산업경영시스템학회지 (Journal of Korean Society of Industrial and Systems Engineering)

  • 제47권2호
  • /
  • Pages.1-9
  • /
  • 2024
  • /
  • 2005-0461(pISSN)
  • /
  • 2287-7975(eISSN)
한국산업경영시스템학회 (Society of Korea Industrial and System Engineering)
권호 표지
DOI QR코드

DOI QR Code

진동 아날로그 신호 기반의 이상상황 탐지를 위한 기계학습 모형의 성능지표 향상

Improving the Performance of Machine Learning Models for Anomaly Detection based on Vibration Analog Signals

  • 김재훈 (국립창원대학교 산업시스템공학과) ;
  • 엄상천 (부산대학교 산업공학과) ;
  • 박철순 (국립창원대학교 산업시스템공학과)
  • Jaehun Kim (Department of Industrial & Systems Engineering, Changwon National University) ;
  • Sangcheon Eom (Department of Industrial Engineering, Pusan National University) ;
  • Chulsoon Park (Department of Industrial & Systems Engineering, Changwon National University)
  • 투고 : 2024.03.08
  • 심사 : 2024.04.17
  • 발행 : 2024.06.30
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

New motor development requires high-speed load testing using dynamo equipment to calculate the efficiency of the motor. Abnormal noise and vibration may occur in the test equipment rotating at high speed due to misalignment of the connecting shaft or looseness of the fixation, which may lead to safety accidents. In this study, three single-axis vibration sensors for X, Y, and Z axes were attached on the surface of the test motor to measure the vibration value of vibration. Analog data collected from these sensors was used in classification models for anomaly detection. Since the classification accuracy was around only 93%, commonly used hyperparameter optimization techniques such as Grid search, Random search, and Bayesian Optimization were applied to increase accuracy. In addition, Response Surface Method based on Design of Experiment was also used for hyperparameter optimization. However, it was found that there were limits to improving accuracy with these methods. The reason is that the sampling data from an analog signal does not reflect the patterns hidden in the signal. Therefore, in order to find pattern information of the sampling data, we obtained descriptive statistics such as mean, variance, skewness, kurtosis, and percentiles of the analog data, and applied them to the classification models. Classification models using descriptive statistics showed excellent performance improvement. The developed model can be used as a monitoring system that detects abnormal conditions of the motor test.

키워드

과제정보

This research was supported by Changwon National University in 2023~2024.

참고문헌

  1. Choi, Y.U., Yoon, D.U., Choi, J.H., and Byun, J.M., Hyperparameter Search for Facies Classification with Bayesian Optimization, Geophysics and Geophysical Exploration, 2020, Vol. 23, No. 3, pp. 157-167.
  2. Gustavo, A., Lujan-Moreno, P.R., Howard, O.G., and Rojas, D.C., Montgomery, Design of experiments and response surface methodology to tune machine learning hyperparameters, with a random forest case-study, Expert Systems with Applications, 2018, Vol. 109, pp. 195-205. https://doi.org/10.1016/j.eswa.2018.05.024
  3. Han, J.H., Choi, D.J., Hong, S.K., and Kim, H.S., Motor Fault Diagnosis Using CNN Based Deep Learning Algorithm Considering Motor Rotating Speed, 2019 IEEE 6th International Conference on Industrial Engineering and Applications (ICIEA), Tokyo, Japan, 2019, pp. 440-445.
  4. Hwang, H.E., Cho, Y.S., Hwang, S.C., and Kim, S.B., Optimal Tire Design Using Machine Learning and Bayesian Optimization, Journal of the Korean Institute of Industrial Engineers, 2022, Vol. 48, No. 4, pp. 433-440. https://doi.org/10.7232/JKIIE.2022.48.4.433
  5. Hwang, H.S., Hybrid Machine Learning Model for Predicting the Direction of KOSPI Securities, Journal of the Korea Convergence Society, 2021, Vol. 12, No. 6, pp. 9-16. https://doi.org/10.15207/JKCS.2021.12.6.009
  6. Iqbal, S., Qureshi, A.N., Khursheed, K., Alhussein, M., Haider, S.I., and Rida, I., AMIAC: adaptive medical image analyzes and classification, a robust self-learning framework, Neural Comput & Applic, 2023. https://doi.org/10.1007/s00521-023-09209-1.
  7. Jung, H. and Kim, J.W.. A Machine Learning Approach for Mechanical Motor Fault Diagnosis, Journal of Society of Korea Industrial and Systems Engineering, Mar. 31, 2017.
  8. Kim, S.W. and Lee, S.M., Fault Detection Technique for PVDF Sensor Based on Support Vector Machine, The Journal of the Korea Institute of Electronic Communication Sciences, 2023, Vol. 18, No. 5, pp. 785-796. https://doi.org/10.13067/JKIECS.2023.18.5.785
  9. Kwon, K.B., Choi, H.S., Oh, J.Y., and Kim, D.K., A Study on EPB Shield TBM Face Pressure Prediction Using Machine Learning Algorithms, Journal of Korean Tunnelling and Underground Space Association, 2022, Vol. 24, No. 2, pp. 217-230. https://doi.org/10.9711/KTAJ.2022.24.2.217
  10. Lee, J.E., Kim, Y.B., and Kim, J.N., Hyperparameter Optimization for Image Classification in Convolutional Neural Network, Journal of the Institute of Convergence Signal Processing, 2020, Vol. 21, No. 3, pp. 148-153.
  11. Lee, S.H., Ko, S.K., and Lee, S.A., Fault Classification Model Based on Deep Learning Using Vibration Data of Mechanical Equipment, The Journal of Korean Institute of Next Generation Computing, 2022, Vol. 18, No. 2, pp. 36-46.
  12. Lee, T.C., Cho, H.J., and Choi. J.S., A Study on Single Motor failure Diagnosis and Fault Tolerant Control by Using Open Source and Machine Learning, Journal of The Korean Society for Aeronautical & Space Sciences, Nov. 2022, pp. 810-812.
  13. Lee, Y.K., Hong, J.K., and Hong, S.C., A study on the anomaly prediction system of drone using big data, Journal of Internet Computing and Services (JICS), Vol. 21, No. 2, Apr. 2020, pp. 27-37.
  14. Lee, Y.S., Yoo, J.H., and Kim, B.H., Multi-grid Application and Optimal Hyper-parameter Selection in AI Urban Flood Prediction, Korean Society of Civil Engineers, 2022, Vol. 2022, No. 10, pp. 301-302.
  15. Park, M.J., Lee, Y.S., and Kim, J.Y., Comparing Accuracy of Predicting PM2.5 Concentration according to Hyperparameter Optimization Methods of Long Short-Term Memory (LSTM) Model, Proceeding of Korean Society for Atmospheric Environment, 2021, Vol. 2021 No. 10, pp. 371-371.
  16. Snoek, J., Larochelle, H., and Adams, R.P., Practical bayesian optimization of machinelearning algorithms, Proceeding of NeuralInformation Processing Systems, 2012, pp. 2951-2959.
  17. Vermesan, O., Coppola, M., Bahr, R., Bellmann, R.O., Martinsen, J.E., Kristoffersen, A., Hjertaker, T., Breiland, J., Andersen, K., Sand, H.E., and Lindberg, D., An Intelligent Real-Time Edge Processing Maintenance System for Industrial Manufacturing, Control, and Diagnostic, Frontiers in Chemical Engineering, 2022.
  18. Vulpe-Grigorasi, A. and Grigore, O., Convolutional Neural Network Hyperparameters optimization for Facial Emotion Recognition, 2021, 12th International Symposium on Advanced Topics in Electrical Engineering (ATEE), Bucharest, Romania, pp. 1-5.
  19. Won, J.J., Shin, J.M., Kim, J.H., and Lee, J.W., A Survey on Hyperparameter Optimization in Machine Learning, The Journal of Korean Institute of Communications and Information Sciences, 2023, Vol. 48, No. 6, pp. 733-747. https://doi.org/10.7840/kics.2023.48.6.733
  20. Yang, L. and Abdallah, S., On hyperparameter optimization of machine learning algorithms: Theory and practice, Neurocomputing, 2020, Vol. 415, pp. 295-316. https://doi.org/10.1016/j.neucom.2020.07.061
  21. You, H.J. and Kim, C.H., Machine Learning Modeling and Hyper-Parameter Optimization for Weld Nugget Formation and Failure Behavior of Resistance Spot Welds, Journal of Welding and Joining, 2021, Vol. 39, No. 6, pp. 658-665. https://doi.org/10.5781/JWJ.2021.39.6.11
  22. Yuk, D.G. and Sohn, J.W., Motor Anomaly Detection Using Deep Learning, Korean Society of Mechanical Engineers, Nov. 2022, pp. 421-422.
  23. Yu, S.H., A Study on the Method of Diagnosing Failure and Detecting Failure Signals of the Motor and Inverter of Electric Vehicle Using Machine Learning, Graduate School of Soonchunhyang University, Master's thesis, 2023.