Title: Assessment of features from multiple sensors in monitoring titanium milling

Authors: John H. Navarro-Devia; Yun Chen; Dzung Viet Dao; Huaizhong Li

Addresses: School of Engineering and Built Environment, Griffith University, Gold Coast 4222, Australia ' Department of Mechanical and Electrical Engineering, Xiamen University, Xiamen 361005, China ' Mechanical Engineering, School of Engineering and Built Environment, Griffith University, Gold Coast 4222, Australia ' Mechanical Engineering, School of Engineering and Built Environment, Griffith University, Gold Coast 4222, Australia

Abstract: Multi-sensor approach has become a topic of interest in the development of monitoring systems in manufacturing. This paper studies assessing some key statistical features in the time domain. During titanium milling, cutting forces, accelerations, and acoustic emission-RMS signals were extracted under different cutting parameters. A novel method for the automatic extraction of data corresponding to tool engagement is proposed, using the AE-RMS signal as an indicator of tool-workpiece interaction. The results indicate that common monitoring features are also affected by variation in machining parameters, and not only by the tool state, showing that RMS and spectral entropy have a higher sensibility. It justifies the use of multi-sensor for providing additional information, which might be undiscovered in a single-sensor configuration. The use of each type of sensor, the multi-sensor approach, and statistical features extraction are discussed. The findings may aid to expand the knowledge on titanium machining and data-driven monitoring.

Keywords: multi-sensor; monitoring; milling; titanium; signal processing; features extraction; Ti6Al4V; data-driven manufacturing; statistical analysis.

DOI: 10.1504/IJMMM.2022.122785

International Journal of Machining and Machinability of Materials, 2022 Vol.24 No.1/2, pp.16 - 47

Received: 22 Jul 2021
Accepted: 17 Oct 2021
Published online: 10 May 2022 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article