Design of inner and outer gray-box models to predict molten steel temperature in Tundish

Iftikhar Ahmad, Manabu Kano, Shinji Hasebe, Hiroshi Kitada, Noboru Murata

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Abstract

    In order to realize stable production in the steel industry, it is important to control molten steel temperature in a continuous casting process. The present work aims to develop a gray-box model that predicts the molten steel temperature in the tundish (TD temp). In the proposed approach, two parameters in the first-principle model, i.e., overall heat transfer coefficients of ladle and tundish, are optimized for each past batch separately, then the relationship between the two parameters and measured process variables is modeled through random forests (RF). In this inner gray-box model, the statistical models update the physical parameters according to the operating condition. To enhance the accuracy of TD temp estimation, another RF model is developed which compensates errors of the inner gray-box. The proposed approach was validated through its application to real operation data at a steel work.

    Original languageEnglish
    Title of host publicationIFAC Proceedings Volumes (IFAC-PapersOnline)
    PublisherIFAC Secretariat
    Pages744-749
    Number of pages6
    Volume10
    EditionPART 1
    DOIs
    Publication statusPublished - 2013
    Event10th IFAC Symposium on Dynamics and Control of Process Systems, DYCOPS 2013 - Mumbai, India
    Duration: 2013 Dec 182013 Dec 20

    Other

    Other10th IFAC Symposium on Dynamics and Control of Process Systems, DYCOPS 2013
    CountryIndia
    CityMumbai
    Period13/12/1813/12/20

    Fingerprint

    Molten materials
    Steel
    Temperature
    Iron and steel industry
    Continuous casting
    Heat transfer coefficients

    ASJC Scopus subject areas

    • Control and Systems Engineering

    Cite this

    Ahmad, I., Kano, M., Hasebe, S., Kitada, H., & Murata, N. (2013). Design of inner and outer gray-box models to predict molten steel temperature in Tundish. In IFAC Proceedings Volumes (IFAC-PapersOnline) (PART 1 ed., Vol. 10, pp. 744-749). IFAC Secretariat. https://doi.org/10.3182/20131218-3-IN-2045.00089

    Design of inner and outer gray-box models to predict molten steel temperature in Tundish. / Ahmad, Iftikhar; Kano, Manabu; Hasebe, Shinji; Kitada, Hiroshi; Murata, Noboru.

    IFAC Proceedings Volumes (IFAC-PapersOnline). Vol. 10 PART 1. ed. IFAC Secretariat, 2013. p. 744-749.

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Ahmad, I, Kano, M, Hasebe, S, Kitada, H & Murata, N 2013, Design of inner and outer gray-box models to predict molten steel temperature in Tundish. in IFAC Proceedings Volumes (IFAC-PapersOnline). PART 1 edn, vol. 10, IFAC Secretariat, pp. 744-749, 10th IFAC Symposium on Dynamics and Control of Process Systems, DYCOPS 2013, Mumbai, India, 13/12/18. https://doi.org/10.3182/20131218-3-IN-2045.00089
    Ahmad I, Kano M, Hasebe S, Kitada H, Murata N. Design of inner and outer gray-box models to predict molten steel temperature in Tundish. In IFAC Proceedings Volumes (IFAC-PapersOnline). PART 1 ed. Vol. 10. IFAC Secretariat. 2013. p. 744-749 https://doi.org/10.3182/20131218-3-IN-2045.00089
    Ahmad, Iftikhar ; Kano, Manabu ; Hasebe, Shinji ; Kitada, Hiroshi ; Murata, Noboru. / Design of inner and outer gray-box models to predict molten steel temperature in Tundish. IFAC Proceedings Volumes (IFAC-PapersOnline). Vol. 10 PART 1. ed. IFAC Secretariat, 2013. pp. 744-749
    @inproceedings{03f2d002d73744c68912b357d81aeba6,
    title = "Design of inner and outer gray-box models to predict molten steel temperature in Tundish",
    abstract = "In order to realize stable production in the steel industry, it is important to control molten steel temperature in a continuous casting process. The present work aims to develop a gray-box model that predicts the molten steel temperature in the tundish (TD temp). In the proposed approach, two parameters in the first-principle model, i.e., overall heat transfer coefficients of ladle and tundish, are optimized for each past batch separately, then the relationship between the two parameters and measured process variables is modeled through random forests (RF). In this inner gray-box model, the statistical models update the physical parameters according to the operating condition. To enhance the accuracy of TD temp estimation, another RF model is developed which compensates errors of the inner gray-box. The proposed approach was validated through its application to real operation data at a steel work.",
    author = "Iftikhar Ahmad and Manabu Kano and Shinji Hasebe and Hiroshi Kitada and Noboru Murata",
    year = "2013",
    doi = "10.3182/20131218-3-IN-2045.00089",
    language = "English",
    volume = "10",
    pages = "744--749",
    booktitle = "IFAC Proceedings Volumes (IFAC-PapersOnline)",
    publisher = "IFAC Secretariat",
    edition = "PART 1",

    }

    TY - GEN

    T1 - Design of inner and outer gray-box models to predict molten steel temperature in Tundish

    AU - Ahmad, Iftikhar

    AU - Kano, Manabu

    AU - Hasebe, Shinji

    AU - Kitada, Hiroshi

    AU - Murata, Noboru

    PY - 2013

    Y1 - 2013

    N2 - In order to realize stable production in the steel industry, it is important to control molten steel temperature in a continuous casting process. The present work aims to develop a gray-box model that predicts the molten steel temperature in the tundish (TD temp). In the proposed approach, two parameters in the first-principle model, i.e., overall heat transfer coefficients of ladle and tundish, are optimized for each past batch separately, then the relationship between the two parameters and measured process variables is modeled through random forests (RF). In this inner gray-box model, the statistical models update the physical parameters according to the operating condition. To enhance the accuracy of TD temp estimation, another RF model is developed which compensates errors of the inner gray-box. The proposed approach was validated through its application to real operation data at a steel work.

    AB - In order to realize stable production in the steel industry, it is important to control molten steel temperature in a continuous casting process. The present work aims to develop a gray-box model that predicts the molten steel temperature in the tundish (TD temp). In the proposed approach, two parameters in the first-principle model, i.e., overall heat transfer coefficients of ladle and tundish, are optimized for each past batch separately, then the relationship between the two parameters and measured process variables is modeled through random forests (RF). In this inner gray-box model, the statistical models update the physical parameters according to the operating condition. To enhance the accuracy of TD temp estimation, another RF model is developed which compensates errors of the inner gray-box. The proposed approach was validated through its application to real operation data at a steel work.

    UR - http://www.scopus.com/inward/record.url?scp=84896336743&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84896336743&partnerID=8YFLogxK

    U2 - 10.3182/20131218-3-IN-2045.00089

    DO - 10.3182/20131218-3-IN-2045.00089

    M3 - Conference contribution

    AN - SCOPUS:84896336743

    VL - 10

    SP - 744

    EP - 749

    BT - IFAC Proceedings Volumes (IFAC-PapersOnline)

    PB - IFAC Secretariat

    ER -