Generating Input Data for Microstructure Modelling: A Deep Learning Approach Using Generative Adversarial Networks.

Advanced Search

Felix Pütz, Manuel Henrich, Niklas Fehlemann, Andreas Roth, Sebastian Münstermann

Author Information

Felix Pütz: Integrity of Materials and Structures, RWTH Aachen University, 52062 Aachen, Germany. ORCID
Manuel Henrich: Integrity of Materials and Structures, RWTH Aachen University, 52062 Aachen, Germany. ORCID
Niklas Fehlemann: Integrity of Materials and Structures, RWTH Aachen University, 52062 Aachen, Germany.
Andreas Roth: Integrity of Materials and Structures, RWTH Aachen University, 52062 Aachen, Germany.
Sebastian Münstermann: Integrity of Materials and Structures, RWTH Aachen University, 52062 Aachen, Germany. ORCID

PMID: 32977556 DOI: 10.3390/ma13194236

For the generation of representative volume elements a statistical description of the relevant parameters is necessary. These parameters usually describe the geometric structure of a single grain. Commonly, parameters like area, aspect ratio, and slope of the grain axis relative to the rolling direction are applied. However, usually simple distribution functions like log normal or gamma distribution are used. Yet, these do not take the interdependencies between the microstructural parameters into account. To fully describe any metallic microstructure though, these interdependencies between the singular parameters need to be accounted for. To accomplish this representation, a machine learning approach was applied in this study. By implementing a Wasserstein generative adversarial network, the distribution, as well as the interdependencies could accurately be described. A validation scheme was applied to verify the excellent match between microstructure input data and synthetically generated output data.

deep learning dp-steel machine learning microstructure modelling representative volume elements wasserstein gan

Ultramicroscopy. 2011 Dec;111(12):1720-33 [PMID: 22094374]
PLoS Comput Biol. 2016 Mar 28;12(3):e1004845 [PMID: 27018908]
Materials (Basel). 2018 May 09;11(5): [PMID: 29747417]
Materials (Basel). 2020 Apr 17;13(8): [PMID: 32316429]
Sci Rep. 2013 Sep 30;3:2810 [PMID: 24077117]

278868966/Deutsche Forschungsgemeinschaft
390621612/Deutsche Forschungsgemeinschaft

Journal Article

Generating mobility networks with generative adversarial networks.Independent brain F-FDG PET attenuation correction using a deep learning approach with Generative Adversarial Networks.A deep learning approach to private data sharing of medical images using conditional generative adversarial networks (GANs).Cancer diagnosis using generative adversarial networks based on deep learning from imbalanced data.GluGAN: Generating Personalized Glucose Time Series Using Generative Adversarial Networks.Microstructure synthesis using style-based generative adversarial networks.Adversarial Patch Attacks on Deep-Learning-Based Face Recognition Systems Using Generative Adversarial Networks.Generating Synthetic MR Spectroscopic Imaging Data with Generative Adversarial Networks to Train Machine Learning Models.Oncological Applications of Deep Learning Generative Adversarial Networks.EMG Data Augmentation for Grasp Classification Using Generative Adversarial Networks.

DRAGen - A deep learning supported RVE generator framework for complex microstructure models.FEM-GAN: A Physics-Supervised Deep Learning Generative Model for Elastic Porous Materials.

OpenLB
Open Library of Bioscience