• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to previous page

Development of the data-driven thermal-hydraulic analysis using machine learning

Research Project

Project/Area Number 23K23271
Project/Area Number (Other) 22H02003 (2022-2023)
Research Category

Grant-in-Aid for Scientific Research (B)

Allocation TypeMulti-year Fund (2024)
Single-year Grants (2022-2023)
Section一般
Review Section Basic Section 31010:Nuclear engineering-related
Research InstitutionThe University of Tokyo

Principal Investigator

Miwa Shuichiro  東京大学, 大学院工学系研究科(工学部), 准教授 (00705288)

Co-Investigator(Kenkyū-buntansha) 原 聡  電気通信大学, 大学院情報理工学研究科, 教授 (40780721)
Pellegrini Marco  東京大学, 大学院工学系研究科(工学部), 特任准教授 (50741360)
岡本 孝司  東京大学, 大学院工学系研究科(工学部), 教授 (80204030)
武居 昌宏  千葉大学, 大学院工学研究院, 教授 (90277385)
Project Period (FY) 2024-04-01 – 2025-03-31
Project Status Completed (Fiscal Year 2024)
Budget Amount *help
¥17,420,000 (Direct Cost: ¥13,400,000、Indirect Cost: ¥4,020,000)
Fiscal Year 2024: ¥4,550,000 (Direct Cost: ¥3,500,000、Indirect Cost: ¥1,050,000)
Fiscal Year 2023: ¥5,460,000 (Direct Cost: ¥4,200,000、Indirect Cost: ¥1,260,000)
Fiscal Year 2022: ¥7,410,000 (Direct Cost: ¥5,700,000、Indirect Cost: ¥1,710,000)
Keywords原子力工学 / 熱流動 / 機械学習 / 混相流 / 熱流体工学 / 凝縮 / 人工知能 / 気液二相流 / 深層学習
Outline of Research at the Start

次世代原子炉をはじめとした熱流動解析においては、保存則の解法に必要な構成方程式に大きく依存しており、近年重要視されている高詳細・高解像度解析を効率的に実施するためにも、モデルの精度向上はこれまで以上に重要な課題となっている。本研究提案においては機械学習をはじめとしたAI技術を熱流体解析に融合させ、最新鋭の計測機器より構築されるビッグデータから気液界面形状に関する情報を帰納的に抽出することで、気液界面構造の内在的な現象メカニズムを明らかにし、既存の構成方程式を高精度化・高度化することを目的とする。

Outline of Final Research Achievements

In this study, we aimed to establish a foundational framework for data-driven thermal-fluid analysis using machine learning, with a particular focus on fundamental approaches involving AI-based pattern recognition, feature extraction, and data generation. As the first initiative, a model was developed that combines high-resolution two-phase flow images obtained via high-speed cameras with a convolutional neural network-based object detection algorithm, enabling the extraction of gas-liquid interface features within two-phase flow fields.In the second initiative, a generalized PINNs framework for incompressible fluids was constructed, and a novel method integrating three key improvements was proposed.Finally, two-phase flow image datasets were generated using generative AI technologies, thereby demonstrating the applicability of AI for data synthesis.

Academic Significance and Societal Importance of the Research Achievements

人工知能技術(AI)は、これまでとは異なる新たなアプローチであるA駆動型手法をもたらした。本研究提案では、本手法の原子力熱流動への適用を検討し、従来のシミュレーションや実験で収集されたデータセットや、既に構築された物理モデルを、AIモデルの訓練データや損失関数としての利用を試みた。この枠組みにより、パラメータ予測、分類、クラスタリング、物体検出、さらに経験的相関式や機構論モデルを組み込んだ物理インフォームドニューラルネットワーク(PINNs)といった、データ駆動型あるいは物理駆動型のAIモデルが可能となり、今後の原子力熱流動分野における貢献が期待される。

Report

(4 results)
  • 2024 Annual Research Report   Final Research Report ( PDF )
  • 2023 Annual Research Report
  • 2022 Annual Research Report
  • Research Products

    (9 results)

All 2024 2023 2022

All Journal Article (3 results) (of which Int'l Joint Research: 1 results,  Peer Reviewed: 1 results) Presentation (6 results) (of which Int'l Joint Research: 3 results,  Invited: 2 results)

  • [Journal Article] Self-adaptive and time divide-and-conquer physics-informed neural networks for two-phase flow simulations using interface tracking methods2024

    • Author(s)
      Zhou Wen、Miwa Shuichiro、Okamoto Koji
    • Journal Title

      Physics of Fluids

      Volume: 36 Issue: 7 Pages: 120971-120971

    • DOI

      10.1063/5.0214646

    • Related Report
      2024 Annual Research Report
  • [Journal Article] Advancing fluid dynamics simulations: A comprehensive approach to optimizing physics-informed neural networks2024

    • Author(s)
      Zhou Wen、Miwa Shuichiro、Okamoto Koji
    • Journal Title

      Physics of Fluids

      Volume: 36 Issue: 1 Pages: 120971-120971

    • DOI

      10.1063/5.0180770

    • Related Report
      2024 Annual Research Report 2023 Annual Research Report
  • [Journal Article] Spatio-temporal void fraction visualization in air-water two-phase flow regime transitions by combination of convolutional neural network and long short-term memory implemented into multiple current-voltage (MCV-CNN_LSTM)2024

    • Author(s)
      Daisuke Saito, Yosephus Ardean Kurnianto Prayitno, Prima Asmara Sejati, Shuichiro Miwa, Masahiro Takei
    • Journal Title

      Flow Measurement and Instrumentation

      Volume: 97 Pages: 102593-102593

    • DOI

      10.1016/j.flowmeasinst.2024.102593

    • Related Report
      2023 Annual Research Report
    • Peer Reviewed / Int'l Joint Research
  • [Presentation] Physics-Informed Neural Networks for Two-Phase Flow Simulations: An Integrated Approach with Advanced Interface Tracking Methods2024

    • Author(s)
      W. Zhou, S. Miwa, K. Okamoto
    • Organizer
      日本原子力学会、2024年春の年会、近畿大学
    • Related Report
      2023 Annual Research Report
  • [Presentation] 垂直・傾斜・水平配管における気液二相流の, 多電極電流電圧装置と機械学習を用いた, 時空間ボイド率の可視化2023

    • Author(s)
      齊藤大輔, ヨセフス・アルディーノ, クルニアント・プライトノ, 三輪修一郎, 武居昌宏
    • Organizer
      動力・エネルギーシンポジウム 2023
    • Related Report
      2023 Annual Research Report
  • [Presentation] The Role of Nuclear Thermal hydraulics towards Carbon Neutrality: from Drift flux Model to Deep Learning2023

    • Author(s)
      S. Miwa
    • Organizer
      11th International Conference on Multiphase Flow (ICMF 2023)
    • Related Report
      2023 Annual Research Report
    • Int'l Joint Research / Invited
  • [Presentation] 2D SPATIAL/ 1D TEMPORAL VOID FRACTION ANALYSIS OF VERTICAL UPWELLING FLOW USING MULTIPLE CURRENT-VOLTAGE AND CONVOLUTIONAL,2023

    • Author(s)
      S. Y. Choobak, D. Saito, Y. A. K. Prayitno, S. Miwa, M. Takei
    • Organizer
      30th International Conference on Nuclear Engineering (ICONE30)
    • Related Report
      2023 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Two-Phase Flow-Induced Vibration in Various Flow Regimes2023

    • Author(s)
      S. Miwa
    • Organizer
      20th International Topical Meeting on Nuclear Reactor Thermal Hydraulics (NURETH-20)
    • Related Report
      2023 Annual Research Report
    • Int'l Joint Research / Invited
  • [Presentation] ディープラーニングによる物体検出・認識技術を適用した 気液二相流画像解析手法の開発2022

    • Author(s)
      江口 航平,三輪 修一郎,澤 和弘
    • Organizer
      日本原子力学会春の年会
    • Related Report
      2022 Annual Research Report

URL: 

Published: 2022-04-19   Modified: 2026-01-16  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi