Developing a Learning Algorithm-Generated Empirical Relaxer
- Univ. of Colorado, Boulder, CO (United States). Dept. of Applied Math
- Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
One of the main difficulties when running Arbitrary Lagrangian-Eulerian (ALE) simulations is determining how much to relax the mesh during the Eulerian step. This determination is currently made by the user on a simulation-by-simulation basis. We present a Learning Algorithm-Generated Empirical Relaxer (LAGER) which uses a regressive random forest algorithm to automate this decision process. We also demonstrate that LAGER successfully relaxes a variety of test problems, maintains simulation accuracy, and has the potential to significantly decrease both the person-hours and computational hours needed to run a successful ALE simulation.
- Research Organization:
- Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States)
- Sponsoring Organization:
- USDOE
- DOE Contract Number:
- AC52-07NA27344
- OSTI ID:
- 1248278
- Report Number(s):
- LLNL-TR-687141
- Country of Publication:
- United States
- Language:
- English
Similar Records
Preparing Dense Net for Automated HYDRA Mesh Management via Reinforcement Learning
Adaptive node techniques for Maxwell's equations
Pairwise Difference Regression: A Machine Learning Meta-algorithm for Improved Prediction and Uncertainty Quantification in Chemical Search
Technical Report
·
Sun Dec 15 00:00:00 EST 2019
·
OSTI ID:1248278
+5 more
Adaptive node techniques for Maxwell's equations
Conference
·
Sat Apr 01 00:00:00 EST 2000
·
OSTI ID:1248278
Pairwise Difference Regression: A Machine Learning Meta-algorithm for Improved Prediction and Uncertainty Quantification in Chemical Search
Journal Article
·
Wed Aug 04 00:00:00 EDT 2021
· Journal of Chemical Information and Modeling
·
OSTI ID:1248278
+4 more