Improvement of generalization performance of diagnostic system for drill bit abnormality in rotary percussion drilling with grad-CAM

被引：0
作者：
Yuna Nakazawa [1 ]
Natsuo Okada [1 ]
Jo Sasaki [2 ]
Lesego Senjoba [3 ]
Yoko Ohtomo [4 ]
Youhei Kawamura [4 ]
机构：
[1] Division of Sustainable Resources Engineering, Graduate School of Engineering, Hokkaido University, Kita-13, Nishi-8, Sapporo
[2] MMC RYOTEC Co. Ltd., 1528 Yokoi Nakashinden, Kobe-Cho, Anpachi-Gun, Gifu
[3] Graduate School of International Resource Sciences, Akita University, 1-1 Tegata-Gakuenmachi, Akita
[4] Division of Sustainable Resources Engineering, Faculty of Engineering, Hokkaido University, Kita-13, Nishi-8, Kita-Ku, Sapporo
来源：
Discover Applied Sciences | / 7卷 / 4期
关键词：
Convolutional neural network; Drill bit abnormality; Frequency filter; Grad-CAM; Mining automation; Rotary percussion drill;
D O I：
10.1007/s42452-025-06796-7
中图分类号：
学科分类号：
摘要：
Rotary percussion drills, specifically top hammer types, are essential in resource extraction and exploration. These drills frequently suffer damage at the drill bit, necessitating effective decision-making to address issues. Traditionally, on-site operators relied on intuition, but recent advancements promote automated diagnostic systems to enhance safety and failure detection in challenging mining environments, such as those with high dust levels and intense vibrations. This study uses Convolutional Neural Networks (CNN) to develop an automated diagnostic system for detecting drill bit abnormalities. Previous models classified five distinct bit conditions but faced challenges in generalizing performance due to variations in the frequency bands used for judgments. This study refines the approach by simplifying the system to detect two states: normal and abnormal. Using Gradient-weighted Class Activation Mapping (Grad-CAM), the system identifies critical frequency bands, enabling a unified frequency filter. By training the model to recognize differences in rock types, it adapts to untrained drill bit types, achieving an accuracy of 83.3%, significantly higher than the 44.8% in previous models. This innovation allows the system to generalize across different rock and bit conditions, improving drilling efficiency and minimizing downtime in diverse environments. © The Author(s) 2025.
引用
收藏
相关论文
共 30 条
  • [1] Franca L.F.P., A bit-rock interaction model for rotary-percussive drilling, Int J Rock Mech Min Sci, 48, 5, pp. 827-835, (2011)
  • [2] Kawamura Y., Jang H., Hettiarachchi D.S., Takarada Y., Okawa H., Shibuya T., A case study of assessing button bits failure through wavelet transform using rock drilling induced noise signals, J Powder Metall Min, 6, 1, (2017)
  • [3] Senjoba L., Sasaki J., Kosugi Y., Toriya H., Hisada M., Kawamura Y., One-dimensional convolutional neural network for drill bit failure detection in rotary percussion drilling, Mining, 1, 3, pp. 297-314, (2021)
  • [4] Yoshino K., Hisatoshi T., Mahdi S., Tsuyoshi A., Hajime I., Youhei K., Deep learning as an early detection system for rotary percussion drilling malfunctions, Int J Soc Mater Eng Resou, 25, 2, pp. 205-211, (2022)
  • [5] Wei J., He Z., Wang J., Wang D., Zhou X., Fault detection based on multi-dimensional kde and jensen–shannon divergence, Entropy, 23, 3, pp. 1-24, (2021)
  • [6] Reddy G.T., Et al., Analysis of dimensionality reduction techniques on big data, IEEE Access, 8, pp. 54776-54788, (2020)
  • [7] Hussain F., Hussain R., Hassan S.A., Hossain E., Machine learning in IoT security: current solutions and future challenges, IEEE Commun Surv Tutor, 22, 3, pp. 1686-1721, (2020)
  • [8] Tran T., Pham N.T., Lundgren J., A deep learning approach for detecting drill bit failures from a small sound dataset, Sci Rep, 12, 1, (2022)
  • [9] Wang H., Liu Z., Peng D., Zuo M.J., Interpretable convolutional neural network with multilayer wavelet for noise-robust machinery fault diagnosis, Mech Syst Signal Process, 195, (2023)
  • [10] Chai S., Fei X., Wang Y., Dai L., Sui M., Deep learning-based lung medical image recognition, Int J Innov Res Comput Sci Technol, 12, 3, pp. 100-105, (2024)
123