Design & Manufacturing

Korean Society of Die & Mold Engineering (한국금형공학회)
권호 표지

State recognition of fine blanking stamping dies through vibration signal machine learning

진동신호 기계학습을 통한 프레스 금형 상태 인지

  • Seok-Kwan Hong (Department of Molding & Metal Forming R&D, Korea Institute of Industrial Technology) ;
  • Eui-Chul Jeong (Department of Molding & Metal Forming R&D, Korea Institute of Industrial Technology) ;
  • Sung-Hee Lee (Department of Molding & Metal Forming R&D, Korea Institute of Industrial Technology) ;
  • Ok-Rae Kim (Department of Molding & Metal Forming R&D, Korea Institute of Industrial Technology) ;
  • Jong-Deok Kim (Technology Laboratory, DAESUNG FINE TEC Co., LTD.)
  • 홍석관 (한국생산기술연구원 금형성형연구부문) ;
  • 정의철 (한국생산기술연구원 금형성형연구부문) ;
  • 이성희 (한국생산기술연구원 금형성형연구부문) ;
  • 김옥래 (한국생산기술연구원 금형성형연구부문) ;
  • 김종덕 ((주)대성파인텍 기술연구소)
  • Received : 2022.12.09
  • Accepted : 2022.12.31
  • Published : 2022.12.31
Download PDF
⟨ Previous Next ⟩

Abstract

Fine blanking is a press processing technology that can process most of the product thickness into a smooth surface with a single stroke. In this fine blanking process, shear is an essential step. The punches and dies used in the shear are subjected to impacts of tens to hundreds of gravitational accelerations, depending on the type and thickness of the material. Therefore, among the components of the fine blanking mold (dies), punches and dies are the parts with the shortest lifespan. In the actual production site, various types of tool damage occur such as wear of the tool as well as sudden punch breakage. In this study, machine learning algorithms were used to predict these problems in advance. The dataset used in this paper consisted of the signal of the vibration sensor installed in the tool and the measured burr size (tool wear). Various features were extracted so that artificial intelligence can learn effectively from signals. It was trained with 5 features with excellent distinguishing performance, and the SVM algorithm performance was the best among 33 learning models. As a result of the research, the vibration signal at the time of imminent tool replacement was matched with an accuracy of more than 85%. It is expected that the results of this research will solve problems such as tool damage due to accidental punch breakage at the production site, and increase in maintenance costs due to prediction errors in punch exchange cycles due to wear.

Keywords

Acknowledgement

본 논문은 한국생산기술연구원 기관주요사업 "Add-on 모듈 탑재를 통한 지능형 뿌리공정 기술개발(KITECH EO-22-0005)" 의 지원으로 수행한 연구입니다.

References

  1. Zheng, Q., Zhuang, X., & Zhao, Z. State-of-the-art and future challenge in fine-blanking technology. Production engineering, 13(1), pp. 61-70. 2019. https://doi.org/10.1007/s11740-018-0839-7
  2. Lee, J. H., Shin, J., & Realff, M. J. Machine learning: Overview of the recent progresses and implications for the process systems engineering field. Computers & Chemical Engineering, 114, pp. 111-121. 2018. https://doi.org/10.1016/j.compchemeng.2017.10.008
  3. Achouch, M., Dimitrova, M., Ziane, K., Sattarpanah Karganroudi, S., Dhouib, R., Ibrahim, H., & Adda, M. On predictive maintenance in industry 4.0: Overview, models, and challenges. Applied Sciences, 12(16), pp. 8081. 2022.
  4. Hambli, R., Soulat, D., & Chamekh, A. Finite element prediction of blanking tool cost caused by wear. The International Journal of Advanced Manufacturing Technology, 44(7), pp. 648-656. 2009. https://doi.org/10.1007/s00170-008-1859-9
  5. Sari, D. Y., Wu, T. L., & Lin, B. T. Preliminary study for online monitoring during the punching process. The International Journal of Advanced Manufacturing Technology, 88(5), pp. 2275-2285, 2017. https://doi.org/10.1007/s00170-016-8956-y
  6. Ubhayaratne, I., Pereira, M. P., Xiang, Y., & Rolfe, B. F. Audio signal analysis for tool wear monitoring in sheet metal stamping. Mechanical Systems and Signal Processing, 85, pp. 809-826. 2017.