An integrated real-time tool wear monitoring and prediction framework with autonomous in-situ machine vision-based labeling
Source abstract
Abstract
Tool wear in milling operations significantly impacts machining efficiency, surface quality and downtime, with flank wear (VBmax) serving as a critical failure indicator. Traditional tool condition monitoring (TCM) methods suffer from offline measurements, labor-intensive labeling and disjointed monitoring-prediction workflows. This study proposes an integrated real-time TCM framework with autonomous in-situ machine vision-based labeling to address these limitations. An automated data acquisition platform collects multi-sensor signals and high-resolution tool images during machining without interruptions, using CNC integration and FOCAS protocol for precise control. A lightweight U-Net model, enhanced with depthwise separable convolutions (DSConv) and Ghost modules, performs semantic segmentation of worn flank regions. Subsequent geometric contour analysis automatically measures precise VBmax values, thereby providing zero-human-intervention ground-truth labels for data-driven monitoring and prediction models. Multi-domain features from preprocessed signals are refined using Pearson correlation for input to a Kolmogorov-Arnold Network (KAN) for instantaneous wear regression. Accumulated KAN outputs feed a Reformer model for sequential wear prediction. Experimental results demonstrate automatic labeling with error below 5 μm, monitoring performance with RMSE and MAE consistently below 1.5 μm and 1 μm across public and experimental datasets, and superior prediction performance with RMSE and MAE below 0.6 μm and 0.5 μm compared to competitive benchmarks.
Source: openalex_inverted_index · Quality: baseline
Research context
OpenAlex topics: Advanced machining processes and optimization · Surface Roughness and Optical Measurements · Robot Manipulation and Learning
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Mechanical Science & Engineering, Huazhong University of Science and Technology
School of Electrical & Information Engineering, Hubei University of Automotive Technology
School of Management, Xihua University
Dongfeng Motor Corporation
Funding
Confidential · Development and Evaluation of Methods and Systems for Energy and Resource Efficiency Improvements in Sustainable Manufacturing
References (53)
Show source-provided references
1. 10.1007/s40430-017-0762-410.1007/s40430-017-0762-4 ↗
2. 10.1016/j.ijmachtools.2006.09.00710.1016/j.ijmachtools.2006.09.007 ↗
3. 10.3390/s1803082310.3390/s18030823 ↗
4. 10.1007/s10845-019-01526-410.1007/s10845-019-01526-4 ↗
5. 10.1016/j.jclepro.2025.14621010.1016/j.jclepro.2025.146210 ↗
6. 10.1007/s00170-018-2571-z10.1007/s00170-018-2571-z ↗
7. 10.1016/j.jmsy.2021.12.00210.1016/j.jmsy.2021.12.002 ↗
8. 10.1007/s00170020008010.1007/s001700200080 ↗
9. 10.1016/j.jmapro.2023.05.02710.1016/j.jmapro.2023.05.027 ↗
10. 10.3390/s2117598410.3390/s21175984 ↗
11. 10.1016/j.engfailanal.2022.10616410.1016/j.engfailanal.2022.106164 ↗
12. Han K.; Wang Y.; Tian Q.; et al. (2020) Ghostnet: More features from cheap operations[C]//Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 1580–1589.10.1109/CVPR42600.2020.00165 ↗
13. He, Y.; Ling, J.; Wang Y.; et al. (2020) In-process tool wear monitoring model based on LSTM-CNN. China Mechanical Engineering, 31(16): 1959.
14. 10.1016/j.measurement.2022.11071910.1016/j.measurement.2022.110719 ↗
15. Hong Y. (2021) Deep dual-reso-lution networks for real-time and accurate semantic seg-mentation of road scenes. arXiv Preprint arXiv210106085.
16. 10.1016/j.jmapro.2021.01.01610.1016/j.jmapro.2021.01.016 ↗
17. 10.1109/ISCTIS51085.2021.0006510.1109/ISCTIS51085.2021.00065 ↗
18. 10.1007/s00170-022-09058-710.1007/s00170-022-09058-7 ↗
19. 10.1109/ACCESS.2019.295833010.1109/ACCESS.2019.2958330 ↗
20. 10.1088/1757-899X/1305/1/01203710.1088/1757-899X/1305/1/012037 ↗
21. 10.1016/j.procir.2020.03.14010.1016/j.procir.2020.03.140 ↗
22. Kitaev N.; Levskaya A. (2020) Reformer: The efficient transformer. arXiv Preprint arXivŁ200104451.
23. 10.1016/j.wear.2018.05.01210.1016/j.wear.2018.05.012 ↗
24. 10.1109/TIM.2019.295247610.1109/TIM.2019.2952476 ↗
25. 10.1016/j.measurement.2015.10.02910.1016/j.measurement.2015.10.029 ↗
26. 10.1109/CVPR52733.2024.0264010.1109/CVPR52733.2024.02640 ↗
27. 10.1007/s10845-022-01954-910.1007/s10845-022-01954-9 ↗
28. Liu Z.; Wang Y.; Vaidya S.; et al. (2024) Kan: Kolmogorov-arnold networks. arXiv preprint arXiv240419756.
29. Lu, Z.; Zhu, H.; Wu, J. (2024) Prediction of tool wear value based on meta-learning and multi-head attention. Computer Integrated Manufacturing System, 30(10): 3588.
30. 10.1177/0954405423120288910.1177/09544054231202889 ↗
31. 10.1007/s40430-023-04189-810.1007/s40430-023-04189-8 ↗
32. 10.3390/s2312543110.3390/s23125431 ↗
33. 10.1016/j.ijmachtools.2014.10.01110.1016/j.ijmachtools.2014.10.011 ↗
34. 10.1177/095440541143238110.1177/0954405411432381 ↗
35. PHM Society. PHM data challenge 2010[DB/OL]. https://www. phmsociety. org/competition/phm/10.
36. 10.3390/s2217668610.3390/s22176686 ↗
37. 10.1016/j.rcim.2024.10282010.1016/j.rcim.2024.102820 ↗
38. 10.1016/S1003-6326(10)60272-510.1016/S1003-6326(10)60272-5 ↗
39. 10.1016/j.ijmachtools.2004.05.00310.1016/j.ijmachtools.2004.05.003 ↗
40. 10.1016/j.ymssp.2022.10985110.1016/j.ymssp.2022.109851 ↗
41. Vaswani, A.; Shazeer, N.; Parmar, N.; et al. (2017) Attention is all you need. Advances in Neural Information Processing Systems. 30.
42. 10.1016/j.measurement.2019.02.00410.1016/j.measurement.2019.02.004 ↗
43. 10.1109/TPAMI.2020.298368610.1109/TPAMI.2020.2983686 ↗
44. 10.1109/CVPR52729.2023.0138510.1109/CVPR52729.2023.01385 ↗
45. Xie, E.; Wang, W.; Yu, Z.; et al. (2021) SegFormer: Simple and efficient design for semantic segmentation with transformers. Advances in Neural Information Processing Systems, 34: 12077–12090.
46. 10.3390/ma1405131110.3390/ma14051311 ↗
47. 10.1007/s11263-021-01515-210.1007/s11263-021-01515-2 ↗
48. 10.1007/s00530-024-01509-310.1007/s00530-024-01509-3 ↗
49. 10.1080/10910344.2025.247235010.1080/10910344.2025.2472350 ↗
50. 10.1109/TIE.2017.273343810.1109/TIE.2017.2733438 ↗
51. 10.1007/978-3-030-87878-810.1007/978-3-030-87878-8 ↗
52. 10.1109/TIM.2021.307799510.1109/TIM.2021.3077995 ↗
53. 10.3390/sym1308143810.3390/sym13081438 ↗