Neural network representability of fully ionized plasma fluid model closures

Romit Maulik, Nathan A. Garland, Joshua W. Burby, Xian Zhu Tang, Prasanna Balaprakash

Research output: Contribution to journalArticlepeer-review

17 Scopus citations

Abstract

The closure problem in fluid modeling is a well-known challenge to modelers aiming to accurately describe their systems of interest. Over many years, analytic formulations in a wide range of regimes have been presented but a practical, generalized fluid closure for magnetized plasmas remains an elusive goal. In this study, as a first step toward constructing a novel data-based approach to this problem, we apply ever-maturing machine learning methods to assess the capability of neural network architectures to reproduce crucial physics inherent in known magnetized plasma closures. We find encouraging results, indicating the applicability of neural networks to closure physics, but also arrive at recommendations on how one should choose appropriate network architectures for the given locality properties dictated by the underlying physics of the plasma.

Original languageEnglish
Article number072106
JournalPhysics of Plasmas
Volume27
Issue number7
DOIs
StatePublished - Jul 1 2020
Externally publishedYes

Funding

We acknowledge the productive discussions with Dr. Sandeep Madireddy and Dr. Bethany Lusch for this article. This material is based upon work supported by the U.S. Department of Energy (DOE), Office of Science, Office of Advanced Scientific Computing Research, under Contract No. DE-AC02–06CH11357. This research was funded in part and used resources of the Argonne Leadership Computing Facility, which is a DOE Office of Science User Facility supported under Contract No. DE-AC02–06CH11357. Part of this work was performed at the Los Alamos National Laboratory (Contract No. 89233218CNA000001) with support from the SciDAC project on Tokamak Disruption Simulation (TDS) by the Office of Fusion Energy Science and the Office of Advanced Scientific Computing and by the Los Alamos National Laboratory LDRD program under Project No. 20180756PRD4.

FundersFunder number
DOE Office of Science
Office of Fusion Energy Science
U.S. Department of Energy
Office of Science
Advanced Scientific Computing ResearchDE-AC02–06CH11357
Los Alamos National Laboratory89233218CNA000001, 20180756PRD4

    Fingerprint

    Dive into the research topics of 'Neural network representability of fully ionized plasma fluid model closures'. Together they form a unique fingerprint.

    Cite this