Advances in machine learning- and artificial intelligence-assisted material design of steels

Abstract

With the rapid development of artificial intelligence technology and increasing material data, machine learning- and artificial intelligence-assisted design of high-performance steel materials is becoming a mainstream paradigm in materials science. Machine learning methods, based on an interdisciplinary discipline between computer science, statistics and material science, are good at discovering correlations between numerous data points. Compared with the traditional physical modeling method in material science, the main advantage of machine learning is that it overcomes the complex physical mechanisms of the material itself and provides a new perspective for the research and development of novel materials. This review starts with data preprocessing and the introduction of different machine learning models, including algorithm selection and model evaluation. Then, some successful cases of applying machine learning methods in the field of steel research are reviewed based on the main theme of optimizing composition, structure, processing, and performance. The application of machine learning methods to the performance-oriented inverse design of material composition and detection of steel defects is also reviewed. Finally, the applicability and limitations of machine learning in the material field are summarized, and future directions and prospects are discussed.

Keywords

machine learning / data-driven design / new research paradigm / high-performance steel

References

[1] Olson GB. Genomic materials design: The ferrous frontier. Acta Mater., 2013, 61(3): 771.

[2] Hart GLW, Mueller T, Toher C, Curtarolo S. Machine learning for alloys. Nat. Rev. Mater., 2021, 6(8): 730.

[3] T. Lookman, P.V. Balachandran, D.Z. Xue, and R.H. Yuan, Active learning in materials science with emphasis on adaptive sampling using uncertainties for targeted design, Npj Comput. Mater., 5(2019), art. No. 21.

[4] C. Chen, Y.X. Zuo, W.K. Ye, X.G. Li, Z. Deng, and S.P. Ong, A critical review of machine learning of energy materials, Adv. Energy Mater., 10(2020), No. 8, art. No. 1903242.

[5] R. Ramprasad, R. Batra, G. Pilania, A. Mannodi-Kanakkithodi, and C. Kim, Machine learning in materials informatics: Recent applications and prospects, Npj Comput. Mater., 3(2017), art. No. 54.

[6] Batra R, Song L, Ramprasad R. Emerging materials intelligence ecosystems propelled by machine learning. Nat. Rev. Mater., 2020, 6(8): 655.

[7] Rickman JM, Lookman T, Kalinin SV. Materials informatics: From the atomic-level to the continuum. Acta Mater., 2019, 168, 473.

[8] S. Feng, H.D. Fu, H.Y. Zhou, Y. Wu, Z.P. Lu, and H.B. Dong, A general and transferable deep learning framework for predicting phase formation in materials, Npj Comput. Mater., 7(2021), art. No. 10.

[9] Chakraborty S, Chattopadhyay PP, Ghosh SK, Datta S. Incorporation of prior knowledge in neural network model for continuous cooling of steel using genetic algorithm. Appl. Soft Comput., 2017, 58, 297.

[10] J. Schmidt, M.R.G. Marques, S. Botti, and M.A.L. Marques, Recent advances and applications of machine learning in solid-state materials science, Npj Comput. Mater., 5(2019), art. No. 83.

[11] Nosengo N, Ceder G. Can artificial intelligence create the next wonder material. Nature, 2016, 533(7601): 22.

[12] X.Y. Zhou, J.H. Zhu, Y. Wu, X.S. Yang, T. Lookman, and H.H. Wu, Machine learning assisted design of FeCoNiCrMn high-entropy alloys with ultra-low hydrogen diffusion coefficients, Acta Mater., 224(2022), art. No. 117535.

[13] Z.L. Song, X.W. Chen, F.B. Meng, et al., Machine learning in materials design: Algorithm and application, Chin. Phys. B, 29(2020), No. 11, art. No. 116103.

[14] Liu YL, Niu C, Wang Z, et al. Machine learning in materials genome initiative: A review. J. Mater. Sci. Technol., 2020, 57, 113.

[15] Chen YM, Wang SZ, Xiong J, et al. Identifying facile material descriptors for Charpy impact toughness in low-alloy steel via machine learning. J. Mater. Sci. Technol., 2023, 132, 213.

[16] K. Tsutsui, H. Terasaki, K. Uto, et al., A methodology of steel microstructure recognition using SEM images by machine learning based on textural analysis, Mater. Today Commun., 25(2020), art. No. 101514.

[17] X. Feng, Q. Gao, and M.Y. Liu, Roller parameters prediction of steel tube based on principal component analysis and BP neural network, [in] 2018 Chinese Control and Decision Conference (CCDC), Shenyang, 2018, p. 4627.

[18] Widener A. Materials genome initiative. Chem. Eng. News Archive, 2013, 91(31): 25.

[19] Hachmann J, Olivares-Amaya R, Atahan-Evrenk S, et al. The Harvard clean energy project: Large-scale computational screening and design of organic photovoltaics on the world community grid. J. Phys. Chem. Lett., 2011, 2(17): 2241.

[20] Hachmann J, Olivares-Amaya R, Jinich A, et al. Lead candidates for high-performance organic photovoltaics from high-throughput quantum chemistry—The Harvard Clean Energy Project. Energy Environ. Sci., 2014, 7(2): 698.

[21] A. Jain, S.P. Ong, G. Hautier, et al., Commentary: The Materials Project: A materials genome approach to accelerating materials innovation, APL Mater., 1(2013), No. 1, art. No. 011002.

[22] Saal JE, Kirklin S, Aykol M, Meredig B, Wolverton C. Materials design and discovery with high-throughput density functional theory: The open quantum materials database (OQMD). JOM, 2013, 65(11): 1501.

[23] S. Kirklin, J.E. Saal, B. Meredig, et al., The Open Quantum Materials Database (OQMD): Assessing the accuracy of DFT formation energies, Npj Comput. Mater., 1(2015), art. No. 15010.

[24] Curtarolo S, Setyawan W, Wang SD, et al. AFLOWLIB. ORG: A distributed materials properties repository from high-throughput ab initio calculations. Comput. Mater. Sci., 2012, 58, 227.

[25] Allen FH. The Cambridge Structural Database: A quarter of a million crystal structures and rising. Acta Crystallogr. B, 2002, 58(3): 380.

[26] Kalidindi SR, De Graef M. Materials data science: Current status and future outlook. Annu. Rev. Mater. Res., 2015, 45, 171.

[27] Sanchez-Lengeling B, Aspuru-Guzik A. Inverse molecular design using machine learning: Generative models for matter engineering. Science, 2018, 361(6400): 360.

[28] Taylor RH, Rose F, Toher C, et al. A RESTful API for exchanging materials data in the AFLOWLIB.org consortium. Comput. Mater. Sci., 2014, 93, 178.

[29] S.H. Lu, Q.H. Zhou, Y.X. Ouyang, Y.L. Guo, Q. Li, and J.L. Wang, Accelerated discovery of stable lead-free hybrid organic-inorganic perovskites via machine learning, Nat. Commun., 9(2018), No. 1, art. No. 3405.

[30] Raccuglia P, Elbert KC, Adler PDF, et al. Machine-learning-assisted materials discovery using failed experiments. Nature, 2016, 533(7601): 73.

[31] Oliynyk AO, Antono E, Sparks TD, et al. High-throughput machine-learning-driven synthesis of full-Heusler compounds. Chem. Mater., 2016, 28(20): 7324.

[32] Li SY, de Werk M, St-Pierre L, Kumral M. Dimensioning a stockpile operation using principal component analysis. Int. J. Miner. Metall. Mater., 2019, 26(12): 1485.

[33] Aspuru-Guzik A, Lindh R, Reiher M. The matter simulation (R)evolution. ACS Cent. Sci., 2018, 4(2): 144.

[34] P.B. Jørgensen, M.N. Schmidt, and O. Winther, Deep generative models for molecular science, Mol. Inform., 37(2018), No. 1–2, art. No. 1700133.

[35] Jordan MI, Mitchell TM. Machine learning: Trends, perspectives, and prospects. Science, 2015, 349(6245): 255.

[36] N. Wagner and J.M. Rondinelli, Theory-guided machine learning in materials science, Front. Mater., 3(2016), art. No. 28.

[37] 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, 29(4): 635.

[38] Cortes C, Vapnik V. Support-vector networks. Mach. Learn., 1995, 20(3): 273.

[39] S.J. Han, Q.B. Cao, and M. Han, Parameter selection in SVM with RBF kernel function, [in] World Automation Congress, Puerto Vallarta, 2012, p. 1.

[40] Kireeva N, Pervov VS. Materials space of solid-state electrolytes: Unraveling chemical composition-structure-ionic conductivity relationships in garnet-type metal oxides using cheminformatics virtual screening approaches. Phys. Chem. Chem. Phys., 2017, 19(31): 20904.

[41] Rumelhart DE, Widrow B, Lehr MA. The basic ideas in neural networks. Commun. ACM, 1994, 37(3): 87.

[42] Bhadeshia HKDH. Neural networks in materials science. ISIJ Int., 1999, 39(10): 966.

[43] Ghatak A, Robi PS. Prediction of creep curve of HP40Nb steel using artificial neural network. Neural Comput. Appl., 2018, 30(9): 2953.

[44] Feng S, Zhou H, Dong H. Using deep neural network with small dataset to predict material defects. Mater. Des., 2019, 162, 300.

[45] Jain AK, Murty MN, Flynn PJ. Data clustering: A review. ACM Comput. Surv., 1999, 31(3): 264.

[46] G. Stein, B. Chen, A.S. Wu, and K.A. Hua, Decision tree classifier for network intrusion detection with GA-based feature selection, [in] Proceedings of the 43rd Annual Southeast Regional Conference, 2(2005), p. 136.

[47] Zhang P. Model selection via multifold cross validation. Ann. Statist., 1993, 21(1): 299.

[48] Meijer RJ, Goeman JJ. Efficient approximate k-fold and leave-one-out cross-validation for ridge regression. Biom. J., 2013, 55(2): 141.

[49] Bishop CM, Nasrabadi NM. Pattern Recognition and Machine Learning, 2006, New York, Springer.

[50] Ajioka F, Wang ZL, Ogawa T, Adachi Y. Development of high accuracy segmentation model for microstructure of steel by deep learning. ISIJ Int., 2020, 60(5): 954.

[51] Shen C, Wang C, Huang M, Xu N, van der Zwaag S, Xu W. A generic high-throughput microstructure classification and quantification method for regular SEM images of complex steel microstructures combining EBSD labeling and deep learning. J. Mater. Sci. Technol., 2021, 93, 191.

[52] DeCost BL, Lei B, Francis T, Holm EA. High throughput quantitative metallography for complex microstructures using deep learning: A case study in ultrahigh carbon steel. Microsc. Microanal., 2019, 25(1): 21.

[53] Hecht MD, Webler BA, Picard YN. Digital image analysis to quantify carbide networks in ultrahigh carbon steels. Mater. Charact., 2016, 117, 134.

[54] Otsu N. A threshold selection method from gray-level histograms. IEEE Trans. Syst. Man Cybern., 1979, 9(1): 62.

[55] Serre T, Kreiman G, Kouh M, Cadieu C, Knoblich U, Poggio T. A quantitative theory of immediate visual recognition. Prog. Brain Res., 2007, 165, 33.

[56] C. Kunselman, S. Sheikh, M. Mikkelsen, V. Attari, and R. Arróyave, Microstructure classification in the unsupervised context, Acta Mater., 223(2022), art. No. 117434.

[57] S.W. Kim, S.H. Kang, S.J. Kim, and S. Lee, Estimating the phase volume fraction of multi-phase steel via unsupervised deep learning, Sci. Rep., 11(2021), No. 1, art. No. 5902.

[58] X.Y. Huang, H. Wang, W.H. Xue, et al., A combined machine learning model for the prediction of time-temperature-transformation diagrams of high-alloy steels, J. Alloys Compd., 823(2020), art. No. 153694.

[59] Geng XX, Cheng Z, Wang SZ, et al. A data-driven machine learning approach to predict the hardenability curve of boron steels and assist alloy design. J. Mater. Sci., 2022, 57(23): 10755.

[60] Jiang X, Jia BR, Zhang GF, et al. A strategy combining machine learning and multiscale calculation to predict tensile strength for pearlitic steel wires with industrial data. Scr. Mater., 2020, 186, 272.

[61] Q. Lu, S. Liu, W. Li, and X. Jin, Combination of thermodynamic knowledge and multilayer feedforward neural networks for accurate prediction of M_S temperature in steels, Mater. Des., 192(2020), art. No. 108696.

[62] Wang C, Zhu K, Hedström P, Li Y, Xu W. A generic and extensible model for the martensite start temperature incorporating thermodynamic data mining and deep learning framework. J. Mater. Sci. Technol., 2022, 128, 31.

[63] Dehghani K, Shafiei A. Predicting the bake hardenability of steels using neural network modeling. Mater. Lett., 2008, 62(2): 173.

[64] X.Y. Huang, H. Wang, W.H. Xue, et al., Study on time-temperature-transformation diagrams of stainless steel using machine-learning approach, Comput. Mater. Sci., 171(2020), art. No. 109282.

[65] Kang HC, Park BJ, Jang JH, Jang KS, Lee KJ. Determination of the continuous cooling transformation diagram of a high strength low alloyed steel. Met. Mater. Int., 2016, 22(6): 949.

[66] Chen R, Zheng Z, Li N, Li J, Feng F. In-situ investigation of phase transformation behaviors of 300M steel in continuous cooling process. Mater. Charact., 2018, 144, 400.

[67] L. Qiao, J.C. Zhu, and Y. Wang, Machine learning-aided process design: Modeling and prediction of transformation temperature for pearlitic steel, Steel Res. Int., 93(2022), No. 1, art. No. 2100267.

[68] Z. Jančíková, V. Roubíček, and D. Juchelková, Application of artificial intelligence methods for prediction of steel mechanical properties, Metalurgija, 47(2008), No. 4, art. No. 339.

[69] X. Wei, S. van der Zwaag, Z. Jia, C. Wang, and W. Xu, On the use of transfer modeling to design new steels with excellent rotating bending fatigue resistance even in the case of very small calibration datasets, Acta Mater., 235(2022), art. No. 118103.

[70] Du JL, Feng YL, Zhang M. Construction of a machine-learning-based prediction model for mechanical properties of ultra-fine-grained Fe-C alloy. J. Mater. Res. Technol., 2021, 15, 4914.

[71] Li J, Cheng JH, Shi JY, Huang F. Brief introduction of back propagation (BP) neural network algorithm and its improvement. Advances in Computer Science and Information Engineering, 2012, Berlin, Heidelberg, Springer, 553.

[72] Diao Y, Yan L, Gao K. A strategy assisted machine learning to process multi-objective optimization for improving mechanical properties of carbon steels. J. Mater. Sci. Technol., 2022, 109, 86.

[73] Lee Rodgers J, Nicewander WA. Thirteen ways to look at the correlation coefficient. Am. Stat., 1988, 42(1): 59.

[74] Chen TY, He LF, Cullison MH, et al. The correlation between microstructure and nanoindentation property of neutron-irradiated austenitic alloy D9. Acta Mater., 2020, 195, 433.

[75] Guo Z, Sha W, Vaumousse D. Microstructural evolution in a PH13-8 stainless steel after ageing. Acta Mater., 2003, 51(1): 101.

[76] I.D. Jung, D.S. Shin, D. Kim, et al., Artificial intelligence for the prediction of tensile properties by using microstructural parameters in high strength steels, Materialia, 11(2020), art. No. 100699.

[77] Wang ZL, Adachi Y. Property prediction and properties-to-microstructure inverse analysis of steels by a machine-learning approach. Mater. Sci. Eng. A, 2019, 744, 661.

[78] Y. Adachi, N. Sato, M. Ojima, M. Nakayama, and Y.T. Wang, Development of fully automated serial-sectioning 3D microscope and topological approach to pearlite and dual-phase microstructure in steels, [in] Proceedings of the 1st International Conference on 3D Materials Science, Pennsylvania, 2012, p. 37.

[79] Dietterich T. Overfitting and undercomputing in machine learning. ACM Comput. Surv., 1995, 27(3): 326.

[80] Golnabi H, Asadpour A. Design and application of industrial machine vision systems. Robotics Comput. Integr. Manuf., 2007, 23(6): 630.

[81] F.A. Saiz, I. Serrano, I. Barandiarán, and J.R. Sánchez, A robust and fast deep learning-based method for defect classification in steel surfaces, [in] 2018 International Conference on Intelligent Systems (IS), Funchal, 2018, p. 455.

[82] Song K, Yan Y. A noise robust method based on completed local binary patterns for hot-rolled steel strip surface defects. Appl. Surf. Sci., 2013, 285, 858.

[83] Krizhevsky A, Sutskever I, Hinton GE. ImageNet classification with deep convolutional neural networks. Commun. ACM, 2017, 60(6): 84.

[84] A. Zunger, Inverse design in search of materials with target functionalities, Nat. Rev. Chem., 2(2018), art. No. 0121.

[85] Butler KT, Davies DW, Cartwright H, Isayev O, Walsh A. Machine learning for molecular and materials science. Nature, 2018, 559(7715): 547.

[86] Xie JX, Su YJ, Xue DZ, Jiang X, Fu HD, Huang HY. Machine learning for materials research and development. Acta Metall. Sin., 2021, 57(11): 1343.

[87] Baluc N, Gelles DS, Jitsukawa S, et al. Status of reduced activation ferritic/martensitic steel development. J. Nucl. Mater., 2007, 367–370, 33.

[88] X. Li, M. Zheng, X. Yang, P. Chen, and W. Ding, A property-oriented design strategy of high-strength ductile RAFM steels based on machine learning, Mater. Sci. Eng. A, 840(2022), art. No. 142891.

[89] Chaudhary N, Abu-Odeh A, Karaman I, Arróyave R. A data-driven machine learning approach to predicting stacking faulting energy in austenitic steels. J. Mater. Sci., 2017, 52(18): 11048.

[90] Nenchev B, Tao Q, Dong ZH, et al. Evaluating data-driven algorithms for predicting mechanical properties with small datasets: A case study on gear steel hardenability. Int. J. Miner. Metall. Mater., 2022, 29(4): 836.

[91] S.L. Liu, Y.J. Su, H.Q. Yin, et al., An infrastructure with user-centered presentation data model for integrated management of materials data and services, Npj Comput. Mater., 7(2021), art. No. 88.

[92] Agarwal R, Dhar V. Big data, data science, and analytics: The opportunity and challenge for IS research. Inf. Syst. Res., 2014, 25(3): 443.

[93] B.Y. Ma, X.Y. Wei, C.N. Liu, et al., Data augmentation in microscopic images for material data mining, Npj Comput. Mater., 6(2020), art. No. 125.