機械学習のページ

• 科研費基盤S（2021〜2025年度，課題番号 21H05007）
  「機械学習を活用した革新的流れ制御パラダイムの創出と実践」
• 科研費基盤A（2018〜2020年度，課題番号 18H03758）
  「機械学習による乱流ビッグデータの特徴抽出手法の構築」
  基盤Aのみのページは こちら

K. Fukami, K. Hasegawa, T. Nakamura, M. Morimoto, and K. Fukagata, SN Comput. Sci. 2, 467 (2021).

研究体制

• 研究代表者：深潟 康二（慶大）
• 研究分担者：山本 誠 （東理大），長谷川 洋介（東大），岩本 薫（農工大），塚原 隆裕（東理大），福島 直哉（東海大），守 裕也（電通大）

研究目的

• 流れの制御に機械学習を用いることにより，新たな流れの制御手法構築の方法論を提案したい．

主な研究成果

レビュー論文

1. K. Fukagata, K. Iwamoto, and Y. Hasegawa,
   "Turbulent drag reduction by streamwise traveling waves of wall-normal forcing,"
   Annu. Rev. Fluid Mech. 56 (2024 to appear).

2. K. Fukagata,
   "Reduced order modeling of fluid flows using convolutional neural networks,"
   J. Fluid Sci. Technol. 18, JFST0002 (2023).

未出版のプレプリント

1. N. Moriya, K. Fukami, Y. Nabae, M. Morimoto, T. Nakamura, and K. Fukagata,
   "Inserting machine-learned virtual wall velocity for large-eddy simulation of turbulent channel flows,"
   arXiv preprint, arXiv:2106.09271 [physics.flu-dyn].

2. M. Matsuo, T. Nakamura, M. Morimoto, K. Fukami, and K. Fukagata,
   "Supervised convolutional network for three-dimensional fluid data reconstruction from sectional flow fields with adaptive super-resolution assistance,"
   arXiv preprint, arXiv:2103.09020 [physics.flu-dyn].

   • Sample code: Available on GitHub

研究論文

1. K. Fukami, K. Fukagata, and K. Taira,
   "Super-resolution analysis via machine learning: A survey for fluid flows,"
   Theor. Comput. Fluid Dyn. (to appear).
   (Preprint, arXiv:2301.10937 [physics.flu-dyn])

2. T. Sonoda, Z. Liu, T. Itoh, and Y. Hasegawa,
   "Reinforcement learning of control strategies for reducing skin friction drag in a fully developed turbulent channel flow,"
   J. Fluid Mech. 930, A30 (2023).

3. H. Wang and Y. Hasegawa,
   "Multi-objective optimization of actuation waveform for high-precision drop-on-demand inkjet printing,"
   Phys. Fluids 35, 013318 (2023).

4. M. Atzori, F. Mallor, R. Pozuelo, K. Fukagata, R. Vinuesa, and P. Schlatter,
   "A new perspective on skin-friction contributions in adverse-pressure-gradient turbulent boundary layers,"
   Int. J. Heat Fluid Flow 101, 109117 (2023).

5. T. Ishigami, M. Irikura, and T. Tsukahara,
   "Applicability of convolutional neural network for estimation of turbulent diffusion distance from source point,"
   Processes 10, 860 (2022).

6. T. Ishigami, M. Irikura, T. Tsukahara,
   "Machine learning to estimate the mass-diffusion distance from a point source under turbulent conditions,"
   Processes 10, 860 (2022).

7. M. Morimoto, K. Fukami, R. Maulik, R. Vinuesa, and K. Fukagata,
   "Assessments of epistemic uncertainty using Gaussian stochastic weight averaging for fluid-flow regression,"
   Physica D 440, 133454 (2022).
   (Preprint, arXiv:2109.08248 [physics.flu-dyn])

8. Y. Nabae and K. Fukagata,
   "Drag reduction effect of streamwise traveling wave-like wall deformation with spanwise displacement variation in turbulent channel flow,"
   Flow Turbul. Combust. 109, 1175–1194 (2022).

9. T. Nakamura and K. Fukagata,
   "Robust training approach of neural networks for fluid flow state estimations,"
   Int. J. Heat Fluid Flow 96, 108977 (2022).
   (Preprint, arXiv:2112.02751 [physics.flu-dyn])

10. T. Nakamura, K. Fukami, and K. Fukagata,
    "Identifying key differences between linear stochastic estimation and neural networks for fluid flow regressions,"
    Sci. Rep. 12, 3726 (2022).
    (Preprint: arXiv:2105.00913 [physics.flu-dyn]).

11. Y. Morita, S. Rezaeiravesh, N. Tabatabaei, R. Vinuesa, K. Fukagata, and P. Schlatter,
    "Applying Bayesian optimization with Gaussian process regression to computational fluid dynamics problems,"
    J. Comput. Phys. 449 110788 (2022).
    

    • Preprint: arXiv:2101.09985 [physics.flu-dyn]

12. M. Morimoto, K. Fukami, K. Zhang, and K. Fukagata,
    "Generalization techniques of neural networks for fluid flow estimation,"
    Neural Comput. Appl. 34, 3647–3669 (2022).

    • Preprint: arXiv:2011.11911 [physics.flu-dyn]
    • Sample code for Grad-CAM: Available on GitHub

13. Y. Nabae and K. Fukagata,
    "Bayesian optimization of traveling wave-like wall deformation for friction drag reduction in turbulent channel flow,"
    J. Fluid Sci. Technol. 16, JFST0024 (2021).

14. K. Fukami, K. Hasegawa, T. Nakamura, M. Morimoto, and K. Fukagata,
    "Model order reduction with neural networks: Application to laminar and turbulent flows,"
    SN Comput. Sci. 2, 467 (2021).
    

15. M. Morimoto, K. Fukami, and K. Fukagata,
    "Experimental velocity data estimation for imperfect particle images using machine learning,"
    Phys. Fluids 33, 087121 (2021). Editor's pick

    • Sample code: Available on GitHub

16. K. Fukami, T. Murata, K. Zhang, and K. Fukagata,
    "Sparse identification of nonlinear dynamics with low-dimensionalized flow representations,"
    J. Fluid Mech. 926, A10 (2021).

    • Sample code: Available on GitHub

17. K. Fukami, R. Maulik, N. Ramachandra, K. Fukagata, and K. Taira,
    "Global field reconstruction from sparse sensors with Voronoi tessellation-assisted deep learning,"
    Nat. Mach. Intell. 3, 945-951 (2021).
    

    • Preprint: arXiv:2101.00554 [physics.flu-dyn]
    • Sample code: Available on GitHub

18. M. Morimoto, K. Fukami, K. Zhang, A. G. Nair, and K. Fukagata,
    "Convolutional neural networks for fluid flow analysis: toward effective metamodeling and low-dimensionalization,"
    Theor. Comput. Fluid Dyn. 35, 633-658 (2021).
    

    • Sample code: Available on GitHub

19. W. Kobayashi, T. Shimura, A. Mitsuishi, K. Iwamoto, and A. Murata,
    "Prediction of the drag reduction effect of pulsating pipe flow based on machine learning,"
    Int. J. Heat Fluid Flow 88, 108783 (2021).

20. A. J. Kaithakkal, Y. Kametani, Y. Hasegawa,
    "Dissimilar heat transfer enhancement in a fully developed laminar channel flow subject to a traveling wave-like wall blowing and suction,"
    Int. J. Heat Mass Transfer 164, 120485 (2021).

21. 志村 敬彬，光石 暁彦，岩本 薫，
    「機械学習による円管内脈動乱流の予測」
    日本燃焼学会誌 63, 52-59 (2021).

22. T. Nakamura, K. Fukami, K. Hasegawa, Y. Nabae, and K. Fukagata,
    "Convolutional neural network and long short-term memory based reduced order surrogate for minimal turbulent channel flow,"
    Phys. Fluids 33, 025116 (2021). Editor's pick
    

    • Sample code: Available on GitHub

23. K. Fukami, K. Fukagata, and K. Taira,
    "Machine-learning-based spatio-temporal super resolution reconstruction of turbulent flows,"
    J. Fluid Mech. 909, A9 (2021).
    

24. K. Hasegawa, K. Fukami, T. Murata, and K. Fukagata,
    "CNN-LSTM based reduced order modeling of two-dimensional unsteady flows around a circular cylinder at different Reynolds numbers,"
    Fluid Dyn. Res. 52, 065501 (2020). Highlights of 2020

25. R. Maulik, K. Fukami, N. Ramachandra, K. Fukagata, and K. Taira,
    "Probabilistic neural networks for fluid flow surrogate modeling and data recovery,"
    Phys. Rev. Fluids 5, 104401 (2020).
    

26. K. Fukami, T. Nakamura, and K. Fukagata,
    "Convolutional neural network based hierarchical autoencoder for nonlinear mode decomposition of fluid field data,"
    Phys. Fluids 32, 095110 (2020).
    
    • Sample code: HierarchicalAE_fig3.py

27. K. Hasegawa, K. Fukami, T. Murata, and K. Fukagata,
    "Machine-learning-based reduced order modeling for unsteady flows around bluff bodies of various shapes,"
    Theor. Comput. Fluid Dyn. 34, 367-383 (2020).

    • Sample code: Available on GitHub

28. K. Fukami, K. Fukagata, and K. Taira,
    "Assessment of supervised machine learning methods for fluid flows,"
    Theor. Comput. Fluid Dyn. 34, 497–519 (2020).

29. T. Murata, K. Fukami, and K. Fukagata,
    "Nonlinear mode decomposition with convolutional neural networks for fluid dynamics,"
    J. Fluid Mech. 882, A13 (2020).

    • Sample code: MD-CNN-AE.py

30. K. Fukami, Y. Nabae, K. Kawai, and K. Fukagata,
    "Synthetic turbulent inflow generator using machine learning,"
    Phys. Rev. Fluids 4, 064603 (2019).
    

    • Sample code: Available on GitHub
    • Animation: Machine-Learned Turbulence Generator, Ver. 2 (MLTG2)

31. Q. Wang, Y. Hasegawa, and T. Zaki,
    "Spatial reconstruction of steady scalar sources from remote measurements in turbulent flow,"
    J. Fluid Mech. 870, 316-352 (2019)

32. K. Fukami, K. Fukagata, and K. Taira,
    "Super-resolution reconstruction of turbulent flows with machine learning,"
    J. Fluid Mech. 870, 106-120 (2019).

    • Sample code: Available at UCLA Taira Lab

解説・特集記事など

1. 塚原 隆裕,
   「粘弾性流体や乱流拡散を対象とした深層学習」，
   日本機械学会流体工学部門ニュースレター「流れ」, 2022年11月号, Art. 5 (2022).

2. 深潟 康二，
   「畳み込みニューラルネットワークを用いた流体場の低次元化と欠損情報推定」，
   日本風工学会誌 47(3), 215-220 (2022)．

3. 深潟 康二，
   「基礎的な流れ場に対する機械学習の応用」，
   日本ガスタービン学会誌 50(3), 179-184 (2022)．

4. 長谷川 洋介，
   「計測とシミュレーションの融合による熱流動場の推定」，
   機械の研究 73(12), 911-918 (2021).

5. 深潟 康二，
   「乱流の機械学習と制御」，
   フルードパワーシステム 52(6), 237-241 (2021)．

6. 深潟 康二，深見 開，
   「機械学習縮約モデルを用いた革新的流れ制御に向けて」，
   伝熱 60(253), 12-15 (2021)．

7. 深潟 康二，
   「機械学習の乱流への応用」，
   日本機械学会誌 124(1232), 10-13 (2021).

8. 深潟 康二，深見 開，
   「機械学習を用いた乱流ビッグデータ解析に向けて」，
   計測と制御 59(8), 571-576 (2020).

9. 森本 将生，深見 開，長谷川 一登，村田 高彬，村上 光，深潟 康二，
   〔特集〕注目研究 in CFD33：「機械学習に基づくデータ拡張によるPIV の精度向上」，
   ながれ 39, 84-87 (2020).

10. 深見 開，深潟 康二，平 邦彦，
    「チャネル乱流における機械学習3次元超解像解析」，
    日本機械学会流体工学部門ニュースレター「流れ」，2020年2月号, Art. 4 (2020)．

11. 深見 開，深潟 康二，平 邦彦，
    〔特集〕注目研究 in 年会2019：「2次元流れ場への機械学習超解像の応用」，
    ながれ 38, 395-398 (2019).

12. 光石 暁彦，志村 敬彬，岩本 薫，
    〔特集〕機械学習の流体力学研究への応用：「壁乱流制御の効率的最適化に向けた機械学習の応用」，
    ながれ 38, 329-336 (2019).

13. 塚原 隆裕，川口 靖夫，
    〔特集〕機械学習の流体力学研究への応用：「乱流物質拡散源推定に向けた CNN 画像認識」，
    ながれ 38, 337-343 (2019).

14. 長谷川 一登，深見 開，村田 高彬，深潟 康二，
    〔特集〕注目研究 in CFD32：「機械学習を用いた円柱周り流れのレイノルズ数依存性の予測」，
    ながれ 38, 81-84 (2019).

15. 深潟 康二，山本 誠，岩本 薫，長谷川 洋介，塚原 隆裕，福島 直哉，守 裕也，青木 義満，
    〔特集〕注目研究 in 年会2018：「機械学習を用いた乱流の特徴抽出手法の構築に向けて」，
    ながれ 37, 524-527 (2018).