摘要
在总结国内外焦炭质量预测模型的基础上 ,采用改进的GMDH方法 ,以大量SCO(SimulatedCokeOven)炉炼焦试验数据为基础 ,建立了适用于宝钢的SCO炉焦炭质量预测模型 ,再校正到生产焦炉。实践表明 ,该模型能很好地预测大生产焦炭质量 ,指导宝钢炼焦配煤生产。预测平均偏差DI1 501 5为 0 1 6 % ,CRI 0 65 % ,CSR 0 82 %。
To satisfy the need of large volume blast furnace, a target value of 66% or higher for coke strength after reaction (CSR), 26% or lower for coke reactivity index (CRI) and 87% or higher for JIS drum strength (DI 150 15 ) have been set in Baosteel Co , LTD, In order to keep up with this target and control the coke making cost, coke quality prediction models for Baosteel Co circumstance were required Summarizing the international and domestic coke quality prediction methods and using evolved GHMP (group method data handle) method, the Baosteel Co coke quality prediction model was established on the basis of 16 coals and 64 blends carbonization results on simulated coke oven (SCO), then it is adjusted to the commercial coke oven One of the important contributions of the model is that the introduction of the mineral catalysis index (MCI) as a main factor has increased the accuracy of the model for prediction of the CSR and CRI The MCI takes into account not only the quantity of the mineral composition but also the catalysis ability including both positive and negative Application has proved that the derived model can predict the coke quality and conduct the blending of coal effectively The mean deviations of predicted values versus the measured ones of DI, CRI and CSR are 0 16%, 0 65% and 0 82% respectively
出处
《燃料化学学报》
EI
CAS
CSCD
北大核心
2002年第4期300-305,共6页
Journal of Fuel Chemistry and Technology
