본문 바로가기
KERD
논문KCI 등재

딥러닝 기반 기계음 분석을 통한 환경 소음에 강건한 베어링 샤프트 시스템의 실시간 고장 감지

Real-time Fault Detection in a Bearing-shaft System Through Deep Learning-Based Machine Sound Analysis Robust to Environmental Noises

오수정|이건창|백수정|이민희
한국산업경영시스템학회 (KOREAN SOCIETY OF INDUSTRIAL AND SYSTEMS ENGINEERING)|2024.12|한국산업경영시스템학회지 (Journal of Korean Society of Industrial and Systems Engineering)|제47권 제4호|pp. 132-144

0

조회

0

다운로드

0

인용

0

참고문헌

키워드

Fault DetectionConvolution Neural NetworkBearing-shaft SystemDenoising AutoencoderMachine Sound
주제분류산업공학

초록

Bearing-shaft systems are essential components in various automated manufacturing processes, primarily designed for the efficient rotation of a main shaft by a motor. Accurate fault detection is critical for operating manufacturing processes, yet challenges remain in sensor selection and optimization regarding types, locations, and positioning. Sound signals present a viable solution for fault detection, as microphones can capture mechanical sounds from remote locations and have been traditionally employed for monitoring machine health. However, recordings in real industrial environments always contain non-negligible ambient noise, which hampers effective fault detection. Utilizing a high-performance microphone for noise cancellation can be cost-prohibitive and impractical in actual manufacturing sites, therefore to address these challenges, we proposed a convolution neural network-based methodology for fault detection that analyzes the mechanical sounds generated from the bearing-shaft system in the form of Log-mel spectrograms. To mitigate the impact of environmental noise in recordings made with commercial microphones, we also developed a denoising autoencoder that operates without requiring any expert knowledge of the system. The proposed DAE-CNN model demonstrates high performance in fault detection regardless of whether environmental noise is included(98.1%) or not(100%). It indicates that the proposed methodology effectively preserves significant signal features while overcoming the negative influence of ambient noise present in the collected datasets in both fault detection and fault type classification.

Abstract

Bearing-shaft systems are essential components in various automated manufacturing processes, primarily designed for the efficient rotation of a main shaft by a motor. Accurate fault detection is critical for operating manufacturing processes, yet challenges remain in sensor selection and optimization regarding types, locations, and positioning. Sound signals present a viable solution for fault detection, as microphones can capture mechanical sounds from remote locations and have been traditionally employed for monitoring machine health. However, recordings in real industrial environments always contain non-negligible ambient noise, which hampers effective fault detection. Utilizing a high-performance microphone for noise cancellation can be cost-prohibitive and impractical in actual manufacturing sites, therefore to address these challenges, we proposed a convolution neural network-based methodology for fault detection that analyzes the mechanical sounds generated from the bearing-shaft system in the form of Log-mel spectrograms. To mitigate the impact of environmental noise in recordings made with commercial microphones, we also developed a denoising autoencoder that operates without requiring any expert knowledge of the system. The proposed DAE-CNN model demonstrates high performance in fault detection regardless of whether environmental noise is included(98.1%) or not(100%). It indicates that the proposed methodology effectively preserves significant signal features while overcoming the negative influence of ambient noise present in the collected datasets in both fault detection and fault type classification.

참고문헌 (32)

  1. "Dataset for AI Research"[URL]
  2. "Dataset for AI Research"[URL]
  3. Baek, S., "Supervised discretization of multivariate sensor data for fault detection in bearing shaft systems", Trans. Korean Soc. Mech. Eng. A, 44(1), pp.63~69 (2020)
  4. Baek, W., "In-process noise inspection system for product fault detection in a loud shop-floor environment", IEEE Transactions on Instrumentation &Measurement, 70, pp.3512111~ (2021)
  5. Chandra, N. H., "Fault detection in rotor bearing systems using time frequency techniques", Mechanical Systems and Signal Processing, 72-73, pp.105~133 (2016)
  6. Hou, L., "An inter-shaft bearing fault diagnosis datasets from an aero-engine system", Journal of Dynamics, Monitoring and Diagnostics, 2, pp.228~242 (2023)
  7. Im, J., "Optimal path planning for spatiotemporal space with grid search algorithm", Proceedings of the Korean Society for Aeronautical and Space Sciences Conference, pp.1065~1066 (2019)
  8. Jo, H., "A framework for diagnosis of abnormal noise and vibration of vehicle", Transaction of the Korean Society of Automotive Engineers, 26(4), pp.457~463 (2018)
  9. Jung, D., "Sensor selection for fault diagnosis in uncertain systems", International Journal of Control, 93(3), pp.629~639 (2020)
  10. Jung, H., "A machine learning approach for mechanical motor fault diagnosActual health statesis", Journal of Korean Society of Industrial and Systems Engineering, 40(1), pp.57~64 (2017)
  11. Kang, D.H., "Machine learning-based fault diagnosis of high-speed rotating machinery supporting by tilting pad bearings", Master, pp.2017
  12. Kang, D. H., "Fault diagnosis of high-speed rotating machinery considering metal temperature of oil lubricated berarings", Proceedings of 2017 Joint Conference by KSNVE, ASK, and KSME, pp.217~ (2017)
  13. Kang, K. -W., "CNN-based automatic machine fault diagnosis method using spectrogram images", The Journal of Korea Institute of Convergence Signal Processing, 21(3), pp.121~126 (2020)
  14. Kim, B. S., "Real-time fault detection system of a pneumatic cylinder via deep-learning model considering time-variant characteristic of sensor data", Journal of Korean Society of Industrial and Systems Engineering, 47(2), pp.10~20 (2024)
  15. Kim, G. -D., "Group action recognition through grid search and transformer", Proceedings of the Korean Society of Computer Information Conference, pp.513~515 (2023)
  16. Kim, H. C., "Comparison of Hyper-Parameter Optimization Methods for Deep Neural Networks", Journal of IKEEE, 24(4), pp.49~54 (2020)
  17. Kim, S. H., "The fault prognostic system for the logistics rotational equipment", Journal of Korean Society of Industrial and Systems Engineering, 46(2), pp.168~175 (2023)
  18. Kulkarni, A., "Sensor selection framework for designing fault diagnostics system", Sensors, 21(19), pp.6470~ (2021)
  19. Lee, J. -H., "Failure diagnostics of camera image sensor for vehicle using CNN", Journal of the Korea Academia-Industrial cooperation Society, 23(12), pp.877~884 (2022)
  20. Lee, M., "Deep Learning Based Fault Prediction of Bearing-Shaft System Using Multivariate Sensor Signals", Proceedings of IEEM2024 (2024)
  21. O, G. -U., "Bearing fault diagnosis in induction motor", Proceedings of 2023 Fall Conference of the Korean Institute of Electrical Engineers, pp.158~159 (2023)
  22. Park, J. H., "Analysis of accuracy and computation complexity of bearing fault diagnosis methods using CNN-based deep learning", The Journal of Korean Institute of Next Generation Computing, 18(1), pp.7~18 (2022)
  23. Park, J. H., "Computation reduction method for CNN-based bearing fault diagnosis", Journal of Digital Contents Society, 24(6), pp.1373~1379 (2023)
  24. Rowaihy, H., "A survey of sensor selection schemes in wireless sensor network", Unattended Ground, Sea, and Air Sensor Technologies and Applications IX, 6562, pp.1~13 (2007)
  25. Saha, D. K., "Development of intelligent fault diagnosis technique of rotary machine element bearing : A machine learning approach", Sensors, 22(3), pp.1~15 (2022)
  26. Seong, J., "A novel denoising autoencoder based acoustic communication system design with audio frequency noise, hardware impairment", Proceedings of the Korean Institute of Communication Sciences Conference, pp.90~91 (2022)
  27. Tran, T., "Drill fault diagnosis based on the scalogram and mel spectrogram of sound signals using artificial intelligence", IEEE Access, 8, pp.203655~203666 (2020)
  28. Vincent, P., "Extracting and composing robust features with denoising autoencoders", Proceedings of the 25th International Conference on Machine Learning, pp.1096~1103 (2008)
  29. Wen, B., "Effect of ball bearing misalignment on dynamic characteristics of rotor system", IOP Conf. Series : Materials Science and Engineering, 1081, pp.1~10 (2021)
  30. Won, J., "A survey on hyperparameter optimization in machine learning", The Journal of Korean Institute of Communications and Information Sciences, 48(6), pp.733~747 (2023)
  31. Yang, S., "Improving False Positive Rate of Extended Learned Bloom Filters Using Grid Search", Journal of KIISE, 49(1), pp.78~88 (2022)
  32. Yeom, D. -J., "Remote fault detection in conveyor system using drone based on audio FFT analysis", Journal of Convergence for Information Technology, 9(10), pp.101~107 (2019)

관련 논문AI추천

An approach for anomaly detection of motor rotating shaft alignment using deep learning김재훈, 박철순한국마린엔지니어링학회 · 한국마린엔지니어링학회지2025 · 제49권 제6호MFCCs를 이용한 입력 변환과 CNN 학습에 기반한운영 환경 변화에 강건한 베어링 결함 진단 방법서양진한국정보처리학회 · 정보처리학회 논문지2022 · 제11권 제4호CNN 기반 딥러닝을 이용한 베어링 고장 진단의 정확도 및 계산 복잡도 분석김철홍, 박재현한국차세대컴퓨팅학회 · 한국차세대컴퓨팅학회 논문지2022 · 제18권 제1호오토인코더를 이용한 기어박스 진동신호의잡음 제거 방법에 관한 연구김지혁, 이제동한국IT정책경영학회 · 한국IT정책경영학회 논문지2020 · 제12권 제5호시계열 데이터에 대한 전이 학습 기반 이상 탐지 방법 연구김재훈, 박철순한국산업경영시스템학회 · 한국산업경영시스템학회지2025 · 제48권 제4호스펙트로그램 이미지를 이용한 CNN 기반 자동화 기계 고장 진단 기법강경원, 이경민한국융합신호처리학회 · 융합신호처리학회 논문지2020 · 제21권 제3호

이 간행물의 다른 논문 (20건)

다단계 베르누이 라인을 위한 성능 모델: 다품종 제품과 WIP 및 순환 정책박경수pp. 1-11다중구역 제조시스템을 위한 에이전트 기반 디스패칭 시스템신문수, 김민정pp. 12-21가치 사슬에서 스마트 IT 시스템 활용이 제조 중소기업의 경쟁력에 미치는 영향이준희, 조인수pp. 22-28BAGAN-GP와 오버샘플링 기법들을 이용한 혼합 오버샘플링이동주pp. 29-38부분 버퍼 공유 정책을 가지는 공간-시간 우선순위 대기행렬김길환pp. 39-55
원문 보기