journal article·2023-10Open access

Multi-objective optimization for high-performance Fe-based metallic glasses via machine learning approach

Yu-Xing Zhang, She-Juan Xie, Wei Guo, Jun Ding, Leong Hien Poh, Zhen-Dong Sha

Abstract

Fe-based metallic glasses (MGs) are a class of promising soft magnetic materials that have received great attention in transformer industries. However, it is challenging to achieve a balance between saturation magnetization ( B s), glass-forming ability and plasticity due to their contradictory correlations in Fe-based MGs, which severely hinders the development of new Fe-based MGs with advanced performances. Inspired by the significant development in machine learning technology, we herein propose a multi-objective optimization strategy to search for Fe-based MGs with optimal combinations of critical casting size ( D max), B s , and plasticity. The objective functions are built in combination with neural network models for predicting D max and B s, as well as empirical formula for plasticity. The effect of number of hidden layers is investigated and the dropout regularization method employed to improve the prediction performance. Our results show that the predictions of B s and D max by using alloy composition as the sole input perform well, as evidenced by their r 2 values of 0.963 and 0.874, respectively. Multi-objective optimization based on the genetic algorithm is executed to obtain the Pareto front and Pareto-optimal solutions. The Pareto-optimal alloys predicted for the Fe 83C 1B xSi yP 16-x-y and Fe xCo yNi 72-x-yB 19.2Si 4.8Nb 4 systems are in good agreement with those reported in experiments. This work thus showcases potential applications for the design of high-performance Fe-MGs against conflicting objectives.
Source: expertdata · Quality: candidate unverified · Source record ↗

Research context

OpenAlex topics: Metallic Glasses and Amorphous Alloys · Magnetic Properties and Applications · Magnetic properties of thin films

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

References (75)

Show source-provided references
1. Bulk metallic glasses with functional physical properties10.1002/adma.200901053 ↗
2. Bulk metallic glasses: At the cutting edge of metals research10.1557/mrs2007.121 ↗
3. Metallic glasses as structural materials10.1016/j.scriptamat.2005.09.051 ↗
4. Bulk glassy alloys: Historical development and current research10.15302/J-ENG-2015038 ↗
5. New Fe‐based soft magnetic alloys composed of ultrafine grain structure10.1063/1.342149 ↗
6. Research progress and application prospect of Fe-based soft magnetic amorphous/nanocrystalline alloys
7. Soft magnetic properties of nanocrystalline bcc Fe‐Zr‐B and Fe‐M‐B‐Cu (M=transition metal) alloys with high saturation magnetization10.1063/1.350006 ↗
8. Amorphous and nanocrystalline materials for applications as soft magnets10.1016/S0079-6425(99)00002-X ↗
9. Glass-forming ability of alloys10.1016/0022-3093(93)90003-G ↗
11. A brief overview of bulk metallic glasses10.1038/asiamat.2011.30 ↗
12. The complementary effects of Fe and metalloids on the saturation magnetization of Fe-based amorphous alloys10.1016/j.intermet.2021.107116 ↗
13. Fe-based bulk metallic glasses: Glass formation, fabrication, properties and applications10.1016/j.pmatsci.2019.01.003 ↗
14. Effect of silicon addition on the magnetic properties of Fe‐B‐C amorphous alloys10.1063/1.326860 ↗
15. Effect of B to P concentration ratio on glass-forming ability and soft-magnetic properties in [(Fe0.5Ni0.5)0.78B0.22−xPx]97Nb3 glassy alloys10.1016/j.intermet.2011.08.020 ↗
16. Effect of P addition on microstructure and soft magnetic properties of Fe(73.5−x)Nb3Cu1PxSi13.5B9 alloys10.1016/j.jmmm.2006.01.160 ↗
17. A novel FeNi-based bulk metallic glass with high notch toughness over 70 MPa m1/2 combined with excellent soft magnetic properties10.1016/j.matdes.2020.108597 ↗
18. Effects of Ni and Si additions on mechanical properties and serrated flow behavior in FeMoPCB bulk metallic glasses10.1016/j.jallcom.2018.12.331 ↗
19. Role of Ni and Co in tailoring magnetic and mechanical properties of Fe84Si2B13P1 metallic glass10.1016/j.jallcom.2019.152549 ↗
20. Effect of the substitution of C for Si on microstructure, magnetic properties and bending ductility in high Fe content FeSiBCuPC alloy ribbons10.1016/j.jallcom.2017.08.181 ↗
21. Effect of P addition on glass forming ability and soft magnetic properties of melt-spun FeSiBCuC alloy ribbons10.1016/j.jmmm.2016.05.103 ↗
22. Effect of P on glass forming ability, magnetic properties and oxidation behavior of FeSiBP amorphous alloys10.1016/j.intermet.2016.12.024 ↗
23. Effect of Cu and P on the crystallization behavior of Fe-rich hetero-amorphous FeSiB alloy10.2320/matertrans.M2009206 ↗
24. Machine learning in materials genome initiative: A review10.1016/j.jmst.2020.01.067 ↗
25. Machine learning for molecular and materials science10.1038/s41586-018-0337-2 ↗
26. Property-oriented material design based on a data-driven machine learning technique10.1021/acs.jpclett.0c00665 ↗
27. Accelerated discovery of metallic glasses through iteration of machine learning and high-throughput experiments10.1126/sciadv.aaq1566 ↗
28. A machine learning approach for engineering bulk metallic glass alloys10.1016/j.actamat.2018.08.002 ↗
29. A general-purpose machine learning framework for predicting properties of inorganic materials10.1038/npjcompumats.2016.28 ↗
30. Machine learning approach for prediction and understanding of glass-forming ability10.1021/acs.jpclett.7b01046 ↗
31. Compositional optimization of glass forming alloys based on critical dimension by using artificial neural network10.1016/S1003-6326(14)63213-1 ↗
32. Selecting doping elements by data mining for advanced magnets10.1021/acs.chemmater.9b03379 ↗
33. Interpretable machine-learning strategy for soft-magnetic property and thermal stability in Fe-based metallic glasses10.1038/s41524-020-00460-x ↗
34. Cluster-formula-embedded machine learning for design of multicomponent β-Ti alloys with low Young’s modulus10.1038/s41524-020-00372-w ↗
35. Design of Fe-based bulk metallic glasses for maximum amorphous diameter (Dmax) using machine learning models10.1016/j.commatsci.2020.110230 ↗
36. Machine learning prediction of glass-forming ability in bulk metallic glasses10.1016/j.commatsci.2021.110362 ↗
37. Discovery of novel quaternary bulk metallic glasses using a developed correlation-based neural network approach10.1016/j.commatsci.2020.110025 ↗
38. Designing metallic glasses with optimal combinations of glass-forming ability and mechanical properties10.1016/j.jmst.2020.08.028 ↗
39. Optimizing the design of airfoil and optical buffer problems using spotted hyena optimizer10.3390/designs2030028 ↗
40. Simultaneous optimization of thermal and electrical conductivity of high density polyethylene-carbon particle composites by artificial neural networks and multi-objective genetic algorithm10.1016/j.commatsci.2021.110956 ↗
41. Approximation by superpositions of a sigmoidal function10.1007/BF02551274 ↗
42. V. Nair, G.E. Hinton, Rectified linear units improve restricted boltzmann machines, in: Haifa, 2010: pp. 807–814.
43. X. Glorot, A. Bordes, Y. Bengio, Deep sparse rectifier neural networks, in: AISTATS, 2010: p. 10.
44. Dropout: a simple way to prevent neural networks from overfitting
45. Rawlings · 1998
46. Intrinsic plasticity or brittleness of metallic glasses10.1080/09500830500080474 ↗
47. Intrinsic and extrinsic toughening of metallic glasses10.1016/j.scriptamat.2005.10.010 ↗
48. The elastic properties, elastic models and elastic perspectives of metallic glasses10.1016/j.pmatsci.2011.07.001 ↗
49. Precisely predicting and designing the elasticity of metallic glasses10.1063/1.4880337 ↗
50. Jazzbin, Geatpy: The genetic and evolutionary algorithm toolbox with high performance in python, (2020). 〈http://geatpy.com〉 (accessed December 22, 2021).
51. A fast and elitist multiobjective genetic algorithm: NSGA-II10.1109/4235.996017 ↗
52. An evolutionary many-objective optimization algorithm using reference-point based nondominated sorting approach, part II: handling constraints and extending to an adaptive approach10.1109/TEVC.2013.2281534 ↗
53. Towards more uniform deformation in metallic glasses: The role of Poisson’s ratio10.1016/j.msea.2012.09.096 ↗
54. Effect of microalloying on the toughness of metallic glasses10.1063/1.4769997 ↗
55. A new criterion for predicting the glass-forming ability of alloys based on machine learning10.1016/j.commatsci.2020.110259 ↗
56. Intrinsic correlation between fragility and bulk modulus in metallic glasses10.1103/PhysRevB.76.054204 ↗
57. The novel Ti-based metallic glass with excellent glass forming ability and an elastic constant dependent glass forming criterion10.1016/j.mtla.2019.100433 ↗
58. Machine learning prediction of elastic properties and glass-forming ability of bulk metallic glasses10.1557/mrc.2019.44 ↗
59. A new criterion for evaluating the glass-forming ability of bulk glass forming alloys10.1016/j.jallcom.2007.05.037 ↗
60. Relationship between melting enthalpy ΔHm and critical cooling rate Rc for bulk metallic glasses10.1016/j.msea.2007.01.140 ↗
61. Ductile bulk metallic glass10.1103/PhysRevLett.93.255506 ↗
62. Thulium-based bulk metallic glass10.1063/1.2908047 ↗
63. Tough Fe-based bulk metallic glasses10.1063/1.2890489 ↗
64. Kawazoe · 1997
65. A valence electron concentration criterion for glass-formation ability of metallic liquids10.1063/1.1571984 ↗
66. A two-step fused machine learning approach for the prediction of glass-forming ability of metallic glasses10.1016/j.jallcom.2021.160040 ↗
67. Critical feature space for predicting the glass forming ability of metallic alloys revealed by machine learning10.1016/j.chemphys.2020.110898 ↗
68. A machine-learning approach to predicting and understanding the properties of amorphous metallic alloys10.1016/j.matdes.2019.108378 ↗
69. Machine learning prediction of magnetic properties of Fe-based metallic glasses considering glass forming ability10.1016/j.jmst.2021.05.076 ↗
70. Formation and structure-property correlation of new bulk Fe–B–Si–Hf metallic glasses10.1016/j.matdes.2016.05.102 ↗
71. Generalized slater-pauling curve for transition-metal magnets10.1109/TMAG.1983.1062706 ↗
72. Fe-based amorphous alloys for wide ribbon production with high Bs and outstanding amorphous forming ability10.1016/j.jallcom.2015.01.056 ↗
73. Relevance of Fe atomic volumes for the magnetic properties of Fe-rich metallic glasses10.1016/j.jmmm.2012.07.003 ↗
74. Developments and applications of bulk glassy alloys in late transition metal base system10.2320/matertrans.47.1275 ↗
75. Origin of large plasticity and multiscale effects in iron-based metallic glasses10.1038/s41467-018-03744-5 ↗