A Decision-making Strategy Of Blast Furnace Burden Distribution For Energy Consumption Index Optimization

Burden distribution is a common adjustment method for the iron-making process,and is an important means to maintain the stable condition of the blast furnace.Blast furnace is a complex "black box" system with few key detection information and complicated mechanism,the relationship between the burden distribution and conditions is unknown.It is difficult to adjust the blast furnace condition by adjusting the burden distribution parameters to optimize the production index.Therefore,analysis of the characteristics of the production process,extraction of production data characteristics,and quantitative relationship between blast furnace burden distribution parameters and production indexes are of great significance for steel smelting.Blast furnace is a device of iron and steel industry,which is the main source of energy consumption and pollution.Burden distribution is the most active and the most critical means during ironmaking process,which ensures steady,effective and energyefficient running of blast furnace.Since blast furnace is a highly complex ‘black box’ which lacks internal detection,clear mechanism,and accurate models,it is fully hard to clear the relationship of burden distribution,state of blast furnace and production indexes.Thus,burden distribution is scarcely adjusted scientifically.Deep analysis of the characteristics of blast furnace site data and clear quantitative relationship among burden distribution,blast furnace condition and production indexes to optimize burden distribution reasonably are of significant importance for iron and steel enterprises and the whole society.First,the principle of blast furnace ironmaking is analyzed and gas utilization rate is selected as an energy consumption index.Mutual information method is used to choose important condition variables that are closely related to the gas utilization rate.Then,the burden distribution process and its characteristics are analyzed,and a simplified expression based on calculating ore coke ratio is designed.Then,a classification standard of blast furnace condition is established,and fuzzy clustering method is used to classify the blast furnace condition.Based on the above analysis,a burden distribution decisionmaking strategy for energy consumption index optimization is designed.Second,aiming at the unclear relationship between the burden distribution parameters and the gas utilization rate,support vector regression and case-matching method are used to build models for predicting the cooling temperature and the gas utilization.Based on the proposed models,a burden distribution decision-making method based on ore coke ratio is designed.This method determines the burden distribution parameters by two steps.The first step is to make a decision on the ore coke ratio that obtained from the edge of the blast furnace,and the second step is to determine the central ore coke ratio according to the gas utilization rate.The simulation experiments with field data show that the models can predict the gas utilization rate and the cooling temperature,which verify the effectiveness of the proposed method.Finally,a blast furnace simulation experiment system based on the industrial data is developed,a monitoring configuration environment is constructed,and a burden distribution decision-making software is designed.The design of human-computer interaction interface module,communication interface module,data management module and algorithm module are completed to develop the function of the system software.The system is applied in the industrial field,and the process monitoring and burden distribution control are realized by collecting field operation data.The results show that the system can make effective decision and control on burden distribution,which lays a foundation for further optimization of the blast furnace operation.