# Prediction Model of End-Point Phosphorus Content in EAF Steelmaking Based on BP Neural Network with Periodical Data Optimization

^{*}

## Abstract

**:**

## 1. Introduction

## 2. Model Structure and Method

## 3. Modeling

#### 3.1. Selection of Input-Output Variable of Model

#### 3.2. Establishment of Artificial Neural Network Model

#### 3.3. Improvement of Prediction Model

- (1)
- Optimization of the model with consumption of oxygen divided into stages.

- (2)
- Optimization of the model with the time of the first addition of lime.

## 4. Discussion

## 5. Conclusions

## Author Contributions

## Funding

## Data Availability Statement

## Acknowledgments

## Conflicts of Interest

## References

- Toulouevski, Y.N.; Zinurov, I.Y. (Eds.) Innovation in Electric Arc Furnaces: Scientific Basis for Selection; Springer: Berlin/Heidelberg, Germany, 2013; pp. 1–24. [Google Scholar]
- Yang, L.-Z.; Hu, H.; Yang, Z.-S.; Xue, B.-T.; Guo, Y.-F.; Wang, S. A review on bath fluid flow stirring technologies in EAF steelmaking. J. Iron Steel Res. Int.
**2021**, 28, 1341–1351. [Google Scholar] [CrossRef] - Bloom, T.A.; Fosnacht, D.R.; Haezebrouck, D.M. Influence of phosphorus on the properties of sheet steel products and methods used to control steel phosphorus levels in steel product Manufacturing. Part I Iron Steelmak.
**1990**, 17, 35–41. [Google Scholar] - Chunlin, C.; Ling, Z.; Lehmann, J. Thermodynamic Modelling of Phosphorus in Steelmaking Slags. High Temp. Mater. Process.
**2013**, 32, 237–246. [Google Scholar] - Wang, Y.; Tsutsumi, S.; Kawakubo, T.; Fujii, H. Effects of phosphorus content on fatigue performance of friction stir welded mild steels. Constr. Build. Mater.
**2022**, 324, 126682. [Google Scholar] [CrossRef] - Rodrigues, C.; Bandeira, R.M.; Duarte, B.B.; Tremiliosi-Filho, G.A.M.J.; Jorge, A.M., Jr. Effect of phosphorus content on the mechanical, microstructure and corrosion properties of supermartensitic stainless steel. Mater. Sci. Eng. A
**2016**, 650, 75–83. [Google Scholar] [CrossRef] - Basu, S.S. Studies on Dephosphorisation during Steelmaking; Royal Institute of Technology: Stockholm, Sweden, 2007. [Google Scholar]
- Xue, M. The blue brittleness of 1Cr17Ni2 steel submarine motor shaft. Mater. Lett.
**2002**, 57, 369–373. [Google Scholar] [CrossRef] - He, F.; Zhang, L.Y. Prediction model of end-point phosphorus content in BOF steelmaking process based on PCA and BP neural network. J. Process. Control
**2018**, 66, 51–58. [Google Scholar] [CrossRef] - Hay, T.; Visuri, V.-V.; Aula, M.; Echterhof, T. A Review of Mathematical Process Models for the Electric Arc Furnace Process. Steel Res. Int.
**2020**, 92, 2000395. [Google Scholar] [CrossRef] - Morales, R.D. A mathematical model for the reduction kinetics of iron oxide in electric furnace slags by graphite injection. ISIJ Int.
**1997**, 37, 1072–1080. [Google Scholar] [CrossRef] - Bekker, J.G. Modeling and simulation of Electric Arc Furnace. ISIJ Int.
**1999**, 39, 23–32. [Google Scholar] [CrossRef] - Zhang, J.J. Terminal adaptive prediction and expert directing operation for the steelmaking process of Electric Arc Furnace. Acta Autom. Sin.
**1993**, 19, 463–467. [Google Scholar] - Zhang, X.G. Introduction to statistics learning theory and support vector machines. Acta Autom. Sin.
**2000**, 26, 34–42. [Google Scholar] - Reimann, A.; Hay, T.; Echterhof, T.; Kirschen, M.; Pfeifer, H. Application and Evaluation of Mathematical Models for Prediction of the Electric Energy Demand Using Plant Data of Five Industrial-Size EAFs. Metals
**2021**, 11, 1348. [Google Scholar] [CrossRef] - Wang, Z.; Xie, F.M.; Wang, B.; Liu, Q.; Lu, X.C.; Hu, L.N.; Cai, F. The control and prediction of End-Point Phosphorus Content during BOF Steelmaking Process. Steel Res. Int.
**2014**, 4, 85. [Google Scholar] [CrossRef] - Chang, S.C.; Zhao, C.H.; Li, Y.L.; Zhou, M.; Fu, C.; Qiao, H.L. Multi-Channel Graph Convolutional Network based End-Point Element Composition Prediction of Converter Steelmaking. IFAC-Pap. OnLine
**2021**, 54, 152–157. [Google Scholar] [CrossRef] - Liu, H.; Wang, B.; Xiong, X. Basic oxygen furnace steelmaking end-point prediction based on computer vision and general regression neural network. Opt. Int. J. Light Electron Opt.
**2014**, 125, 5241–5248. [Google Scholar] [CrossRef] - Lai, Z.Y.; Xie, Z.; Meng, H.J. Prediction model of End-point Phosphorous in Converter Based on Cluster Analysis and Gray Theory. In Proceedings of the 2008 7th World Congress on Intelligent Control and Automation, Chongqing, China, 25–27 June 2008; pp. 2988–2991. [Google Scholar]
- Yuan, P.; Mao, Z.Z.; Wang, F.L. Endpoint Prediction of EAF Based on Multiple Support Vector Machines. J. Iron Steel Res. Int.
**2007**, 14, 20–24. [Google Scholar] [CrossRef] - Zhou, K.X.; Lin, W.H.; Sun, J.K.; Zhang, J.S.; Zhang, D.Z.; Feng, X.M.; Liu, Q. Prediction model of end-point phosphorus content for BOF based on monotone-constrained BP neural network. J. Iron Steel Res. Int.
**2021**, 1–10. [Google Scholar] [CrossRef] - Wang, H.B.; Xu, A.J.; Ai, L.X.; Tian, N.Y. Prediction of endpoint phosphorus content of Molten Steel in BOF using weighted K-Means and GMDH neural network. J. Iron Steel Res. Int.
**2012**, 19, 11–16. [Google Scholar] [CrossRef] - Li, C.R.; Hao, H.W.; Xie, X. Prediction of end-point phosphorus content for BOF based on LM BP neural network. Iron Steel
**2011**, 46, 23. [Google Scholar] - Liu, J.; Yuan, S.Q.; Li, J.G. Prediction of end phosphorus in converter steelmaking using optimized BP network based on SPR. J. Iron Steel Res. Int.
**2005**, 17, 65. [Google Scholar] - Xie, S.M.; Tao, J.; Chai, T.Y. Intelligent method for BOF endpoint phosphorus estimation. Control. Theory Appl.
**2003**, 20, 555. [Google Scholar] - Yang, H.; Zhao, C.; Du, X.; Wang, J. Risk Prediction of City Distribution Engineering Based on BP. Syst. Eng. Procedia
**2012**, 5, 55–60. [Google Scholar] [CrossRef] - Cissé, Y.; Kinouchi, Y.; Nagashino, H.; Akutagawa, M. Identification of homeostatic dynamics for a circadian signal source using BP neural networks. ITBM-RBM
**2000**, 21, 24–32. [Google Scholar] [CrossRef] - Liu, Y.; Zhu, J.C.; Cao, Y. Modeling effects of alloying elements and heat treatment parameters on mechanical properties of hot die steel with back-propagation artificial neural network. J. Iron Steel Res. Int.
**2017**, 24, 1254–1260. [Google Scholar] [CrossRef] - He, F.; He, D.F.; Xu, A.J.; Wang, H.B.; Tian, N.Y. Hybrid model of molten steel temperature prediction based on ladle heat status and artificial neural network. J. Iron Steel Res. Int.
**2014**, 21, 182–191. [Google Scholar] [CrossRef] - Sun, Y.-J.; Zhang, S.; Miao, C.-X.; Li, J.-M. Improved BP Neural Network for Transformer Fault Diagnosis. J. China Univ. Min. Technol.
**2007**, 17, 138–142. [Google Scholar] [CrossRef] - Qu, Y. Principles of Steelmaking, 3rd ed.; Metallurgical Industry Press: Beijing, China, 1987. [Google Scholar]
- Huang, X.G. Principles of Iron and Steel Metallurgy, 2nd ed.; Metallurgical Industry Press: Beijing, China, 2002. [Google Scholar]
- Healy, G.W. New look at phosphorus distribution. J. Iron Steel Inst.
**1970**, 208, 664–668. [Google Scholar] - Suito, H.; Inoue, R.; Takada, M. Phosphorus distribution between liquid iron and MgO saturated slags of the system CaO-MgO-FeOx-SiO2. Tetsu-to-Hagane
**1981**, 67, 2645–2654. [Google Scholar] [CrossRef] - Camdali, U.; Tunc, M. Calculation of Chemical Reaction Energy in an Electric Arc Furnace and Ladle Furnace System. Metallurgist
**2016**, 60, 669–675. [Google Scholar] [CrossRef] - Xia, Y.; Li, J.; Fan, D.; Hou, G. Effects of Interfacial Oxygen Potential and Slag Phase Changing during Slag Formation Process on Dephosphorization Behavior. ISIJ Int.
**2019**, 59, 1519–1526. [Google Scholar] [CrossRef] - Winkler, T.B.; Chipman, J. An equilibrium study of the distribution of phosphorus between liquid iron and basic slags. Trans. AIME
**1946**, 167, 111–133. [Google Scholar] - Hüsken, R.; Fechner, R.; Cappel, J. Use of hot metal with high phosphorous content in combined blowing BOF converters. Iron Steel Technol.
**2011**, 8, 46–58. [Google Scholar] - Liu, Q.; Wang, B.; Wang, Z.; Wang, B.; Xie, F.; Chang, J. Fine Production in Steelmaking Plants. Mater. Today Proc.
**2015**, 2, S348–S357. [Google Scholar] [CrossRef] - Nawi, N.M.; Khan, A.; Rehman, M.Z. A new levenberg marquardt based back propagation algorithm trained with cuckoo search. Procedia Technol.
**2013**, 11, 18–23. [Google Scholar] [CrossRef] - Elkoumy, M.M.; Fathy, A.M.; Megahed, G.M.; El-Mahallawi, I.; Ahmed, H.; El-Anwar, M. Empirical Model for Predicting Process Parameters during Electric Arc Furnace Refining Stage Based on Real Measurements. Steel Res. Int.
**2019**, 90, 1900208. [Google Scholar] [CrossRef] - Manocha, S.; Ponchon, F. Management of Lime in Steel. Metals
**2018**, 8, 686. [Google Scholar] [CrossRef] - Hu, H.; Yang, L.; Guo, Y.; Chen, F.; Wang, S.; Zheng, F.; Li, B. Numerical Simulation of Bottom-Blowing Stirring in Different Smelting Stages of Electric Arc Furnace Steelmaking. Metals
**2021**, 11, 799. [Google Scholar] [CrossRef] - Yugov, P.I. Mechanism and Kinetics of Optimized Slag Formation in an Oxygen Converter. Metallurgist
**2005**, 49, 307–310. [Google Scholar] [CrossRef] - Assis, A.A.; Tayeb, A.M.; Sridhar, S.; Fruehan, J.R. Phosphorus Equilibrium Between Liquid Iron and CaO-SiO2-MgO-Al2O3-FeO-P2O5 Slags: EAF Slags, the Effect of Alumina and New Correlation. Metals
**2019**, 9, 116. [Google Scholar] [CrossRef] [Green Version] - Notman, G.K. Developments in BOS process control at British steel. Steel Times
**1991**, 7, 360. [Google Scholar]

**Figure 4.**Results of model for prediction of end-point phosphorus (P) content in EAF steelmaking process.

**Figure 8.**Frequency of absolute deviation of P content in the prediction model optimized with consumption of oxygen divided into stages.

**Figure 11.**Frequency of absolute deviation in the prediction model optimized with time of the first addition of lime.

**Figure 12.**Frequency of absolute deviation of P content in the prediction model before and after optimization.

**Table 1.**Preliminary selection for influencing factors and reasons for end-point phosphorus (P) content in the electric arc furnace (EAF) steelmaking process.

Influencing Factors | Symbol for Factors | Reason for the Selection |
---|---|---|

Weight of scrap | ${x}_{1}$ | Main material of EAF, staple source of P |

Weight of hot metal | ${x}_{2}$ | |

C content in hot metal | ${x}_{3}$ | Elements in molten steel affecting dephosphorization |

Si content in hot metal | ${x}_{4}$ | |

Mn content in hot metal | ${x}_{5}$ | |

P content in hot metal | ${x}_{6}$ | |

Temperature of hot metal | ${x}_{7}$ | Affecting the temperature of the bath |

Consumption of power | ${x}_{8}$ | Factors of steelmaking |

Consumption of oxygen | ${x}_{9}$ | Oxidant |

Consumption of lime | ${x}_{10}$ | Dephosphorization agent |

− | ${\mathit{x}}_{1}$ | ${\mathit{x}}_{2}$ | ${\mathit{x}}_{3}$ | ${\mathit{x}}_{4}$ | ${\mathit{x}}_{5}$ | ${\mathit{x}}_{6}$ | ${\mathit{x}}_{7}$ | ${\mathit{x}}_{8}$ | ${\mathit{x}}_{9}$ | ${\mathit{x}}_{10}$ |
---|---|---|---|---|---|---|---|---|---|---|

${x}_{1}$ | 1.00 | - | - | - | - | - | - | - | - | - |

${x}_{2}$ | −0.05 | 1.00 | - | - | - | - | - | - | - | - |

${x}_{3}$ | −0.03 | −0.15 | 1.00 | - | - | - | - | - | - | - |

${x}_{4}$ | 0.02 | −0.03 | 0.05 | 1.00 | - | - | - | - | - | - |

${x}_{5}$ | 0.07 | 0.10 | 0.17 | −0.18 | 1.00 | - | - | - | - | - |

${x}_{6}$ | −0.01 | 0.14 | 0.02 | 0.06 | 0.22 | 1.00 | - | - | - | - |

${x}_{7}$ | −0.07 | 0.19 | −0.11 | 0.02 | 0.51 | 0.49 | 1.00 | - | - | - |

${x}_{8}$ | 0.07 | −0.30 | 0.08 | −0.05 | −0.06 | −0.08 | 0.08 | 1.00 | - | - |

${x}_{9}$ | −0.09 | −0.16 | 0.08 | −0.01 | −0.05 | 0.03 | −0.02 | 0.12 | 1.00 | - |

${x}_{10}$ | 0.03 | 0.35 | 0.19 | 0.11 | 0.07 | −0.12 | 0.09 | 0.05 | 0.24 | 1.00 |

$Y$ | 0.18 | 0.19 | 0.06 | −0.04 | 0.02 | −0.05 | 0.14 | 0.01 | −0.09 | 0.06 |

Influence Factors | Units | Training Set | Test Set | ||
---|---|---|---|---|---|

Mean | Standard Deviation | Mean | Standard Deviation | ||

Weight of scrap | t | 54.44 | 13.06 | 52.73 | 11.37 |

Weight of hot metal | t | 64.47 | 11.80 | 63.45 | 7.68 |

C content in hot metal | wt% | 4.42 | 0.04 | 4.42 | 0.04 |

Si content in hot metal | wt% | 0.32 | 0.13 | 0.26 | 0.15 |

Mn content in hot metal | wt% | 0.33 | 0.04 | 0.28 | 0.06 |

P content in hot metal | wt% | 0.11 | 0.01 | 0.10 | 0.01 |

Temperature of hot metal | °C | 1329.44 | 32.90 | 1329.01 | 38.10 |

Consumption of power | kw·h | 11,514.65 | 3380.36 | 7745.28 | 2036.53 |

Consumption of oxygen | Nm^{3} | 4504.15 | 571.21 | 4147.13 | 604.35 |

Consumption of lime | kg | 2988.23 | 795.67 | 2789.43 | 1046.72 |

Total Oxygen Consumption (m^{3}) | Oxygen Consumption in Stage 1 (m^{3}) | Oxygen Consumption in Stage 2 (m^{3}) | Oxygen Consumption in Stage 3 (m^{3}) | Oxygen Consumption in Stage 4 (m^{3}) | |
---|---|---|---|---|---|

Mean | 4514 | 709 | 1431 | 1325 | 1050 |

Maximum | 6250 | 1557 | 2097 | 1975 | 2869 |

Minimum | 3058 | 231 | 518 | 463 | 0 |

Standard deviation | 528 | 170 | 255 | 239 | 515 |

Mean (min) | Maximum (min) | Minimum (min) | Standard Deviation (min) | |
---|---|---|---|---|

Time of lime first added | 27.36 | 48.00 | 17.00 | 10.20 |

Heat No. | Actual Value (wt%) | Predicted Value (wt%) | Absolute Deviation (wt%) | Heat No. | Actual Value (wt%) | Predicted Value (wt%) | Absolute Deviation (wt%) |
---|---|---|---|---|---|---|---|

1 | 0.0141 | 0.0120 | 0.0021 | 16 | 0.0057 | 0.0076 | −0.0019 |

2 | 0.0123 | 0.0096 | 0.0027 | 17 | 0.0088 | 0.0068 | 0.0020 |

3 | 0.0113 | 0.0085 | 0.0028 | 18 | 0.0085 | 0.0113 | −0.0028 |

4 | 0.0057 | 0.0079 | −0.0022 | 19 | 0.0077 | 0.0070 | 0.0007 |

5 | 0.0101 | 0.0106 | −0.0005 | 20 | 0.0096 | 0.0099 | −0.0003 |

6 | 0.0114 | 0.0125 | −0.0011 | 21 | 0.0086 | 0.0107 | −0.0021 |

7 | 0.0172 | 0.0163 | 0.0009 | 22 | 0.0121 | 0.0093 | 0.0028 |

8 | 0.0064 | 0.0085 | −0.0021 | 23 | 0.0106 | 0.0103 | 0.0003 |

9 | 0.0049 | 0.0074 | −0.0025 | 24 | 0.0074 | 0.0096 | −0.0022 |

10 | 0.0061 | 0.0086 | −0.0025 | 25 | 0.0092 | 0.0102 | −0.0010 |

11 | 0.0064 | 0.0109 | −0.0045 | 26 | 0.0135 | 0.0171 | −0.0036 |

12 | 0.0037 | 0.0058 | −0.0021 | 27 | 0.0179 | 0.0114 | 0.0065 |

13 | 0.0059 | 0.0087 | −0.0028 | 28 | 0.0098 | 0.0132 | −0.0034 |

14 | 0.0070 | 0.0107 | −0.0037 | 29 | 0.012 | 0.0147 | −0.0027 |

15 | 0.0119 | 0.0086 | 0.0033 | 30 | 0.0066 | 0.0101 | −0.0035 |

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |

© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Zou, Y.; Yang, L.; Li, B.; Yan, Z.; Li, Z.; Wang, S.; Guo, Y.
Prediction Model of End-Point Phosphorus Content in EAF Steelmaking Based on BP Neural Network with Periodical Data Optimization. *Metals* **2022**, *12*, 1519.
https://doi.org/10.3390/met12091519

**AMA Style**

Zou Y, Yang L, Li B, Yan Z, Li Z, Wang S, Guo Y.
Prediction Model of End-Point Phosphorus Content in EAF Steelmaking Based on BP Neural Network with Periodical Data Optimization. *Metals*. 2022; 12(9):1519.
https://doi.org/10.3390/met12091519

**Chicago/Turabian Style**

Zou, Yuchi, Lingzhi Yang, Bo Li, Zefan Yan, Zhihui Li, Shuai Wang, and Yufeng Guo.
2022. "Prediction Model of End-Point Phosphorus Content in EAF Steelmaking Based on BP Neural Network with Periodical Data Optimization" *Metals* 12, no. 9: 1519.
https://doi.org/10.3390/met12091519