Journal of the Korean Society of Fisheries and Ocean Technology (수산해양기술연구)

The Korean Society of Fisheries and Ocean Technology (한국수산해양기술학회)
권호 표지
DOI QR코드

DOI QR Code

Outlier detection of main engine data of a ship using ensemble method

앙상블 기법을 이용한 선박 메인엔진 빅데이터의 이상치 탐지

  • KIM, Dong-Hyun (Autonomous Ship Technology Center, Korea Marine Equipment Research Institute) ;
  • LEE, Ji-Hwan (Division of Systems Management and Engineering, Pukyong National University) ;
  • LEE, Sang-Bong (LAB021) ;
  • JUNG, Bong-Kyu (Marine Industry Research Center, Gyeongsang National University)
  • 김동현 (한국조선해양기자재연구원) ;
  • 이지환 (부경대학교 시스템경영공학부) ;
  • 이상봉 (랩오투원) ;
  • 정봉규 (경상대학교 해양산업연구소)
  • Received : 2020.08.10
  • Accepted : 2020.11.09
  • Published : 2020.11.30
Download PDF
⟨ Previous Next ⟩

Abstract

This paper proposes an outlier detection model based on machine learning that can diagnose the presence or absence of major engine parts through unsupervised learning analysis of main engine big data of a ship. Engine big data of the ship was collected for more than seven months, and expert knowledge and correlation analysis were performed to select features that are closely related to the operation of the main engine. For unsupervised learning analysis, ensemble model wherein many predictive models are strategically combined to increase the model performance, is used for anomaly detection. As a result, the proposed model successfully detected the anomalous engine status from the normal status. To validate our approach, clustering analysis was conducted to find out the different patterns of anomalies the anomalous point. By examining distribution of each cluster, we could successfully find the patterns of anomalies.

Keywords

References

  1. Aggarwal, C.C. 2013. Outlier ensembles: position paper. ACM SIGKDD Explorations Newsletter, 2013, 14, 49-58. https://doi.org/10.1145/2481244.2481252.
  2. Aldous L, Smith T, Bucknall R and Thompson P. 2015. Uncertainty analysis in ship performance monitoring. Ocean Engineering 110, 29-38. https://doi.org/10.1016/j.oceaneng.2015.05.043.
  3. Bae YM, Kim MJ, Kim KJ, Jun CH, Byeon SS and Park KM. 2018. A case study on the establishment of upper control limit to detect vessel's main engine failures using multivariate control chart. Journal of the Society of Naval Architects of Korea 55, 505-513. https://doi.org/10.3744/SNAK.2018.55.6.505.
  4. Deris S, Omatu S, Ohta H, Kutar LCS and Abd Samat P. 1999. Ship maintenance scheduling by genetic algorithm and constraint-based reasoning. European Journal of Operational Research 112, 489-502. https://doi.org/10.1016/S0377-2217(97)00399-8.
  5. Kim KY, Kim YI and Yu YH. 2011. The Fault Diagnosis of Marine Diesel Engines Using Correlation Coefficient for Fault Detection. The Journal of Advanced Navigation Technology 15, 18-24. https://doi.org/10.12673/jant.2011.15.1.018.
  6. Lazakis I, Turan O and Aksu S. 2010. Increasing ship operational reliability through the implementation of a holistic maintenance management strategy. Ships and Offshore Structures, 5, 337-357. https://doi.org/10.1080/17445302.2010.480899.
  7. Lundberg SM and Lee SI. 2017. A unified approach to interpreting model predictions. In Proceedings of advances in neural information processing systems 2017, 4765-4774.
  8. Mak L, Sullivan M, Kuczora A and Millan J. 2014. Ship performance monitoring and analysis to improve fuel efficiency. In Proceedings of 2014 Oceans St. John's 1-10. https://doi.org/10.1109/OCEANS.2014.7003300.
  9. Perera LP and Mo B. 2017. Marine engine-centered data analytics for ship performance monitoring. Journal of Offshore Mechanics and Arctic Engineering 139, 1-28. https://doi.org/10.1115/1.4034923.
  10. Phaladiganon P, Kim SB, Chen VC, Baek JG and Park SK. 2011. Bootstrap-based T2 multivariate control charts. Communications in Statistics - Simulation and Computation 40, 645-662. https://doi.org/10.1080/03610918.2010.549989.
  11. Soner O, Akyuz E and Celik M. 2018. Use of tree based methods in ship performance monitoring under operating conditions. Ocean Engineering 166, 302-310. https://doi.org/10.1016/j.oceaneng.2018.07.061