烧结系统智能制造与大数据技术应用探讨  

Discussion on intelligent manufacturing of sintering system and application of big data technology

作　　者：刘颂 赵亚迪 甘丽[1] 冯伟 李福民 吕庆 LIU Song;ZHAO Ya-di;GAN Li;FENG Wei;LI Fu-min;LÜQing(Department of Computer Science and Technology,Tangshan College,Tangshan 063000,Hebei,China;College of Metallurgy and Energy,North China University of Science and Technlogy,Tangshan 063210,Hebei,China)

机构地区：[1]唐山学院计算机科学与技术系,河北唐山063000 [2]华北理工大学冶金与能源学院,河北唐山063210

出　　处：《钢铁》2021年第10期54-64,共11页Iron and Steel

基　　金：河北省教育厅科学技术研究资助项目(BJ2021099)。

摘　　要：为了提升烧结工序的智能制造水平,系统总结了近几十年来烧结系统模型的研究进展。针对当前烧结终点预报、烧结矿成分和质量预报以及烧结配料优化模型存在的问题,开展了基于大数据、集成学习和深度学习等技术的烧结系统参数预报与优化研究,并着重介绍了模型在预报精度及泛化能力提升方面取得的成效。基于上述烧结系统参数预报模型,提出了现场应用烧结过程参数预报与优化系统系统的硬件结构设计和软件结构设计方法。最后从钢铁行业需求出发,剖析了先进信息化技术与工业自动化装备深度融合是提升烧结系统智能制造水平的重要途径,并探讨了大数据及人工智能技术在铁前烧结领域的研究方向和应用前景。In order to improve the intelligent manufacturing level of the sintering process,the research progress about the sintering system model in recent decades was systematically summarized.For the current problems in the models of sintering endpoint prediction,composition and quality prediction of sinter ore,and batching optimization,the prediction and optimization of the parameter in the sintering system were investigated by using big data,integrated learning and deep learning.Accordingly,the remarkable results in terms of the improvement in the prediction accuracy and the generalization ability were also emphatically introduced.Moreover,based on the parameter prediction model mentioned above,the hardware and software structure design methods of the parameter prediction and optimization system for on-site application were put forward.Finally,starting from the needs of the iron and steel industry,the point of view,the further integration of advanced information technology and industrial automation equipment was an important way to improve the level of intelligent manufacturing of sintering systems was analyzed,and the research direction and application prospects of big data and artificial intelligence technology in the sintering were also discussed.

关 键 词：烧结 预测模型 大数据 集成学习 深度学习 

分 类 号：TF046.4[冶金工程—冶金物理化学] TP311.13[自动化与计算机技术—计算机软件与理论]