A multidimensional continuous conditional generative adversarial network to link composition-process-microstructure-properties in Mg alloys
Source abstract
Abstract
This investigation presents a novel multidimensional continuous conditional generative adversarial network that is able to create alloy microstructures directly from composition-process variables in a single forward pass. The model is applied to discover ternary Mg-Al-Sn alloys manufactured by casting with high ultimate tensile strength and low corrosion rate. From the target performance indices, the model provides the candidate alloy compositions together with their corresponding process parameters, and the resulting microstructures. Experimental validation reveals deviations of only 1.4 % for the strength and 7.5 % for the corrosion rate with respect to the predictions while the errors in the fraction of each phase are below 0.4 %. By coupling continuous microstructure generation with multi-objective optimization, this work provides a scalable artificial intelligence-based inverse-design paradigm for magnesium and broader engineering alloys.
Source: openalex_inverted_index · Quality: baseline
Research context
OpenAlex topics: Machine Learning in Materials Science · Magnesium Alloys: Properties and Applications · Microstructure and mechanical properties
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (56)
Show source-provided references
1. Machine-learning design of ductile FeNiCoAlTa alloys with high strength10.1038/s41586-025-09160-2 ↗
2. Closed-loop inverse design of high entropy alloys using symbolic regression-oriented optimization10.1016/j.mattod.2025.06.033 ↗
3. Language models for materials discovery and sustainability: progress, challenges, and opportunities10.1016/j.pmatsci.2025.101495 ↗
4. Design biomedical β-Ti alloys with exceptional strength-ductility balance via domain knowledge-based machine learning10.1016/j.actamat.2025.121550 ↗
5. From creep-life prediction to ultra-creep-resistant steel design: an uncertainty-informed machine learning approach10.1016/j.actamat.2025.121073 ↗
6. Towards high entropy alloy with enhanced strength and ductility using domain knowledge constrained active learning10.1016/j.matdes.2022.111186 ↗
7. Machine learning pipeline for structure–Property modeling in Mg-alloys using microstructure and texture descriptors10.1016/j.actamat.2025.121132 ↗
8. Explainable machine learning-enabled dual-objective design of γ' phase characteristic parameters in γ'-strengthened Co-based superalloys10.1038/s41524-025-01797-x ↗
9. Multi-objective materials bayesian optimization with active learning of design constraints: design of ductile refractory multi-principal-element alloys10.1016/j.actamat.2022.118133 ↗
10. Bayesian optimization and explainable machine learning for high-dimensional multi-objective optimization of biodegradable magnesium alloys10.1016/j.jmst.2025.02.059 ↗
11. Accelerated design of high-performance Mg-Mn-based magnesium alloys based on novel bayesian optimization10.1016/j.jma.2024.01.005 ↗
12. Discovering novel lead-free solder alloy by multi-objective bayesian active learning with experimental uncertainty10.1038/s41524-024-01480-7 ↗
13. Spatiotemporal microstructure evolution during martensitic transformation in titanium alloys using deep learning10.1016/j.actamat.2025.121603 ↗
14. Multiscale computational framework linking alloy composition to microstructure evolution via machine learning and nanoscale analysis10.1038/s41524-025-01730-2 ↗
15. Multi-scale simplified residual convolutional neural network model for predicting compositions of binary magnesium alloys10.1016/j.jma.2025.06.005 ↗
16. Application of machine learning to assess the influence of microstructure on twin nucleation in Mg alloys10.1038/s41524-024-01212-x ↗
17. Active learning framework to optimize process parameters for additive-manufactured Ti-6Al-4V with high strength and ductility10.1038/s41467-025-56267-1 ↗
18. Deep learning-driven innovation in metallic materials: a comprehensive review on microstructure analysis, property prediction, and inverse design10.1016/j.jmst.2026.02.007 ↗
19. Deep learning in shock response of metastable high entropy alloy: from prediction to inverse microstructure design10.1016/j.matdes.2025.115159 ↗
20. A review of deep learning in metal additive manufacturing: impact on process, structure, and properties10.1016/j.pmatsci.2025.101587 ↗
21. Fitting-free mechanical response prediction in dual-phase steels by crystal plasticity theory guided deep learning10.1016/j.actamat.2025.120936 ↗
22. From microstructure to mechanical properties: image-based machine learning prediction for AZ80 magnesium alloy10.1016/j.jma.2025.07.019 ↗
23. FIP-GNN: graph neural networks for scalable prediction of grain-level fatigue indicator parameters10.1016/j.scriptamat.2024.116407 ↗
24. Materials fatigue prediction using graph neural networks on microstructure representations
25. Machine learning of microstructure–property relationships in materials leveraging microstructure representation from foundational vision transformers10.1016/j.actamat.2025.121217 ↗
26. Microstructural characterization and hardness prediction of AlCuCrFeNi high entropy alloys using Transformer-based Physics-informed Neural networks (T-PINN)10.1016/j.jmrt.2025.08.021 ↗
27. PCS: property-composition-structure chain in Mg-Nd alloys through integrating sigmoid fitting and conditional generative adversarial network modeling10.1016/j.scriptamat.2025.116762 ↗
28. Plastic deformation and damage modeling of AA7075 synthetic 3D microstructure created using generative AI10.1016/j.actamat.2024.120431 ↗
29. A machine learning method to quantitatively predict alpha phase morphology in additively manufactured Ti-6Al-4V, npj10.1038/s41524-023-01152-y ↗
30. Analysis of the strength–ductility balance of dual-phase steel using a combination of generative adversarial networks and finite element method10.1016/j.commatsci.2024.113143 ↗
31. High-fidelity grain growth modeling: leveraging deep learning for fast computations10.1016/j.actamat.2025.121486 ↗
32. Towards inverse microstructure-centered materials design using generative phase-field modeling and deep variational autoencoders10.1016/j.actamat.2023.119204 ↗
33. Exploring descriptors for titanium microstructure via digital fingerprints from variational autoencoders10.1016/j.commatsci.2024.112992 ↗
34. Denoising diffusion probabilistic models for generative alloy design
35. 3D Denoising diffusion probabilistic models for 3D microstructure image generation of fuel cell electrodes10.1016/j.commatsci.2024.113596 ↗
36. A call to elevate the role of processing in AI-driven materials design10.1038/s41578-025-00846-7 ↗
37. Recent applications of machine learning in alloy design: a review10.1016/j.mser.2023.100746 ↗
38. PSP-GEN: stochastic inversion of the process–Structure–Property chain in materials design through deep, generative probabilistic modeling10.1016/j.actamat.2024.120600 ↗
39. Enhanced strength-ductility of a Mg-Al-Sn alloy by the synergy of solute segregation and nano-sized precipitates10.1016/j.jmst.2025.02.094 ↗
40. Microstructural tuning enabled by friction stir processing for enhanced corrosion and discharge performance of a Mg-Al-Sn anode10.1016/j.est.2026.120442 ↗
41. Discharge and corrosion behavior of Mg-Al-Sn anodes: an in-depth study of the morphological impact of Mg2Sn phase10.1016/j.jpowsour.2024.235935 ↗
42. A novel neural network-based alloy design strategy: gated recurrent unit machine learning modeling integrated with orthogonal experiment design and data augmentation10.1016/j.actamat.2022.118420 ↗
43. Rapid design of high-end copper alloy processes combining orthogonal experiments, machine learning, and Pareto analysis10.1016/j.jmrt.2025.03.119 ↗
44. Descriptor selection for predicting interfacial thermal resistance by machine learning methods10.1038/s41598-020-80795-z ↗
45. Predictive microstructure image generation using denoising diffusion probabilistic models10.1016/j.actamat.2023.119406 ↗
46. Uncertainty-aware machine learning framework for predicting dislocation plasticity and stress–strain response in metallic alloys, Part I : FCC systems10.1016/j.actamat.2025.121610 ↗
47. Uncertainty propagation in a multiscale CALPHAD-reinforced elastochemical phase-field model10.1016/j.actamat.2019.11.031 ↗
48. Design and properties of ultra-hard high-entropy alloys based on machine learning and statistical feature selection10.1016/j.jallcom.2025.184356 ↗
49. Interpretable machine learning excavates a low-alloyed magnesium alloy with strength-ductility synergy based on data augmentation and reconstruction10.1016/j.jma.2025.01.003 ↗
50. Composition optimization and mechanical properties of Mg-Al-Sn-Mn alloys by orthogonal design10.3390/ma11081424 ↗
51. Influence of solute elements (Sn and Al) on microstructure evolution of Mg alloys: an experimental and simulation study10.1016/j.jcrysgro.2018.09.032 ↗
52. Effects of heat treatment on the corrosion behavior and mechanical properties of biodegradable Mg alloys10.1016/j.jma.2022.04.010 ↗
53. Effect of solution treatment on microstructure and corrosion resistance of AZ91HP magnesium alloy
54. Influences of metallurgical factors on the corrosion behaviour of extruded binary Mg–Sn alloys10.1016/j.corsci.2014.01.035 ↗
55. The influence of Mg17Al12 phase volume fraction on the corrosion behaviour of AZ91 magnesium alloy
56. Influence of β-Mg17Al12 and Al-Mn intermetallic compounds on the corrosion behaviour of cast and solution treated Mg-Al-Zn-Mn alloys10.1016/j.corsci.2023.111363 ↗