DOE OSTI · 1878014
Enabling Predictive Scale-Bridging Simulations through Active Learning (Institutional Computing Annual Report (Project w21_alscalebridging)) [Slides]
Abstract
The goal of this project was to develop, demonstrate, and provide a new capability to achieve greater physical fidelity in large-scale simulations, rather than the usual brute-force increases in the number of mesh elements or particles. Transport in nanoporous media, critical to hydrocarbon extraction from tight shale formations, is affected by molecular-level interactions. Several coarse-scale Lattice Boltzmann model (LBM) parameters cannot be directly computed, so instead we calibrate them to molecular dynamics (MD) simulations, by building emulators that mimic MD and LBM behavior and then training an upscaler. The resulting machine learning (ML)-based scale-bridging framework is up to 7 orders of magnitude faster than direct MD.
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
Germann, Timothy Clark. 2022-07-22. Enabling Predictive Scale-Bridging Simulations through Active Learning (Institutional Computing Annual Report (Project w21_alscalebridging)) [Slides]. https://doi.org/10.2172/1878014. https://searchnasa.com/record/osti%3A1878014/enabling-predictive-scale-bridging-simulations-through-active-learning-institutional-computing-annua
Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite Search NASA by Indiciae: BibTeX · RIS.