摘要
为改进对铁矿粉基础特性预测的加权平均估算方法,本文选取15种常用的烧结铁矿粉进行同化性和液相流动性试验,并进行热力学计算;进而采用Levenberg-Marquardt和通用全局优化算法对铁矿粉的基础特性值与化学成分分别进行一维和二维数学模型拟合;最后采用随机的混矿方案对模型预测结果进行验证。结果表明:最低同化温度-开始熔化温度以及液相流动性指数-有效液相量之间有着较好的吻合程度;二维数学模型的预测在Pearson相关系数上有着更优的相关性,且预测值的平均误差也显著低于一维数学模型。在一定数量的矿种范围内,基于化学成分的二维数学模型对混匀矿基础特性的预测可以认为是有效的。
In order to improve the weighted average estimation method for the basis properties of iron ore powder,15 kinds of commonly used sintered iron ore powder are selected for assimilative and liquid phase mobility tests and thermodynamic calculations are conducted;then,Levenberg-Marquardt and generalized global optimization algorithms are used to fit one-and two-dimensional mathematical models for the basis properties values and chemical composition of iron ore powder,respectively;finally,the model prediction results are validated by using a randomized blending scheme.The results show that,the minimum assimilation temperature-onset melting temperature and the liquid fluidity index-effective liquid phase have a good agreement;the prediction of the two-dimensional mathematical model has a better Pearson correlation coefficient,and the average error of the predicted value is also significantly lower than that of the one-dimensional mathematical model.Within a certain range of minerals,the two-dimensional mathematical model based on chemical composition can be considered effective for predicting the basic properties of mixed ore.
作者
李占国
张建良
刘征建
王耀祖
牛乐乐
LI Zhanguo;ZHANG Jianliang;LIU Zhengjian;WANG Yaozu;NIU Lele(School of Metallurgical and Ecological Engineering,University of Science&Technology Beijing,Beijing 100083,China;Institute of Artificial Intelligence,University of Science&Technology Beijing,Beijing 100083,China)
出处
《烧结球团》
北大核心
2022年第2期38-45,共8页
Sintering and Pelletizing
基金
广东省区域联合基金-青年基金项目(2020A1515111008)
中央高校基本科研业务费(06500170)。
关键词
烧结基础特性
二维数学模型
化学成分
预测
basic properties of sintering
two-dimensional mathematical model
chemical composition
prediction
