论文部分内容阅读
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 the study of the main influencing factors on the temperature of hot metal during the blast furnace - converter interface, the parameters influencing the temperature during hot metal - blast furnace interface transportation were determined. The temperature of hot metal - blast furnace interface based on Levenberg - Marquardt (LM) Prediction model of temperature drop in hot metal process. The model was trained with 100-pack molten iron in Shagang and the 50-pack hot metal data was used for on-site forecasting. The results showed that when the absolute error │X│≤20 ℃, the hot metal temperature hit in the blast furnace - converter interface When the absolute error │X│≤40 ℃, the hit rate of hot metal is 100%, and the hot metal hit rate is 92%. The forecast model can meet the actual production in the field Demand for steel production has a good guiding significance.