Research On Collaborative Optimization Of Blast Furnace Fuel Ratio Based On Energy Balance

Blast furnace is one of the most important industrial objects in the steel industry,and it is also the main source of high energy consumption and high emissions in China’s steel industry.Facing the severe situation of energy conservation and consumption reduction in the iron and steel industry,the development of blast furnace ironmaking towards high efficiency,low consumption and automation is imperative.Blast furnace coal injection technology is a technology widely used in modern blast furnace ironmaking production,and has become one of the indispensable important means for the regulation of the lower part of the blast furnace.Replacing expensive and increasingly scarce coke with low-cost coal powder not only reduces the coke ratio of blast furnace ironmaking,saves costs,but also reduces energy consumption and pollution in coking production.Therefore,increasing the amount of coal injection to replace part of the coke is an effective measure for blast furnace production to reduce energy consumption and save costs.At present,the blast furnace smelting coal injection operation still uses the manual mode,that is,the operator decides the coal injection setting value based on the process indicators and smelting knowledge,with the accumulated experience,and increases or decreases the coal injection amount according to the furnace temperature and furnace condition.However,due to the complexity,hysteresis and variability of the blast furnace smelting process,the coal injection settings are artificially given by the furnace chief,and there are problems such as blindness and roughness,which cannot accurately determine the best injection under the current furnace conditions.Coal settings.At the same time,in the blast furnace smelting process,each unit is operated in isolation,lack of collaborative optimization,which seriously affects the efficiency,safety and reliability of the operation of large blast furnaces,making it difficult to achieve the optimization control goals of low consumption,high output and high quality.Therefore,using the expert knowledge and process data of the blast furnace smelting process to establish an operation optimization control model is a hot issue in the field of metallurgy and control research,and it is also a difficult problem to be solved urgently.In response to the above problems,this paper mainly conducts research on collaborative optimization decision-making of blast furnace fuel ratio based on energy balance.The main work is as follows:(1)Description and research plan of blast furnace smelting problems.Through reading and studying a lot of literature and theoretical knowledge,the influence of blast furnace smelting technology and coal injection technology on blast furnace smelting is reviewed.The control mode of the blast furnace smelting process is described.According to the characteristics of the control mode,the layered optimization method is used to decouple the optimizing the operation of the blast furnace.By analyzing the energy balance and collaborative optimization in the optimizing process of blast furnace operation,a research plan for collaborative optimization decision of BF fuel ratio based on energy balance is proposed.(2)Data analysis and processing of blast furnace smelting process.Because the blast furnace production data collected on site has the characteristics of large data volume,different sampling periods,and large influences from external factors,data processing is indispensable.The blast furnace production process data should be taken as the research object,and the blast furnace smelting process data analysis and processing should be carried out from three aspects: variable selection,data preprocessing,and data dimensionality reduction.Among them,the variable selection is determined according to the production status and expert experience;data preprocessing includes outlier detection,missing value repair and data normalization,focusing on the methods used in the process of data preprocessing;data reduction mainly uses correlation The analysis method and correlation analysis of the blast furnace process data after data preprocessing.(3)The optimization model of the setting value of coal injection is established.Using the blast furnace production data collected by a steel plant,a radial basis neural network based on Kmeans clustering is used to establish an optimized target correlation model(fuel ratio,coal ratio prediction model),through which the coal injection amount and other production process parameters Associated with optimization goals(fuel ratio,coal ratio).A time series-based radial basis neural network is used to establish a furnace temperature prediction model(a model for predicting molten iron temperature and silicon content),and the furnace temperature prediction index is obtained through the model.(4)Optimization of the setting value of pulverized coal injection amount based on the optimal fuel ratio.With the optimal fuel ratio(minimum fuel ratio,maximum coal ratio)as the optimization goal,the optimization target correlation model as the objective function,the furnace temperature prediction index as the constraint condition,and the amount of coal injection as the decision variables,single-objective and multi-objective optimization are established respectively.The model is optimized by NSGA-Ⅱ algorithm.After that,the optimization effects of singleobjective and multi-objective optimization are compared,and after comparison,it is concluded that multi-objective optimization is more suitable for optimization of the setting value of the blast furnace coal injection amount based on the optimal fuel ratio.(5)Multi-objective optimization based on DE and NSGA-Ⅱ hybrid evolutionary algorithms.The differential evolution strategy is introduced into the NSGA-Ⅱ algorithm to form the DE and NSGA-Ⅱ hybrid evolution algorithm(DE-NSGA-Ⅱ algorithm).Multi-objective optimization based on the optimal fuel ratio(the smallest fuel ratio and the largest coal ratio)is performed by using the established multi-objective optimization model and the DE-NSGA-Ⅱ algorithm.Then,the optimization effects of DE-NSGA-Ⅱ algorithm and NSGA-Ⅱ algorithm are compared,and the comparison shows that: DE-NSGA-Ⅱ algorithm is used to optimize the optimization of the blast furnace coal injection setting value based on the optimal fuel ratio.Better results.Therefore,a multi-objective optimization method based on the DE-NSGA-Ⅱ algorithm should be used to optimize the coal injection volume setting value,and the optimal coal injection setting value under the current furnace condition should be determined according to the optimization target.This subject takes a large blast furnace of a steel plant as the research object,and proposes a research plan for collaborative optimization of blast furnace fuel ratio based on energy balance.The optimal fuel ratio(the smallest fuel ratio and the largest coal ratio)is used as the optimization goal to perform blast furnace coal injection.Optimization of the setting value of the furnace,the optimal coal injection setting value under the current furnace conditions was determined,and the purposes of reducing energy consumption and saving costs were achieved.It has important scientific significance and broad significance for the research on the optimization control of the blast furnace smelting system.Application prospects.