基于数据驱动与工艺机理的高炉炉热预测与反馈

石泉; 唐珏; 储满生

doi:10.1007/s12613-023-2693-7

Volume 31 Issue 6

Jun. 2024

Turn off MathJax

图(13) / 表(6)

数据统计

计量

文章访问数: 580
HTML全文浏览量: 321
PDF下载量: 59
被引次数: 0

文章导航 > 矿物冶金与材料学报（英文） > 2024 > 31(6): 1228-1240

Quan Shi, Jue Tang, and Mansheng Chu, Process metallurgy and data-driven prediction and feedback of blast furnace heat indicators, Int. J. Miner. Metall. Mater., 31(2024), No. 6, pp. 1228-1240. https://doi.org/10.1007/s12613-023-2693-7

Cite this article as:

Quan Shi, Jue Tang, and Mansheng Chu, Process metallurgy and data-driven prediction and feedback of blast furnace heat indicators, Int. J. Miner. Metall. Mater., 31(2024), No. 6, pp. 1228-1240. https://doi.org/10.1007/s12613-023-2693-7

Citation:

PDF( 2462 KB)

引用本文 PDF XML SpringerLink

研究论文

基于数据驱动与工艺机理的高炉炉热预测与反馈

1)
东北大学冶金学院, 沈阳 110819, 中国
2)
东北大学低碳钢铁前沿技术研究院, 沈阳 110819, 中国
3)
低碳炼钢前沿技术教育部工程研究中心, 沈阳 110819, 中国

通讯作者:
唐珏 E-mail: tangj@smm.neu.edu.cn

储满生 E-mail: chums@smm.neu.edu.cn

文章亮点

(1) 基于大数据和高炉工艺机理，开发了高炉炉热智能预测与反馈模型。

(2) 实现提前1 h对铁水温度、铁水Si含量的高精度预测与操作建议的同步反馈。

(3) 成功在线应用，实现了大数据技术在高炉炼铁的深度应用。

中文摘要

对于复杂、难控制的小时级延迟的高炉系统，炉热指标的预测与控制对改善高炉炉热水平和炉况顺行具有重要意义。本文提出了一种基于数据驱动与高炉工艺融合的炉热指标预测与反馈模型，在建模过程中，全面分析了原燃料、工艺操作、冶炼状态和渣铁排放整个高炉工序的数据，共171个变量，9223组数据。并结合高炉工艺知识提出了一种新的炉热指标时滞性分析方法，提取的时滞性变量在建模中起到了重要作用。相较于传统单一机器学习算法，本文采用的遗传算法与集成学习结合的方法模型性能有明显提升，铁水温度在正负10°C误差范围内的命中率为92.4%，铁水[Si]含量在正负0.1wt%误差范围内的命中率93.3%。并通过与其他5种机器学习算法的对比，验证了所提方法的性能。并在炉热预测模型的基础上融合专家经验建立了一种符合高炉工艺的炉热操作反馈模型，为稳定炉热水平定量推送操作建议，得到了高炉操作者的高度认可。本模型成功实现了工业在线应用，应用期间炉温水平明显改善，炉温稳定率由54.9%提升至84.9%，取得了显著的经济效益。

关键词:

Research Article

Process metallurgy and data-driven prediction and feedback of blast furnace heat indicators

+ Author Affiliations

Abstract

Abstract

The prediction and control of furnace heat indicators are of great importance for improving the heat levels and conditions of the complex and difficult-to-operate hour-class delay blast furnace (BF) system. In this work, a prediction and feedback model of furnace heat indicators based on the fusion of data-driven and BF ironmaking processes was proposed. The data on raw and fuel materials, process operation, smelting state, and slag and iron discharge during the whole BF process comprised 171 variables with 9223 groups of data and were comprehensively analyzed. A novel method for the delay analysis of furnace heat indicators was established. The extracted delay variables were found to play an important role in modeling. The method that combined the genetic algorithm and stacking efficiently improved performance compared with the traditional machine learning algorithm in improving the hit ratio of the furnace heat prediction model. The hit ratio for predicting the temperature of hot metal in the error range of ±10°C was 92.4%, and that for the chemical heat of hot metal in the error range of ±0.1wt% was 93.3%. On the basis of the furnace heat prediction model and expert experience, a feedback model of furnace heat operation was established to obtain quantitative operation suggestions for stabilizing BF heat levels. These suggestions were highly accepted by BF operators. Finally, the comprehensive and dynamic model proposed in this work was successfully applied in a practical BF system. It improved the BF temperature level remarkably, increasing the furnace temperature stability rate from 54.9% to 84.9%. This improvement achieved considerable economic benefits.
- blast furnace;
- furnace heat;
- genetic algorithm;
- stacking;
- prediction and feedback

FullText(HTML)

Supplements(0)

References(35)

References

[1]	D. Pan, Z.H. Jiang, Z.P. Chen, W.H. Gui, Y.F. Xie, and C.H. Yang, Temperature measurement method for blast furnace molten iron based on infrared thermography and temperature reduction model, Sensors, 18(2018), No. 11, art. No. 3792. doi: 10.3390/s18113792
[2]	X. Liu, W.J. Zhang, Q. Shi, and L. Zhou, Operation parameters optimization of blast furnaces based on data mining and cleaning, J. Northeastern Univ. Nat. Sci., 41(2020), No. 8, p. 1153.
[3]	Z.N. Li, M.S. Chu, Z.G. Liu, G.J. Ruan, and B.F. Li, Furnace heat prediction and control model and its application to large blast furnace, High Temp. Mater. Process., 38(2019), p. 884. doi: 10.1515/htmp-2019-0049
[4]	X.J. Liu, Y. Deng, X. Li, L.Y. Hao, E.H. Liu, and Q. Lyu, Prediction of silicon content in hot molten of blast furnace based on bid data technology, China Metall., 31(2021), No. 2, p. 10.
[5]	M.S. Chu, J. Yagi, and F. Shen, Modelling on Blast Furnace Process and Innovative Ironmaking Technologies, Northeastern University Press, Shenyang, 2006, p.36.
[6]	X.G. Liu and F. Liu, Blast Furnace Ironmaking Process Optimization and Intelligent Control system, Metallurgy Industry Press, Beijing, 2003, p. 90.
[7]	Q. Shi, J. Tang, and M.S. Chu, Key issues and progress of industrial big data-based intelligent blast furnace ironmaking technology, Int. J. Miner. Metall. Mater., 30(2023), No. 9, p. 1651. doi: 10.1007/s12613-023-2636-3
[8]	R.Y. Yin, Review on the study of metallurgical process engineering, Int. J. Miner. Metall. Mater., 28(2021), No. 8, p. 1253. doi: 10.1007/s12613-020-2220-z
[9]	Q. Shi, J. Tang, and M.S. Chu, Evaluation, prediction, and feedback of blast furnace hearth activity based on data-driven analysis and process metallurgy, Steel Res. Int., 95 (2024), art. No. 2300385. doi: 10.1002/srin.202300385
[10]	G.F. Pan, F.Y. Wang, C.L. Shang, et al., Advances in machine learning- and artificial intelligence-assisted material design of steels, Int. J. Miner. Metall. Mater., 30(2023), No. 6, p. 1003. doi: 10.1007/s12613-022-2595-0
[11]	R.H. Zhang and J. Yang, State of the art in applications of machine learning in steelmaking process modeling, Int. J. Miner. Metall. Mater., 30(2023), No. 11, . 2055. doi: 10.1007/s12613-023-2646-1
[12]	C.L. Huang, Y.L. Tang, X.F. Zhang, and Y.Z. Chu, Prediction and simulation of silicon content in blast furnace for PCA and PSO–ELM, Comput. Simul., 37(2020), No. 2, p. 398.
[13]	M. Yuan, P. Zhou, M.L. Li, R.F. Li, H. Wang, and T.Y. Chai, Intelligent multivariable modeling of blast furnace molten iron quality based on dynamic AGA-ANN and PCA, J. Iron Steel Res. Int., 22(2015), No. 6, p. 487. doi: 10.1016/S1006-706X(15)30031-5
[14]	P. Zhou, M. Yuan, H. Wang, Z. Wang, and T.Y. Chai, Multivariable dynamic modeling for molten iron quality using online sequential random vector functional-link networks with self-feedback connections, Inf. Sci., 325(2015), p. 237. doi: 10.1016/j.ins.2015.07.002
[15]	Z.Y. Wang, D.H. Jiang, X.D. Wang, J.L. Zhang, Z.J. Liu, and B.J. Zhao, Prediction of blast furnace hot metal temperature based on support vector regression and extreme learning machine, Chin. J. Eng., 43(2021), No. 4, p. 569. doi: 10.13374/j.issn2095-9389.2020.05.28.001
[16]	J.P. Li, C.C. Hua, and X.P. Guan, Inputs screening of hot metal silicon content model on blast furnace, [in] 2017 Chinese Automation Congress (CAC ), Jinan, 2017, p. 3747.
[17]	Y. Deng and Q. Lyu, Establishment of evaluation and prediction system of comprehensive state based on big data technology in a commercial blast furnace, ISIJ Int., 60(2020), No. 5, p. 898. doi: 10.2355/isijinternational.ISIJINT-2019-545
[18]	P. Zhou, P. Dai, H.D. Song, and T.Y. Chai, Data-driven recursive subspace identification based online modelling for prediction and control of molten iron quality in blast furnace ironmaking, IET Control Theory Appl., 11(2017), No. 14, p. 2343. doi: 10.1049/iet-cta.2016.1474
[19]	Y.R. Li and C.J. Yang, Domain knowledge based explainable feature construction method and its application in ironmaking process, Eng. Appl. Artif. Intell., 100(2021), art. No. 104197. doi: 10.1016/j.engappai.2021.104197
[20]	K. Jiang, Z.H. Jiang, Y.F. Xie, D. Pan, and W.H. Gui, Prediction of multiple molten iron quality indices in the blast furnace ironmaking process based on attention-wise deep transfer network, IEEE Trans. Instrum. Meas., 71(2022), art. No. 2512114. doi: 10.1109/TIM.2022.3185325
[21]	J.P. Li, C.C. Hua, Y.N. Yang, and X.P. Guan, A novel MIMO T–S fuzzy modeling for prediction of blast furnace molten iron quality with missing outputs, IEEE Trans. Fuzzy Syst., 29(2021), No. 6, p. 1654. doi: 10.1109/TFUZZ.2020.2983667
[22]	J.P. Li, C.C. Hua, and Y.N. Yang, A novel multiple-input–multiple-output random vector functional-link networks for predicting molten iron quality indexes in blast furnace, IEEE Trans. Ind. Electron., 68(2021), No. 11, p. 11309. doi: 10.1109/TIE.2020.3031525
[23]	J.P. Li, C.C. Hua, J.L. Qian, and X.P. Guan, Low-rank based Multi-Input Multi-Output Takagi-Sugeno fuzzy modeling for prediction of molten iron quality in blast furnace, Fuzzy Sets Syst., 421(2021), p. 178. doi: 10.1016/j.fss.2020.08.012
[24]	Z.N. Li, Prediction and Optimization of Key Process Parameters of Large Blast Furnace Based on Big Data Mining [Dissertation], Northeastern University, Shenyang, 2020, p. 27.
[25]	Y. Han, Z.B. Hu, A.M. Yang, J. Li, and Y.Z. Zhang, Intelligent recommendation model for reducing silicon deviation fluctuation of hot metal in BF and application, Iron Steel, 58(2023), p. 30. doi: 10.13228/j.boyuan.issn0449-749x.20220667
[26]	H.Y. Li, X.P. Bu, X.J. Liu, et al., Evaluation and prediction of blast furnace status based on big data platform of ironmaking and data mining, ISIJ Int., 61(2021), No. 1, p. 108. doi: 10.2355/isijinternational.ISIJINT-2020-249
[27]	J.L. Bai, J.L. Zhang, H.W. Guo, S. Du, and Y.J. Cao, Basic mathematical models in blast furnace expert system, J. Wuhan Univ. Sci. Technol., 36(2013), No. 5, p. 331.
[28]	J.L. Zhang, X.D. Jiang, H.B. Zuo, and Z.J. Liu, Heat state judgment for calcium carbide furnaces based on heat index calculation and furnace temperature prediction, Chin. J. Eng., 35(2013), No. 9, p. 1131. doi: 10.13374/j.issn1001-053x.2013.09.002
[29]	L. Wei, S.S. Yang, F. Zhang, and Q. Bai, A Mathematical model on prediction of hot metal silicon content and temperature using blast furnace hearth thermal state parameters, [in] Metallurgical Research Center 2005 Metallurgical Engineering Science Forum, Beijing, 2005, p. 62.
[30]	X.Q. Niu, Q.W. Ye, Y. Zhou, and X.D. Wang, Autoregressive model electroencephalogram signal identification based on feature selection of genetic algorithm, Comput. Eng., 42(2016), No. 3, p. 283.
[31]	Z.Q. Li, J.Q. Du, B. Nie, W.P. Xiong, C.Y. Huang, and H. Li, Summary of feature selection methods, Comput. Eng. Appl., 55(2019), No. 24, p. 10. doi: 10.3778/j.issn.1002-8331.1909-0066
[32]	H.B. Yu, Q.N. Zhu, L. Kang, G.Z. Qiao, and J.C. Zeng, A Multi-operator collaborative particle swarm optimization algorithm with biased roulette, Control Decis., 39(2024), No. 4, p. 1167. doi: 10.13195/j.kzyjc.2022.1486
[33]	J.W. Xu and Y. Yang, A survey of ensemble learning approaches, J. Yunnan Univ. Nat. Sci. Ed., 40(2018), No. 6, p. 1082. doi: 10.7540/j.ynu.20180455
[34]	Z.N. Li, M.S. Chu, Z.G. Liu, and B.F. Li, Prediction and optimization of blast furnace parameters based on machine learning and genetic algorithm, J. Northeastern Univ. Nat. Sci. Ed., 41(2020), No. 9, p. 1262. doi: 10.12068/j.issn.1005-3026.2020.09.008
[35]	Q. Feng, Q. Li, W. Quan, and X.M. Pei, Overview of multiobjective particle swarm optimization algorithm, Chin. J. Eng., 43(2021), No. 6, p. 745. doi: 10.13374/j.issn2095-9389.2020.10.31.001