摘要
为了更好地检测炉缸工作状态的变化,运用AdaBoost集成算法建立了炉芯温度提前预测模型。通过提前预测炉芯温度,可以捕捉炉缸状态的变化趋势,便于操控者及时调整。收集了承钢4号高炉相关参数的实时数据,以时间为轴进行整合对应,使用拉依达准则进行粗大异常值的检验,线性插值法进行空缺值的填补,再利用相关性系数对输入参数进行冗余剔除,最后运用AdaBoost集成树模型进行预测,发现其预测效果较之单棵决策树模型要更加精准。
In order to better detect changes of the working condition of hearth, the AdaBoost integrated algorithm was used to establish a model for predicting the furnace core temperature in advance. By predicting the core temperature in advance, the trend of the state of the hearth is captured, which is convenient for the operator to adjust in time. The real-time data of the relevant parameters of Chenggang No.4 blast furnace were collected, and the time was used as the axis to integrate and match the parameters. The PauTa criterion was used to test the large outliers, and the linear interpolation method was used to fill the gaps, and the input parameters were redundantly eliminated by the correlation coefficient. Using the AdaBoost model for prediction, it is found that the prediction effect is more accurate than that of the single decision tree model.
作者
王坤
刘小杰
刘二浩
李宏杨
刘颂
吕庆
WANG Kun;LIU Xiao-jie;LIU Er-hao;LI Hong-yang;LIU Song;Lü Qing(College of Metallurgy and Energy,Key Laboratory for Advanced Metallurgy Technology of Ministry of Education,North China University of Science and Technology,Tangshan 063009,Hebei,China;Technical Centre,Chengde Iron and Steel Group Co.,Ltd.,Chengde 067000,Hebei,China)
出处
《钢铁研究学报》
CAS
CSCD
北大核心
2020年第5期363-369,共7页
Journal of Iron and Steel Research
基金
河北省高等学校技术研究资助项目(QN2019200)
河北省高端钢铁冶金联合研究基金资助项目(E2019209314)。
