Prediction of Ship Roll Motion using Machine Learning-based Surrogate Model

기계학습기반의 근사모델을 이용한 선박 횡동요 운동 예측

  • Kim, Young-Rong (Graduate School of Korea Maritime and Ocean University) ;
  • Park, Jun-Bum (Division of Navigation Science, Korea Maritime and Ocean University) ;
  • Moon, Serng-Bae (Division of Navigation Science, Korea Maritime and Ocean University)
  • 김영롱 (한국해양대학교 대학원) ;
  • 박준범 (한국해양대학교 항해학부) ;
  • 문성배 (한국해양대학교 항해학부)
  • Received : 2018.09.04
  • Accepted : 2018.12.06
  • Published : 2018.12.31


Seakeeping safety module in Korean e-Navigation system is one of the ship remote monitoring services that is employed to ensure the safety of ships by monitoring the ship's real time performance and providing a warning in advance when the abnormal conditions are encountered in seakeeping performance. In general, seakeeping performance has been evaluated by simulating ship motion analysis under specific conditions for its design. However, due to restriction of computation time, it is not realistic to perform simulations to evaluate seakeeping performance under real-time operation conditions. This study aims to introduce a reasonable and faster method to predict a ship's roll motion which is one of the factors used to evaluate a ship's seakeeping performance by using a machine learning-based surrogate model. Through the application of various learning techniques and sampling conditions on training data, it was observed that the difference of roll motion between a given surrogate model and motion analysis was within 1%. Therefore, it can be concluded that this method can be useful to evaluate the seakeeping performance of a ship in real-time operation.

한국형 e-Navigation의 내항성 안전 모듈은 운항 중인 선박을 실시간으로 모니터링하고 내항성의 이상 상태를 사전에 경고함으로써 선박의 안정성을 확보하는 선내 원격 모니터링 서비스 중 하나이다. 일반적으로 선박설계를 위한 내항성능은 주어진 조건에서 선체 운동 시뮬레이션을 수행하여 평가하여 왔다. 하지만 운항 중 선박의 내항성능을 실시간으로 평가하기 위해 이러한 시뮬레이션을 실제 운항조건에 맞추어 수행하는 것은 계산시간의 한계로 인해 현실적이지 않다. 본 연구에서는 기계학습 기반의 근사모델을 활용하여 선박의 내항성능 평가 요소들 중 하나인 횡동요 운동특성을 합리적으로 보다 빠르게 예측하는 방법을 소개하고자 한다. 다양한 학습 기법과 데이터의 샘플링 조건을 적용하여, 얻어진 근사모델의 결과와 운동해석 결과의 오차가 거의 1% 내로 일치함을 보였다. 따라서 이러한 방법을 활용하면 선박의 실시간 내항성능을 평가하는데 효율적으로 사용할 수 있을 것으로 판단된다.


GHMHD9_2018_v42n6_395_f0001.png 이미지

Fig. 1 Hull panel of the model ship

GHMHD9_2018_v42n6_395_f0002.png 이미지

Fig. 2 R2 of TGP model by heading sampling interval

GHMHD9_2018_v42n6_395_f0003.png 이미지

Fig. 3 R2 of PLA model by heading sampling interval

GHMHD9_2018_v42n6_395_f0004.png 이미지

Fig. 4 R2 of TGP model by speed sampling interval

GHMHD9_2018_v42n6_395_f0005.png 이미지

Fig. 5 R2 of PLA model by speed sampling interval

GHMHD9_2018_v42n6_395_f0006.png 이미지

Fig. 6 Example of full factorization

GHMHD9_2018_v42n6_395_f0007.png 이미지

Fig. 7 Error rate by ship’s speed and heading

GHMHD9_2018_v42n6_395_f0008.png 이미지

Fig. 8 Comparison of actual roll RAO and predicted roll RAO at maximum error rate

GHMHD9_2018_v42n6_395_f0009.png 이미지

Fig. 9 Average error rate of alternative models by the number of training data set

GHMHD9_2018_v42n6_395_f0010.png 이미지

Fig. 10 Maximum error rate of alternative models by the number of training data set

GHMHD9_2018_v42n6_395_f0011.png 이미지

Fig. 11 Error rate by ship’s speed and heading(Case 6)

GHMHD9_2018_v42n6_395_f0012.png 이미지

Fig. 12 Comparison of actual roll RAO and predicted roll RAO at maximum Error rate(Case 6)

GHMHD9_2018_v42n6_395_f0013.png 이미지

Fig. 13 Histogram of peak roll RAO

GHMHD9_2018_v42n6_395_f0014.png 이미지

Fig. 14 ITTC wave spectrum

GHMHD9_2018_v42n6_395_f0015.png 이미지

Fig. 15 Comparison of actual roll response and predicted roll response at maximum error rate

GHMHD9_2018_v42n6_395_f0016.png 이미지

Fig. 16 Comparison of actual m0 and predicted m0 by ship’s heading at 18knot

GHMHD9_2018_v42n6_395_f0017.png 이미지

Fig. 17 Comparison of actual m2 and predicted m2 by ship’s heading at 18knot

Table 1 Principal particulars of the model ship

GHMHD9_2018_v42n6_395_t0001.png 이미지

Table 2 Simulation cases for test data set

GHMHD9_2018_v42n6_395_t0002.png 이미지

Table 3 Classification guidances for ship‘s motion & strength analysis

GHMHD9_2018_v42n6_395_t0003.png 이미지

Table 4 Sampling intervals for training data set

GHMHD9_2018_v42n6_395_t0004.png 이미지

Table 5 Prediction accuracy of top four techniques

GHMHD9_2018_v42n6_395_t0005.png 이미지

Table 6 Sampling intervals for training data set

GHMHD9_2018_v42n6_395_t0006.png 이미지

Table 7 Prediction accuracy of TGP & PLA models

GHMHD9_2018_v42n6_395_t0007.png 이미지

Table 8 Minimum sampling interval of classification guidances

GHMHD9_2018_v42n6_395_t0008.png 이미지

Table 9 Prediction accuracy of model with minimum sampling interval

GHMHD9_2018_v42n6_395_t0009.png 이미지

Table 10 Sampling intervals for training data set of alternative models

GHMHD9_2018_v42n6_395_t0010.png 이미지

Table 11 Prediction accuracy of alternative models

GHMHD9_2018_v42n6_395_t0011.png 이미지

Table 12 Comparison of actual values and predicted values for m0, m2, Qp at maximum error rate

GHMHD9_2018_v42n6_395_t0012.png 이미지


Grant : IMO 차세대 해양안전 종합관리체계 기술개발

Supported by : 한국해양과학기술진흥원


  1. American Bureau of Shipping(2005), "Guidance Notes on 'Safehull-Dynamic Loading Approach' for Container carriers".
  2. Belyaev, M., Burnaev, E. and Kapushev, Y.(2015), "Gaussian process regression for structured data sets", Lecture Notes in Artificial Intelligence, Proceedings of SLDS, Vol. 9047, pp. 106-115.
  3. Belyaev, M., Burnaev, E., Kapushev, E., Panov, M., Prikhodko, P., Vetrov, D. and Yarotsky, D.(2016a), "GTApprox: surrogate modeling for industrial design", Advances in Engineering Software, Vol. 102, pp. 29-39.
  4. Belyaev, M., Burnaev, E. and Kapushev, Y.(2016b), "Computationally efficient algorithm for Gaussian Process regression in case of structured samples", Computational Mathematics and Mathematical Physics, Vol. 56, No. 4, pp. 499-513.
  5. Burnaev, E., Zaytsev, A., Panov, M., Prihodko, P. and Yanovich, Y.(2011), "Modeling of nonstationary covariance function of gaussian process using decomposition in dictionary of nonlinear functions", Information Technologies and Systems-2011, pp. 355-362.
  6. Burnaev, E. V., Panov, M. E. and Zaytsev, A. A.(2016), "Regression on the basis of nonstationary Gaussian processes with Bayesian regularization", Journal of communications technology and electronics, Vol. 61, No. 6, pp. 661-671.
  7. Choi, K. S., Park, M. K., Lee, J. H. and Park, G. I. (2007), "A study on the optimum navigation route safety assessment system using real time weather forecasting", Journal of the Korean Society of Marine Environment & Safety, Vol. 13, No. 2, pp. 133-140.
  8. Cressie, N.(1992), "Statistics for spatial data", Terra Nova, Vol. 4, No. 5, pp. 613-617.
  9. Det Norske Veritas(2013), "Classification Notes: CSA - Direct Analysis of Ship Structures".
  10. International Maritime Organization(2005), "Development of an e-Navigation Strategy submitted by Japan, Marshall Islands, the Netherlands, Norway, Singapore, the United Kingdom and the United States", MSC 81/23/10.
  11. International Maritime Organization(2008a), "Strategy for the Development and Implementation of e-Navigati on - Report of the Maritime Safety Committee on its Eighty-fifth Session", MSC 85/26, Annex 20.
  12. International Maritime Organization(2008b), "Adoption of the international code on intact stability, 2008 (2008 IS Code)", IMO Resolutiuon MSC 267(85).
  13. International Maritime Organization(2014), "Draft e-Navigation Strategy Implementation Plan - Report to the Maritime Safety Committee", NCSR 1/28, Annex 7.
  14. Korea Ministry of Maritime Affairs and Fisheries(2016), "IMO Next-Generation Marine Safety Integrated Management System Development", Korean e-Navigation Project Notice No. 2016-01.
  15. Korean Maritime Safety Tribunal(2017), Statistical annual reports from 2013 to 2017, Status of marine accidents' causes as per types and years,
  16. Korea Research Institute of Ships & Ocean Engineering(2018), "Implementation of Core Technology for Korean e-Navigation Services", Annual performance plan report.
  17. Lee, J. H., Choi, K. S., Park, G. I., Kim, M. S. and Bang, C. S.(2005), "Basic Research of Optimum Routing Assessment System for Safe and Efficient Voyage", Journal of the Society of Naval Architects of Korea, Vol. 42, No. 1, pp. 57-63.
  18. Lloyd's register of Shipping(2009), "Ship Right FDA level 3 procedures manual", technical planning and development department.
  19. Rasmussen, C. E. and Williams, C. K.(2006), "Gaussian process for machine learning", MIT press.