Abstract
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.
Highlights
For modern applications, the microstructures and properties of steels have become exceedingly complex, utilising multiple phases and alloying concepts to improve mechanical properties for the specific use case
A significant improvement of the fit is visible, where epoch 200 is an unoptimised guess, while the network improves over time with the training and is able to represent the distributions of the input parameters at the best fit epoch accurately
This study presented a solution on how to generate input for microstructure modelling that is true to the real microstructure
Summary
The microstructures and properties of steels have become exceedingly complex, utilising multiple phases and alloying concepts to improve mechanical properties for the specific use case. For the damage behaviour of a material grain, shape and size both play a very important role, modern RVE generation algorithms take many parameters into account that describe the grain shape, like slope or elongation [12,13,14,15]. The input for these parameters are separate, independent distribution functions of the applied parameters. This allowed the generation of a virtually unlimited amount of input data, which follows the distribution functions of the individual parameters, and exactly reflects the dependencies between individual parameters
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
