journal article·2023-06Open access

Explainable machine learning for predicting the mechanical properties in bainitic steels

Marc Ackermann, Deniz Iren, Yao Yao

Materials & Design· Volume 230

Abstract

Microstructures in steel can be understood as hierarchical structures holding information on various length scales (e.g. from nano to macro scale). In electron microscopy images, the microscope resolution allows extraction of meso-scale information of grain interiors and boundaries encrypted in morphological features. So far, experts design experiments to extract and select important microstructural features based on their experience and evidence from literature. These features often serve as indicator for interpretation of corresponding mechanical properties. But such approaches commonly capture only a few features while others remain hidden in the microstructure. This becomes evident for complex bainitic microstructures in contrast to other microstructures in steel. In this study, we use a machine learning approach to correlate the most important morphological features, retrieved from segmented Martensite-Austenite (M−A) islands in microscopy images of bainite, with the impact energy from mechanical testing. The used open-source framework selects the most important features and explains their global and individual contributions on the model output. We test the learned correlation on generalizability of these microstructure-property linkages on bainitic steels with different chemical compositions and applied cooling regimes.
Source: openalex_inverted_index · Quality: baseline

Research context

OpenAlex topics: Microstructure and Mechanical Properties of Steels · Hydrogen embrittlement and corrosion behaviors in metals · Force Microscopy Techniques and Applications

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

References (55)

Show source-provided references
1. Hall-Petch relation and boundary strengthening10.1016/j.scriptamat.2004.06.002 ↗
2. Hall-Petch Behavior in Ultra-Fine-Grained AISI 301LN Stainless Steel10.1007/s11661-007-9143-4 ↗
3. On the Hall-Petch relationship and substructural evolution in type 316L stainless steel10.1016/0956-7151(95)00110-H ↗
4. Influence of bainite morphology on impact toughness of continuously cooled cementite free bainitic steels10.1179/1743284710Y.0000000047 ↗
5. Influence of martensite-austenite (MA) on impact toughness of X80 line pipe steels10.1016/j.msea.2016.03.095 ↗
6. Microstructure of martensite–austenite constituents in heat affected zones of high strength low alloy steel welds in relation to toughness properties10.1179/136217100101538164 ↗
7. Characterisation and Quantification of Complex Bainitic Microstructures in High and Ultra-High Strength Linepipe Steels10.4028/www.scientific.net/MSF.500-501.387 ↗
8. Chemistry and three-dimensional morphology of martensite-austenite constituent in the bainite structure of low-carbon low-alloy steels10.1016/j.actamat.2017.11.036 ↗
9. Formation of Martensite Austenite Constituent in Continuously Cooled Nb-Bearing Low Carbon Steels10.4028/www.scientific.net/MSF.638-642.3080 ↗
10. Effect of Molybdenum and Cooling Regime on Microstructural Heterogeneity in Bainitic Steel Wires10.1002/srin.201900663 ↗
11. Structural characterization of “carbide-free” bainite in a Fe–0.2C–1.5Si–2.5Mn steel10.1016/j.matchar.2015.02.020 ↗
12. Advanced microstructure classification by data mining methods10.1016/j.commatsci.2018.03.004 ↗
13. Objective microstructure classification by support vector machine (SVM) using a combination of morphological parameters and textural features for low carbon steels10.1016/j.commatsci.2019.01.006 ↗
14. A machine learning–based classification approach for phase diagram prediction10.1016/j.matdes.2022.110497 ↗
15. Advanced Steel Microstructural Classification by Deep Learning Methods10.1038/s41598-018-20037-5 ↗
16. Microstructural classification of unirradiated LiAlO2 pellets by deep learning methods10.1016/j.commatsci.2020.109728 ↗
17. A generic high-throughput microstructure classification and quantification method for regular SEM images of complex steel microstructures combining EBSD labeling and deep learning10.1016/j.jmst.2021.04.009 ↗
18. Application of deep learning to shale microstructure classification10.1016/j.marpetgeo.2022.105842 ↗
19. Author Correction: Unsupervised microstructure segmentation by mimicking metallurgists' approach to pattern recognition10.1038/s41598-021-88173-z ↗
20. Microstructure classification in the unsupervised context10.1016/j.actamat.2021.117434 ↗
21. X. Zhong, B. Gallagher, S. Liu, B. Kailkhura, A. Hiszpanski, T.Y.-J. Han, Explainable machine learning in materials science, npj Comput Mater 8 (2022). Doi: 10.1038/s41524-022-00884-7.10.1038/s41524-022-00884-7 ↗
22. Autonomous characterization of grain size distribution using nonlinear Lamb waves based on deep learning10.1121/10.0014289 ↗
23. Explainable artificial intelligence approach for yield strength prediction in as-cast multi-principal element alloys10.1016/j.mtla.2022.101628 ↗
24. Explainable machine learning for understanding and predicting geometry and defect types in Fe-Ni alloys fabricated by laser metal deposition additive manufacturing10.1016/j.jmrt.2022.11.137 ↗
25. Prediction compressive strength of cement-based mortar containing metakaolin using explainable Categorical Gradient Boosting model10.1016/j.engstruct.2022.114768 ↗
26. Interpretable machine learning workflow for evaluation of the transformation temperatures of TiZrHfNiCoCu high entropy shape memory alloys10.1016/j.matdes.2022.111513 ↗
27. Elucidating the auxetic behavior of cementitious cellular composites using finite element analysis and interpretable machine learning10.1016/j.matdes.2021.110341 ↗
28. Ackermann, Deniz Iren, Sebastian Wesselmecking, Ulrich Krupp, Deekshith Shetty, M-A island segmentation of bainite microstructures in SEM image data, figshare, 2022.
29. Automated segmentation of martensite-austenite islands in bainitic steel10.1016/j.matchar.2022.112091 ↗
30. Design of Novel High-Strength Bainitic Steels10.4028/www.scientific.net/MSF.426-432.1337 ↗
31. Retained Austenite Characteristics and Tensile Properties in a TRIP Type Bainitic Sheet Steel10.2355/isijinternational.40.902 ↗
32. Design of cold rolled and continuous annealed carbide-free bainitic steels for automotive application10.1016/j.matdes.2013.02.046 ↗
33. Methods to Classify Bainite in Wire Rod Steel10.1002/srin.202000454 ↗
34. Materials Knowledge Systems in Python - A Data Science Framework for Accelerated Development of Hierarchical Materials10.1007/s40192-017-0089-0 ↗
35. A. Moez, PyCaret: An open source, low-code machine learning library in Python, 1st ed., 2020.
36. Calculation of phase fraction in steel microstructure images using random forest classifier10.1049/iet-ipr.2017.1154 ↗
37. Automatic recognition of SEM microstructure and phases of steel using LBP and random decision forest operator10.1016/j.measurement.2019.107224 ↗
38. Automated image analysis for quantification of materials microstructure evolution10.1088/1361-651X/abfd1a ↗
39. Predicting microstructure-dependent mechanical properties in additively manufactured metals with machine- and deep-learning methods10.1016/j.commatsci.2020.109599 ↗
40. H. Sabnis, J. Angel Arul Jothi, A.M. Deva Prasad, Microstructure Image Classification of Metals Using Texture Features and Machine Learning, in: K.K. Patel, G. Doctor, A. Patel, P. Lingras (Eds.), Soft Computing and its Engineering Applications, Springer International Publishing, Cham, 2022, pp. 235–248.10.1007/978-3-031-05767-0_19 ↗
41. A.V. Dorogush, V. Ershov, A. Gulin, CatBoost: gradient boosting with categorical features support, 2018.
42. T. Chen, C. Guestrin, XGBoost: A scalable tree boosting system, Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining 785–794. Doi: 10.1145/2939672.2939785.10.1145/2939672.2939785 ↗
43. Random search for hyper-parameter optimization
44. A unified approach to interpreting model predictions
45. Random forests10.1023/A:1010933404324 ↗
46. Greedy Function Approximation: A Gradient Boosting Machine10.1214/aos/1013203451 ↗
47. Effect of morphologies of martensite–austenite constituents on impact toughness in intercritically reheated coarse-grained heat-affected zone of HSLA steel10.1016/j.msea.2017.10.079 ↗
48. Microstructural Classification of Bainitic Subclasses in Low-Carbon Multi-Phase Steels Using Machine Learning Techniques10.3390/met11111836 ↗
49. Role of retained austenite with different morphologies on sub-surface fatigue crack initiation in advanced bainitic steels10.1016/j.scriptamat.2020.03.036 ↗
50. Mechanical properties of carbide-free lower bainite in complex-alloyed constructional steel: Effect of bainitizing treatment parameters10.4149/km_2020_2_129 ↗
51. Multiphase microstructure formation and its effect on fracture behavior of medium carbon high silicon high strength steel10.1016/j.jmst.2020.09.034 ↗
52. Effect of Ti content and martensite–austenite constituents on microstructure and mechanical property10.1080/13621718.2017.1404694 ↗
53. Effect of martensite-austenite (MA) distribution on mechanical properties of inter-critical Reheated Coarse Grain heat affected zone in X80 linepipe steel10.1016/j.msea.2019.138301 ↗
54. The correlation between martensite-austenite islands evolution and fatigue behavior of SA508-IV steel10.1016/j.ijfatigue.2020.105776 ↗
55. Characterization of martensite-austenite constituents and micro-hardness in intercritical reheated and coarse-grained heat affected zones of API X70 HSLA steel10.1016/j.matchar.2018.05.057 ↗