Related Experiment Video
Updated: Jul 29, 2026
08:22
A Novel Single Animal Motor Function Tracking System Using Simple, Readily Available Software
Published on: August 31, 2018
6.6K
A Novel Multivariate Cutting Force-Based Tool Wear Monitoring Method Using One-Dimensional Convolutional Neural
Xu Yang1,2, Rui Yuan1,2, Yong Lv1,2
1Key Laboratory of Metallurgical Equipment and Control Technology, Ministry of Education, Wuhan University of Science and Technology, Wuhan 430081, China.
Sensors (Basel, Switzerland)
|November 11, 2022
Summary
This study introduces a new method for monitoring tool wear using cutting force signals and a one-dimensional convolutional neural network (1D CNN). The approach accurately detects tool wear conditions in precision manufacturing.
Area of Science:
- Manufacturing Engineering
- Signal Processing
- Machine Learning
Background:
- Tool wear monitoring is critical in precision manufacturing for maintaining product quality and process efficiency.
- Cutting force signals contain valuable dynamic information about tool wear states.
- Existing methods may struggle with the nonlinear and nonstationary characteristics of tool wear signals.
Purpose of the Study:
- To propose a novel multivariate cutting force-based method for tool wear condition monitoring.
- To leverage the power of one-dimensional convolutional neural networks (1D CNN) for enhanced monitoring accuracy and stability.
- To develop a robust system for real-time tool wear assessment in machining processes.
Main Methods:
- Multivariate variational mode decomposition (MVMD) was applied to process multivariate cutting force signals, extracting multivariate band-limited intrinsic mode functions (BLIMFs).
- Modified multiscale permutation entropy (MMPE) was utilized to quantify the complexity and extract condition indicators from the BLIMFs.
- One-dimensional feature vectors were constructed from the entropy values and fed into a 1D CNN for classification of tool wear conditions.
Main Results:
- The MVMD successfully decomposed cutting force signals into informative BLIMFs, capturing nonlinear and nonstationary tool wear characteristics.
- MMPE effectively generated condition indicators reflecting the complexity of the extracted BLIMFs across multiple scales.
- The 1D CNN model achieved accurate and stable tool wear condition monitoring using the derived feature vectors.
Conclusions:
- The proposed method, integrating MVMD, MMPE, and 1D CNN, offers a promising approach for effective tool wear monitoring.
- This technique demonstrates significant potential for improving precision manufacturing through reliable condition monitoring.
- The developed methodology provides a robust framework for analyzing complex cutting force dynamics for tool wear assessment.
Related Concept Videos
01:20Two-Dimensional Force System
A two-dimensional system in mechanical engineering involves the analysis of motion and forces in a plane. A two-dimensional force vector can be resolved into its components as:
01:30Three-Dimensional Force System
In mechanical engineering, a three-dimensional force system is a system of forces acting in three dimensions, with forces applied along the x, y, and z coordinate axes. The three-dimensional force system is an important concept in mechanical engineering, as it allows engineers to understand and analyze the behavior of objects and structures in three dimensions. By understanding the forces acting on a system, engineers can design more efficient and effective mechanical systems that can withstand...