Study On Prediction Of Parameters And Optimization Control Model Of Blast Furnace Iron Making

With the influence of international environmental change and the economy crisis,there is huge change in the stainless steel.Energy saving has been the key of the stainless steel industry.In the industry,the iron front system,as the large energy consumption,takes 70% of the total energy consumption of iron and steel enterprises,while blast furnace takes the 60%of the energy of consumption of iron front system.Therefore,the blast furnace takes larger percentage of the stain steel industry consumption.Hence,it is the most effective of the key work of energy saving to begin with blast furnace in the industry.The paper will have a comprehensive analysis of blast furnace iron making process and mechanism of blast furnace.Based on it and combined with qualitative analysis method and quantitative analysis method,the particle swarm optimization algorithm should be applied to build up the least squares support vector machine algorithm as the main body of the blast furnace coke ratio prediction model and a generalized regression neural network algorithm for the main body of the blast furnace temperature prediction model of molten iron as well as prediction model of hot metal production in blast furnace respectively.And then do the prediction and analysis of the into the blast furnace coke rate,temperature of molten iron in blast furnace and production of hot metal in blast furnace.Finally,the multi objective optimization control model for blast furnace iron making is designed according to the prediction model of blast furnace.The optimal control model makes the model of taking into the blast furnace coke rate prediction and the model of blast furnace production as objective function.The temperature constraint of molten iron in blast furnace and upper and lower bounds of decision variables are constraint condition and have a comprehensive optimization of multiple targets of blast furnace production.Improved multi objective particle swarm optimization algorithm is used to search and choose the solution set of multi objective optimization control model for blast furnace iron making to ensure the correctness of optimal solution set of multi objective optimization control model for blast furnace iron making.Finally,the effectiveness and correctness of into the blast furnace coke rate prediction model and the prediction model of blast furnace production model,blast furnace temperature through simulation analysis.The built prediction model of blast furnace has important reference significance to the energy management of blast furnace iron making process and production planning.Besides,it has designed the multi objective optimization control of blast furnace iron making and prove the feasibility of the method from theory,which has great meaning in the energy saving and optimal scheduling in the blast furnace production process.