In an integrated steelworks the Basic Oxygen Steelmaking (BOS) process is required for the refining of molten iron from a
blast furnace to produce steel. The development of an expert diagnostic system is considered in the context of the initial
phase of BOS operation, the loading and operation of the BOS vessel. In this part of the steelmaking process the BOS
vessel is charged with the molten iron, scrap metal and fluxes which are there to facilitate the capture of impurities by
forming slag. The nature of the elements added requires knowledge of the steelmaking process, the actual state of the
contents of the vessel and the available process management options. The expert system produced to oversee this process
exhibits the capability of dealing with both continuous and batch data, combining the two together to aid effective decision
making and management. Fuzzy inference is used in the main diagnostic system due to the large rule base required to
diagnose faults and infer a process state. The operation of the system and its use by the process operators and the
application of this approach into other areas of the steelworks is considered in this paper.
[1]
Rajesh Ransing,et al.
Application of neural computing in basic oxygen steelmaking
,
2002
.
[2]
Yao Zhao,et al.
Dynamic control model of BOF steelmaking process based on ANFIS and robust relevance vector machine
,
2011,
Expert Syst. Appl..
[3]
Meng Zhang,et al.
A model of basic oxygen furnace (BOF) end-point prediction based on spectrum information of the furnace flame with support vector machine (SVM)
,
2011
.
[4]
W. Lankford.
The Making, shaping, and treating of steel
,
1985
.
[5]
G. Horvath,et al.
A Sparse Robust Model for a Linz–Donawitz Steel Converter
,
2009,
IEEE Transactions on Instrumentation and Measurement.
[6]
Jun Wang,et al.
Applying input variables selection technique on input weighted support vector machine modeling for BOF endpoint prediction
,
2010,
Eng. Appl. Artif. Intell..