摘要
转炉煤气发生量预测为转炉炉口微差压调控、转炉煤气回收质量提升和全厂煤气平衡调度提供重要支撑。以某钢厂吹炼过程转炉煤气实际发生量数据为依据,采用深度学习方法,建立了BP神经网络、LSTM长短记忆神经网络和RBFNN径向基函数神经网络3种转炉煤气发生量的预测模型,对比分析预测步数、输入样本量和隐藏单元数3个参数对预测模型精度和计算效率的影响。研究结果表明,模型的预测精度均随着预测步数的增加而减小,选择30步预测更符合钢厂的实际需求。随着样本输入量的增加,LSTM的精度无明显变化,BP的精度呈降低趋势,RBF的精度先大幅提高后缓慢降低。LSTM的预测效率无明显变化,BP大幅降低,RBF无变化。当3种模型各自在最佳样本输入量和30步预测条件下,随着隐藏单元数的增加,LSTM的精度基本不变,BP先略有升高再缓慢降低,RBF先大幅升高之后保持平稳,后面再大幅降低。LSTM的预测效率小幅降低,BP大幅降低,RBF保持不变。最终,在30步预测的条件下,LSTM、BP、RBF 3种模型的最佳参数条件为,LSTM样本输入量为125,隐藏单元数为135,此时均方根误差ERMS最小为13.38,训练时长为4.7min;BP样本输入量为50,隐藏单元数为60,此时ERMS最小为31.46,训练时长为16.8min;RBF样本输入量为210,隐藏单元数为210,此时ERMS最小为2.07,训练时长为1.2min,与实际数据相比,RBF预测效果最好。采用转炉煤气发生量预测结果调控风机的转速,可以使炉口微差压保持在更稳定的状态,减少吸风量,提高回收煤气热值。
The prediction of the occurrence of converter gas provides important support for the micro differential pressure control at the converter inlet,the improvement of converter gas recovery quality,and the overall gas balance scheduling of the plant.Based on the actual occurrence data of converter gas during the blowing process of a certain steel plant,a deep learning method was used to establish three prediction models for converter gas occurrence:BP neural network,LSTM long short memory neural network,and RBFNN radial basis function neural network.The effects of three parameters,namely prediction steps,input sample size,and hidden unit number,on the accuracy and computational efficiency of the prediction model were compared and analyzed.The research results indicate that the prediction accuracy of the model decreases with the increase of prediction steps,and choosing 30step prediction is more in line with the actual needs of steel mills.As the sample input increases,there is no significant change in the accuracy of LSTM,while the accuracy of BP shows a decreasing trend.The accuracy of RBF first increases significantly and then slowly decreases.The prediction efficiency of LSTM showed no significant change,BP significantly decreased,and RBF remained unchanged.When the three models are under the optimal sample input and 30step prediction conditions,the accuracy of LSTM remains basically unchanged as the number of hidden units increases.BP first slightly increases and then slowly decreases,while RBF first increases significantly and then remains stable,and then decreases significantly.The prediction efficiency of LSTM has slightly decreased,BP has significantly decreased,and RBF remains unchanged.Finally,under the condition of 30step prediction,the optimal parameter conditions for LSTM,BP,and RBF models are as follows:LSTM sample input quantity is 125,hidden unit number is 135,ERMSminimum is 13.38,and training duration is 4.7min;The input amount of BP samples is 50,the number of hidden units is 60,and the minimumERMSi
作者
包向军
陈凯
郦秀萍
杨筱静
刘骁
陈光
BAO Xiangjun;CHEN Kai;LI Xiuping;YANG Xiaojing;LIU Xiao;CHEN Guang(School of Energy and Environment,Anhui University of Technology,Maanshan 243000,Anhui,China;Steel Industry Green and Intelligent Manufacturing Technology Center,China Iron and Steel Research Institute Group Co.,Ltd.,Beijing 100081,China;Iron and Steel Research Institute Co.,Ltd.,Beijing 100081,China)
出处
《钢铁》
CAS
CSCD
北大核心
2024年第1期67-74,共8页
Iron and Steel
基金
国家重点研发计划资助项目(2020YFB1711101)。
关键词
吹炼过程
转炉煤气
发生量预测
深度学习
模型对比分析
blowing process
converter gas
prediction of occurrence volume
deep-learning
comparative analysis of model