Microstructure Generation via Generative Adversarial Network for Heterogeneous, Topologically Complex 3D Materials

Using a large-scale, experimentally captured 3D microstructure data set, we implement the generative adversarial network (GAN) framework to learn and generate 3D microstructures of solid oxide fuel cell electrodes. The generated microstructures are visually, statistically, and topologically realisti...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:JOM (1989) Jg. 73; H. 1; S. 90 - 102
Hauptverfasser: Hsu, Tim, Epting, William K., Kim, Hokon, Abernathy, Harry W., Hackett, Gregory A., Rollett, Anthony D., Salvador, Paul A., Holm, Elizabeth A.
Format: Journal Article
Sprache:Englisch
Veröffentlicht: New York Springer US 01.01.2021
Springer Nature B.V
Springer
Schlagworte:
Algorithms
Augmenting Physics-based Models in ICME with Machine Learning and Uncertainty Quantification
Chemistry/Food Science
Datasets
Earth Sciences
Electrochemical analysis
Engineering
Environment
Finite element method
Generative adversarial networks
Machine learning
MATERIALS SCIENCE
Mathematical models
Microstructure
Neural networks
Particle size
Physics
Simulation
Solid oxide fuel cells
Tortuosity
ISSN:1047-4838, 1543-1851
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Using a large-scale, experimentally captured 3D microstructure data set, we implement the generative adversarial network (GAN) framework to learn and generate 3D microstructures of solid oxide fuel cell electrodes. The generated microstructures are visually, statistically, and topologically realistic, with distributions of microstructural parameters, including volume fraction, particle size, surface area, tortuosity, and triple-phase boundary density, being highly similar to those of the original microstructure. These results are compared and contrasted with those from an established, grain-based generation algorithm (DREAM.3D). Importantly, simulations of electrochemical performance, using a locally resolved finite element model, demonstrate that the GAN-generated microstructures closely match the performance distribution of the original, while DREAM.3D leads to significant differences. The ability of the generative machine learning model to recreate microstructures with high fidelity suggests that the essence of complex microstructures may be captured and represented in a compact and manipulatable form.
Bibliographie:SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 14
89243318CFE000003; CMMI-1826218; FA8650-19-2-5209
USDOE Office of Fossil Energy (FE)
National Science Foundation (NSF)
ISSN:1047-4838
1543-1851
DOI:10.1007/s11837-020-04484-y