Prediction of dynamic recrystallization behavior of SAE52100 large section bearing steel based on machine learning

Abstract

This paper investigates the dynamic recrystallization characteristics of SAE52100 large section bearing steel under hot compression, focusing on both the center and surface. Using data from thermal simulation experiments the physical models were developed. Four machine learning algorithms including support vector regression, k-nearest neighbors, random forest, and extreme gradient boosting were then employed to develop dynamic recrystallization prediction models based on the experimental data and inferred values from the physical model. The results show that the machine learning methods provide a better numerical description of the model, provided these are fed with extensive data. To enhance the scope of application, we obtained data from the dynamic recrystallization models for both the center and surface of SAE52100 steel in the as-cast state, as well as extrapolated values from the literature regarding the hot-rolled condition. When the SHAP method was introduced to reveal the mechanism of the influence of each input feature on the prediction results of the machine learning model, it was found that the test results of the Cr element did not match the theory, mainly because of the small scale of Cr elemental data and the strong dependence on grain size and secondary dendrite spacing.

Keywords

dynamic recrystallization / large section bearing steel / machine learning / recrystallization kinetics

References

1	Ghosh S, Mula S. Thermomechanical processing of low carbon Nb-Ti stabilized microalloyed steel: microstructure and mechanical properties. Mater Sci Eng A-Struct Mater Prop Microstruct Process 2015;646:218-233.
2	Mohebbi MS, Rezayat M, Parsa MH, Nagy S, Nosko M. The impact of Nb on dynamic microstructure evolution of an Nb-Ti microalloyed steel. Mater Sci Eng A-Struct Mater Prop Microstruct Process 2018;723:194-203.
3	Hayoune A, Hamana D. A dilatometric and high-temperature X-ray diffraction study of cold deformation effect on the interaction between precipitation, recovery and recrystallization reactions in Al-12 wt.%Mg alloy. Mater Sci Eng A-Struct Mater Prop Microstruct Process 2010;527(27–28):7261-7264.
4	Dai QS, Deng YL, Wang Y, Huang WH. Dynamic recrystallization critical conditions and a physically-based constitutive model of Al-4.8Mg alloy under hot working conditions. Materials. 2020;13(21):19.
5	Liu W, Zhang H, Liu Z, et al. Hot-working characteristics and dynamic recrystallization behavior of hot isostatically pressed FGH4096 superalloy. Metall Mater Trans A. 2023;54(3):962-982
6	Varin RA, Tangri K. On the possible changes in a grain boundary structure during recrystallization and grain growth. Scr Metall (USA). 1980;14(3):337-340.
7	Belyakov A, Miura H, Sakai T. Dynamic recrystallization in ultra finegrained 304 stainless steel. Scripta Mater. 2000;43(1):21-26.
8	Sakai T, Miura H, Muramatsu N. Effect of small amount addition of Co on dynamic recrystallization of Cu-Be alloy. J Jpn Inst Metals. 1994;58(1):15-22.
9	Sakai T, Ohashi M. Dislocation substructures developed during dynamic recrystallization in polycrystalline nickel. Mater Sci Technol 1990;6(12):1251-1257.
10	Cao RZ, Wang W, Ma S, Yan H, Mu Z, Zhang S. Arrhenius constitutive model and dynamic recrystallization behavior of 18CrNiMo7-6 steel. J Mater Res Technol. 2023;24:6334-6347.
11	Zhao GH, Tian Y, Li H, Ma L, Li Y, Li J. Microstructure evolution and dynamic recrystallization mechanisms of 316L stainless steel during hot deformation. Arch Civ Mech Eng. 2024;24(1):19.
12	Feng H, Wu M, Li H, et al. Elucidating the dual effect of vanadium microalloying on hot deformation behavior of high nitrogen martensitic stainless steel. J Mater Res Technol. 2024;29:4902-4916.
13	Song CN, Cao J, Xiao J, Zhao Q, Sun S, Xia W. High-temperature constitutive relationship involving phase transformation for nonoriented electrical steel based on PSO-DNN approach. Mater Today Commun 2023;34:11.
14	Xu Y, Hu LX, Sun Y. Deformation behaviour and dynamic recrystallization of AZ61 magnesium alloy. J Alloys Compd 2013;580:262-269.
15	Yue CX, Zhang L, Liao S, et al. Research on the dynamic recrystallization behavior of GCr15 steel. Mater Sci Eng A-Struct Mater Prop Microstruct Process 2009;499(1–2):177-181.
16	Xiao-Gang Lu YH, Zheng W, Zheng W. Design of advanced steels by integrated computational materials engineering. Mater Genome Eng Adv. 2024;2(2):e36.
17	Shaojie Li ZD, Jin J, Pan H, Hu Z, Hou R, Qin G. Optimal design of high-performance rare-earth-free wrought magnesium alloys using machine learning. Mater Genome Eng Adv. 2024;2:45.
18	Siyu Han CW, Zhang Y, Xu W, Di H, Di H. Employing deep learning in non-parametric inverse visualization of elastic-plastic mechanisms in dual-phase steels. Mater Genome Eng Adv. 2024;2(1):29.
19	Li X, Zhou XG, Cao GM, Xu SH, Wang Y, Liu ZY. Machine learning hot deformation behavior of Nb micro-alloyed steels and its extrapolation to dynamic recrystallization kinetics. Metall Mater Trans A. 2021;52(7):3171-3181.
20	Li HY, Wang X, Song Y, Li Y, Li X, Ji Y. Physical metallurgy guided machine learning to predict hot deformation mechanism of stainless steel. Mater Today Commun 2023;36:13.
21	Chen XW, Sun J, Yang Y, Liu B, Zhou J. Finite element analysis of dynamic recrystallization model and microstructural evolution for GCr15 bearing steel warm-hot deformation process. Materials. 2023;16(13):21.
22	Liu ZL. GCr15 Bearing Steel Heat Distortion Behavior Study;2016.
23	Mandal S, Sivaprasad AV, Dube RK. Modeling microstructural evolution during dynamic recrystallization of alloy D9 using artificial neural network. J Mater Eng Perform. 2007;16(6):672-679.
24	Zhang ZH. Introduction to machine learning: k-nearest neighbors. Ann Transl Med. 2016;4(11):218.
25	Kavousi-Fard A, Samet H, Marzbani F. A new hybrid modified firefly algorithm and support vector regression model for accurate short term load forecasting. Expert Syst Appl. 2014;41(13):6047-6056.
26	Zhou J, Huang S, Zhou T, Armaghani DJ, Qiu YG. Employing a genetic algorithm and grey wolf optimizer for optimizing RF models to evaluate soil liquefaction potential. Artif Intell Rev. 2022;55(7):5673-5705.
27	Qian Qiao, QL , J Pu, H Shi, et al. A comparative study of machine learning in predicting the mechanical properties of the deposited AA6061 alloys via additive friction stir deposition. MGE Adv. 2024;2(1):e31.
28	Burman P, Chow E, Nolan DA. Cross-validatory method for dependent data. Biometrika. 1994;81(2):351-358.
29	Poliak EI, Jonas JJ. Initiation of dynamic recrystallization in constant strain rate hot deformation. ISIJ Int. 2003;43(5):684-691.
30	Sellars CM, Whiteman JA. Recrystallization and grain growth in hot rolling. Metal Sci. 1979;13(3–4):187-194.
31	Karhausen K, Kopp R, Desouza MM. Numerical-simulation method for designing thermomechanical treatments, illustrated by bar rolling. Scand J Metall. 1991;20(6):351-363.