期刊文献+

综合炉料熔滴行为的影响机制与预测模型的构建 被引量:2

THE INFLUENCE MECHANISM OF CHARGE DROPLET BEHAVIOR AND THE CONSTRUCTION OF PREDICTION MODEL WERE SYNTHESIZED
在线阅读 下载PDF
导出
摘要 利用FactSage软件,模拟计算了综合炉料熔融过程中生成的矿物质各相态组成,得出综合炉料在该温度下的渣铁比,通过绘制综合炉料渣铁含量比与温度的关系图,构建了炉料渣相生成温度、渣相完全熔化温度、渣相温度区间、易熔难熔物交接温度四种新的表征综合炉料的熔滴性能。从热力学角度分析MgO/Al_(2)O_(3)、碱度、还原度对综合炉料熔滴性能的影响机理,发现综合炉料MgO/Al_(2)O_(3)增加和还原度提高,其综合炉料熔滴性能提高,适当的碱度有利于改善综合炉料的熔滴特性。在此基础上开展了机器学习模型建立,发现基于SVM算法构建的模型能对综合炉料熔滴性能实现较好的预测效果,模型预测渣相生成温度时RMSE为2.38、R^(2)为0.87;预测渣相完全熔化温度时RMSE为2.32、R^(2)为0.97。 Using FactSage software,the phase composition of minerals generated during the melting process of the comprehensive charge is simulated and calculated,and the slag-iron ratio of the comprehensive charge at this temperature is obtained,and the slag phase generation temperature,the slag phase complete melting temperature,the slag phase temperature range,and the fusible refractory melt transfer temperature are constructed by drawing the relationship between the slag iron content ratio and the temperature of the comprehensive charge.From the thermodynamic point of view,the influence mechanism of MgO/Al_(2)O_(3),alkalinity and reduction degree on the droplet performance of the comprehensive charge was analyzed,and it was found that the increase of MgO/Al_(2)O_(3) and the reduction degree of the comprehensive charge increased and the droplet performance of the comprehensive charge was improved,and the appropriate alkalinity was conducive to improving the droplet characteristics of the comprehensive charge.On this basis,the machine learning model is established,and it is found that the model constructed based on SVM algorithm can achieve a good prediction effect on the droplet performance of the comprehensive charge,and the RMSE is 2.38 and R^(2) is 0.87 when the model predicts the slag phase generation temperature,and the RMSE is 2.32 and R^(2) is 0.97 when the slag phase is completely melted temperature.
作者 周新富 王振阳 张建良 江德文 张松 Zhou Xinfu;Wang Zhenyang;Zhang Jianliang;Jiang Dewen;Zhang Song(School of Metallurgical and Ecological Engineering,University of Science and Technology Beijing,Beijing 100083;School of Chemical Engineering,University of Queensland,Saint Lucia 4072)
出处 《河北冶金》 2023年第8期13-19,34,共8页 Hebei Metallurgy
关键词 熔滴行为 MgO/Al_(2)O_(3) 碱度 还原度 机器学习 SVM算法 RMSE R^(2) droplet behavior MgO/Al_(2)O_(3) alkalinity degree of reduction machine learning SVM algorithm RMSE R^(2)
作者简介 周新富(1999-),男,硕士,从事低碳炼铁技术研究,E-mail:M202110263@xs.ustb.edu.cn;通讯作者:王振阳(1989-),男,博士,副教授,现主要从事低碳炼铁技术研究,E-mail:wangzhenyang@ustb.edu.cn。
  • 相关文献

参考文献19

二级参考文献225

共引文献173

同被引文献22

引证文献2

二级引证文献2

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部