The Journal of Bigdata (한국빅데이터학회지)

Korea Bigdata Society (한국빅데이터학회)
권호 표지
DOI QR코드

DOI QR Code

A Study on the Design of Supervised and Unsupervised Learning Models for Fault and Anomaly Detection in Manufacturing Facilities

제조 설비 이상탐지를 위한 지도학습 및 비지도학습 모델 설계에 관한 연구

  • 오민지 (충북대학교 대학원 빅데이터학과) ;
  • 최은선 (충북대학교 대학원 빅데이터학과) ;
  • 노경우 (충북대학교 경영정보학과) ;
  • 김재성 (충북대학교 대학원 빅데이터학과) ;
  • 조완섭 (충북대학교 경영정보학과)
  • Received : 2021.07.02
  • Accepted : 2021.08.23
  • Published : 2021.08.31
Download PDF
⟨ Previous Next ⟩

Abstract

In the era of the 4th industrial revolution, smart factories have received great attention, where production and manufacturing technology and ICT converge. With the development of IoT technology and big data, automation of production systems has become possible. In the advanced manufacturing industry, production systems are subject to unscheduled performance degradation and downtime, and there is a demand to reduce safety risks by detecting and reparing potential errors as soon as possible. This study designs a model based on supervised and unsupervised learning for detecting anomalies. The accuracy of XGBoost, LightGBM, and CNN models was compared as a supervised learning analysis method. Through the evaluation index based on the confusion matrix, it was confirmed that LightGBM is most predictive (97%). In addition, as an unsupervised learning analysis method, MD, AE, and LSTM-AE models were constructed. Comparing three unsupervised learning analysis methods, the LSTM-AE model detected 75% of anomalies and showed the best performance. This study aims to contribute to the advancement of the smart factory by combining supervised and unsupervised learning techniques to accurately diagnose equipment failures and predict when abnormal situations occur, thereby laying the foundation for preemptive responses to abnormal situations. do.

제4차 산업혁명 선언 이후 생산 제조 기술과 정보통신기술(ICT)이 융합된 스마트 팩토리가 큰 주목을 받고 사물인터넷(IoT) 기술 및 빅데이터 기술 등이 발전하면서 생산 시스템의 자동화가 가능해졌다. 고도화된 제조 산업에서 생산 시스템에는 예정되지 않은 성능 저하 및 가동 중지 발생 가능성이 존재하며, 가능한 한 빨리 잠재적인 오류를 감지하여 이를 복구해 안전 위험을 줄여나가야 한다는 요구가 있다. 본 연구는 유압 시스템에 부착된 다중 센서 데이터를 기반으로 장비의 고장 예측과 이상 발생 시점 예측을 결합하여 제조 설비 이상탐지를 위한 지도학습 및 비지도학습 모델을 설계한다. 지도학습 분석 방법으로 XGBoost, LightGBM, CNN 모델의 정확도를 비교하였다. 혼동행렬 기반의 평가지표를 통해 LightGBM의 예측력이 97%로 가장 우수한 것을 확인하였다. 또한 비지도 학습 분석 방법으로 MD, AE, LSTM-AE 모델을 구축하여 각 모델을 비교 분석한 결과 LSTM-AE 모델이 이상패턴을 75% 감지하여 가장 우수한 성능을 보였다. 본 연구는 지도학습과 비지도학습 기법을 종합하여 설비의 고장여부를 정확하게 진단하고 이상상황이 발생하는 시점을 예측함으로써 이상상황에 대한 선제대응을 할 수 있는 기반을 마련하여 스마트 팩토리 고도화에 기여하고자 한다.

Keywords

Acknowledgement

이 연구는 산업 통상 자원부 (MOTIE)와 한국산업기술진흥원 (KIAT)에서 국제 협력 R & D 프로그램(프로젝트 ID : P0011880)을 통해 재정적으로 지원되었습니다.

References

  1. Schwab, K. (2017). The fourth industrial revolution. Currency.
  2. Lee Sang Hyun, Jang Yoon Jong, Kim Sang Hun,. (2018). Germany's Industry 4.0 Strategy Spreading Development Trend and Policy Implications (2018-327). Korea Institute for Industrial Economics & Trade.
  3. Chang Youl Choi. (2017). The Fourth Industrial Revolution and Development Strategies and Implications of e-Business Companies. The e-Business Studies, 18(3), 39-54. https://doi.org/10.15719/GEBA.18.3.201706.39
  4. Seunghee Lee, Jun Geol Baek. (2021). Manufacturing Process Anomaly Detection Using Adversarial Autoencoder with Multiple Discriminator. Journal of the Korea Institute of Industrial Engineers, 47(2), 217-223. https://doi.org/10.7232/JKIIE.2021.47.2.217
  5. Gaddam, A., Wilkin, T., Angelova, M., & Gaddam, J. (2020). Detecting sensor faults, anomalies and outliers in the internet of things: A survey on the challenges and solutions. Electronics, 9(3), 511. https://doi.org/10.3390/electronics9030511
  6. Sehyug Kwon. (2020). Anomaly Detection of Big Time Series Data Using Machine Learning. Journal of Society of Korea Industrial and Systems Engineering, 43(2), 33-38. https://doi.org/10.11627/jkise.2020.43.2.033
  7. Mihui Kim, Gihun Lee. (2020). Anomaly Data Detection Using Machine Learning in Crowdsensing System. Journal of IKEEE, 24(2), 104-114.
  8. Geonkyo Hong, Jeonghoon Choi, Dongjun Suh. (2020). A Study on the Design of Time Series Data-based Deep Learning Model for Detecting Machine Abnormalities. Proceeding of Symposium of the Korean Institute of communications and Information Sciences, 887-888.
  9. Jooyeon Lee, Ki Yong Lee. (2021). An Anomalous Sequence Detection Method Based on An Extended LSTM Autoencoder. The Journal of Society for e-Business Studies, 26(1), 127-140. https://doi.org/10.7838/JSEBS.2021.26.1.127
  10. Jooyeon Lee, Ki Yong Lee. (2021). An Effective Detection Method of Anomalous Sequences Considering the Occurrence Order and Time Interval of the Elements. Journal of KIISE, 48(4), 469-478. https://doi.org/10.5626/JOK.2021.48.4.469
  11. Jin, S., Zhang, Z., Chen, G., Chakrabarty, K., & Gu, X. (2016). Anomaly-Detection-Based Failure Prediction in a Core Router System. In Proc. International Conference on Advances in System Testing and Validation Lifecycle (VALID).
  12. Tan, P. N., Steinbach, M., & Kumar, V. (2005). Introduction to data mining.1st.
  13. Chen, T., & Guestrin, C. (2016, August). Xgboost: A scalable tree boosting system. In Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining (pp. 785-794).
  14. Guolin Ke, Qi Meng, Thomas Finley, Taifeng Wang, Wei Chen, Weidong Ma, Qiwei Ye, Tie-Yan Liu. (2017). Lightgbm: A highly efficient gradient boosting decision tree. Advances in neural information processing systems, 30, 3146-3154.
  15. Cho Jaeyoung, Joo Jihwan, Han Ingoo. (2021). The Prediction of Export Credit Guarantee Accident using Machine Learning. Journal of Intelligence and Information Systems, 20(1), 83-102.
  16. Wonwoo Nam, Byung Wook Kim. (2020). Deep Learning based Depression Classification using Environmental Factor Selection. The transaction of The Korean Institute of Electrical Engineers, 69(7), 1102-1110. https://doi.org/10.5370/kiee.2020.69.7.1102
  17. Seong Chul Lee, Soo Yong Lee. (2020). Is Deep Learning Useful for Noise and Vibration Fields?. Noise.Vibration, 30(1), 7-13.
  18. Seunghwan Jung, Sungshin Kim. (2021). Fault Detection Method for Multivariate Process using Mahalanobis Distance and ICA. Journal of Korea Institute of Information, Electronics, and Communication Technology, 14(1), 22-28. https://doi.org/10.17661/JKIIECT.2021.14.1.22
  19. Kyuhan Kim, Heemyung Jeong, Heungseok Lee, Hwaseok Lee, Hyungsu Kim, June Ho Park. (2020). Fault Detection Using Signal Reconstruction Model Based on Autoencoder in Thermal Power Plant. The transactions of The Korean Institute of Electrical Engineers, 69(6), 800-807. https://doi.org/10.5370/kiee.2020.69.6.800
  20. A. Geron. (2018). Hands-On Machine Learning with Scikit-Learn & TensorFlow. Hanbit Media.
  21. Ghrib, Z., Jaziri, R., and Romdhane, R. (2020). Hybrid approach for Anomaly Detection in Time Series Data. International Joint Conference on Neural Networks(IJCNN). 1-7.