Prediction of the end-point carbon content and temperature of AOD furnace based on LAOA-DeepSCNs

The end-point carbon content and temperature in the steelmaking process of AOD furnace are important factors affecting product quality, and the traditional algorithms have the problems of slow convergence, being easy to fall into the local optimal solution and lack a unified parameter selection criterion, which leads to the problems of slow convergence and low prediction accuracy. To solve the above problems, the research group adopts the improved arithmetic optimization algorithm and deep stochastic configuration networks (LAOA-DeepSCNs) to predict the end-point carbon content and temperature of the AOD furnace. First, correlation analysis was performed with SPSS to identify the seven factors as model inputs. Second, to verify the prediction effect of the model, the algorithm is compared with three typical algorithms: BP, RBF and SCN. The results show that LAOA-DeepSCNs have the fastest convergence speed, the highest prediction accuracy, and the strongest generalization ability. Finally, the model was applied to the actual production of a steel mill, and the results showed that the hit rate is 90.8%, 86.4%; and 92.6%, 88.1% for refining end-point carbon content and end-point temperature error within ±0.015%, ±0.01%; and ±10 °C, ±5 °C, respectively. Which can well meet the practical needs of a steel mill. It also provides theoretical guidance for the control of carbon content and temperature at the end-point of the AOD furnace.