DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Self-consistent core-pedestal transport simulations with neural network accelerated models

Abstract

Fusion whole device modeling simulations require comprehensive models that are simultaneously physically accurate, fast, robust, and predictive. In this paper we describe the development of two neural-network (NN) based models as a means to perform a snon-linear multivariate regression of theory-based models for the core turbulent transport fluxes, and the pedestal structure. Specifically, we find that a NN-based approach can be used to consistently reproduce the results of the TGLF and EPED1 theory-based models over a broad range of plasma regimes, and with a computational speedup of several orders of magnitudes. These models are then integrated into a predictive workflow that allows prediction with self-consistent core-pedestal coupling of the kinetic profiles within the last closed flux surface of the plasma. Finally, the NN paradigm is capable of breaking the speed-accuracy trade-off that is expected of traditional numerical physics models, and can provide the missing link towards self-consistent coupled core-pedestal whole device modeling simulations that are physically accurate and yet take only seconds to run.

Authors:
 [1];  [1];  [1];  [1];  [1];  [1];  [1];  [1];  [1];  [2];  [3]; ORCiD logo [4]
  1. General Atomics, San Diego, CA (United States)
  2. Politecnico di Torino, Torino (Italy)
  3. Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
  4. Princeton Plasma Physics Lab. (PPPL), Princeton, NJ (United States)
Publication Date:
Research Org.:
General Atomics, San Diego, CA (United States)
Sponsoring Org.:
USDOE Office of Nuclear Energy (NE)
OSTI Identifier:
1374389
Grant/Contract Number:  
FG02-95ER54309
Resource Type:
Accepted Manuscript
Journal Name:
Nuclear Fusion
Additional Journal Information:
Journal Volume: 57; Journal Issue: 8; Journal ID: ISSN 0029-5515
Publisher:
IOP Science
Country of Publication:
United States
Language:
English
Subject:
70 PLASMA PHYSICS AND FUSION TECHNOLOGY; neural networks; transport; pedestal; tokamak

Citation Formats

Meneghini, Orso, Smith, Sterling P., Snyder, Philip B., Staebler, Gary M., Candy, Jeffrey, Belli, E., Lao, Lang L., Kostuk, Mark, Luce, Timothy C., Luda, Teobaldo, Park, Jin Myung, and Poli, F. Self-consistent core-pedestal transport simulations with neural network accelerated models. United States: N. p., 2017. Web. doi:10.1088/1741-4326/aa7776.
Meneghini, Orso, Smith, Sterling P., Snyder, Philip B., Staebler, Gary M., Candy, Jeffrey, Belli, E., Lao, Lang L., Kostuk, Mark, Luce, Timothy C., Luda, Teobaldo, Park, Jin Myung, & Poli, F. Self-consistent core-pedestal transport simulations with neural network accelerated models. United States. https://doi.org/10.1088/1741-4326/aa7776
Meneghini, Orso, Smith, Sterling P., Snyder, Philip B., Staebler, Gary M., Candy, Jeffrey, Belli, E., Lao, Lang L., Kostuk, Mark, Luce, Timothy C., Luda, Teobaldo, Park, Jin Myung, and Poli, F. Wed . "Self-consistent core-pedestal transport simulations with neural network accelerated models". United States. https://doi.org/10.1088/1741-4326/aa7776. https://www.osti.gov/servlets/purl/1374389.
@article{osti_1374389,
title = {Self-consistent core-pedestal transport simulations with neural network accelerated models},
author = {Meneghini, Orso and Smith, Sterling P. and Snyder, Philip B. and Staebler, Gary M. and Candy, Jeffrey and Belli, E. and Lao, Lang L. and Kostuk, Mark and Luce, Timothy C. and Luda, Teobaldo and Park, Jin Myung and Poli, F.},
abstractNote = {Fusion whole device modeling simulations require comprehensive models that are simultaneously physically accurate, fast, robust, and predictive. In this paper we describe the development of two neural-network (NN) based models as a means to perform a snon-linear multivariate regression of theory-based models for the core turbulent transport fluxes, and the pedestal structure. Specifically, we find that a NN-based approach can be used to consistently reproduce the results of the TGLF and EPED1 theory-based models over a broad range of plasma regimes, and with a computational speedup of several orders of magnitudes. These models are then integrated into a predictive workflow that allows prediction with self-consistent core-pedestal coupling of the kinetic profiles within the last closed flux surface of the plasma. Finally, the NN paradigm is capable of breaking the speed-accuracy trade-off that is expected of traditional numerical physics models, and can provide the missing link towards self-consistent coupled core-pedestal whole device modeling simulations that are physically accurate and yet take only seconds to run.},
doi = {10.1088/1741-4326/aa7776},
journal = {Nuclear Fusion},
number = 8,
volume = 57,
place = {United States},
year = {Wed Jul 12 00:00:00 EDT 2017},
month = {Wed Jul 12 00:00:00 EDT 2017}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record

Citation Metrics:
Cited by: 58 works
Citation information provided by
Web of Science

Save / Share:

Works referenced in this record:

A theory-based transport model with comprehensive physics
journal, May 2007

  • Staebler, G. M.; Kinsey, J. E.; Waltz, R. E.
  • Physics of Plasmas, Vol. 14, Issue 5
  • DOI: 10.1063/1.2436852

Development and validation of a predictive model for the pedestal height
journal, May 2009

  • Snyder, P. B.; Groebner, R. J.; Leonard, A. W.
  • Physics of Plasmas, Vol. 16, Issue 5
  • DOI: 10.1063/1.3122146

Tokamak profile prediction using direct gyrokinetic and neoclassical simulation
journal, June 2009

  • Candy, J.; Holland, C.; Waltz, R. E.
  • Physics of Plasmas, Vol. 16, Issue 6
  • DOI: 10.1063/1.3167820

30 years of adaptive neural networks: perceptron, Madaline, and backpropagation
journal, January 1990

  • Widrow, B.; Lehr, M. A.
  • Proceedings of the IEEE, Vol. 78, Issue 9
  • DOI: 10.1109/5.58323

Integrated modeling applications for tokamak experiments with OMFIT
journal, July 2015


Multi-scale gyrokinetic simulations: Comparison with experiment and implications for predicting turbulence and transport
journal, April 2016

  • Howard, N. T.; Holland, C.; White, A. E.
  • Physics of Plasmas, Vol. 23, Issue 5
  • DOI: 10.1063/1.4946028

The first transport code simulations using the trapped gyro-Landau-fluid model
journal, May 2008

  • Kinsey, J. E.; Staebler, G. M.; Waltz, R. E.
  • Physics of Plasmas, Vol. 15, Issue 5
  • DOI: 10.1063/1.2889008

Predicting temperature and density profiles in tokamaks
journal, May 1998

  • Bateman, Glenn; Kritz, Arnold H.; Kinsey, Jon E.
  • Physics of Plasmas, Vol. 5, Issue 5
  • DOI: 10.1063/1.872848

A gyro-Landau-fluid transport model
journal, July 1997

  • Waltz, R. E.; Staebler, G. M.; Dorland, W.
  • Physics of Plasmas, Vol. 4, Issue 7
  • DOI: 10.1063/1.872228

Physics basis of Multi-Mode anomalous transport module
journal, March 2013

  • Rafiq, T.; Kritz, A. H.; Weiland, J.
  • Physics of Plasmas, Vol. 20, Issue 3
  • DOI: 10.1063/1.4794288

Gyro-Landau fluid equations for trapped and passing particles
journal, October 2005

  • Staebler, G. M.; Kinsey, J. E.; Waltz, R. E.
  • Physics of Plasmas, Vol. 12, Issue 10
  • DOI: 10.1063/1.2044587

An Eulerian gyrokinetic-Maxwell solver
journal, April 2003


Kinetic calculation of neoclassical transport including self-consistent electron and impurity dynamics
journal, July 2008


An Eulerian method for the solution of the multi-species drift-kinetic equation
journal, June 2009


Fully electromagnetic gyrokinetic eigenmode analysis of high-beta shaped plasmas
journal, November 2010

  • Belli, E. A.; Candy, J.
  • Physics of Plasmas, Vol. 17, Issue 11
  • DOI: 10.1063/1.3495976

Ion temperature profile stiffness: non-linear gyrokinetic simulations and comparison with experiment
journal, February 2014


Gyrokinetic studies of core turbulence features in ASDEX Upgrade H-mode plasmas
journal, April 2015

  • Navarro, A. Bañón; Happel, T.; Görler, T.
  • Physics of Plasmas, Vol. 22, Issue 4
  • DOI: 10.1063/1.4919022

Nonlinear gyrokinetic simulations of the I-mode high confinement regime and comparisons with experimenta)
journal, May 2015

  • White, A. E.; Howard, N. T.; Creely, A. J.
  • Physics of Plasmas, Vol. 22, Issue 5
  • DOI: 10.1063/1.4921150

Noncircular, finite aspect ratio, local equilibrium model
journal, April 1998

  • Miller, R. L.; Chu, M. S.; Greene, J. M.
  • Physics of Plasmas, Vol. 5, Issue 4
  • DOI: 10.1063/1.872666

Real-time capable first principle based modelling of tokamak turbulent transport
journal, July 2015


Validation studies of gyrofluid and gyrokinetic predictions of transport and turbulence stiffness using the DIII-D tokamak
journal, July 2013


Stable numeric scheme for diffusion equation with a stiff transport
journal, October 2008


A first-principles predictive model of the pedestal height and width: development, testing and ITER optimization with the EPED model
journal, August 2011


Integrated fusion simulation with self-consistent core-pedestal coupling
journal, April 2016

  • Meneghini, O.; Snyder, P. B.; Smith, S. P.
  • Physics of Plasmas, Vol. 23, Issue 4
  • DOI: 10.1063/1.4947204

Variational moment solutions to the Grad–Shafranov equation
journal, January 1981


Edge localized modes and the pedestal: A model based on coupled peeling–ballooning modes
journal, May 2002

  • Snyder, P. B.; Wilson, H. R.; Ferron, J. R.
  • Physics of Plasmas, Vol. 9, Issue 5
  • DOI: 10.1063/1.1449463

Numerical studies of edge localized instabilities in tokamaks
journal, April 2002

  • Wilson, H. R.; Snyder, P. B.; Huysmans, G. T. A.
  • Physics of Plasmas, Vol. 9, Issue 4
  • DOI: 10.1063/1.1459058

Stability and dynamics of the edge pedestal in the low collisionality regime: physics mechanisms for steady-state ELM-free operation
journal, August 2007


Exploration of the Super H-mode regime on DIII-D and potential advantages for burning plasma devices
journal, March 2016

  • Solomon, W. M.; Snyder, P. B.; Bortolon, A.
  • Physics of Plasmas, Vol. 23, Issue 5
  • DOI: 10.1063/1.4944822

The Design and Implementation of the SWIM Integrated Plasma Simulator
conference, February 2010

  • Elwasif, Wael R.; Bernholdt, David E.; Shet, Aniruddha G.
  • 2010 18th Euromicro International Conference on Parallel, Distributed and Network-Based Processing (PDP), 2010 18th Euromicro Conference on Parallel, Distributed and Network-based Processing
  • DOI: 10.1109/PDP.2010.63

Multi-level concurrency in a framework for integrated loosely coupled plasma simulations
conference, December 2011

  • Elwasif, Wael R.; Bernholdt, David E.; Foley, Samantha S.
  • 2011 9th IEEE/ACS International Conference on Computer Systems and Applications (AICCSA)
  • DOI: 10.1109/AICCSA.2011.6126601

Parameter Sweep and Optimization of Loosely Coupled Simulations Using the DAKOTA Toolkit
conference, December 2012

  • Elwasif, Wael R.; Bernholdt, David E.; Pannala, Sreekanth
  • 2012 IEEE 15th International Conference on Computational Science and Engineering (CSE)
  • DOI: 10.1109/ICCSE.2012.24

Chapter 1: Overview and summary
journal, December 1999

  • Editors, ITER Physics Basis; Co-Chairs, ITER Physics Expert Group Chairs an; Unit, ITER Joint Central Team and Physics
  • Nuclear Fusion, Vol. 39, Issue 12
  • DOI: 10.1088/0029-5515/39/12/301

ELMs and constraints on the H-mode pedestal: peeling–ballooning stability calculation and comparison with experiment
journal, January 2004


Theory of plasma transport in toroidal confinement systems
journal, April 1976


Effect of finite aspect ratio on the neoclassical ion thermal conductivity in the banana regime
journal, January 1982


Works referencing / citing this record:

Integrated Tokamak modeling: When physics informs engineering and research planning
journal, May 2018


Predict-first experimental analysis using automated and integrated magnetohydrodynamic modeling
journal, May 2018

  • Lyons, B. C.; Paz-Soldan, C.; Meneghini, O.
  • Physics of Plasmas, Vol. 25, Issue 5
  • DOI: 10.1063/1.5025838

Propagation of input parameter uncertainties in transport models
journal, October 2018

  • Vaezi, P.; Holland, C.; Grierson, B. A.
  • Physics of Plasmas, Vol. 25, Issue 10
  • DOI: 10.1063/1.5053906

Predict-first experiments and modeling of perturbative cold pulses in the DIII-D tokamak
journal, June 2019

  • Rodriguez-Fernandez, P.; White, A. E.; Howard, N. T.
  • Physics of Plasmas, Vol. 26, Issue 6
  • DOI: 10.1063/1.5096800

Machine-learning assisted steady-state profile predictions using global optimization techniques
journal, October 2019


Fast modeling of turbulent transport in fusion plasmas using neural networks
journal, February 2020

  • van de Plassche, K. L.; Citrin, J.; Bourdelle, C.
  • Physics of Plasmas, Vol. 27, Issue 2
  • DOI: 10.1063/1.5134126

Progress and challenges in understanding core transport in tokamaks in support to ITER operations
journal, December 2019

  • Mantica, P.; Angioni, C.; Bonanomi, N.
  • Plasma Physics and Controlled Fusion, Vol. 62, Issue 1
  • DOI: 10.1088/1361-6587/ab5ae1

Real-time capable modeling of neutral beam injection on NSTX-U using neural networks
journal, March 2019


Deep neural network Grad–Shafranov solver constrained with measured magnetic signals
journal, December 2019


Investigation of a Multiple-Timescale Turbulence-Transport Coupling Method in the Presence of Random Fluctuations
journal, July 2018


Fast modeling of turbulent transport in fusion plasmas using neural networks
text, January 2019


Fast modeling of turbulent transport in fusion plasmas using neural networks
text, January 2019


Fast modeling of turbulent transport in fusion plasmas using neural networks
text, January 2019


Investigation of a Multiple-Timescale Turbulence-Transport Coupling Method in the Presence of Random Fluctuations
text, January 2018