Journal of the Society of Naval Architects of Korea (대한조선학회논문집)

The Society of Naval Architects of Korea (대한조선학회)
권호 표지
DOI QR코드

DOI QR Code

The Study of Failure Mode Data Development and Feature Parameter's Reliability Verification Using LSTM Algorithm for 2-Stroke Low Speed Engine for Ship's Propulsion

선박 추진용 2행정 저속엔진의 고장모드 데이터 개발 및 LSTM 알고리즘을 활용한 특성인자 신뢰성 검증연구

  • Jae-Cheul Park (Digitalization Team, Digital Technology Center, R&D Division, Korean Register) ;
  • Hyuk-Chan Kwon (Digitalization Team, Digital Technology Center, R&D Division, Korean Register) ;
  • Chul-Hwan Kim (NEXT Engineering) ;
  • Hwa-Sup Jang (Digitalization Team, Digital Technology Center, R&D Division, Korean Register)
  • 박재철 ((사)한국선급 연구본부 디지털기술원 디지털라이제이션팀) ;
  • 권혁찬 ((사)한국선급 연구본부 디지털기술원 디지털라이제이션팀) ;
  • 김철환 (넥스트엔지니어링) ;
  • 장화섭 ((사)한국선급 연구본부 디지털기술원 디지털라이제이션팀)
  • Received : 2022.10.31
  • Accepted : 2023.03.07
  • Published : 2023.04.20
Download PDF
⟨ Previous Next ⟩

Abstract

In the 4th industrial revolution, changes in the technological paradigm have had a direct impact on the maintenance system of ships. The 2-stroke low speed engine system integrates with the core equipment required for propulsive power. The Condition Based Management (CBM) is defined as a technology that predictive maintenance methods in existing calender-based or running time based maintenance systems by monitoring the condition of machinery and diagnosis/prognosis failures. In this study, we have established a framework for CBM technology development on our own, and are engaged in engineering-based failure analysis, data development and management, data feature analysis and pre-processing, and verified the reliability of failure mode DB using LSTM algorithms. We developed various simulated failure mode scenarios for 2-stroke low speed engine and researched to produce data on onshore basis test_beds. The analysis and pre-processing of normal and abnormal status data acquired through failure mode simulation experiment used various Exploratory Data Analysis (EDA) techniques to feature extract not only data on the performance and efficiency of 2-stroke low speed engine but also key feature data using multivariate statistical analysis. In addition, by developing an LSTM classification algorithm, we tried to verify the reliability of various failure mode data with time-series characteristics.

Keywords

Acknowledgement

본 연구는 2020년도 산업통산자원부 및 산업기술평가관리원 연구비 지원으로 수행된 '자율운항선박 기술개발사업(20011164, 자율운항선박 핵심 기관시스템 상태 모니터링 및 고장예측 진단기술 개발)'의 연구결과입니다.

References

  1. Bae, J.H, 2021. Fault detection of aircraft turbofan engine system using a fault detection filter. Institute of Korean Electrical and Electronics Engineers, 25(2), pp.330-336.
  2. Benkamin, L., Timo, M., Hannes, V., Nasser, J. and Michael, W., 2021. A survey on long short-term memory networks for time series prediction, 14th CIRP 99, pp.650-655.
  3. Byun, S.I. and Lee, D.I., 2022. Health monitoring for autonomous underwater vehicles using fault tree analysis. Journal of Institute of Control, Robotics and Systems, 28(5), pp. 398-405. https://doi.org/10.5302/J.ICROS.2022.22.0021
  4. Chae, S.G., Kim, G.R., Bae, B.Y. and Bae, S.J., 2021. Failure diagnosis and prediction for a thermal power plant generator using fastICA. Journal of Applied Reliability, 21(4), pp.341-351. https://doi.org/10.33162/JAR.2021.12.21.4.341
  5. Cho, H.J., Choi, H.S., Kim, H.J., Nam, K.S., Ryu, J.D. and Ha, K. N., 2022. Feature selection for unmanned surface vehicle fault diagnosis research and experimental verification. Journal of Institute of Control, Robotics and Systems, 28(6), pp.542-550. https://doi.org/10.5302/J.ICROS.2022.22.0046
  6. Hwang, S.Y., Heo, J.Y., Hong, K.T. and Lee, J.H., 2018. Time series data analysis and fault diagnosis of plant process equipment using statistical machine learning method. Korean Journal of Computational Design and Engineering, 23(3), pp. 193-201. https://doi.org/10.7315/CDE.2018.193
  7. Kim, S.H, 2022. A study of machine learning technique for noise-based engine fault diagnosis. Journal of the KNST, 5(1), pp.16-19. https://doi.org/10.31818/JKNST.2022.03.5.1.16
  8. Kim, J.Y., Lee, T.H., Lee, S.H., Lee, J.J., Shin, D.M., Lee, W.K. and Kim, Y.J., 2022. A study on the development of a failure simulation database for condition based maintenance of marine engine system auxiliary equipment. Journal of the Society of Naval Architects of Korea, 59(4), pp.200-206. https://doi.org/10.3744/SNAK.2022.59.4.200
  9. Kim, S.I., Noh, Y.J., Kang, Y.J., Park, S.H. and Ahn, B.H., 2021. Fault classification model based on time domain feature extraction vibration data. Journal of the Computational Structural Engineering Institute of Korea, 34(I), pp.25-33. https://doi.org/10.7734/COSEIK.2021.34.1.25
  10. Lee, J.H., 2021. Experimental study on application of an anomaly detection algorithm electric current datasets generated from marine air compressor with time-series features. Journal of the Korean Society of Marine Environment & Safety, 27(1), pp.127-134. https://doi.org/10.7837/kosomes.2021.27.1.127
  11. Ugo, C., Carlo, C. and Raphael, Z., 2018. Marine gas turbine monitoring and diagnostics by simulation and pattern recognition. International Journal of Naval Architecture and Ocean Engineering, 10, pp.617-628. https://doi.org/10.1016/j.ijnaoe.2017.09.012
  12. Zhengping, C., Sanjay, P., Cho, K.H., David, S. and Yan, L., 2018. Recurrent neural networks for multivariate time series with missing values, Scientific Reports, 8:6085.