联合大数据和神经网络的高炉透气性预报模型

doi:10.13228/j.boyuan.issn0449-749x.20220790

Abstract
Figure/Table
References
Related Citation (15)

Download: PDF (2462 KB) HTML (1 KB)
Export: BibTeX | EndNote (RIS)

Abstract Blast furnace (BF) permeability is vital to smooth operation, but related studies are limited, especially for BFs toward low-carbon operations. In the BF production practice, the permeability index (PI) is often obtained by a posterior estimate, resulting in a challenge to make reasonable adjustments for the BF′s stable production and response to abnormal events in time. In addition, it is not possible to effectively obtain the permeability under the design stage of low-carbon BFs. Thus, a priori PI prediction model is established to circumvent these problems. Specifically, combining big data from the practical process, all variables affecting BF PI were analyzed based on both approaches of Pearson correlation and heat map of GRA (Grey Relation Analysis) coefficients, and 44 variables were identified as the primary parameters for predicting PI. Then, big data consisting of these parameters were cleaned to avoid the problems raised due to the existence of incomplete, abnormal, order-of-magnitude difference values, and further were normalized. Thus, the resultant data set for predicting the PI was found. Furthermore, a neural network model for PI prediction of BFs (PI-Net), which consists of an input layer, a 3-layer hidden layer and an output layer, was established and trained based on those big data. The results show that the mean square error of PI-Net is 9.6×10^-5, the root mean square error is 9.78×10^-3, the mean absolute error is 7.6×10^-3, and the linear regression coefficient of determination is 0.979 2, indicating that the established model has reasonable accuracy, robustness and generalization ability. Finally, PI-Net was applied to evaluate the PI of several typical BF designed toward low carbon operations and explore the feasibility of these proposals due to being restricted by PI.

Key words： blast furnace permeability big data neural network model low-carbon blast furnace prediction model

Received: 20 December 2022

	Service

	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	WANG Shuai
	LI Qiang

Cite this article:

WANG Shuai,LI Qiang. Prediction model of blast furnace permeability by combining big data with neural network[J]. Iron and Steel, 2023, 58(7): 46-53.

URL:

http://www.chinamet.cn/Jweb_gt/EN/10.13228/j.boyuan.issn0449-749x.20220790 OR http://www.chinamet.cn/Jweb_gt/EN/Y2023/V58/I7/46

[1] CASTRO J,NOGAMI H,YAGI J I. Three-dimensional multiphase mathematical modeling of the blast furnace based on the multifluid model[J]. ISIJ International,2002,42(1):44.
[2] DONG X F, YU A B, CHEW S J, et al. Modeling of blast furnace with layered cohesive zone[J]. Metallurgical and Materials Transactions B,2010,41(2):330.
[3] KUANG S B,LI Z Y,YU A B. Review on modeling and simulation of blast furnace[J]. Steel Research International,2018,89(1):1700071.
[4] HSU K W, KO Y C. Analysis of operation performance of blast furnace with machine learning methods[M]//Utilizing Big Data Paradigms for Business Intelligence.IGI Global, 2019: 242.
[5] PARK T J,MIN S K,KIM H,et al. Observation of the effect of various operating factors on cohesive zone using the blast furnace irregularity simulator[J]. Steel Research International,2021,92(2):2000315.
[6] LI Q,TANG R,WANG S,et al. A coupled LES-LBM-IMB-DEM modeling for evaluating pressure drop of a heterogeneous alternating-layer packed bed[J]. Chemical Engineering Journal,2022,433(2):133529.
[7] LI Q,GUO S,WANG S,et al. CFD-DEM investigation on pressure drops of heterogeneous alternative-layer particle beds for low-carbon operating blast furnaces[J]. Metals,2022,12(9):1507.
[8] 张军,严铁军,席玮城. 焦炭质量提升对高炉冶炼的影响分析[J]. 冶金能源,2022,41(4):15.(ZHANG J,YAN T J,XI W C. Influence analysis of coke quality improvement on blast furnace smelting[J]. Energy for Metallurgical Industry,2022,41(4):15.)
[9] 曾宇,姜喆,张建良,等. 大喷煤条件下鞍钢高炉喷吹煤粉的炉内利用率[J]. 中国冶金,2022,32(3):61.(ZENG Y,JIANG Z,ZHANG J L,et al. Utilization rate of coal injection for blast furnace in Ansteel under the large injection condition[J]. China Metallurgy,2022,32(3):61.)
[10] 武吉,周鹏,侯士彬,等. 炼焦配煤与焦炭质量评价的新认识[J]. 冶金能源,2021,40(5):8.(WU J,ZHOU P,HOU S B,et al. New understanding of coking coal blending and coke quality evaluation[J]. Energy for Metallurgical Industry,2021,40(5):8.)
[11] 潘林辉,黄胜,时峰,等. 气煤替代1/3焦煤的配煤方案优化试验[J]. 中国冶金,2022,32(1):90.(PAN L H,HUANG S,SHI F,et al. Optimization test for coal blending scheme of replacing 1/3 coking coal with gas coal[J]. China Metallurgy,2022,32(1):90.)
[12] 李伟华. 新钢2 500 m³高炉燃料结构合理优化[J]. 冶金能源,2022,41(6):41.(LI W H. Production practice of rational fuel structure for 2 500 m³ blast furnace of Xingang[J]. Energy for Metallurgical Industry,2022,41(6):41.)
[13] 张寿荣,姜曦. 中国大型高炉生产现状分析及展望[J]. 钢铁,2017,52(2):1.(ZHANG S R,JIANG X. Production and development of large blast furnaces in China[J]. Iron and Steel,2017,52(2):1.)
[14] 王新东,金永龙. 高炉使用高比例球团的战略思考与球团生产的试验研究[J]. 钢铁,2021,56(5):7.(WANG X D,JIN Y L. Strategy analysis and testing study of high ratio of pellet utilized in blast furnace[J]. Iron and Steel,2021,56(5):7.)
[15] 王海风,平晓东,周继程,等. 中国钢铁工业绿色发展回顾及展望[J]. 钢铁,2023,58(2):8.(WANG H F,PING X D,ZHOU J C,et al. Review and prospect of green development for Chinese steel industry[J]. Iron and Steel,2023,58(2):8.)
[16] SU X,ZHANG S,YIN Y,et al. Prediction model of permeability index for blast furnace based on the improved multi-layer extreme learning machine and wavelet transform[J]. Journal of the Franklin Institute,2018,355(4):1663.
[17] PAN Y Z,ZUO H B,WANG J S,et al. Review on improving gas permeability of blast furnace[J]. Journal of Iron and Steel Research International,2020,27(2):121.
[18] LECUN Y,BENGIO Y,HINTON G. Deep learning[J]. Nature,2015,521:436.
[19] XU G W,LIU M,JIANG Z F,et al. Online fault diagnosis method based on transfer convolutional neural networks[J]. IEEE Transactions on Instrumentation and Measurement,2020,69(2):509.
[20] HOLM E A,COHN R,GAO N,et al. Overview:Computer vision and machine learning for microstructural characterization and analysis[J]. Metallurgical and Materials Transactions A,2020,51(12):5985.
[21] 刘然,赵伟光,刘颂,等. 高炉冶炼智能化的发展与探讨[J]. 钢铁,2023,58(5):1.(LIU R,ZHAO W G,LIU S,et al. Development and discussion of intelligent blast furnace smelting[J]. Iron and Steel,2023,58(5):1.)
[22] 张峥,仲兆准,李阳,等. 基于深度学习的带钢精轧过程自由宽展预测[J]. 中国冶金,2022,32(11):121.(ZHANG Z,ZHONG Z Z,LI Y,et al. Prediction of lateral spread for hot strip finishing mill based on deep learning[J]. China Metallurgy,2022,32(11):121.)
[23] 郝良元,刘福龙,牛佳星,等. 提高高炉智能控制水平的分析与思考[J]. 冶金能源,2021,40(5):55.(HAO L Y,LIU F L,NIU J X,et al. Analysis and thinking on improving intelligent control level of BF[J]. Energy for Metallurgical Industry,2021,40(5):55.)
[24] 邱华东,田建艳,王书宇,等. 模糊神经网络融合建模方法及其在轧制力控制中的应用[J]. 中国冶金,2021,31(1):52.(QIU H D,TIAN J Y,WANG S Y,et al. Modeling method of fuzzy neural network and its application in rolling force control[J]. China Metallurgy,2021,31(1):52.)
[25] 崔桂梅,李静,张勇,等. 基于T-S模糊神经网络模型的高炉铁水温度预测建模[J]. 钢铁,2013,48(11):11.(CUI G M,LI J,ZHANG Y,et al. Prediction modeling study for blast furnace hot metal temperature based on T-S fuzzy neural network model[J]. Iron and Steel,2013,48(11):11.)
[26] ZHOU P,YUAN M,WANG H,et al. Data-driven dynamic modeling for prediction of molten iron silicon content using ELM with self-feedback[J]. Mathematical Problems in Engineering,2015,2015:326160.
[27] WANG Y T,HUANG P,YANG G. A visual PCI blockage detection in blast furnace raceway[J]. ISIJ International,2019,60(3):519.
[28] 朱宏平. 基于卷积神经网络的钢表面缺陷检测方法[J]. 工业控制计算机,2020,33(8):83.(ZHU H P. Detection method of steel surface defects based on convolution neural network[J]. Journal of Industrial Control Computer,2020,33(8):83.)
[29] 姚煜,刘漫贤,张兆杰,等. 基于多传感器融合的钢包裸浇检测[J]. 中国冶金,2021,31(5):104.(YAO Y,LIU M X,ZHANG Z J,et al. Naked pouring detection of ladle based on multi-sensor fusion[J]. China Metallurgy,2021,31(5):104.)
[30] CHEN L. Development and application of blast furnace expert system with self-learning function based on pattern recognition[J]. Journal of Southeast University(Natural ence Edition),2012,42:117.
[31] LUO S H,GAO C H,ZENG J S,et al. Blast furnace system modeling by multivariate phase space reconstruction and neural networks[J]. Asian Journal of Control,2013,15(2):553.
[32] 李壮年. 基于大数据挖掘的大型高炉关键工艺参数预测和优化[D]. 沈阳:东北大学,2020.(LI Z N. Prediction and optimization of Key Process Parameters of Large Blast Furnace Based on Big Data Mining[D]. Shenyang:Northeastern University,2020.)
[33] SCHOBER P,BOER C,SCHWARTE L A. Correlation coefficients:Appropriate use and interpretation[J]. Anesthesia and Analgesia,2018,126(5):1.
[34] KUO Y,YANG T,HUANG G W. The use of grey relational analysis in solving multiple attribute decision-making problems[J]. Computers and Industrial Engineering,2008,55(1):80.
[35] MEI K,LIU J,ZHANG X,et al. Performance analysis on machine learning-based channel estimation[J]. IEEE Transactions on Communications,2021,69(8):5183.
[36] CHICCO D,WARRENS M J,JURMAN G. The coefficient of determination R-squared is more informative than SMAPE,MAE,MAPE,MSE and RMSE in regression analysis evaluation[J]. PeerJ Computer Science,2021, 7:e623.
[37] 张晓辉. 高炉复合喷吹煤粉与天然气的可行性研究[D]. 沈阳:东北大学,2022.(ZHANG X H. Feasibility Study on Co-injection of Pulverized Coal and Natural Gas in Blast Furnace[D]. Shenyang:Northeastern University,2022.)