한국해양학회지:바다 (The Sea:JOURNAL OF THE KOREAN SOCIETY OF OCEANOGRAPHY)

  • 제28권4호
  • /
  • Pages.133-142
  • /
  • 2023
  • /
  • 1226-2978(pISSN)
  • /
  • 2671-8820(eISSN)
한국해양학회 (The Korean Society of Oceanography)
권호 표지
DOI QR코드

DOI QR Code

Lorenz 시스템의 역학 모델과 자료기반 인공지능 모델의 특성 비교

Comparison of the Characteristics between the Dynamical Model and the Artificial Intelligence Model of the Lorenz System

  • 김영호 (부경대학교 지구환경시스템과학부(해양학전공) ) ;
  • 임나경 (부경대학교 지구환경시스템과학부(해양학전공) ) ;
  • 김민우 (부경대학교 지구환경시스템과학부(해양학전공)) ;
  • 정재희 (부경대학교 지구환경시스템과학부(해양학전공)) ;
  • 정은서 (부경대학교 지구환경시스템과학부(해양학전공))
  • YOUNG HO KIM (Division of Earth Environmental System Science (Major of Oceanography), Pukyong National University) ;
  • NAKYOUNG IM (Division of Earth Environmental System Science (Major of Oceanography), Pukyong National University) ;
  • MIN WOO KIM (Division of Earth Environmental System Science (Major of Oceanography), Pukyong National University) ;
  • JAE HEE JEONG (Division of Earth Environmental System Science (Major of Oceanography), Pukyong National University) ;
  • EUN SEO JEONG (Division of Earth Environmental System Science (Major of Oceanography), Pukyong National University)
  • 투고 : 2023.08.29
  • 심사 : 2023.11.16
  • 발행 : 2023.11.30
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

이 논문에서는 RNN (Recurrent Neural Networks)-LSTM (Long Short-Term Memory) 을 적용하여 Lorenz 시스템을 예측하는 자료 기반 인공지능 모델을 구축하고, 이 모델이 미분방정식을 차분화하여 해를 구하는 역학 모델을 대체할 수 있는지 가능성을 진단하였다. 구축된 자료기반 모델이 초기 조건의 작은 교란이 근본적으로 다른 결과를 만들어내는 Lorenz 시스템의 카오스적인 특성을 반영한다는 것과, 시스템의 안정적인 두 개의 닻을 중심으로 운동하면서 전이 과정을 반복하는 특성, "결정론적 불규칙 흐름"의 특성, 분기 현상을 모사한다는 것을 확인하였다. 또한, 적분 시간 간격을 조절함으로써 전산자원을 절감할 수 있는 자료기반 모델의 장점을 보였다. 향후 자료기반 모델의 정교화와 자료기반 모델을 위한 자료동화 기법의 연구를 통해 자료기반 인공지능 모델의 활용성을 확대할 수 있을 것으로 기대한다.

In this paper, we built a data-driven artificial intelligence model using RNN-LSTM (Recurrent Neural Networks-Long Short-Term Memory) to predict the Lorenz system, and examined the possibility of whether this model can replace chaotic dynamic models. We confirmed that the data-driven model reflects the chaotic nature of the Lorenz system, where a small error in the initial conditions produces fundamentally different results, and the system moves around two stable poles, repeating the transition process, the characteristic of "deterministic non-periodic flow", and simulates the bifurcation phenomenon. We also demonstrated the advantage of adjusting integration time intervals to reduce computational resources in data-driven models. Thus, we anticipate expanding the applicability of data-driven artificial intelligence models through future research on refining data-driven models and data assimilation techniques for data-driven models.

키워드

과제정보

이 논문은 부경대학교 자율창의학술연구비(2021년)에 의하여 연구되었음.

참고문헌

  1. Abarbanel, H.D., P.J. Rozdeba and S. Shirman, 2018. Machine Learning: Deepest learning as statistical data assimilation problems. Neural Computation, 30(8): 2025-2055. DOI: https://doi.org/10.1162/neco_a_01094.
  2. Brajard, J., A. Carrassi, M. Bocquet and L. Bertino, 2019. Combining data assimilation and machine learning to emulate a dynamical model from sparse and noisy observations: A case study with the Lorenz 96 model. Geoscientific Model Development Discussions, 44: 1-21. DOI: https://doi.org/10.1016/j.jocs.2020.101171.
  3. Chattopadhyay, A., P. Hassanzadeh and D. Subramanian, 2020a. Data-driven predictions of a multiscale Lorenz 96 chaotic system using machine-learning methods: reservoir computing, artificial neural network, and long short-term memory network, Nonlin. Processes Geophys., 27(3): 373-389, DOI: https://doi.org/10.5194/npg-27-373-2020.
  4. Chattopadhyay, A., A. Subel and P. Hassanzadeh, 2020b. Data-driven super-parameterization using deep learning: Experimentation with multi-scale Lorenz 96 systems and transfer-learning, arXiv [preprint], arXiv:2002.11167.
  5. Dubois, P., T. Gomez, L. Planckaert and L. Perret, 2020. Data-driven predictions of the Lorenz system. Physica D: Nonlinear Phenomena, 408: 132495. DOI: 10.1016/j.physd.2020.132495.
  6. Erichson, N.B., L. Mathelin, Z. Yao, S.L. Brunton, M.W. Mahoney and J.N. Kutz, 2019. Shallow learning for fluid flow reconstruction with limited sensors and limited data, arXiv preprint arXiv:1902.07358.
  7. Ham, Y.-G., J.-H. Kim and J.-J. Luo, 2019. Deep learning for multi-year ENSO forecasts. Nature, 573: 568-572. DOI: 10.1038/s41586-019-1559-7.
  8. Kingma, D.P. and J. Ba, 2014. Adam: A method for stochastic optimization, arXiv preprint arXiv:1412.6980.
  9. Kutz, J.N., 2013. Data-driven modeling & scientific computation: methods for complex systems & big data, Oxford University Press, 656 pp.
  10. Lorenz, E.N., 1963. Deterministic nonperiodic flow. Journal of the atmospheric sciences, 20(2): 130-141. https://doi.org/10.1175/1520-0469(1963)020<0130:DNF>2.0.CO;2
  11. Scher, S. and G. Messori, 2019. Generalization properties of feed-forward neural networks trained on Lorenz systems. Nonlin. Processes Geophys., 26(4): 381-399. DOI: https://doi.org/10.5194/npg-26-381-2019.
  12. Wang, J. and Y. Li, 2018. Multi-step ahead wind speed prediction based on optimal feature extraction, long short term memory neural network and error correction strategy. Applied Energy, 230: 429-443. https://doi.org/10.1016/j.apenergy.2018.08.114
  13. Yu, R., S. Zheng and Y. Liu, 2017. Learning chaotic dynamics using tensor recurrent neural networks. in: Proceedings of the ICML, Vol. 17.
  14. Zhang, J.-S. and X.-C. Xiao, 2000. Predicting chaotic time series using recurrent neural network. Chinese Physics Letters, 17(2): 88.