Control Engineering Practice 78 (2018) 186–195
Contents lists available at ScienceDirect
Control Engineering Practice
journal homepage: www.elsevier.com/locate/conengprac
An energy efficient decision-making strategy of burden distribution for
blast furnace
Min Wu
a,b,
*, Kexin Zhang
a,b
, Jianqi An
a,b
, Jinhua She
a,b,c
, Kang-Zhi Liu
a,b,d
a
School of Automation, China University of Geosciences, Wuhan 430074, China
b
Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems, Wuhan 430074, China
c
School of Engineering, Tokyo University of Technology, Hachioji, Tokyo 192-0982, Japan
d
Department of Electrical and Electronic Engineering, Chiba University, Chiba 263-8522, Japan
A R T I C L E I N F O
Keywords:
Burden distribution
Blast furnace
Decision-making strategy
Support vector regression
Case-matching model
A B S T R A C T
Burden distribution plays an important role in the optimization of energy-consuming index of a blast furnace (BF).
However, due to the complex mechanism of the burden distribution process and the poor understanding of this
process, it is difficult for operators to make a suitable decision for burden distribution parameters. In this paper,
a decision-making strategy of burden distribution parameters is devised for improving energy-consuming index,
where carbon-monoxide utilization rate (CMUR) is taken as the energy-consuming index. Firstly, the support
vector regression is used to build a model between the burden distribution parameters and BF state variables
that are closely related to CMUR. Then, a probability-based case-matching model is built by the historical data
to predict the trend of CMUR’s change. Finally, based on the trend of CMUR’s change, a decision for burden
distribution parameters is obtained. Simulation results based on industrial data show that the devised decision-
making strategy provides a good guide on making a suitable decision for burden distribution parameters.
1. Introduction
A blast furnace (BF) is a complex metallurgical reactor that converts
iron ore into liquid pig iron through a series of chemical and physical
changes. Energy consumption is one of the most important factors in
ironmaking, and is closely related to the state of the BF. Here the
state is the combination of the variables that are used to describe the
chemical and physical changes in the BF, and these variables include top
temperature, permeability index, etc. A suitable distribution of gas flow
in the BF creates a smooth state, and is fundamental for the reduction of
energy consumption. In general, the distribution of gas flow is closely
related to the burden distribution. Therefore, a correct decision of the
burden distribution has great significance for the reduction of energy
consumption.
In order to reduce the energy consumption in a BF, many researchers
focused on building models to analyze energy consumption from the
viewpoint of carbon efficiency. The proposed methods for reducing
the energy consumption in a BF were mainly through adjusting the
burden distribution or improving the properties of raw materials. For
example, various types of iron ore were taken as the pre-reduced input
materials to reduce the carbon dioxide emissions (Yilmaz & Turek,
2017). Carbon-monoxide utilization rate (CMUR) was often used as
*
Corresponding author at: School of Automation, China University of Geosciences, Wuhan 430074, China.
E-mail address: wumin@cug.edu.cn (M. Wu).
an energy-consuming index because it reflects the degree of energy
exchange inside a BF and has real-time character suitable for monitoring
the state of a BF. For example, a chaotic analysis method was presented
to study the characteristics of CMUR (Xiao, An, He, & Wu, 2017).
Watakabe, Takeda, Nishimura, Goto, Nishimura, Uchida, and Kiguchi
(2006) used a high-ratio ore-mixed-coke charging technique to reduce
the coke ratio.
For the decision-making of the burden distribution parameters,
some researchers analyzed the burden distribution process and made
a decision of burden distribution. A mathematical charging model was
used to determine the best charging pattern according to the overall
geometry stock line (Radhakrishnan & Ram, 2001). Kuang, Li, Yan,
Qi, and Yu (2014) carried out a numerical study of multiphase flow,
heat and mass transfer, showing that the hot charging improved the
productivity and reduced the comprehensive coke ratio and carbon
dioxide emission. Focusing on the nonuniform descending speed in the
burden distribution process, a mathematical model for estimating the
burden distribution was developed (Fu, Chen, & Zhou, 2015). Zhou,
Li, Shi, and Zhou (2016) proposed a two-dimensional temperature
distribution model to analyze the burden distribution process. Wu, Kou,
Xu, Guo, Du, Shen, and Sun (2013); Yu and Saxén (2010) used a discrete
https://doi.org/10.1016/j.conengprac.2018.06.019
Received 3 January 2018; Received in revised form 23 April 2018; Accepted 27 June 2018
Available online 6 July 2018
0967-0661/© 2018 Elsevier Ltd. All rights reserved.