Machine learning of turbulent transport in fusion plasmas with neural network

Li, Hui; Fu, Yanlin; Li, Jiquan; Wang, Jialei

doi:10.1088/2058-6272/ac15ec

Cited by 16 publications

(15 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Li provided details of a new machine learning-based model of turbulent transport. Simulation data from the extended fluid code (ExFC) which include both ITG and TEM turbulence was used to train a neural network model, ExFC-NN [93]. It was demonstrated that ExFC-NN could successfully reproduce the dominant turbulence instability, turbulence flux magnitudes, and radial profiles of fluctuations and fluxes.…”

Section: Development Coupling and Application Of Reduced Core Turbule...mentioning

confidence: 99%

Joint meeting of 9th Asia Pacific-Transport Working Group (APTWG) & EU-US Transport Task Force (TTF) workshop

Ida¹,

McDermott²,

Holland³

et al. 2022

Nucl. Fusion

View full text Add to dashboard Cite

This conference report summarizes the contributions to, and discussions at the joint meeting of the 9th Asia Pacific-Transport Working Group (APTWG) & EU-US Transport Task Force (TTF) workshop held online, hosted by Kyushu University, Japan, during 6-9 July 2021. The topics of the meeting were organized under five main topics: 1)Isotope effect on transport and physics on isotope mixture plasma, 2)Turbulence spreading and coupling in core-edge-SOL, 3)Interplay between MHD topology/instability and turbulent transport, 4)Interaction between energetic particle driven instability and transport, 5)Model reduction and experiments for validation.

show abstract

Section: Development Coupling and Application Of Reduced Core Turbule...mentioning

confidence: 99%

Joint meeting of 9th Asia Pacific-Transport Working Group (APTWG) & EU-US Transport Task Force (TTF) workshop

Ida¹,

McDermott²,

Holland³

et al. 2022

Nucl. Fusion

View full text Add to dashboard Cite

show abstract

“…Machine learning is a tool that is gaining increasing usage in the nuclear fusion community and can be useful in this type of situation where it is desirable to quickly synthesize data into interpretable outputs. Some examples include its usage in profile predictions [1][2][3], turbulent transport [4], identifying scalings laws [5,6], and predicting instabilities and disruptions [7][8][9][10][11][12][13].…”

Section: Introductionmentioning

confidence: 99%

Neural net modeling of equilibria in NSTX-U

Wai,

Boyer,

Kolemen

2022

Preprint

View full text Add to dashboard Cite

Neural networks (NNs) offer a path towards synthesizing and interpreting data on faster timescales than traditional physics-informed computational models. In this work we develop two neural networks relevant to equilibrium and shape control modeling, which are part of a suite of tools being developed for the National Spherical Torus Experiment-Upgrade (NSTX-U) for fast prediction, optimization, and visualization of plasma scenarios. The networks include Eqnet, a freeboundary equilibrium solver trained on the EFIT01 reconstruction algorithm, and Pertnet, which is trained on the Gspert code and predicts the non-rigid plasma response, a nonlinear term that arises in shape control modeling. The equilibrium neural network is trained with different combinations of inputs and outputs in order to offer flexibility in use cases. In particular, the NN can use magnetic diagnostics as inputs for equilibrium prediction thus acting as a reconstruction code, or can use profiles and external currents as inputs to act as a traditional free-boundary Grad-Shafranov solver. We report strong performance for both networks indicating that these models could reliably be used within closed-loop simulations. Some limitations regarding generalizability and noise are discussed.

show abstract

“…Further details of the equations and the simulation code were shown in our recent work. [35,36] The ITG and TEM instabilities are involved in this multi-mode model, including plasma density 𝑛e, electron temperature 𝑇e, vorticity 𝛺, parallel ion velocity 𝜐 ‖ and ion temperature 𝑇i:…”

mentioning

confidence: 99%

Global Effects on Drift Wave Microturbulence in Tokamak Plasmas

Li 李,

Wang 王

2023

Chinese Phys. Lett.

View full text Add to dashboard Cite

In this paper, microturbulence excited by ion temperature gradient (ITG)-dominant and trapped electron mode (TEM)-dominant instabilities is investigated employing an extended fluid code (ExFC) based on the so-called Landau-Fluid model, which includes the trapped electron dynamics. Firstly, the global effect is emphasized through direct comparison of ITG and TEM instability domains based on local and global simulations. The global effect makes differences in both linear instability and nonlinear transport, including the fluxes and the structure of zonal flow (ZF). The transitions between ITG, TEM and ITG & TEM instabilities/turbulence depend not only on the three key drive forces (R/Ln ,R/LTe ,R/LTi ) but also on their global (profile) effects. Secondly, a lot of electrostatic linear gyrofluid simulations are concluded to obtain a distribution of the instability.

show abstract

Machine learning of turbulent transport in fusion plasmas with neural network

Cited by 16 publications

References 38 publications

Joint meeting of 9th Asia Pacific-Transport Working Group (APTWG) & EU-US Transport Task Force (TTF) workshop

Joint meeting of 9th Asia Pacific-Transport Working Group (APTWG) & EU-US Transport Task Force (TTF) workshop

Neural net modeling of equilibria in NSTX-U

Global Effects on Drift Wave Microturbulence in Tokamak Plasmas

Contact Info

Product

Resources

About