摘要
通过研究高炉-转炉界面铁水运输过程温度的主要影响因素,确定了影响高炉-转炉界面铁水运输过程温度的参数,建立了基于Levenberg-Marquardt(LM)算法BP神经网络的高炉-转炉界面铁水温度及铁水过程温降的预报模型。用沙钢100包铁水数据进行模型训练,50包铁水数据进行现场预报,结果表明:在高炉-转炉界面"一包到底"模式下,当绝对误差│X│≤20℃时,铁水温度命中率为94%,铁水温降命中率为78%;当绝对误差│X│≤40℃时,铁水温度命中率为100%,铁水温降命中率为92%,该预报模型能够满足现场实际生产需求,对炼钢生产有很好的指导意义。
Through studying the main influencing factor of hot metal transportation process temperature for BF-BOF interface,the main parameters affecting temperature of hot metal transportation process for BF-BOF interface was determined,and a prediction model of hot metal temperature for BF-BOF interface was established based on Levenberg-Marquardt(LM) algorithm of BP neural network.The data of 100 ladles were used to training the model and the other 50 ladles were selected as the predictive samples.It is shown that: under the model of 'one hot metal ladle going through process' for BF-BOF interface,when the absolute error│X│≤20 ℃,the temperature of hot metal is shooting 94%,the hit rate of temperature drop of hot metal is 78%;when the absolute error│X│≤40 ℃,the temperature of hot metal is shooting 100%,the hit rate of temperature drop of hot metal is 92%,this prediction model can meet the actual production needs and can provide a very good guide to steel-making production.
出处
《钢铁》
CAS
CSCD
北大核心
2012年第9期40-42,49,共4页
Iron and Steel
关键词
温度
BP神经网络
LM算法
预报模型
temperature
BP neural network
LM algorithm
predictive model
