Software sensor for slab reheating furnace

Zhihua Xiong, Guohong Huang, Huihe Shao

分享

计量
  • 文章访问数:  221
  • HTML全文浏览量:  54
  • PDF下载量:  7
  • 被引次数: 0

目录

    Cite this article as:

    Zhihua Xiong, Guohong Huang, and Huihe Shao, Software sensor for slab reheating furnace, J. Univ. Sci. Technol. Beijing , 12(2005), No. 2, pp.123-127.
    Zhihua Xiong, Guohong Huang, and Huihe Shao, Software sensor for slab reheating furnace, J. Univ. Sci. Technol. Beijing , 12(2005), No. 2, pp.123-127.
    引用本文 PDF XML SpringerLink
    Metallurgy

    Software sensor for slab reheating furnace

    基金项目: 

    This work was financially supported by the National High-Tech Research and Development Program of China(No.2002AA4120I0).

      通信作者:

      Zhihua Xiong E-mail: zhxiong@sjtu.edu.cn

    It has long been thought that a reheating furnace, with its inherent measurement difficulties and complex dynamics, posed almost insurmountable problems to engineers in steel plants. A novel software sensor is proposed to make more effective use of those measurements that are already available, which has great importance both to slab quality and energy saving. The proposed method is based on the mixtures of Gaussian processes (GP) with the expectation maximization (EM) algorithm employed for parameter estimation of the mixture of models. The mixture model can alleviate the computational complexity of GP and also accords with the changes of operating condition in practical processes. It is demonstrated by on-line estimation of the furnace gas temperature in 1580 reheating furnace in Baosteel Corporation (Group).

     

    Metallurgy

    Software sensor for slab reheating furnace

    Author Affilications
    • Funds: 

      This work was financially supported by the National High-Tech Research and Development Program of China(No.2002AA4120I0).

    • Received: 07 June 2004;
    It has long been thought that a reheating furnace, with its inherent measurement difficulties and complex dynamics, posed almost insurmountable problems to engineers in steel plants. A novel software sensor is proposed to make more effective use of those measurements that are already available, which has great importance both to slab quality and energy saving. The proposed method is based on the mixtures of Gaussian processes (GP) with the expectation maximization (EM) algorithm employed for parameter estimation of the mixture of models. The mixture model can alleviate the computational complexity of GP and also accords with the changes of operating condition in practical processes. It is demonstrated by on-line estimation of the furnace gas temperature in 1580 reheating furnace in Baosteel Corporation (Group).

     

    /

    返回文章
    返回