Title: Tool wear condition monitoring method based on graph neural network with a single sensor

Authors: Chen Gao; Jie Zhou; Yuan Yang; Yukun Fang; Gaofeng Zhi; Bintao Sun

Addresses: Department of Electromechanical and Transportation, Jiaxing Nanyang Polytechnic Institute, Jiaxing City, Zhejiang Province, China ' College of Mechanical and Electrical Engineering, Wenzhou University, Wenzhou City, Zhejiang Province, China ' College of Mechanical and Electrical Engineering, Wenzhou University, Wenzhou City, Zhejiang Province, China ' College of Mechanical and Electrical Engineering, Wenzhou University, Wenzhou City, Zhejiang Province, China ' College of Mechanical and Electrical Engineering, Wenzhou University, Wenzhou City, Zhejiang Province, China ' College of Mechanical and Electrical Engineering, Wenzhou University, Wenzhou City, Zhejiang Province, China

Abstract: Tool wear condition monitoring (TCM) is an important part of machining automation. In recent years, deep learning (DL)-based TCM methods have been widely researched. However, almost all DL-based methods need sufficient learning samples to obtain good accuracy, which is hard for TCM in terms of cost and time. In order to enhance the recognition accuracy of DL-based TCM under small samples, this paper proposed a new improved multi-scale edge-labelling graph neural network (MEGNN). Firstly, the signal of cutting force sensor is expanded to multi-dimensional data through phase space reconstruction. Secondly, these multi-dimensional data are encoded into a recurrence plot (RP). Then, the RP is input to multi-scale EGNN (MEGNN) to extract features. Finally, the tool condition is estimated through the updated edge labels using a weighted voting method. Our milling TCM experiments demonstrate the proposed MEGNN-based TCM method outperforms three DL-based methods under small samples.

Keywords: tool wear condition monitoring; TCM; small samples; recurrence plot; MEGNN; phase space reconstruction; PSR.

DOI: 10.1504/IJMMM.2022.125197

International Journal of Machining and Machinability of Materials, 2022 Vol.24 No.3/4, pp.199 - 214

Received: 25 Sep 2021
Accepted: 04 Nov 2021

Published online: 01 Sep 2022 *

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