한국전산구조공학회논문집 (Journal of the Computational Structural Engineering Institute of Korea)

  • 제36권4호
  • /
  • Pages.221-232
  • /
  • 2023
  • /
  • 1229-3059(pISSN)
  • /
  • 2287-2302(eISSN)
한국전산구조공학회 (Computational Structural Engineering Institute of Korea)
권호 표지
DOI QR코드

DOI QR Code

초기 볼트풀림 상태의 볼트 체결력 예측을 위한 주파수응답 유사성 기반의 합성곱 신경망

Convolutional Neural Network-based Prediction of Bolt Clamping Force in Initial Bolt Loosening State Using Frequency Response Similarity

  • 이제현 (국립안동대학교 기계설계공학과 ) ;
  • 한정삼 (국립안동대학교 기계로봇공학과)
  • Jea Hyun Lee (Department of Mechanical Design Engineering, Andong National University) ;
  • Jeong Sam Han (Department of Mechanical & Robotics Engineering, Andong National University)
  • 투고 : 2023.06.02
  • 심사 : 2023.07.03
  • 발행 : 2023.08.31
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

본 논문에서는 볼트로 체결된 구조체에 대하여 초기 볼트풀림 상태에서의 볼트 체결력 예측 합성곱 신경망 훈련 방법을 제시한다. 8개의 볼트의 체결력이 변경된 상태에서 계산한 주파수응답들을 완전 체결된 상태의 초기 모델과의 크기 및 모양 유사성을 표현하는 유사성 지도로 생성한다. 주파수응답 데이터들의 생성에는 크리로프 부공간법 기반의 모델차수축소법을 적용하여 효율적인 방법으로 수행할 수 있도록 한다. 합성곱 신경망 모델은 회귀 출력 계층을 사용하여 볼트의 체결력을 예측하도록 하였으며, 훈련 데이터의 개수와 합성곱 신경망 계층의 개수를 다르게 준비하여 훈련시킨 네트워크들을 비교하여 그 성능을 평가하였다. 주파수응답에서 파생되는 유사성 지도를 입력 데이터로 사용하여 초기 볼트풀림 영역에서 볼트 체결력의 진단 가능성과 유효성을 제시하였다.

This paper presents a novel convolutional neural network (CNN)-based approach for predicting bolt clamping force in the early bolt loosening state of bolted structures. The approach entails tightening eight bolts with different clamping forces and generating frequency responses, which are then used to create a similarity map. This map quantifies the magnitude and shape similarity between the frequency responses and the initial model in a fully fastened state. Krylov subspace-based model order reduction is employed to efficiently handle the large amount of frequency response data. The CNN model incorporates a regression output layer to predict the clamping forces of the bolts. Its performance is evaluated by training the network by using various amounts of training data and convolutional layers. The input data for the model are derived from the magnitude and shape similarity map obtained from the frequency responses. The results demonstrate the diagnostic potential and effectiveness of the proposed approach in detecting early bolt loosening. Accurate bolt clamping force predictions in the early loosening state can thus be achieved by utilizing the frequency response data and CNN model. The findings afford valuable insights into the application of CNNs for assessing the integrity of bolted structures.

키워드

과제정보

이 논문은 2021학년도 안동대학교 해외파견연구보조금에 의하여 연구되었음.

참고문헌

  1. ANSYS (2022) ANSYS Mechanical APDL Theory Reference, Release 2022 R1, ANSYS. Inc.
  2. Han, J.S. (2010) Direct Design Sensitivity Analysis of Frequency Response Function Using Krylov Subspace Based Model Order Reduction, J. Comput. Struct. Eng. Inst. Korea, 23(2), pp.153~163.
  3. Han, J.S. (2012) Efficient Frequency Response and Its Direct Sensitivity Analyses for Large-Size Finite Element Systems Using Krylov Subspace-Based Model Order Reduction, J. Mech. Sci. & Technol., 26(4), pp.1115~1126. https://doi.org/10.1007/s12206-012-0227-8
  4. Jeong, K.H., Kim, W.S., Kim, N.R., Park. J.H. (2022) Machine Fault Diagnosis based on Vibration Analysis and Convolutional Neural Network, J. Korean Soc. Nondestruct. Test., 42(6), pp.496~502. https://doi.org/10.7779/JKSNT.2022.42.6.496
  5. Jiang, X., Zhu, Y., Hong, J., Chen, X., Zhang, Y. (2013) Investigation Into the Loosening Mechanism of Bolt in Curvic Coupling Subjected to Transverse Loading, Eng. Fail. Anal., 32, pp.360~373. https://doi.org/10.1016/j.engfailanal.2013.04.005
  6. Kang, M.S., Shin, K.H., Han, J.S. (2021) Characteristic Feature based on Frequency Response Analysis for Evaluating Bolt Loosening Location of Multiple-Bolt Gearbox, Trans. Korean Soc. Mech. Eng. A, 45(10), pp.843~852. https://doi.org/10.3795/KSME-A.2021.45.10.843
  7. Kim, D.H., Han, J.S. (2022) Frequency Response Similarity-Based Bolt Clamping Force Prediction Method Using Convolutional Neural Networks, J. Mech. Sci. & Technol., 36(8), pp.3801~3813. https://doi.org/10.1007/s12206-022-0703-8
  8. Liu, J., Ouyang, H., Feng, Z., Cai, Z., Liu, X., Zhu, M. (2017) Study on Self-Loosening of Bolted Joints Excited by Dynamic Sxial Load, Tribol. Int., 115,pp.432~451. https://doi.org/10.1016/j.triboint.2017.05.037
  9. Oman, S., Nagode, M. (2017) Bolted Connection of an End-Plate Cantilever Beam: The Distribution of Operating Force, J. Mech. Eng., 63(11), pp.617~627. https://doi.org/10.5545/sv-jme.2017.4638
  10. Park, J.H., Huynh, T.C., Choi, S.H., Kim, J.T. (2015) Vision-based Technique for Bolt-Loosening Detection in Wind Turbine Tower, Wind Struct, 21(6), pp.709~726. https://doi.org/10.12989/was.2015.21.6.709
  11. Seong, Y.U., Kim, W.S., Kim, J.S., Park. J.H. (2020) A Study on Non-destructive Measurement System for Axial Clamping Force of Bolting Structure Using Structural Wave, J. Korean Soc. Nondestruct. Test., 40(6), pp.413~418. https://doi.org/10.7779/JKSNT.2020.40.6.413
  12. Shin, K.H. (2016) An Alternative approach to Measure Similarity between Two Deterministic Transient Signals, J. Sound & Vibr., 371, pp.434~445. https://doi.org/10.1016/j.jsv.2016.02.037
  13. Sun, Y., Li, M., Dong, R., Chen, W., Jiang, D. (2022) Vision-based Detection of Bolt Loosening using YOLOv5, Sensors, 22(14), p.5184.
  14. The MathWorks, Inc. (2023) MATLAB Getting Started Guide, The MathWorks, Inc.
  15. Won, B.R., Han, J.S. (2014) Comparison of Projection-Based Model Order Reduction for Frequency Response, Trans. Korean Soc. Mech. Eng. A, 38(9), pp.933~941. https://doi.org/10.3795/KSME-A.2014.38.9.933
  16. Yang, M., Jeong, S.M., Lee, S.H., Lim, J.Y. (2022) Experimental Study of Structural Capability Loss of Bolt Clamping Force and Loosening Life of Bolts under Cyclic Loading Conditions, J. Korean Soc. Railw., 25(2), pp.89~95. https://doi.org/10.7782/JKSR.2022.25.2.89