Journal of the Institute of Convergence Signal Processing (융합신호처리학회논문지)

The Korea Institute of Convergence Signal Processing (한국융합신호처리학회)
권호 표지
DOI QR코드

DOI QR Code

Machine Tool State Monitoring Using Hierarchical Convolution Neural Network

계층적 컨볼루션 신경망을 이용한 공작기계의 공구 상태 진단

  • 이경민 (신라대학교 컴퓨터공학부)
  • Received : 2022.06.15
  • Accepted : 2022.06.25
  • Published : 2022.06.30
Download PDF
⟨ Previous Next ⟩

Abstract

Machine tool state monitoring is a process that automatically detects the states of machine. In the manufacturing process, the efficiency of machining and the quality of the product are affected by the condition of the tool. Wear and broken tools can cause more serious problems in process performance and lower product quality. Therefore, it is necessary to develop a system to prevent tool wear and damage during the process so that the tool can be replaced in a timely manner. This paper proposes a method for diagnosing five tool states using a deep learning-based hierarchical convolutional neural network to change tools at the right time. The one-dimensional acoustic signal generated when the machine cuts the workpiece is converted into a frequency-based power spectral density two-dimensional image and use as an input for a convolutional neural network. The learning model diagnoses five tool states through three hierarchical steps. The proposed method showed high accuracy compared to the conventional method. In addition, it will be able to be utilized in a smart factory fault diagnosis system that can monitor various machine tools through real-time connecting.

공작기계 상태 진단은 기계의 상태를 자동으로 감지하는 프로세스이다. 실제로 가공의 효율과 제조공정에서 제품의 품질은 공구 상태에 영향을 받으며 마모 및 파손된 공구는 공정 성능에 보다 심각한 문제를 일으키고 제품의 품질 저하를 일으킬 수 있다. 따라서 적절한 시기에 공구가 교체될 수 있도록 공구 마모 진행 및 공정 중 파손 방지 시스템 개발이 필요하다. 본 논문에서는 공구의 적절한 교체 시기 등을 진단하기 위해 딥러닝 기반의 계층적 컨볼루션 신경망을 이용하여 5가지 공구 상태를 진단하는 방법을 제안한다. 기계가 공작물을 절삭할 때 발생하는 1차원 음향 신호를 주파수 기반의 전력스펙트럼밀도 2차원 영상으로 변환하여 컨볼루션 신경망의 입력으로 사용한다. 학습 모델은 계층적 3단계를 거쳐 5가지 공구 상태를 진단한다. 제안한 방법은 기존의 방법과 비교하여 높은 정확도를 보였고, 실시간 연동을 통해 다양한 공작기계를 모니터링할 수 있는 스마트팩토리 고장 진단 시스템에 활용할 수 있을 것이다.

Keywords

References

  1. Y. J. An and J. Y. Kim, "A Study on Real-time Tool Breakage Monitoring on CNC Lathe Using Fusion Sensor," Tribology and Lubricants, vol. 28, no. 3, pp. 130~135, 2012. https://doi.org/10.9725/KSTLE.2012.28.3.130
  2. J. H. Kim, J. S. Yoon, and D. Y. Lee, "Estimation of Tangential Cutting Force using Spindle Load of CNC Machining Center," Journal of the Korean Society of Manufacturing Technology Engineers, vol. 28 no. 6, pp. 343-349, 2019. https://doi.org/10.7735/ksmte.2019.28.6.343
  3. B. H. Park, Y. J. Lee, and C. W. Lee, "Tool Condition Monitoring Using Deep Learning in Machining Process," Journal of the Korean Society for Precision Engineering, vol. 37, no. 6, pp. 415-420, 2020. https://doi.org/10.7736/JKSPE.020.040
  4. D. Y. Lee and S. J. Yun, "Development of Deburring Spindle up to 80,000 rpm for Precision Finishing," Korean Society for Precision Engineering Conference, pp. 441-442, 2017.
  5. I. S. Yook, D. H. Lee, G. S. Han, W. K. Han, and J. H. Hwang, "Development of the Air Bearing Spindle for Precision Machining for Metal case of the Smart-phone," Korean Society for Precision Engineering Conference, pp. 280-281, 2016.
  6. C. H. Park and T. G. Yoon, "Development of Drilling Spindle for Micro-hole Machining with Magnetic Bearings," Transactions of the Korean Society for Noise and Vibration Engineering, vol. 27, no. 4, pp. 510-517, 2017. https://doi.org/10.5050/KSNVE.2017.27.4.510
  7. M. Saimurugan and K.I. Ramachandran, "Detection The Behavior of Smartphone Users using Time-division Feature Fusion Convolutional Neural Network," Journal of The Korean Institute of Information and Communication Engineering, vol. 24, no. 9, pp. 1224-1230, 2020.
  8. S. J. Yoon, M. Y. Lee, J. H. Lee, S. H. Lee, and J. C. Na, "Fault Diagnosis Using Artificial Intelligence for the Spindle of Machine Tools," Transactions of the Korean Society of Mechanical Engineers, vol. 45, no. 5, pp. 401-408, 2021. https://doi.org/10.3795/KSME-A.2021.45.5.401
  9. J. Shi, G. Si, S. Li, B. Oresanya, and Y. Zhang, "Feature extraction based on the fractional Fourier transform for vibration signals with application to measuring the load of a tumbling mill," Control Engineering Practice, vol. 84, pp. 238-246, 2019. https://doi.org/10.1016/j.conengprac.2018.11.012
  10. K. M. Lee, C. Vununu, K. S. Moon, S. H. Lee, and K. R. Kwon, "Automatic machine fault diagnosis system using discrete wavelet transform and machine learning," Journal of Korea Multimedia Society, vol. 20, no. 8, pp. 1299-1311, 2017. https://doi.org/10.9717/KMMS.2017.20.8.1299
  11. M. Zhao, M. Kang, B. Tang, and M. Pecht, "Deep Residual Networks with Dynamically Weighted Wavelet Coefficients for Fault Diagnosis of Planetary Gearboxes," IEEE Transactions on Industrial Electronics, vol. 65, issue 5, pp. 4290-4300, 2018. https://doi.org/10.1109/TED.2018.2865225
  12. K. M. Lee, "A Machine Fault Diagnosis System based on Convolution Neural Network using Acoustic Power Spectral Density," Ph. D. dissertation, Pukyong National University, Busan, Republic of Korea, 2019.