Hybrid model for BOF oxygen blowing time prediction based on oxygen balance mechanism and deep neural network

Abstract

The amount of oxygen blown into the converter is one of the key parameters for the control of the converter blowing process, which directly affects the tap-to-tap time of converter. In this study, a hybrid model based on oxygen balance mechanism (OBM) and deep neural network (DNN) was established for predicting oxygen blowing time in converter. A three-step method was utilized in the hybrid model. First, the oxygen consumption volume was predicted by the OBM model and DNN model, respectively. Second, a more accurate oxygen consumption volume was obtained by integrating the OBM model and DNN model. Finally, the converter oxygen blowing time was calculated according to the oxygen consumption volume and the oxygen supply intensity of each heat. The proposed hybrid model was verified using the actual data collected from an integrated steel plant in China, and compared with multiple linear regression model, OBM model, and neural network model including extreme learning machine, back propagation neural network, and DNN. The test results indicate that the hybrid model with a network structure of 3 hidden layer layers, 32-16-8 neurons per hidden layer, and 0.1 learning rate has the best prediction accuracy and stronger generalization ability compared with other models. The predicted hit ratio of oxygen consumption volume within the error ±300 m³ is 96.67%; determination coefficient (R²) and root mean square error (RMSE) are 0.6984 and 150.03 m³, respectively. The oxygen blow time prediction hit ratio within the error ±0.6 min is 89.50%; R² and RMSE are 0.9486 and 0.3592 min, respectively. As a result, the proposed model can effectively predict the oxygen consumption volume and oxygen blowing time in the converter.

Keywords

basic oxygen furnace / oxygen consumption / oxygen blowing time / oxygen balance mechanism / deep neural network / hybrid model

References

[1] YinRYTheory and Method of Metallurgical Process Integration, 20161st ed.Beijing, Metallurgical Industry Press: 102

[2] LiuQ, ShaoX, YangJP, ZhangJS. Multiscale modeling and collaborative manufacturing for steelmaking plants. Chin. J. Eng, 2021, 43(12): 1698

[3] YinRY. Review on the study of metallurgical process engineering. Int. J. Miner. Metall. Mater., 2021, 28(8): 1253

[4] XuZJ, ZhengZ, GaoXQ. Operation optimization of the steel manufacturing process: A brief review. Int. J. Miner. Metall. Mater., 2021, 28(8): 1274

[5] M. Iglesias-Escudero, J. Villanueva-Balsera, F. Ortega-Fernandez, and V. Rodriguez-Montequín, Planning and scheduling with uncertainty in the steel sector: A review, Appl. Sci., 9(2019), No. 13, art. No. 2692.

[6] García-MenéndezD, Morán-PalaciosH, Ortega-FernándezF, Díaz-PiloñetaM. Scheduling in continuous steelmaking casting: A systematic review. ISIJ Int., 2020, 60(6): 1097

[7] LiuQ, LiuQ, YangJP, et al.. Progress of research on steel-making-continuous casting production scheduling. Chin. J. Eng., 2020, 42(2): 144

[8] JiangSL, LiuM, LinJH, ZhongHX. A prediction-based online soft scheduling algorithm for the real-world steel-making-continuous casting production. Knowl. Based Syst., 2016, 111: 159

[9] J.Y. Long, Z.Z. Sun, P.M. Pardalos, Y. Bai, S.H. Zhang, and C. Li, A robust dynamic scheduling approach based on release time series forecasting for the steelmaking-continuous casting production, Appl. Soft Comput., 92(2020), art. No. 106271.

[10] YuSP. A prediction method for abnormal condition of scheduling plan with operation time delay in steelmaking and continuous casting production process. ISIJ Int., 2013, 53(6): 1028

[11] YangJP, ZhangJS, GuoWD, GaoS, LiuQ. End-point temperature preset of molten steel in the final refining unit based on an integration of deep neural network and multi-process operation simulation. ISIJ Int., 2021, 61(7): 2100

[12] CoxIJ, LewisRW, RansingRS, LaszczewskiH, BerniG. Application of neural computing in basic oxygen steelmaking. J. Mater. Process. Technol., 2002, 120(1–3): 310

[13] RajeshN, KhareMR, PabiSK. Feed forward neural network for prediction of end-blow oxygen in LD converter steel making. Mater. Res., 2010, 13(1): 15

[14] HanM, ZhaoY. Dynamic control model of BOF steelmaking process based on ANFIS and robust relevance vector machine. Expert Syst. Appl., 2011, 38(12): 14786

[15] WangM, GaoC, AiXG, ZhaiBP, LiSL. Hybrid end-point static control model for 80 tons BOF steelmaking. Trans. Indian Inst. Met., 2022, 75(9): 2281

[16] AiXL, WangYS, TangWM. Prediction of oxyen blow rate in BP neural network based converter refining. Steelmaking, 2013, 29(2): 34

[17] DoganN, BrooksGA, RhamdhaniMA. Comprehensive model of oxygen steelmaking part 1: Model development and validation. ISIJ Int., 2011, 51(7): 1086

[18] ShenCG, WangCC, WeiXL, LiY, van der ZwaagS, XuW. Physical metallurgy-guided machine learning and artificial intelligent design of ultrahigh-strength stainless steel. Acta Mater., 2019, 179: 201

[19] W.Z. Mu, M. Rahaman, F.L. Rios, J. Odqvist, and P. Hedström, Predicting strain-induced martensite in austenitic steels by combining physical modelling and machine learning, Mater. Des., 197(2021), art. No. 109199.

[20] XinZC, ZhangJS, ZhangJG, ZhengJ, JinY, LiuQ. Predicting temperature of molten steel in LF-refining process using IF-ZCA-DNN model. Metall. Mater. Trans. B, 2023, 54(3): 1181

[21] LiY, HanM, JiangLW. Blowing oxygen volume calculation model of BOF steelmaking based on oxygen decarburization efficiency prediction. J. Dalian Univ. Technol., 2012, 52(5): 725

[22] WangZ, LiuQ, XieFM, et al.. Model for prediction of oxygen required in BOF steelmaking. Ironmaking Steelmaking, 2012, 39(3): 228

[23] WuSW, YangJ, CaoGM. Prediction of the Charpy V-notch impact energy of low carbon steel using a shallow neural network and deep learning. Int. J. Miner. Metall. Mater., 2021, 28(8): 1309

[24] MohantyI, BanerjeeR, SantaraA, KunduS, MitraP. Prediction of properties over the length of the coil during thermo-mechanical processing using DNN. Ironmaking Steelmaking, 2021, 48(8): 953

[25] S. Mittal, A survey on modeling and improving reliability of DNN algorithms and accelerators, J. Syst. Archit., 104(2020), art. No. 101689.

[26] HeF, ChaiXY, ZhuZH. Prediction of oxygen-blowing volume in BOF steelmaking process based on BP neural network and incremental learning. High Temp. Mater. Process., 2022, 41(1): 403

[27] LinWH, JiaoSQ, SunJK, LiuM, SuX, LiuQ. Modified exponential model for carbon prediction in the end blowing stage of basic oxygen furnace converter. Chin. J. Eng., 2020, 42(7): 854

[28] LiGH, WangB, LiuQ, et al.. A process model for BOF process based on bath mixing degree. Int. J. Miner. Metall. Mater., 2010, 17(6): 715

[29] HintonGE, OsinderoS, TehYW. A fast learning algorithm for deep belief nets. Neural Comput., 2006, 18(7): 1527

[30] BengioYLearning Deep Architectures for AI, 2009, Boston, Now Foundations and Trends: 44

[31] S. Shamshirband, M. Fathi, A. Dehzangi, A.T. Chronopoulos, and H. Alinejad-Rokny, A review on deep learning approaches in healthcare systems: Taxonomies, challenges, and open issues, J. Biomed. Inform., 113(2021), art. No. 103627.

[32] ZhangYG, XieYL, ZhangY, QiuJB, WuSX. The adoption of deep neural network (DNN) to the prediction of soil liquefaction based on shear wave velocity. Bull. Eng. Geol. Environ., 2021, 80(6): 5053

[33] S. Liu, X.J. Liu, Q. Lyu, and F.M. Li, Comprehensive system based on a DNN and LSTM for predicting sinter composition, Appl. Soft Comput., 95(2020), art. No. 106574.

[34] MyersCA, NakagakiT. Prediction of nucleation lag time from elemental composition and temperature for iron and steelmaking slags using deep neural networks. ISIJ Int., 2019, 59(4): 687

[35] FruehanRJThe Making, Shaping and Treating of Steel: Steelmaking and Refining Volume, 199811th ed.Pittsburgh, The AISE Steel Foundation: 496

[36] XinZC, ZhangJS, ZhengJ, JinY, LiuQ. A hybrid modeling method based on expert control and deep neural network for temperature prediction of molten steel in LF. ISIJ Int., 2022, 62(3): 532

[37] SchmidhuberJ. Deep learning in neural networks: An overview. Neural Netw., 2015, 61: 85

[38] LeCunY, BengioY, HintonG. Deep learning. Nature, 2015, 521(7553): 436

RIGHTS & PERMISSIONS

University of Science and Technology Beijing