PCA和PSO-ELM在高炉铁水硅含量中的预测仿真

炉温的实时预测技术对高炉运转具有重要意义。在高炉炼铁过程中,通常以铁水硅含量来表征高炉热状态。针对硅含量预测效率和精度不足的问题,提出主成分分析和粒子群改进的极限学习机相结合的方法对高炉铁水硅含量进行预测。由于影响铁水硅含量的因素众多,且各因素之间相互影响,通过主成分分析对影响硅含量的输入变量进行降维处理。利用粒子群算法来优化极限学习机的权值和阈值,并以均方根误差作为适应度函数建立预测模型。将提取出的主成分作为模型输入,铁水硅含量作为模型输出。最后比较了极限学习机算法和粒子群改进的极限学习机,实验结果表明改进后的预测模型提高了硅含量预测的准确度,上述方法可为高炉的生产操作提供参考。

Prediction and Simulation of Silicon Content in Blast Furnace for PCA and PSO-ELM

The real-time prediction technology of furnace temperature is of great significance to the operation of blast furnace.In the process of blast furnace ironmaking,the silicon content of molten iron is usually used to characterize the thermal state of the blast furnace.In view of the shortage of efficiency and accuracy in silicon content prediction,this paper presents a method combining principal component analysis with particle swarm optimization based extreme learning machine to predict silicon content in molten iron of blast furnace.Because there are many factors that affect the silicon content in hot metal,and the factors influence each other,the input variables affecting silicon content were reduced through principal component analysis;the weights and thresholds of the extreme learning machine were optimized with particle swarm optimization algorithm,and the prediction model was established with root mean square error as fitness function.The extracted principal component was used as model input and molten iron silicon content as model output.Finally,a comparison between the extreme learning machine algorithm and the particle swarm optimization extreme learning machine was carried out.The experimental results show that the improved prediction model can improve the accuracy of silicon content prediction.This method can provide a reference for the production and operation of blast furnace.