Artificial intelligence-assisted non-metallic inclusion particle analysis in advanced steels using machine learning: A review

Abstract

The detection and characterization of non-metallic inclusions are essential for clean steel production. Recently, imaging analysis combined with high-dimensional data processing of metallic materials using artificial intelligence (AI)-based machine learning (ML) has developed rapidly. This technique has achieved impressive results in the field of inclusion classification in process metallurgy. The present study surveys the ML modeling of inclusion prediction in advanced steels, including the detection, classification, and feature prediction of inclusions in different steel grades. Studies on clean steel with different features based on data and image analysis via ML are summarized. Regarding the data analysis, the inclusion prediction methodology based on ML establishes a connection between the experimental parameters and inclusion characteristics and analyzes the importance of the experimental parameters. Regarding the image analysis, the focus is placed on the classification of different types of inclusions via deep learning, in comparison with data analysis. Finally, further development of inclusion analyses using ML-based methods is recommended. This work paves the way for the application of AI-based methodologies for ultraclean-steel studies from a sustainable metallurgy perspective.

Keywords

machine learning / inclusion classification / image analysis / data analysis / clean steel

References

[1] Mu WZ, Dogan N, Coley KS. Agglomeration of non-metallic inclusions at steel/Ar interface: in-situ observation experiments and model validation. Metall. Mater. Trans. B, 2017, 4852379

[2] Jiao ZB, Liu CT. Ultrahigh-strength steels strengthened by nanoparticles. Sci. Bull., 2017, 62151043

[3] Wang ZL, Bao YP. Development and prospects of molten steel deoxidation in steelmaking process. Int. J. Miner. Metall. Mater., 2024, 31118

[4] Sahai Y. Tundish technology for casting clean steel: A review. Metall. Mater. Trans. B, 2016, 4742095

[5] Duan SC, Lee MJ, Su Y, Mu WZ, Kim DS, Park JH. Evolution of nonmetallic inclusions in 80-t 9CrMoCoB large-scale ingots during electroslag remelting process. Int. J. Miner. Metall. Mater., 2024, 3171525

[6] Vignal V, Krawiec H, Heintz O, Oltra R. The use of local electrochemical probes and surface analysis methods to study the electrochemical behaviour and pitting corrosion of stainless steels. Electrochim. Acta, 2007, 52154994

[7] Krawiec H, Vignal V, Heintz O, Oltra R, Krawiec H, Chauveau E. Dissolution of chromium-enriched inclusions and pitting corrosion of resulfurized stainless steels. Metall. Mater. Trans. A, 2006, 3751541

[8] Pisarek M, Kędzierzawski P, Janik-Czachor M, Kurzydłowski KJ. Effect of hydrostatic extrusion on passivity breakdown on 303 austenitic stainless steel in chloride solution. J. Solid State Electrochem., 2009, 132283

[9] Tan YT, Wijesinghe TLSL, Ng GKL, Blackwood DJ. Investigation into the influence of laser melting on the sulphide inclusions in AISI 416 stainless steel. Corros. Sci., 2011, 53123950

[10] Chiba A, Muto I, Sugawara Y, Hara N. Pit initiation mechanism at MnS inclusions in stainless steel: Synergistic effect of elemental sulfur and chloride ions. J. Electrochem. Soc., 2013, 16010C511

[11] D. Kovalov, C.D. Taylor, H. Heinrich, and R.G. Kelly, Operando electrochemical TEM, ex-situ SEM and atomistic modeling studies of MnS dissolution and its role in triggering pitting corrosion in 304L stainless steel, Corros. Sci., 199(2022), art. No. 110184.

[12] Baker MA, Castle JE. The initiation of pitting corrosion at MnS inclusions. Corros. Sci., 1993, 344667

[13] X. Tan, Y.M. Jiang, Y.Q. Chen, A.Q. Tong, J. Li, and Y.T. Sun, Roles of different components of complex inclusion in pitting of 321 stainless steel: Induction effect of CaS and inhibition effect of TiN, Corros. Sci., 209(2022), art. No. 110692.

[14] Li S, Hu JZ, Zhang J, Ren Y, Zhang LF. Pitting corrosion initiated by Al₂O₃–CaO–CaS inclusions in a 304 stainless steel. Metall. Mater. Trans. B, 2023, 5441784

[15] H. Chen, G.S. Ma, L. Lu, Y.H. Huang, and X.G. Li, Unraveling the effect of Al₂O₃–MnS complex inclusion on the pitting corrosion behavior of dual phase steel, Corros. Sci., 230(2024), art. No. 111938.

[16] Mu WZ, Dogan N, Coley KS. Agglomeration of non-metallic inclusions at the steel/Ar interface: Model application. Metall. Mater. Trans. B, 2017, 4842092

[17] Gu C, Liu WQ, Lian JH, Bao YP. In-depth analysis of the fatigue mechanism induced by inclusions for high-strength bearing steels. Int. J. Miner. Metall. Mater., 2021, 285826

[18] Gu C, Bao YP, Gan P, Wang M, He JS. Effect of main inclusions on crack initiation in bearing steel in the very high cycle fatigue regime. Int. J. Miner. Metall. Mater., 2018, 256623

[19] Mu WZ, Jönsson PG, Nakajima K. Recent aspects on the effect of inclusion characteristics on the intragranular ferrite formation in low alloy steels: A review. High Temp. Mater. Process., 2017, 364309

[20] Ren Y, Zhang LF. In-situ observation of nonmetallic inclusions in steel using confocal scanning laser microscopy: A review. Int. J. Miner. Metall. Mater., 2025, 325975

[21] C.F. Kusche, J.S.K.L. Gibson, M.A. Wollenweber, and S. Korte-Kerzel, On the mechanical properties and deformation mechanisms of manganese sulphide inclusions, Mater. Des., 193(2020), art. No. 108801.

[22] P. Wang, B. Wang, Y. Liu, et al., Effects of inclusion types on the high-cycle fatigue properties of high-strength steel, Scripta Mater., 206(2022), art. No. 114232.

[23] H.W. Fu, J.J. Rydel, A.M. Gola, et al., The relationship between 100Cr6 steelmaking, inclusion microstructure and rolling contact fatigue performance, Int. J. Fatigue, 129(2019), art. No. 104899.

[24] Deng XX, Wang QQ, Ji CXet al.. Three-dimensional distributions of large-sized inclusions in the surface layer of IF steel slabs. Metall. Mater. Trans. B, 2020, 511318

[25] Murakami Y, Endo M. Effects of defects, inclusions and inhomogeneities on fatigue strength. Int. J. Fatigue, 1994, 163163

[26] Meurling F, Melander A, Tidesten M, Westin L. Influence of carbide and inclusion contents on the fatigue properties of high speed steels and tool steels. Int. J. Fatigue, 2001, 233215

[27] Garrison WM, Wojcieszynski AL. A discussion of the effect of inclusion volume fraction on the toughness of steel. Mater. Sci. Eng. A, 2007, 4641–2321

[28] Thornton PA. The influence of nonmetallic inclusions on the mechanical properties of steel: A review. J. Mater. Sci., 1971, 64347

[29] Tomita Y. Development of fracture toughness of ultrahigh strength, medium carbon, low alloy steels for aerospace applications. Int. Mater. Rev., 2000, 45127

[30] Wang P, Zhang P, Wang B, Zhu YK, Xu ZK, Zhang ZF. Fatigue cracking criterion of high-strength steels induced by inclusions under high-cycle fatigue. J. Mater. Sci. Technol., 2023, 154: 114

[31] S.T. Zhou, Z.D. Li, L. Jiang, et al., An investigation into the role of non-metallic inclusions in cleavage fracture of medium carbon pearlitic steels for high-speed railway wheel, Eng. Fail. Anal., 131(2022), art. No. 105860.

[32] Ren CY, Huang CD, Zhang LF, Ren Y. In situ observation of the dissolution kinetics of Al₂O₃ particles in CaO–Al₂O₃–SiO₂ slags using laser confocal scanning microscopy. Int. J. Miner. Metall. Mater., 2023, 302345

[33] Gleinig J, Weidner A, Fruhstorfer J, Aneziris CG, Volkova O, Biermann H. Characterization of nonmetallic inclusions in 18CrNiMo7-6. Metall. Mater. Trans. B, 2019, 501337

[34] Liu Y, Wang SY, Yang ZW, Avdeev M, Shi SQ. Auto-MatRegressor: Liberating machine learning alchemists. Sci. Bull., 2023, 68121259

[35] Zhang D, Xiong YQ, Lu HXet al.. Predicting tremor improvement after MRgFUS thalamotomy in essential tremor from preoperative spontaneous brain activity: A machine learning approach. Sci. Bull., 2024, 69193098

[36] Chen WZ, Gou W, Li YGet al.. Machine learning design of 400 MPa grade biodegradable Zn–Mn based alloys with appropriate corrosion rates. Int. J. Miner. Metall. Mater., 2024, 31122727

[37] Zhang RH, Yang J, Sun H, Yang WK. Prediction of lime utilization ratio of dephosphorization in BOF steelmaking based on online sequential extreme learning machine with forgetting mechanism. Int. J. Miner. Metall. Mater., 2024, 313508

[38] Wang MY, Tang J, Chu MS, Shi Q, Zhang Z. Prediction and optimization of flue pressure in sintering process based on SHAP. Int. J. Miner. Metall. Mater., 2025, 322346

[39] L.F. Fang, F.Y. Su, Z. Kang, and H.J. Zhu, Artificial neural network model for temperature prediction and regulation during molten steel transportation process, Processes, 11(2023), No. 6, art. No. 1629.

[40] Singh MK, Choudhury A, Uikey D, Pal S. Correlation and prediction of molten steel temperature in steel melting shop using reliable machine learning (RML) approach. Trans. Indian Inst. Met., 2023, 76123365

[41] Xin ZC, Zhang JS, Zheng J, Jin Y, Liu Q. A hybrid modeling method based on expert control and deep neural network for temperature prediction of molten steel in LF. ISIJ Int., 2022, 623532

[42] Xin ZC, Zhang JS, Peng KXet al.. Explainable machine learning model for predicting molten steel temperature in the LF refining process. Int. J. Miner. Metall. Mater., 2024, 31122657

[43] J. Kačur, P. Flegner, M. Durdán, and M. Laciak, Prediction of temperature and carbon concentration in oxygen steelmaking by machine learning: A comparative study, Appl. Sci., 12(2022), No. 15, art. No. 7757.

[44] Liu C, Tang LX, Liu JY. A stacked autoencoder with sparse Bayesian regression for end-point prediction problems in steelmaking process. IEEE Trans. Autom. Sci. Eng., 2020, 172550

[45] C.L. Shang, D.X. Zhu, H.H. Wu, et al., A quantitative relation for the ductile–brittle transition temperature in pipeline steel, Scripta Mater., 244(2024), art. No. 116023.

[46] Hu H, Zhao F, Wu DXet al.. Digital model for rapid prediction and autonomous control of die forging force for aluminum alloy aviation components. Int. J. Miner. Metall. Mater., 2025, 3292189

[47] G.W. Song, B.A. Tama, J. Park, et al., Temperature control optimization in a steel-making continuous casting process using a multimodal deep learning approach, Steel Res. Int., 90(2019), No. 12, art. No. 1900321.

[48] Z.Q. Lu, N. Ren, X.W. Xu, et al., Real-time prediction and adaptive adjustment of continuous casting based on deep learning, Commun. Eng., 2(2023), art. No. 34.

[49] Zhao LH, Yang S, Xu YZ, Wang ZL, Liu X, Bao YP. Factor analysis and machine learning for predicting endpoint carbon content in converter steelmaking. Int. J. Miner. Metall. Mater., 2025, 32102469

[50] K. Parul, A.R. Samiraj, and S.S. Hazra, Prediction of end point %C of CONARC® furnace using machine learning methods, Sādhanā, 48(2023), No. 3, art. No. 132.

[51] R.H. Zhang, J. Yang, S.W. Wu, H. Sun, and W.K. Yang, Comparison of the prediction of BOF end-point phosphorus content among machine learning models and metallurgical mechanism model, Steel Res. Int., 94(2023), No. 5, art. No. 2200682.

[52] Pan GF, Wang FY, Shang CLet al.. Advances in machine learning- and artificial intelligence-assisted material design of steels. Int. J. Miner. Metall. Mater., 2023, 3061003

[53] Shi Q, Tang J, Chu MS. Key issues and progress of industrial big data-based intelligent blast furnace ironmaking technology. Int. J. Miner. Metall. Mater., 2023, 3091651

[54] Cao Y, Cao GM, Cui CY, Li X, Wu SW, Liu ZY. Modeling continuous cooling transformations for HSLA steels with physical metallurgy guided hereditary machine learning. Metall. Mater. Trans. A, 2023, 54124891

[55] Wang SW, Wang SZ, Wu HH, Wu Y, Mi ZL, Mao XP. Towards enhanced strength-ductility synergy via hierarchical design in steels: From the material genome perspective. Sci. Bull., 2021, 6610958

[56] Shen CG, Wang CC, Rivera-Díaz-del-Castillo PEJet al.. Discovery of marageing steels: Machine learning vs. physical metallurgical modelling. J. Mater. Sci. Technol., 2021, 87: 258

[57] Shao X, Liu Q, Xin ZC, Zhang JS, Zhou T, Li SS. Hybrid model for BOF oxygen blowing time prediction based on oxygen balance mechanism and deep neural network. Int. J. Miner. Metall. Mater., 2024, 311106

[58] Xu G, He JS, Lü ZM, Li M, Xu JW. Prediction of mechanical properties for deep drawing steel by deep learning. Int. J. Miner. Metall. Mater., 2023, 301156

[59] Yang XJ, Jia JH, Li Qet al.. Stress-assisted corrosion mechanism of 3Ni steel by using gradient boosting decision tree machining learning method. Int. J. Miner. Metall. Mater., 2024, 3161311

[60] Li FF, He AR, Song Yet al.. Deep learning for predictive mechanical properties of hot-rolled strip in complex manufacturing systems. Int. J. Miner. Metall. Mater., 2023, 3061093

[61] Y.X. Zhang, Z.G. Gao, J.C. Sun, and L.L. Liu, Machine-learning algorithms for process condition data-based inclusion prediction in continuous-casting process: A case study, Sensors, 23(2023), No. 15, art. No. 6719.

[62] Cuartas M, Ruiz E, Ferreño D, Setién J, Arroyo V, Gutiérrez-Solana F. Machine learning algorithms for the prediction of non-metallic inclusions in steel wires for tire reinforcement. J. Intell. Manuf., 2021, 3261739

[63] E. Ruiz, D. Ferreño, M. Cuartas, et al., Machine learning methods for the prediction of the inclusion content of clean steel fabricated by electric arc furnace and rolling, Metals, 11(2021), No. 6, art. No. 914.

[64] Deng ZX, Zhang YG, Zhang L, Cong JQ. A transformer and random forest hybrid model for the prediction of non-metallic inclusions in continuous casting slabs. Integr. Mater. Manuf. Innov., 2023, 124466

[65] Wang WJ, Zhang LF, Ren Y, Luo Y, Sun XH, Yang W. Prediction of calcium yield during calcium treatment process performed in steelmaking using neural network. Metall. Mater. Trans. B, 2022, 5311

[66] M. Abdulsalam, N. Gao, B.A. Webler, and E.A. Holm, Prediction of inclusion types from BSE images: RF vs. CNN, Front. Mater., 8(2021), art. No. 754089.

[67] Mu W, Shen C, Xuan Cet al.. A machine learning model to predict non-metallic inclusion dissolution in the metallurgical slag. 12th International Conference of Molten Slags, Fluxes and Salts (MOLTEN 2024) Proceedings, 20241169

[68] X.L. Zhu, W.H. Wan, L. Qian, et al., Research on intelligent identification and grading of nonmetallic inclusions in steels based on deep learning, Micromachines, 14(2023), No. 2, art. No. 482.

[69] Oliveira Filho MFD, Caradec PDB, Calsaverini R, Spinelli JE, Ishikawa TT. Neural network for classification of MnS microinclusions in steels. J. Mater. Res. Technol., 2023, 24: 8522

[70] Abdulsalam M, Zhang TS, Tan J, Webler BA. Automated classification and analysis of non-metallic inclusion data sets. Metall. Mater. Trans. B, 2018, 4941568

[71] Liang Y, Zhu HY, Luo LG, Liu Z, Wang B. Image recognition of non-metallic inclusions in steel based on deep learning. Iron and Steel, 2023, 581262

[72] J. Schmidt, T.F.T. Cerqueira, A.H. Romero, et al., Improving machine-learning models in materials science through large datasets, Mater. Today Phys., 48(2024), art. No. 101560.

[73] Gao WJ, Wang HL, Xu YHet al.. Accurately predicting multiple performance of 3D printing photopolymers using ensemble learning. ACS Appl. Polym. Mater., 2024, 684501

[74] A. Palanivinayagam and R. Damaševičius, Effective handling of missing values in datasets for classification using machine learning methods, Information, 14(2023), No. 2, art. No. 92.

[75] Alabadla M, Sidi F, Ishak Iet al.. Systematic review of using machine learning in imputing missing values. IEEE Access, 2022, 10: 44483

[76] Sanchez-Lengeling B, Aspuru-Guzik A. Inverse molecular design using machine learning: Generative models for matter engineering. Science, 2018, 3616400360

[77] Jadhav A, Pramod D, Ramanathan K. Comparison of performance of data imputation methods for numeric dataset. Appl. Artif. Intell., 2019, 3310913

[78] W.C. Lin, C.F. Tsai, and J.R. Zhong, Deep learning for missing value imputation of continuous data and the effect of data discretization, Knowl. Based Syst., 239(2022), art. No. 108079.

[79] Obatola SO, Tang JJ. A data-driven approach to grid-connected PV system reliability assessment: Combining deep learning and hybrid optimization. Energy Rep., 2024, 12: 5582

[80] Y.J. Xu, L. Jiang, and X. Qi, Machine learning in thermoelectric materials identification: Feature selection and analysis, Comput. Mater. Sci., 197(2021), art. No. 110625.

[81] Zhang ZF, Wen GR, Chen SB. Weld image deep learning-based on-line defects detection using convolutional neural networks for Al alloy in robotic arc welding. J. Manuf. Process., 2019, 45: 208

[82] Dai ZJ, Sun Y, Liu W, Yang SF, Li JS. Smelting stage recognition for converter steelmaking based on the convolutional recurrent neural network. Int. J. Miner. Metall. Mater., 2025, 3292152

[83] D.B. Dai, T. Xu, X. Wei, et al., Using machine learning and feature engineering to characterize limited material datasets of high-entropy alloys, Comput. Mater. Sci., 175(2020), art. No. 109618.

[84] C.C. Wang, X.L. Wei, D. Ren, X. Wang, and W. Xu, High-throughput map design of creep life in low-alloy steels by integrating machine learning with a genetic algorithm, Mater. Des., 213(2022), art. No. 110326.

[85] H.C. Gong, Q.B. Fan, W.Q. Xie, et al., Mining the relationship between the dynamic compression performance and basic mechanical properties of Ti20C based on machine learning methods, Mater. Des., 226(2023), art. No. 111633.

[86] Kaneko H. Interpretation of machine learning models for data sets with many features using feature importance. ACS Omega, 2023, 82523218

[87] He JJ, Sandström R, Zhang J, Qin HY. Application of soft constrained machine learning algorithms for creep rupture prediction of an austenitic heat resistant steel Sanicro 25. J. Mater. Res. Technol., 2023, 22: 923

[88] Fu HD, Zhang HT, Wang CS, Yong W, Xie JX. Recent progress in the machine learning-assisted rational design of alloys. Int. J. Miner. Metall. Mater., 2022, 294635

[89] Jain AK, Murty MN, Flynn PJ. Data clustering: A review. ACM Comput. Surv., 1999, 313264

[90] N.S. Ross, P.M. Mashinini, C. Sherin Shibi, et al., A new intelligent approach of surface roughness measurement in sustainable machining of AM-316L stainless steel with deep learning models, Measurement, 230(2024), art. No. 114515.

[91] H. Singh, T. Alatarvas, A.A. Kistanov, et al., Unveiling interactions of non-metallic inclusions within advanced ultra-high-strength steel: A spectro-microscopic determination and first-principles elucidation, Scripta Mater., 197(2021), art. No. 113791.

[92] Cao L, Han LL, Wang GC, Tao K, Xiao YY. Effects of Mn content on the formation of inclusions in high aluminum steel. Metall. Mater. Trans. B, 2023, 5452680

[93] J.C. Yan, T. Li, Z.Q. Shang, and H. Guo, Three-dimensional characterization of MnS inclusions in steel during rolling process, Mater. Charact., 158(2019), art. No. 109944.

[94] X.P. Guo, M. Tan, T. Li, et al., Formation mechanisms and three-dimensional characterization of composite inclusion of MnS–Al₂O₃ in high speed wheel steel, Mater. Charact., 197(2023), art. No. 112669.

[95] Y. Liu, Z.W. Yang, X.X. Zou, et al., Data quantity governance for machine learning in materials science, Natl. Sci. Rev., 10(2023), No. 7, art. No. nwad125.

[96] Iwayama M, Wu S, Liu C, Yoshida R. Functional output regression for machine learning in materials science. J. Chem. Inf. Model., 2022, 62204837

[97] B. Bayerlein, T. Hanke, T. Muth, et al., A perspective on digital knowledge representation in materials science and engineering, Adv. Eng. Mater., 24(2022), No. 6, art. No. 2101176.

[98] Ren Y, Zhou Q, Hua DPet al.. Wear-resistant CoCrNi multi-principal element alloy at cryogenic temperature. Sci. Bull., 2024, 692227

[99] S. Chittam, B. Gokaraju, Z.G. Xu, J. Sankar, and K. Roy, Big data mining and classification of intelligent material science data using machine learning, Appl. Sci., 11(2021), No. 18, art. No. 8596.

[100] P.L. Zhao, Y.W. Wang, B.Y. Jiang, H.M. Zhang, X.W. Cheng, and Q.B. Fan, Neural network modeling of titanium alloy composition–microstructure–property relationships based on multimodal data, Mater. Sci. Eng. A, 879(2023), art. No. 145202.

RIGHTS & PERMISSIONS

The Author(s)