(1. College of Metallurgy and Energy, North China University of Science and Technology, Tangshan 063009, Hebei, China 2. Key Laboratory for Advanced Metallurgy Technology, Ministry of Education, Tangshan 063009, Hebei, China 3. Technical Centre, Chengde Iron and Steel Group Co., Ltd., Chengde 067000, Hebei, China)
Abstract:Through deploying the big data acquisition platform and using the efficient distributed information transmission technology,the collection and summary for the massive data of sintering production were completed,and data warehouse of the whole production line was established. By combining the process knowledge and data mining technology,the underlying rules of the parameters of raw material property,ore blending theory,process parameters,production quality index and production cost were extracted. Based on the mathematical statistics and machine learning algorithm as the core,the quality intelligent control model based on the big data technology for the whole production line of sintering was further researched. Using the methods of decision tree and the optimization,a perfect decision system was established. Based on the algorithm of gradient boosting decision tree,the prediction model of sintering burn through point was established,and the prediction rate of the model is more than 99%. Compared with the prediction model established in the past,the prediction accuracy and stability of the model were further improved. The research results will promote the development of innovation,automation and intellectualization for the sintering production,stabilize the quality index,reduce the production cost,and have wide application value.
吕 庆,刘 颂,刘小杰,毕中心,李建鹏,. 基于大数据技术的烧结全产线质量智能控制系统[J]. , 2018, 53(7): 1-9.
Lv Qing,,LIU Song,,LIU Xiao-jie,,BI Zhong-xin,LI Jian-peng,. Intelligent control system based on big data technology for whole production line of sintering quality. Iron and Steel, 2018, 53(7): 1-9.
Yen C C, Hsu J S.Pagerank algorithm improvement by page relevance measurement.[J].IEEE International Conference on Fuzzy Systems, 2009, :502-506
[8]
Yen C C, Hsu J S.Pagerank algorithm improvement by page relevance measurement.[J].IEEE International Conference on Fuzzy Systems, 2009, :502-506
[9]
Dean J, Ghemawat S.Simplified data processing on large clusters[J].Communications of the ACM, 2008, 51(1):107-113
[9]
Dean J, Ghemawat S.Simplified data processing on large clusters[J].Communications of the ACM, 2008, 51(1):107-113
[10]
Althebyan Q, ALQudah O, Jararweh Y, Yaseen Q.Multi-threading based Map Reduce tasks scheduling.[J].International Conference on Information and Communication Systems(ICICS), 2014, :1-6
[10]
Althebyan Q, ALQudah O, Jararweh Y, Yaseen Q.Multi-threading based Map Reduce tasks scheduling.[J].International Conference on Information and Communication Systems(ICICS), 2014, :1-6
LAITINEN P.J,SAXEN HA neural network based model of sinter quality and sinter plant performance indices[J].Ironmaking and Steelmaking, 2007, 34(2):109-114
[29]
LAITINEN P.J,SAXEN HA neural network based model of sinter quality and sinter plant performance indices[J].Ironmaking and Steelmaking, 2007, 34(2):109-114
JUNHONG Z, ANGUO X, FENG-MAN S.Multi-Objective Optimization and Analysis Model of Sintering Process Based on BP Neural Network[J].Journal of Iron and Steel Research, International, 2007, 14(2):1-5
[31]
JUNHONG Z, ANGUO X, FENG-MAN S.Multi-Objective Optimization and Analysis Model of Sintering Process Based on BP Neural Network[J].Journal of Iron and Steel Research, International, 2007, 14(2):1-5