DOI QR코드

DOI QR Code

Reduced Order Modeling of Marine Engine Status by Principal Component Analysis

주성분 분석을 통한 선박 기관 상태의 차수 축소 모델링

  • Seungbeom Lee (Department of Naval Architecture and Ocean Engineering, Chungnam National University) ;
  • Jeonghwa Seo (Department of Naval Architecture and Ocean Engineering, Chungnam National University) ;
  • Dong-Hwan Kim (Institute of Advanced Transportation Vehicles, Chungnam National University) ;
  • Sangmin Han (Autonomous Ship Research Center, Samsung Heavy Industries Co., LTD.) ;
  • Kwanwoo Kim (Autonomous Ship Research Center, Samsung Heavy Industries Co., LTD.) ;
  • Sungwook Chung (Autonomous Ship Research Center, Samsung Heavy Industries Co., LTD.) ;
  • Byeongwoo Yoo (Department of Autonomous Vehicle System Engineering, Chungnam National University)
  • 이승범 (충남대학교 선박해양공학과) ;
  • 서정화 (충남대학교 선박해양공학과) ;
  • 김동환 (충남대학교 첨단수송체연구소) ;
  • 한상민 ((주)삼성중공업 조선해양연구소 자율운항연구센터) ;
  • 김관우 ((주)삼성중공업 조선해양연구소 자율운항연구센터) ;
  • 정성욱 ((주)삼성중공업 조선해양연구소 자율운항연구센터) ;
  • 유병우 (충남대학교 자율운항시스템공학과)
  • Received : 2023.10.04
  • Accepted : 2023.12.08
  • Published : 2024.02.20

Abstract

The present study concerns reduced order modeling of a marine diesel engine, which can be used for outlier detection in status monitoring and carbon intensity index calculation. Principal Component Analysis (PCA) is introduced for the reduced order modeling, focusing on the feasibility of detecting and treating nonlinear variables. By cross-correlation, it is found that there are seven non-linear data channels among 23 data channels, i.e., fuel mode, exhaust gas temperature after the turbocharger, and cylinder coolant temperatures. The dataset is handled so that the mean is located at the nominal continuous rating. Polynomial presentation of the dataset is also applied to reflect the linearity between the engine speed and other channels. The first principal mode shows strong effects of linearity of the most data channels to show the linearity of the system. The non-linear variables are effectively explained by other modes. second mode concerns the temperature of the cylinder cooling water, which shows small correlation with other variables. The third and fourth modes correlates the fuel mode and turbocharger exhaust gas temperature, which have inferior linearity to other channels. PCA is proven to be applicable to data given in binary type of fuel mode selection, as well as numerical type data.

Keywords

Acknowledgement

본 연구는 2023년 삼성중공업의 재원으로 충남대학교에서 수행된 '자율운항시스템 핵심 기술 개발'과제 및 22023년도 정부(산업통상자원부)의 재원으로 한국산업기술진흥원의 지원을 받아 수행된 연구임(P0017006, 2023년 산업혁신인재성장지원사업).

References

  1. Alexandris, N., Gupta, S. and Koutsias, N., 2017. Remote sensing of burned areas via PCA, Part 1; centering, scaling and EVD vs SVD. Open Geospatial Data, Software and Standards, 2, Article No. 17.
  2. Bishop, C.M., 2006. Pattern Recognition and Machine Learning. Springer-Verlag.
  3. Kim, S.-H., Lee, Y.-S. and Han, Y.-J., 2003. A study on the design of fault diagnostic system based on PCA. Journal of Korean Institute of Intelligent Systems, 13(5), pp.600-605. https://doi.org/10.5391/JKIIS.2003.13.5.600
  4. Kim, Y.-I., Oh, H.-K. and Yu, Y.-H., 2006. The development of diesel engine room fault diagnosis system using a correlation analysis method. Journal of the Korean Society of Marine Engineering, 30(2), pp.253-259.
  5. Kim, K.-S., Roh, M.-I., Jeong, Y.-J. and Na, G.-J., 2020a. Development of an application for predicting and visualizing the ocean weather. Korean Journal of Computational Design and Engineering, 25(4), pp.434-444. https://doi.org/10.7315/CDE.2020.434
  6. Kim, D., Lee, S. and Lee, J., 2020b. Abnormaly detection of vessel main engine big data using Gaussian mixture model. Journal of the Korean Data Analysis Society, 22(4), pp.1473-1489. https://doi.org/10.37727/jkdas.2020.22.4.1473
  7. Kim, D., Han, Y., Kim, H., Kang, S., Kim, K. and Bae, H., 2022. Outlier detection and labeling of ship main engine using LSTM-AutoEncoder. The Korea Journal of Big Data, 7(1), pp.125-137.
  8. Korean Register, 2021. Guidebook for CII Regulation. Korean Register R&D Division.
  9. Landgraf, A.J. and Lee, Y., 2020. Dimensionality reduction for binary data through the projection of natural parameters. Journal of Multivariate Analysis, 180, Article No. 104668.
  10. Lee, S.B., Han, B.W., Park, D.W., Ahn, Y.W., Go, S.C. and Seo, H.W., 2012. Proper orthogonal decomposition of pressure fluctuations in moonpool. Journal of the Society of Naval Architects of Korea, 49(6), pp.484-490. https://doi.org/10.3744/SNAK.2012.49.6.484
  11. Lee, J.H., 2017. Proper orthogonal decomposition and its application: parametric reduced order models. Computational Structural Engineering, 30(1), pp.29-35.
  12. Leeuw, J., 2006. Principal component analysis of binary data by iterated singular value decomposition. Computational Statistics and Data analysis, 50(1), pp.21-39. https://doi.org/10.1016/j.csda.2004.07.010
  13. Loyola-Gonzalez, O., 2019. Black-box vs white-box: understanding their advantages and weaknesses from a practical point of view. IEEE Access, 7, 154096-154113. https://doi.org/10.1109/ACCESS.2019.2949286
  14. Moeini, B., Avval, T.G., Gallagher, N. and Linford, M.R., 2023. Surface analysis insight note. Principal component analysis (PCA) of an X-ray photoelectron spectroscopy image. The importance of preprocessing. Surface and Interface Analysis, 55(11), pp.798-807. https://doi.org/10.1002/sia.7252
  15. Paik, B.-G., Kim, K.-Y., Kim, K.-S., Lee, J.-Y. and Lee, S.-J., 2010. Analysis of the unstable propeller wake using POD method. Journal of the Society of Naval Architects of Korea, 47(1), pp.20-29. https://doi.org/10.3744/SNAK.2010.47.1.020
  16. Park, J.-H., Jang, M.-K., Lee, G.-H., Oh, E.-K. and Hur, S.-W., 2016. Forecasting algorithm for vessel engine failure. Journal of Korean Institute of Information Technology, 14(11), pp.109-117.
  17. Park, J.-H., Oh, E.-K., Jang, M.-K., Seo, Y.-W. and Hur, S.-W., 2017. Improved forecasting algorithm for vessel engine failure. Journal of Korean Institute of Information Technology, 15(11), pp.175-185. https://doi.org/10.14801/jkiit.2017.15.11.175
  18. Park, J.-C., Kwon, H.-C., Kim, C.-H. and Jang, H.-S., 2023. 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. Journal of the Society of Naval Architects of Korea, 60(2), pp.95-109. https://doi.org/10.3744/SNAK.2023.60.2.95
  19. Savage, N., 2022. Breaking into the black box of artificial intelligence. Nature, DOI:10.1038/d41586-022-00858-1.
  20. Shin, S.-Y., Jung, K.-H., Kang, Y.-D., Suh, S.-B., Kim, J. and An, N.-H., 2017. A study on the effect of large coherent structures to the skin friction by POD analysis. Journal of the Society of Naval Architects of Korea, 54(5), pp.406-414. https://doi.org/10.3744/SNAK.2017.54.5.406