Fuzzy based similarity adjustment of case retrieval process in CBR system for BOF oxygen volume control
Xinzhe Wang, Jie Dong
Abstract
Xinzhe Wang, Jie Dong
Abstract
Oxygen volume is the most important control in BOF (Basic Oxygen Furnace) steelmaking production and the control accuracy affects the quality of liquid steel directly. In this study, a CBR (Case-based Reasoning) method is adopted to calculate the oxygen blowing volume in the second period of BOF steelmaking production. When retrieve the similar cases from the case base, a similarity reward and punish strategy is introduced to make the retrieved similar cases more effective. Similarity reward and punish strategy is based on the fuzzy membership to enhance the similarity of relatively more successful cases. The ultimate goal of introducing the strategy is to retrieve more useful similar cases and improve the model accuracy. Tests are implemented on a practical 180t converter in a steel plant and results show that this CBR system for BOF oxygen volume control is feasible and effective.
OpenAlex reports 7 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Oxygen volume is the most important control in BOF (Basic Oxygen Furnace) steelmaking production and the control accuracy affects the quality of liquid steel directly. In this study, a CBR (Case-based Reasoning) method is adopted to calculate the oxygen blowing volume in the second period of BOF steelmaking production. When retrieve the similar cases from the case base, a similarity reward and punish strategy is introduced to make the retrieved similar cases more effective. Similarity reward and punish strategy is based on the fuzzy membership to enhance the similarity of relatively more successful cases. The ultimate goal of introducing the strategy is to retrieve more useful similar cases and improve the model accuracy. Tests are implemented on a practical 180t converter in a steel plant and results show that this CBR system for BOF oxygen volume control is feasible and effective.
Key concepts: Steelmaking, Volume (thermodynamics), Similarity (geometry), Basic oxygen steelmaking, Process (computing), Computer science, Fuzzy logic, Liquid steel