Development And Optimization Of 1700 Heating Furnace Model Of Tangsteel Company

As one of the important equipments in the iron and steel metallurgy industry,the heating furnace is also the main energy consuming equipment in the steel production process.The improvement of the efficiency of the heating furnace and the reduction of energy consumption are of great significance to the entire steel industry.As a complex controlled object with many variables,time-varying,nonlinear,strong coupling,large inertia and pure hysteresis,the heating furnace contains various processes of thermodynamics,chemistry and physics.Therefore,the modeling of heating furnace is of great significance to the iron and steel metallurgy industry.In this paper,the 1700 production line in Tangsteel' no.1 heating furnace as the actual background,using the actual data,to establish a billet temperature prediction model,and on this basis,the use of optimization algorithm to establish optimal temperature setting model to achieve the actual production of energy conservation and improve the passing rate of product.Compared MATLAB simulation with actual data,the more suitable temperature prediction model and furnace temperature optimization model of furnace billet were obtained.The billet temperature prediction model is mainly established by BP neural network and support vector machine,and simulated by MATLAB.By comparing the predicted value of the two models with the actual value,it can be concluded that under the condition of relatively sufficient actual data,BP neural network's prediction effect of is better than SVM,so this paper chooses the model built by BP neural network as prediction model of billet temperature.In the establishment of the optimal temperature sett model,this paper mainly compares the optimization results of two intelligent algorithms: genetic algorithm and fireworks algorithm.After comparison,the optimal temperature of the genetic algorithm is lower than the actual temperature and the optimized value of the fireworks algorithm under the condition of meeting the heating requirement.After the two models are confirmed,a GUI interface is established by MATLAB to facilitate the manual operation.At present,the billet temperature prediction model and the temperature setting model have been applied in actual production.