• 研究課題をさがす
  • 研究者をさがす
  • KAKENの使い方
  1. 前のページに戻る

Physics Informed Machine Learning for Complex Flows

公募研究

研究領域「学習物理学」の創成-機械学習と物理学の融合新領域による基礎物理学の変革
研究課題/領域番号 23H04508
研究種目

学術変革領域研究(A)

配分区分補助金
審査区分 学術変革領域研究区分(Ⅱ)
研究機関京都大学

研究代表者

MOLINA JOHN  京都大学, 工学研究科, 助教 (20727581)

研究期間 (年度) 2023-04-01 – 2025-03-31
研究課題ステータス 交付 (2024年度)
配分額 *注記
2,340千円 (直接経費: 1,800千円、間接経費: 540千円)
2024年度: 1,170千円 (直接経費: 900千円、間接経費: 270千円)
2023年度: 1,170千円 (直接経費: 900千円、間接経費: 270千円)
キーワードMachine Learning / Stokes Flow / Multi-Scale Simulation / Soft Matter / Multi-Scale Simulations / Polymer Melts / Flow Inference / Gaussian Processes
研究開始時の研究の概要

We will improve and optimize the learning methods we have developed for (A) multi-scale simulations of polymer flows and (B) the inference of Stokes flows with missing and/or noisy data. For the former, we will learn the constitutive relation for the canonical polymer entanglement model (i.e., Doi-Takimoto), and use it to simulate the dynamics of entangled polymer melt flows in 2D/3D. For the latter, we will incorporate hydrodynamics stresses and moving boundaries into the inference framework, to consider experimentally relevant flow problems (e.g., biofluids and colloidal dispersions).

研究実績の概要

For theme A, we have succeeded in extending our Gaussian Process (GP) based learning method, originally developed for non-interacting polymers, to entangled polymer melts relevant for industry. In particular, we have learned the (non-linear) constitutive relation of the dual slip-link model, a coarse-grained entanglement model that can explain many of the rheological properties of polymer melts. Our learned model is able to accurately reproduce the flow behavior of entangled polymers (compared to multi-scale simulations) at a small fraction of the cost. This work was published in Physics of Fluids and selected as "Editor's Pick".

For theme B, we have succeeded in developing a probabilistic framework for solving Stokes flow problems, based on a Physics-Informed Gaussian Process regression. We have validated our method on a non-trivial 2D problem: pressure driven flow through a sinusoidal channel. We are able to accurately solve both forward and inverse problems with a high-degree of accuracy. Our method is capable of inferring velocity/pressure fields from partial and/or noisy data, as well as stresses/forces on boundaries. Furthermore, we have shown that our approach is faster and more robust than alternative Machine-Learning solutions, e.g., Physics-Informed Neural Networks.

現在までの達成度 (区分)
現在までの達成度 (区分)

2: おおむね順調に進展している

理由

Our project is progressing smoothly.
We have made progress on both themes, in line with our original plan.

For theme (A), we have extended our method to entangled polymers. Flow predictions using the learned relations are in good agreement with full-scale multi-scale simulations (at a fraction of the cost), even for complex geometries/flows. For theme (B) we have developed a generalized 2D/3D Stokes flow solver. Our method is able to infer physically meaningful flow solutions given sparse/incomplete data, showing that it is a viable candidate for analyzing experiments.

In addition, we have also explored other areas where Physics-Informed Machine Learning approaches can be used to solve Soft Matter flow problems (e.g., autonomous navigation, inferring molecular weights of polymers).

今後の研究の推進方策

We will continue to develop themes (A) and (B) to be able to study complex 3D flows.

For theme (A), this requires parallelizing our code for high-performance GPU systems (or hybrid GPU/CPU systems). In particular, we aim to develop efficient parallel data structures to handle the creation/destruction of polymer entanglements within the slip-link model. Furthermore, we will also continue to investigate how to implement data-driven / active learning protocols, to improve the accuracy of our predictions.

For theme (B), we will investigate why the Black-Box Matrix-Matrix method, which is the state-of-the-art for Gaussian Process regression, does not yield the expected performance on our custom physics-informed GP problems. If necessary, we will consider alternative methods, e.g., using different pre-conditioners or dense matrix algorithms. Finally, we will apply our method to experimental 3D data.

報告書

(1件)
  • 2023 実績報告書
  • 研究成果

    (20件)

すべて 2024 2023

すべて 雑誌論文 (4件) (うち国際共著 1件、 査読あり 4件、 オープンアクセス 2件) 学会発表 (16件) (うち国際学会 12件)

  • [雑誌論文] Autonomous navigation of smart microswimmers in non-uniform flow fields2024

    • 著者名/発表者名
      Sankaewtong Krongtum、Molina John J.、Yamamoto Ryoichi
    • 雑誌名

      Physics of Fluids

      巻: 36 号: 4

    • DOI

      10.1063/5.0193113

    • 関連する報告書
      2023 実績報告書
    • 査読あり
  • [雑誌論文] Stokesian processes : inferring Stokes flows using physics-informed Gaussian processes2023

    • 著者名/発表者名
      Molina John J、Ogawa Kenta、Taniguchi Takashi
    • 雑誌名

      Machine Learning: Science and Technology

      巻: 4 号: 4 ページ: 045013-045013

    • DOI

      10.1088/2632-2153/ad0286

    • 関連する報告書
      2023 実績報告書
    • 査読あり / オープンアクセス
  • [雑誌論文] Machine-learned constitutive relations for multi-scale simulations of well-entangled polymer melts2023

    • 著者名/発表者名
      Miyamoto Souta、Molina John J、Taniguchi Takashi
    • 雑誌名

      Physics of Fluids

      巻: 35 号: 6

    • DOI

      10.1063/5.0156272

    • 関連する報告書
      2023 実績報告書
    • 査読あり
  • [雑誌論文] Rational social distancing policy during epidemics with limited healthcare capacity2023

    • 著者名/発表者名
      Schnyder Simon K.、Molina John J.、Yamamoto Ryoichi、Turner Matthew S.
    • 雑誌名

      PLOS Computational Biology

      巻: 19 号: 10 ページ: e1011533-e1011533

    • DOI

      10.1371/journal.pcbi.1011533

    • 関連する報告書
      2023 実績報告書
    • 査読あり / オープンアクセス / 国際共著
  • [学会発表] Physics Informed Machine Learning for Soft Matter Flows2024

    • 著者名/発表者名
      John J. Molina (*), Kenta Ogawa, Takashi Taniguchi
    • 学会等名
      アクティブマター研究会 2024
    • 関連する報告書
      2023 実績報告書
  • [学会発表] Bayesian Machine Learning for Multi-Scale Simulations of Polymer Flows2023

    • 著者名/発表者名
      Souta Miyamoto, Yoshiki Ueno, Takashi Taniguchi, John J. Molina (*)
    • 学会等名
      International Congress on Rheology (ICR2023)
    • 関連する報告書
      2023 実績報告書
    • 国際学会
  • [学会発表] Machine Learning for Accelerated Multi-Scale Polymer Flow Simulations2023

    • 著者名/発表者名
      John J. Molina (*), Souta Miyamoto, Yoshiki Ueno, Takashi Taniguchi
    • 学会等名
      10th International Congress on Industrial and Applied Mathematics (ICIAM2023)
    • 関連する報告書
      2023 実績報告書
    • 国際学会
  • [学会発表] Stokesian Processes: Physics Informed Machine Learning for Soft Matter Flows2023

    • 著者名/発表者名
      John J. Molina (*), Kenta Ogawa, Takashi Taniguchi
    • 学会等名
      The 7th International Soft Matter Conference (ISMC2023)
    • 関連する報告書
      2023 実績報告書
    • 国際学会
  • [学会発表] 構成関係の機械学習回帰モデルを用いた高分子溶融体の流動予測シミュレーション2023

    • 著者名/発表者名
      Souta Miyamoto (*), John J. Molina, Takashi Taniguchi
    • 学会等名
      日本流体力学会年会2023
    • 関連する報告書
      2023 実績報告書
  • [学会発表] Flow simulations of well-entangled polymer melts using machine-learned constitutive relations2023

    • 著者名/発表者名
      Souta Miyamoto (*), John J. Molina, Takashi Taniguchi
    • 学会等名
      第71回レオロジー討論会
    • 関連する報告書
      2023 実績報告書
  • [学会発表] Inferring Stokes Flows using Physics-Informed Gaussian Processes2023

    • 著者名/発表者名
      Kenta Ogawa (*), John J. Molina, Takashi Taniguchi
    • 学会等名
      第71回レオロジー討論会
    • 関連する報告書
      2023 実績報告書
  • [学会発表] Machine-learned constitutive relations of entangled polymer melts for multi-scale flow simulations (POSTER)2023

    • 著者名/発表者名
      Souta Miyamoto (*), John J. Molina, Takashi Taniguchi
    • 学会等名
      34th IUPAP Conference on Computational Physics (CCP2023)
    • 関連する報告書
      2023 実績報告書
    • 国際学会
  • [学会発表] Stokes Flow Inference using Physics-Informed Gaussian Processes (POSTER)2023

    • 著者名/発表者名
      Kenta Ogawa (*), John J. Molina, Takashi Taniguchi
    • 学会等名
      28th International Conference on Statistical Physics (Statphys28)
    • 関連する報告書
      2023 実績報告書
    • 国際学会
  • [学会発表] Physics-Informed Gaussian Processes for Stokes Flow Inference (POSTER)2023

    • 著者名/発表者名
      Kenta Ogawa (*), John J. Molina, Takashi Taniguchi
    • 学会等名
      The 7th International Soft Matter Conference (ISMC2023)
    • 関連する報告書
      2023 実績報告書
    • 国際学会
  • [学会発表] Smart microswimmer navigation using hydrodynamical signals (POSTER)2023

    • 著者名/発表者名
      Krongtum Sankawtong (*), John J. Molina, Ryoichi Yamamoto
    • 学会等名
      The 7th International Soft Matter Conference (ISMC2023)
    • 関連する報告書
      2023 実績報告書
    • 国際学会
  • [学会発表] Flow simulations of well-entangled polymer melts using machine-learned constitutive relations (POSTER)2023

    • 著者名/発表者名
      Souta Miyamoto (*), John J. Molina, Takashi Taniguchi
    • 学会等名
      Advanced core-to-core network for the physics of self-organizing active matter
    • 関連する報告書
      2023 実績報告書
    • 国際学会
  • [学会発表] Physics-Informed Gaussian Processes for Stokes Flow Inference (POSTER)2023

    • 著者名/発表者名
      Kenta Ogawa (*), John J. Molina, Takashi Taniguchi
    • 学会等名
      Advanced core-to-core network for the physics of self-organizing active matter
    • 関連する報告書
      2023 実績報告書
    • 国際学会
  • [学会発表] Role of order parameters in learning of fish schooling (POSTER)2023

    • 著者名/発表者名
      Krongtum Sankawtong (*), John J. Molina, Ryoichi Yamamoto
    • 学会等名
      Advanced core-to-core network for the physics of self-organizing active matter
    • 関連する報告書
      2023 実績報告書
    • 国際学会
  • [学会発表] Physics Informed Optimal Control: Inferring Hidden Utilitites form Optimal Behavior (POSTER)2023

    • 著者名/発表者名
      John J. Molina (*), Mark P. Lynch, Simon K. Schnyder, Matthew S. Turner, Ryoichi Yamamoto
    • 学会等名
      Advanced core-to-core network for the physics of self-organizing active matter
    • 関連する報告書
      2023 実績報告書
    • 国際学会
  • [学会発表] Physics Informed Machine Learning for Flow Inference (POSTER)2023

    • 著者名/発表者名
      John J. Molina (*)
    • 学会等名
      International Conference on Machine Learning Physics
    • 関連する報告書
      2023 実績報告書
    • 国際学会

URL: 

公開日: 2023-04-13   更新日: 2024-12-25  

サービス概要 検索マニュアル よくある質問 お知らせ 利用規程 科研費による研究の帰属

Powered by NII kakenhi