Slag, as a by-product of the blast furnace smelting process, plays a crucial role in the operating efficiency and product quality by influencing its fluidity and chemical composition. However, the use of cost-effective raw materials has altered slag performance, presenting new challenges for blast furnace operation. In this study, we focused on the CaO–SiO2–Al2O3–MgO quaternary slag system to analyze changes in slag viscosity, sulfur partition ratio, and melting behavior under various slag compositions. To optimize slagging conditions, we employed random forest, gradient boost regression tree, and artificial neural network to predict slag performance based on composition, evaluating accuracy through metrics like R2, MSE, MAPE, and MAE. The models consistently achieved an R2 above 0.97, indicating reliable predictions. Additionally, a user-friendly software was developed by Python to enable staff to forecast slag performance under different slagging compositions. Experimental results were compared with plant data, with a margin of error of 7%. This research offers a novel approach to predicting slag behavior, providing valuable insights for practical production, mitigating the adverse effects of cost-effective materials, and ensuring the safe and steady operation of blast furnaces.
QiuDWeiSElsherbinyAA, et al.Experimental and numerical simulation study on reduction of pellets based on X-ray micro-computed tomography. Metall Mater Trans B2023; 54: 3299–3311.
2.
IwanagaYTakataniK. Mathematical model analysis for oxidation of coke at high temperature. ISIJ Int1989; 29: 43–48.
3.
GanLLaiC. A general viscosity model for molten blast furnace slag. Metall Mater Trans B2014; 45: 875–888.
4.
JiangDZhangJWangZ, et al.A prediction model of blast furnace slag viscosity based on principal component analysis and k-nearest neighbor regression. JOM2020; 72: 3908–3916.
5.
HenaoHMNexhipCGeorge-KennedyDP, et al.Investigation of liquidus temperatures and phase equilibria of copper smelting slags in the FeO-Fe2O3-SiO2-CaO-MgO-Al2O3 system at PO 2 10− 8 atm. Metall Mater Trans B2010; 41: 767–779.
6.
RoySKNayakDRathSS. A review on the enrichment of iron values of low-grade iron ore resources using reduction roasting-magnetic separation. Powder Technol2020; 367: 796–808.
7.
SunaharaKNakanoKHoshiM, et al.Effect of high Al2O3 slag on the blast furnace operations. ISIJ Int2008; 48: 420–429.
8.
HahaMBLothenbachBLe SaoutG, et al.Influence of slag chemistry on the hydration of alkali-activated blast-furnace slag—part II: effect of Al2O3. Cem Concr Res2012; 42: 74–83.
9.
ShankarAGörnerupMSeetharamanS, et al.Sulfide capacity of high alumina blast furnace slags. Metall Mater Trans B2006; 37: 941–947.
10.
ChenMZhangWZhaoZ, et al.Effects of minor elements to the liquidus temperatures of blast furnace slags. ISIJ Int2016; 56: 2156–2160.
11.
QiuG-xMiaoD-jWeiX-l, et al.Effect of MgO/Al2O3 and CaO/SiO2 on the metallurgical properties of CaO–SiO2–Al2O3–MgO–TiO2 slag. J Non-Cryst Solids2022; 585: 121545.
12.
KimJRLeeYSMinDJ, et al.Influence of MgO and Al2O3 contents on viscosity of blast furnace type slags containing FeO. ISIJ Int2004; 44: 1291–1297.
13.
LiuWZuoH. Effect of MnO and substituting CaO with BaO on the desulfurization ability of blast furnace slag. Metall Mater Trans B2021; 52: 2275–2282.
14.
ShankarAGörnerupMLahiriA, et al.Estimation of viscosity for blast furnace type slags. Ironmak Steelmak2007; 34: 477–481.
15.
WuSWangLKouM, et al.Analysis of operational parameters affecting the sulfur content in hot metal of the COREX process. Metall Mater Trans B2017; 48: 276–285.
16.
BeibeiSLuchuanDGuangY, et al.Mechanical properties prediction of blast furnace slag and fly ash-based alkali-activated concrete by machine learning methods. Constr Build Mater2023; 409: 133933.
17.
RafieiMHKhushefatiWHDemirbogaR, et al.Supervised deep restricted Boltzmann machine for estimation of concrete. ACI Mater J2017; 114: 237.
18.
JovicFJovicAKrmpoticD. Quality control engineering in automated ceramic tile production using a signal information content approach. Adv Eng Inf2013; 27: 93–107.
19.
HuangAHuoYYangJ, et al.Computational modeling and prediction on viscosity of slags by big data mining. Minerals2020; 10: 257.
20.
ZhangSGuMZongY, et al.Improved back propagation neural network method for predicting sulfur content in hot metal. Metall Res Technol2024; 121: 103.
21.
CardosoWdi FeliceR. Prediction of Sulfur in the hot metal based on data mining and artificial neural networks. In: 11th International Conference on Data Science, Technology and Applications (DATA), 2022, pp.400–407. SCITEPRESS.
22.
WangYZhangY. Final sulfur content prediction model in hot metal desulphurization process based on IEA-SVM. In: 2011 Chinese Control and Decision Conference (CCDC), 2011, pp.1684–1687. IEEE.
23.
ZhangXKanoMMatsuzakiS. A comparative study of deep and shallow predictive techniques for hot metal temperature prediction in blast furnace ironmaking. Comput Chem Eng2019; 130: 106575.
24.
HanaoMKawamotoMTanakaT, et al.Evaluation of viscosity of mold flux by using neural network computation. ISIJ Int2006; 46: 346–351.
25.
DuchesneMAMacchiALuDY, et al.Artificial neural network model to predict slag viscosity over a broad range of temperatures and slag compositions. Fuel Process Technol2010; 91: 831–836.
26.
MitchellMW. Bias of the random forest out-of-bag (OOB) error for certain input parameters. Open J Stat2011; 1: 205–211.
27.
BanerjeeINguyenBGarousiV, et al.Graphical user interface (GUI) testing: systematic mapping and repository. Inf Softw Technol2013; 55: 1679–1694.
28.
FengD-CLiuZ-TWangX-D, et al.Failure mode classification and bearing capacity prediction for reinforced concrete columns based on ensemble machine learning algorithm. Adv Eng Inf2020; 45: 101126.
29.
WuXWangLChenB, et al.Multi-objective optimization of shield construction parameters based on random forests and NSGA-II. Adv Eng Inf2022; 54: 101751.
30.
RudenkoAKataevATkachevaOY. Rotational viscometry for studying the viscosity of cryolite melts. Russian Metall (Metally)2023; 2023: 141–146.
31.
XuJFZhangJYJieC, et al.Melting temperature of a selected area in CaO-MgO-Al2O3-SiO2 slag system. Adv Mat Res2011; 194: 169–174.
32.
TovarovskiiIMerkulovA. Analytical study of blast-furnace smelting operations performed with different blast temperatures and oxygen contents. Metallurgist2011; 55: 339.
33.
SongLXiaojieLQingL, et al.Study on the appropriate production parameters of a gas-injection blast furnace. High Temp Mater Processes2020; 39: 10–25.
34.
LiZQiZWangZ, et al.Numerical investigation of coke oven gas (COG) injection into an ironmaking blast furnace (BF). Int J Hydrogen Energy2022; 47: 31109–31128.
