Journal of the Korea Academia-Industrial cooperation Society (한국산학기술학회논문지)

The Korea Academia-Industrial cooperation Society (한국산학기술학회)
권호 표지
DOI QR코드

DOI QR Code

Flaw Evaluation of Bogie connected Part for Railway Vehicle Based on Convolutional Neural Network

CNN 기반 철도차량 차체-대차 연결부의 결함 평가기법 연구

  • Kwon, Seok-Jin (Division of Advanced Railroad Vehicle, Korea Railroad Research Institute) ;
  • Kim, Min-Soo (Division of Advanced Railroad Vehicle, Korea Railroad Research Institute)
  • 권석진 (한국철도기술연구원 차세대철도차량본부) ;
  • 김민수 (한국철도기술연구원 차세대철도차량본부)
  • Received : 2020.10.07
  • Accepted : 2020.11.06
  • Published : 2020.11.30
Download PDF
⟨ Previous Next ⟩

Abstract

The bogies of railway vehicles are one of the most critical components for service. Fatigue defects in the bogie can be initiated for various reasons, such as material imperfection, welding defects, and unpredictable and excessive overloads during operation. To prevent the derailment of a railway vehicle, it is necessary to evaluate and detect the defect of a connection weldment between the car body and bogie accurately. The safety of the bogie weldment was checked using an ultrasonic test, and it is necessary to determine the occurrence of defects using a learning method. Recently, studies on deep learning have been performed to identify defects with a high recognition rate with respect to a fine and similar defect. In this paper, the databases of weldment specimens with artificial defects were constructed to detect the defect of a bogie weldment. The ultrasonic inspection using the wedge angle was performed to understand the detection ability of fatigue cracks. In addition, the convolutional neural network was applied to minimize human error during the inspection. The results showed that the defects of connection weldment between the car body and bogie could be classified with more than 99.98% accuracy using CNN, and the effectiveness can be verified in the case of an inspection.

철도차량의 대차는 열차 주행을 위한 핵심적인 장치이다. 철도차량의 대차에서 피로결함은 운행 중 기대되지 않거나 과도한 하중, 용접결함, 재료 결함 등의 다양한 요인에 의해 발생할 수 있다. 철도차량의 사고를 방지하기 위해서 차체-대차연결부의 손상을 검출하고 발생 결함에 대한 정확한 평가가 요구된다. 이러한 철도차량의 차체-대차 연결부는 초음파 비파괴 검사를 통하여 건전성을 확보하고 있으나 결함 발생에 대한 학습기법을 이용한 판정방법이 필요하다. 최근 미세한 결함이나 유사한 결함을 높은 인식율로 검출하기 위하여 딥러닝 기법에 관한 여러 연구가 진행되고 있다. 본 연구에서는 철도차량의 차체-대차 연결부의 결함 검출능력을 위하여 용접부의 인공결함 시편에 대하여 데이터베이스 구축하였으며. 웨지형 초음파 센서를 이용하여 차체-대차 연결부에 대한 비파괴 검사를 수행하였다. 부가적으로 인적오류를 최소화하기 위하여 결함판단 학습기법인 합성곱 신경망기법(Convolutional Neural Network)을 적용하였다. 그 결과 합성곱 신경망기법 기법을 이용하여 철도차량의 차체-대차 연결 용접부의 균열을 99.98%이상 균열성 결함으로 판별할 수 있었으며 철도차량 차체-대차 연결부의 비파괴검사시 본 연구의 기술이 적용 가능함을 확인할 수 있었다.

Keywords

References

  1. Science technology of high-speed train, The science times, April 19, 2004, Available From: https://www.sciencetimes.co.kr/news
  2. H.K.Jun,S.J.Kwon,D.H.Lee,J.W.Seo, "Fatigue crack analysis in a bolster of a metro train," Engineering Failure Analysis, vol. 76, pp.44-54, 2017 DOI: https://doi.org/10.1016/j.engfailanal.2017.02.005
  3. Railway serious incident investigation report, RI2019-1-I, Japan Transport Safety Board, March 28, 2019, Available From: https://www.mlit.go.jp/jtsb/eng-rail_report/English/
  4. H.K Choi, K.S. Seo, "CNN Based detection of surface defects for electronic parts," Journal of Korean Institute of Intelligent Systems, vol. 27, no. 3, pp.195-200, 2017 DOI: https://doi.org/10.5391/JKIIS.2017.27.3.195
  5. Y. Park, and I. S. Kweon, "Ambiguous surface defect image classification of AMOLED displays in smartphones," IEEE Trans. on Industrial Informatics, vol. 12, no. 2, pp.597-607, 2016 DOI: https://doi.org/10.1109/TII.2016.2522191
  6. R. Girshick, J. Donahue, T. Darrell, J. Malik, "Region-Based Convolutional Networks for accurate object detection and segmentation," IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 38, no. 1, pp.42-158, 2016 DOI: https://doi.org/10.1109/TPAMI.2015.2437384
  7. Y. K. Park, J. G. Park, H. I. On and D. J. Kang, "Convolutional Neural Network-based system for vehicle front-side detection," Journal of Institute of Control, Robotics and Systems, vol. 21, no. 11, pp.1008-1016, 2015 DOI: https://doi.org/10.5302/J.ICROS.2015.15.0163
  8. D. Tsai and H. Tsai, "Low contrast surface inspection of mura defects in liquid crystal displays using optical flow based motion analysis," Machine Vision and Applications, vol. 22, issue 4, pp. 629-649, 2011 DOI: https://doi.org/10.1007/s00138-010-0256-1
  9. M.K.Lee, K.S.Seo, "Comparison of Region-Based CNN methods for defects detection on metal surface," The Transactions of the Korean Institute of Electrical Engineers, vol.67, no.7, pp.865-870, 2018 DOI: https://doi.org/10.5370/KIEE.2018.67.7.865
  10. A.R.Kim,D.H.Kim,Y.S.Byun,S.W.Lee, "Crack detection of concrete structure using deep learning and image processing method in geotechnical engineering," Journal of the Korean Geotechnical Society, vol. 34, no. 12, pp.145-154, 2018 DOI: https://doi.org/10.7843/kgs.2018.34.12.145
  11. N.Munir, H.J.Kim, J.Park, S.J.Song, S.S.Kang, "Convolutional neural network for ultrasonic weldment flaw classification in noisy conditions," Ultrasonics, vol. 94, pp.74-81, 2019 DOI: https://doi.org/10.1016/j.ultras.2018.12.001
  12. N.Munir, H.J.Kim, S.J.Song, S.S.Kang, "Investigation of deep neural network with drop out for ultrasonic flaw classification in weldments," Journal of Mechanical Science and Technology, vol. 32, no. 7, pp.3073-3080, 2018 DOI: https://doi.org/10.1007/s12206-018-0610-1
  13. N.Munir, J.Park, H.J.Kim, S.J.Song, S.S.Kang, "Performance enhancement of convolutional neural network for ultrasonic flaw classification by adopting autoencoder," NDT & E International, vol. 111, pp.1-13, 2020 DOI: https://doi.org/10.1016/j.ndteint.2020.102218