U.S. Department of EnergyOffice of Scientific and Technical Information
Uncovering turbulent plasma dynamics via deep learning from partial observations
Abstract
One of the most intensely studied aspects of magnetic confinement fusion is edge plasma turbulence which is critical to reactor performance and operation. Drift-reduced Braginskii two-fluid theory has for decades been widely applied to model boundary plasmas with varying success. Towards better understanding edge turbulence in both theory and experiment, we demonstrate that a novel multi-network physics-informed deep learning framework constrained by partial differential equations can accurately learn turbulent fields consistent with the two-fluid theory from partial observations of electron pressure which is not otherwise possible using conventional equilibrium models. This technique presents a novel paradigm for the advanced design of plasma diagnostics and validation of magnetized plasma turbulence theories in challenging thermonuclear environments.
- Authors:
-
- OSTI
- Publication Date:
- DOE Contract Number:
- SC0014264; FC02-08ER54966; FG02-04ER54742; AC52-07NA27344
- Research Org.:
- Massachusetts Inst. of Technology (MIT), Cambridge, MA (United States). Plasma Science and Fusion Center; Univ. of Texas, Austin, TX (United States); Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
- Sponsoring Org.:
- USDOE Office of Science (SC), Fusion Energy Sciences (FES); USDOE National Nuclear Security Administration (NNSA)
- Subject:
- 70 PLASMA PHYSICS AND FUSION TECHNOLOGY
- OSTI Identifier:
- 1882600
- DOI:
- https://doi.org/10.7910/DVN/JNMLN2
Citation Formats
Mathews, Abhilash, Francisquez, Manaure, Hughes, Jerry, Hatch, David, Zhu, Ben, and Rogers, Barrett. Uncovering turbulent plasma dynamics via deep learning from partial observations. United States: N. p., 2021.
Web. doi:10.7910/DVN/JNMLN2.
Mathews, Abhilash, Francisquez, Manaure, Hughes, Jerry, Hatch, David, Zhu, Ben, & Rogers, Barrett. Uncovering turbulent plasma dynamics via deep learning from partial observations. United States. doi:https://doi.org/10.7910/DVN/JNMLN2
Mathews, Abhilash, Francisquez, Manaure, Hughes, Jerry, Hatch, David, Zhu, Ben, and Rogers, Barrett. 2021.
"Uncovering turbulent plasma dynamics via deep learning from partial observations". United States. doi:https://doi.org/10.7910/DVN/JNMLN2. https://www.osti.gov/servlets/purl/1882600. Pub date:Mon May 24 04:00:00 UTC 2021
@article{osti_1882600,
title = {Uncovering turbulent plasma dynamics via deep learning from partial observations},
author = {Mathews, Abhilash and Francisquez, Manaure and Hughes, Jerry and Hatch, David and Zhu, Ben and Rogers, Barrett},
abstractNote = {One of the most intensely studied aspects of magnetic confinement fusion is edge plasma turbulence which is critical to reactor performance and operation. Drift-reduced Braginskii two-fluid theory has for decades been widely applied to model boundary plasmas with varying success. Towards better understanding edge turbulence in both theory and experiment, we demonstrate that a novel multi-network physics-informed deep learning framework constrained by partial differential equations can accurately learn turbulent fields consistent with the two-fluid theory from partial observations of electron pressure which is not otherwise possible using conventional equilibrium models. This technique presents a novel paradigm for the advanced design of plasma diagnostics and validation of magnetized plasma turbulence theories in challenging thermonuclear environments.},
doi = {10.7910/DVN/JNMLN2},
journal = {},
number = ,
volume = ,
place = {United States},
year = {2021},
month = {5}
}