanical Engineering at Jiangsu University of Science and Technology in China, who is a seasoned expert in the field of tool wear condition monitoring.

"This includes an in-depth analysis of the physical properties and characteristics of machining audible sound, the generation mechanisms of milling audible sound, and advancements in key technologies such as signal acquisition, noise reduction, feature extraction, and decision-making algorithms, along with potential areas for future research."

"We believe that our work could significantly promote the practical application of tool wear monitoring," noted Li Sun, a Lecturer at the same institution, who also specializes in machining process monitoring. "The use of microphone sensors to collect audible sound signals presents a promising strategy due to their close correlation with tool wear, eliminating the need for additional sensors, ease of installation, adaptability in measurement, and non-interference with the processing environment.

"However, the accuracy of tool wear condition monitoring based on audible sound signals is currently insufficient. Our study offers valuable insights and the latest trends to researchers and manufacturers, potentially facilitating the broader application of tool wear monitoring in practical scenarios."

The research team also includes notable contributions from Xinhang Shang, Lei Yang, and Honggen Zhou from the School of Mechanical Engineering at Jiangsu University of Science and Technology in Zhenjiang, China, as well as Bofeng Fu from Shanxi Diesel Engine Heavy Industry Co., Ltd. in Xingping, China.

More information: Guochao LI et al, Application of audible sound signals in tool wear monitoring: a review, Journal of Advanced Manufacturing Science and Technology (2024). DOI: 10.51393/j.jamst.2025003

Provided by Tsinghua University Press