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

Machine Learning to Estimat the Behavior of a Batch Process

Research Project

Project/Area Number 21K04766
Research Category

Grant-in-Aid for Scientific Research (C)

Allocation TypeMulti-year Fund
Section一般
Review Section Basic Section 27020:Chemical reaction and process system engineering-related
Research InstitutionTokyo University of Agriculture and Technology

Principal Investigator

Yamashita Yoshiyuki  東京農工大学, 工学(系)研究科(研究院), 教授 (60200698)

Project Period (FY) 2021-04-01 – 2024-03-31
Project Status Completed (Fiscal Year 2023)
Budget Amount *help
¥4,160,000 (Direct Cost: ¥3,200,000、Indirect Cost: ¥960,000)
Fiscal Year 2023: ¥520,000 (Direct Cost: ¥400,000、Indirect Cost: ¥120,000)
Fiscal Year 2022: ¥390,000 (Direct Cost: ¥300,000、Indirect Cost: ¥90,000)
Fiscal Year 2021: ¥3,250,000 (Direct Cost: ¥2,500,000、Indirect Cost: ¥750,000)
Keywordsプロセスモニタリング / バッチプロセス / 化学プラント / 機械学習 / モデル化 / 動特性 / 予測モデル
Outline of Research at the Start

多品種の化学製品を作るためにはバッチプロセスが使われている.バッチプロセスの挙動を予測するモデルを作る際には,それぞれの品種ごとの運転・操業データを用いてモデルを作る必要があるが,それぞれの品種ごとの運転データが不足しているために高精度なモデルを作ることが困難であった.本研究は,機械学習の最新手法の考え方と,対象プロセスに対する化学工学的知識(物理モデル)とを併用することによって,高い精度の挙動予測モデルを構築するための実用的手法を開発しようとするものである.

Outline of Final Research Achievements

Batch plants are used in the chemical and pharmaceutical industries. They are important for quality control, efficiency improvement, and optimization. Recently, data-driven modeling techniques have been used to achieve these objectives. However, batch plants often produce many different products in small quantities, which makes it difficult to construct models with sufficient accuracy. We developed a data-driven method that can create highly accurate models with limited data. It uses data from different batches, including data from different products.

Academic Significance and Societal Importance of the Research Achievements

この研究はデータ駆動型機械学習を発展し,学習データの不足という問題を解決する新たなアプローチを提案しています.この手法は、少量のデータからも高精度な予測モデルを構築可能とし,機械学習の理論と実践のギャップを埋める学術的意義の高いものです.
社会的には,この研究成果は化学産業や製薬産業におけるバッチプラントの生産性や効率,品質,安全性を向上させることに直接寄与します.エネルギー消費の削減や原料の使用効率を高め,環境負荷の軽減にも繋がります.また,作業員のリスクを減少させ,より持続可能な製造業の実現を支援します.

Report

(4 results)
  • 2023 Annual Research Report   Final Research Report ( PDF )
  • 2022 Research-status Report
  • 2021 Research-status Report
  • Research Products

    (13 results)

All 2024 2023 2022 2021

All Journal Article (3 results) (of which Peer Reviewed: 3 results) Presentation (10 results) (of which Int'l Joint Research: 3 results,  Invited: 9 results)

  • [Journal Article] Multi-target regression via target combinations using principal component analysis2024

    • Author(s)
      Takafumi Yamaguchi, Yoshiyuki Yamashita
    • Journal Title

      Computers & Chemical Engineering

      Volume: 181 Pages: 108510-108510

    • DOI

      10.1016/j.compchemeng.2023.108510

    • Related Report
      2023 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Online Batch Process Monitoring with a Combination of Normal Operating History Data and Physical Knowledge2022

    • Author(s)
      Xia Junqing、Yamashita Yoshiyuki
    • Journal Title

      JOURNAL OF CHEMICAL ENGINEERING OF JAPAN

      Volume: 55 Issue: 1 Pages: 38-50

    • DOI

      10.1252/jcej.20we158

    • NAID

      130008143105

    • ISSN
      0021-9592, 1881-1299
    • Year and Date
      2022-01-20
    • Related Report
      2022 Research-status Report 2021 Research-status Report
    • Peer Reviewed
  • [Journal Article] Quality prediction for multi-grade batch process using sparse flexible clustered multi-task learning2021

    • Author(s)
      Takafumi Yamaguchi, Yoshiyuki Yamashita
    • Journal Title

      Computers & Chemical Engineering

      Volume: 150 Pages: 107320-107320

    • DOI

      10.1016/j.compchemeng.2021.107320

    • Related Report
      2021 Research-status Report
    • Peer Reviewed
  • [Presentation] バッチプロセスのスモールデータ環境での品質予測制御2024

    • Author(s)
      山口貴史,山下善之
    • Organizer
      化学工学会第89年会
    • Related Report
      2023 Annual Research Report
  • [Presentation] デジタル技術を駆使した 化学プラントの運転・保守2023

    • Author(s)
      山下善之
    • Organizer
      石油化学業界向けDX推進セミナー
    • Related Report
      2023 Annual Research Report
    • Invited
  • [Presentation] Digitalization in chemical plant operation and maintenance2023

    • Author(s)
      Yoshiyuki Yamashita
    • Organizer
      SII 2023
    • Related Report
      2022 Research-status Report
    • Int'l Joint Research / Invited
  • [Presentation] Expectations Toward Next-Generation Chemical Process Manufacturing2022

    • Author(s)
      Yoshiyuki Yamashita
    • Organizer
      SICE 2022 Workshop
    • Related Report
      2022 Research-status Report
    • Int'l Joint Research / Invited
  • [Presentation] 化学工学とデータ科学の融合 ~AI・IoT・DXの視点から~2022

    • Author(s)
      山下善之
    • Organizer
      化学工学会第87年会
    • Related Report
      2022 Research-status Report
    • Invited
  • [Presentation] Smart Factory2022

    • Author(s)
      山下善之
    • Organizer
      ISPE日本本部2022年度年次大会
    • Related Report
      2022 Research-status Report
    • Invited
  • [Presentation] 化学産業のデジタルトランスフォーメーション2022

    • Author(s)
      山下善之
    • Organizer
      化学工学会第87年会
    • Related Report
      2021 Research-status Report
    • Invited
  • [Presentation] Physical-Principle Based Extended Attributes for Process Fault Detection2021

    • Author(s)
      Junqing Xia and Yoshiyuki Yamashita,
    • Organizer
      化学工学会第52回秋季大会
    • Related Report
      2021 Research-status Report
    • Invited
  • [Presentation] Digital Transformation in the Chemical Industry2021

    • Author(s)
      Yoshiyuki Yamashita
    • Organizer
      iCo-CSET 2021
    • Related Report
      2021 Research-status Report
    • Int'l Joint Research / Invited
  • [Presentation] 化学プラントのDXの現状と提言2021

    • Author(s)
      山下善之
    • Organizer
      INCHEM Tokyo 2021 特別講演会
    • Related Report
      2021 Research-status Report
    • Invited

URL: 

Published: 2021-04-28   Modified: 2025-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi