Page 1 - sherwood_2019_poster

Abstract
TAE Technologies, Inc, has an active fusion plasma research program centered around the FRC (Field Reversed Configuration) magnetic topology and the existing C-2W (aka Norman) experiment. The main goals of our current FRC research program are 1) to study confinement of energy at high electron temperatures in the C-2W experiment, and 2) to master plasma control methods needed to increase plasma temperatures to fusion conditions in a future device.
Integrated Modeling – Thermal Equilibrium + Fast Ions
n Full Orbit MC code + Fluid Equilibrium Fluid Equilibrium physics
à Multiple ion fluids; Rotation; Radial, axial force balance
à Radial pressure profile effects on axial separatrix shape
à Magnets and wall shaping + Monte Carlo Fast Ion Physics
à Anisotropic fast ion pressure
à Neutral Beam parameters; energy, angle, impact
parameter
à Fast Ion transport due to classical processes, charge exchange, non-axisymmetric perturbations
Integrated Modeling – Global Transport + Fast Ions
1. Global Stability—con’t
ANC code:
First Simulations of Turbulent Transport in the FRC
• Global nonlinear simulations between mirrors
• Fluctuations spread from SOL to core
• Toroidal wavenumber spectra consistent with experimentally measured spectra (L. Schmitz et al, Nat. Comm, 2016).”
• Simulations @NERSC, Cori & @ALCF, Theta Collaboration with Prof. Zhihong Lin, UCI
Neutrals
Kinetic thermal neutral model
Equilibrium
Multi ion species force balance
Surrogate Models
(Planned Google collaboration “Black Box Optimizer”)
Beams
Kinetic neutral beam and fast ion model
Global Transport
Coupling of fluid, fast ions, neutral fluid; external coils, end biasing; synthetic experimental diagnostics
Perpendicular Transport
Kinetic microturbulence (TAE/UCI collaboration)
RF Heating
ECH – Ray tracing HHFW – Full wave
Global Stability
Parallel Transport
Electron dynamics in SOL and divertor
FUSION POWER
Fusion power R&D
nee Tri Alpha Energy
LIFE SCIENCES BNCT Cancer therapy
MOBILITY
Electric Vehicle (EV) drivetrain platform
BEAMS
Advanced particle accelerators (beams)
TAE is implementing its Integrated Modeling Vision:
• Assembling Hierarchy of Models, including:
• 1st principles turbulence modeling
• Global Transport model as integration framework
• Machine Learning to build surrogate models
• Models used for Interpretation of Norman results Interactions with wider community
A, B, C-1 C-2
C-2U
Plasma Sustainment
à Neutral transport + Fast Ion Physics
à Heating and Current Drive
à Modification of pressure profile and rotation à Charge exchange losses
1. Global Stability
FPIC 3D Hybrid Kinetic/Fluid Particle-in-Cell code
• Uniform Cartesian mesh
• Arbitrary boundary shape
• MPI / OpenMP model (ALCF Theta)
• MPI / CUDA model (OLCF Titan/Summit)
The electrostatic fluctuation #$/& is shown at the exponential '
growing stage (a) and (b) and after saturation (c) and (d).
2. Domain Size Can Affect Numerical Results
GTC code:
Simulation of ITG Instabilities in C-2U geometry
2019 Simulation Plans
Interpretive Modeling
• 1D Global transport modeling and investigation of source terms with Q1D
• 2D neutral modeling: verification of Q2D neutral fluid model; kinetic neutral models • Fast ion transport in MINERVA and Google plasma reconstructions
• Stability analyses of reconstructed plasmas
Computing and Predictive Modeling
• HHFW simulations
• Optimization and acceleration of Q2D global transport code
• Electrostatic turbulence simulation comparison with Doppler Back Scattering on Norman
• Global stability and effect of actuators – fast ions, biasing, field shape
References
1. D. Fulton et al. APS DPP-Meeting PP.11.103 (2018)
2. C. Lau, et al. Nuclear Fusion (2019).
3. J. Bao, et al. PoP (forthcoming) (2019)
4. F. Ceccherini et al. APS DPP-Meeting PP.11.99 (2018)
5. S. Dettrick et al. APS DPP-Meeting PP.11.97 (2018)
Early development First full-scale and science machine
70’
70’
2013- 2015
1998 – 2009- 2000s 2012
0.8 0.6 0.4 0.2
TAE Technologies, Inc
Fusion Power Research at TAE
Full Orbit MC code + 2D Global Fluid Transport
Global Transport physics
à Relaxation of profiles
à Coupled parallel and perpendicular transport
A Hierarchy of Models is Required to Bridge Multiple Scales
E = 3.27
E = 4.52 exp(-3E/S*)
Analysis codes
Simulation codes
• 0D–
• 1D–Q1D
• 2D–LR
• 2D–Q2D • 2D–RF
• 3D – KSOL • 3D – FPIC • 3D–ANC
Power balance; Surrogate Models • fluid/kineticglobaltransport • fluid plasma global dynamics
fluid/kinetic global transport
full wave code
kinetic parallel electron dynamics • fluid/kinetic global stability
kinetic micro-turbulent transport
• • •
Benchmarked against FRC tilt mode Addingactuatorstostudystabilization:
electrode biasing magnetic fields shape neutral beams.
Future: Simulate Control systems
→Motivates further work in ANC/GTC with extended C-
2W geometry
Q1D domain
MHD domain
TypicalPICdomain
0
0.1
0.2
0.3
0.4
0.5
0.6
•
•
J. Bao et al, to appear in Physics of Plasmas 2019 Global linear simulations in extended axial domain
• Domainlengthchangesaveragecurvaturealong field line
• Changesmostunstablemodefromeventoodd parity
Integrated Modeling of Stability and Transport of FRC Plasmas
Ales Necas, Jian Bao, Dan Barnes, Francesco Ceccherini, Sean Dettrick, Dan Fulton, Sangeeta Gupta, Kevin Hubbard, Calvin Lau, Zhihong Lin, Marco Onofri, Toshiki Tajima, and the TAE team
TAE’s current machine
• First plasma July 2017
• One year construction
• On time, on budget
100’
Norman
(aka C-2W)
2017-
I"
• • •
DoE Leadership Computing Facility INCITE awards 2018 and 2019 Collaboration with UCI GTC group,
Collaboration with ORNL part of ECP CODAR team
Interact further with DoE, e.g. through SciDAC ISEP or ECP
KSOLdomain
E/S*
TAE Technologies, Inc., 19631 Pauling, Foothill Ranch, CA 92610 University of California Irvine, Irvine CA 92697, USA
Future: Whole Device Model Coupled by Global Transport Framework
à Electrode biasing effect on rotation and
heating •
Computational cost
γ/γMHD
6. E. Belova et al. Phys. Plasmas 8, 1267 (2001) parallel domain simulations: (a)Z/R0 2 [13.6, 13.6], (b)Z/R0 2 [16.2, 16.2], (c)Z/R0 2
7. Y. Omelchenko et al. 2014 Winter Simulation Conference (WSC2014)
An award of computer time was provided by the Innovative and Novel Computational Impact on Theory and Experiment (INCITE) program. This research used resources of the Argonne Leadership Computing Facility, which is a DOE Office of Science User Facility supported under Contract DE-AC02-06CH11357.
Figure 9: Comparison of 2D poloidal mode structures of ITG instability between di↵erent
[19.2,19.2], and (d)Z/R0 2 [21.0,21.0]. The dashed lines show the flux surfaces with
maximum mode amplitude. The blue solid line is the separatrix.
23
Figure 10: Comparison of 1D parallel mode structures on the diagnostic flux surfaces between
di↵erent parallel domain simulations. L/⇢i is the field line distance normalized by ⇢i at local flux surface. The diagnositic flux surfaces are shown by the dashed lines in Fig.9.