Status of Fusion Theory and Simulation Research in NFRI 6 th J-K Workshop/NIFS/2011 J.Y. Kim
Feb 24, 2016
Status of Fusion Theory and Simulation Research in NFRI
6th J-K Workshop/NIFS/2011
J.Y. Kim
2
Introduction
Major Research Activities and Plan - Turbulent Transport - MHD Equilibrium & Instability - Pedestal & ELM - Integrated Operation Scenario - Divertor & PWI
Summary
3
Major Goal of Theory & Simulation Research
• To develop advanced theoretical models and simulation skills for understanding fusion plasma & material
• To support KSTAR, ITER, and DEMO research programs
• Physics De-sign Valida-tion• Operation Scenario De-velopment
ITER• Experimen-tal Data Analysis• Operation Scenario De-velopment
KSTAR
DEMO • Physics Design & AT Scenario Devel-opment• Advanced Fusion Material Research
4
Projects supporting Theory & Simulation Research
Fusion Simulation Project
• started from 2007 for promoting the simulation research of fusion plasma in NFRI
• annual budget of about 700k$/year (not including the salary of regular staff)
• now supporting mainly the research work of the field, being not covered by WCI project
WCI (World-Class-Institute) Center Project
• government project for promoting the research capability of national institutes
• our center selected as one of the three centers (Fusion, Casncer, Brain)
- Integrated modeling study of turbulent transport in fusion plasmas
• five year project (2009.12- 2014.11), with annual budget of about 2M$/year
• now, about 11 people with 6 foreign and 5 domestic members (Director : Prof. P. Diamond)
KSTAR Project
• mostly for supporting the experimental research and device upgrade of KSTAR
• partial support for the theoretical modeling and data analysis work
5
Turbulent Transport
Intensive study being performed through WCI project
- Intrinsic rotation and momentum transport
- Transport barrier formation and evolution (ITB, ETB)
- Non-local transport & self-organization
- Particle and impurity transport
- MHD & turbulence interaction etc.
Strong emphases on the global transport simulation study
- Global gyro-fluid simulation using TRB code (circular, ES, basis-function)
※ plan to develop a more general global gyrofluid code (D-shape, EM, finite-difference)
- Global gyrokinetic simulation using gKPSP code (delta-f, PIC)
※ a full-f gyrokinetic code being developed using semi-Lagrangian method (cf. GYSELA)
- Recently, a parallel PC-cluster system started its operation (from June 29, 2011)
A close collaboration study with KSTAR experimental group
6
Global Gyro-fluid Simulation
• TRB code imported and upgraded
- global transport simulation study with a reasonable computation time
- extensive study being done of ITB formation, flow generation & transport etc.
• Global non-local transport study planned of various modes (Ohmic, H, Hybrid etc.)
• Upgrade also planned : circular, Electrostatic => Shaped, Electromagnetic
Power ramp simula-tion
Forward transitionBack transition
[Kim et.al. NF’11]
1st APTWG meeting,S.S. Kim et al
7
Global Gyro-kinetic Simulation
7
• gKPSP (delta-f, PIC) code developed
- Neoclassical equilibrium & trapped electron effects included
- Intrinsic rotation, momentum transport, particle transport being studied
• A new code (full-f, semi-Lagrangian) under development for the study of
- Global non-local transport modeling from core to edge
- Barrier formation of ITB and ETB, with turbulence & neoclassical effects etc.
r/R0 r/R0
ηi = 1.0 (TEM)
ηi = 3.1 (ITG)
ηi = 1.0 (TEM)
ηi = 3.1 (ITG)
V|| V||
Immediate after nonlinear saturation After nonlinear saturation
1st APTWG meeting,J.M. Kwon et al
8
A Parallel PC-Cluster System
HP ProLiant Linix Cluster
(AMD 64bit 12-core 2.2GHz)
Total 576 CPUs (x 12 = 6912
cores) with infiniband QDR in-
terconnection
60TF theoretical peak perfor-
mance, 40 TF HPL
2GBytes/core memory (14
TBytes total)
160 TBytes disk storage
Ranked 423-th in the top 500
(2011. 6)
9
MHD Equilibrium and Instabilities
Calculation of 3D error field and perturbed magnetic field for KSTAR
- A detailed calculation of error field performed using OPERA/MAFLO codes
- 3D perturbed field from KSTAR FEC/ELM/RWM coil under calculation using IPEC etc.
Simulation study of Sawtooth, LM, NTM etc.
- Reduced MHD simulation using the 4-field code (in collaboration with Dr. Aydmir in IFS)
- Extended MHD simulation using NIMROD (in collaboration with NIMROD group)
- The codes installed and simulation study just started
Disruption Simulation for KSTAR
- Initial study performed for KSTAR design using TSC code
- Re-calculation being performed with the modified KSTAR passive plate model
Stability Study of Energetic Particle Mode
- NOVA-K imported and a preliminary study started
10
Calculation of Field error & 3D Perturbed Field for KSTAR
q=2/1
※ for vacuum (under calculation with response)
11
Disruption Simulation for a new KSTAR Model pre-disrup. TQ CQ
KSTAR coil geometry & plasma boundary
Eddy current
(toroidal)
JT X Bp force
12
Modeling of ELM Mitigation by RMP Method
- in collaboration with SciDAC team (CPES)
- Iterative 3D perturbed field calculation with plasma response using M3D, XGC0
- Plasma transport calculation using XGC0
Modeling of ELM Mitigation by Pellet Pace-making Method
- Initial study performed using ASTRA code with an approximate PPM model
- A more self-consistent modeling started using M3D code
Pedestal & ELM Control
Nonlinear simulation of ELM using BOUT++ code
- Recently started in collaboration with Dr. X. Xu in LLNL
- for a more detailed study of ELM itself and benchmark with M3D results
Modeling of NTV phenomena with 3D perturbed field
- in collaboration with Dr. K. Shaing in Taiwan
- both of analytic calculation and simulation study using XGC0 code
13
• Modeling codes: XGC0 (kinetic transport), M3D (nonlinear MHD)• Numerical q95 scan (BT scan) for low collsionality DIII-D case reveals sensi-
tive magnetic stochasticity behavior around the experimental q95 ELM sup-pression window “Vacuum Chirikov is only a necessary condition”
• Current research focus Clarify effects of collisionality & density on ELM suppression mechanism Experimental validation in various tokamak RMP experiments including
KSTAR
Vacuum Chirikov is similar
Plasma-responded
Chirikov is different
3.623.52resonant win-
dow
Time (msec)
Simulation Study of RMP Control by RMP
14
Simulation result for density distri-butionusing M3D code
Initial condition of a simulation for pellet injection
30 30 30 1010 10 10 1010
0 0 0 3030 10 10 1010
0 0 0 00 0 30 30 5
0 0 0 00 0 0 5 5
0 0 0 00 0 0 5 5
0 0 0 00 5 5 5 50 0 0 55 5 5 5 5
0
Pressure gradient, max
Cur
rent
dens
ity<j
//>m
ax.
3 3.5 4 4.5 50.6
0.7
0.8
0.9
1
1.1
1.2
1.3
6.5E-026.0E-025.5E-025.0E-024.5E-024.0E-023.5E-023.0E-022.5E-022.0E-021.5E-021.0E-025.0E-033.0E-032.6E-031.7E-031.0E-030.0E+00
edge pressure and current diagram for peeling-ballooning instability using ELITE code
The linear stability analysis using an ideal MHD stability code (ELITE)
Simulation Study of ELM & its Control by PPM
Nonlinear simulation study of ELM & its control by pellet pace-making method
(cf. talk by Dr. H.S. Hahn)
15
Study of NTV with 3D Perturbed Magnetic Field
Substantial Analytic and Numerical Studies performed - As is well-known, symmetry breaking components in the magnetic field configuration increase
the viscosity in the toroidal direction (so called “Neoclassical Toroidal Viscosity”).
- In the last few years, significant studies on NTV performed and in each collisional regime the
theory has been established solidly.
- Recently, the general solution of NTV has been obtained by solving drift kinetic equation
(DKE)
numerically and the results are in good agreement with the analytic calculations.
A Particle Simulation Study planned - for a more exact study in a more general magnetic field configuration and comparison with the
analytic and DKE-based numerical results
- with EFIT EQDSK files and the actual 3D field component of KSTAR
- will use the codes XGC0 and (gKPSP1)
- in collaboration with G. Y. Park, K. Shaing (Taiwan) (cf. talk by Dr. J. C. Seol)
16
ASTRA Simulation for Operation Scenario Modeling
- KSTAR operation scenarios in the 2nd operation phase (2013-2017)
- NTM control simulation with a self-consistent calculation of plasma evolution
※ in collaboration with SNU theory group (Prof. Y.S. Na)
NBI Heating & CD Simulation
- A more detailed study performed using NUBEAM for KSTAR NBI system model
Integrated Operation Scenario (with heating & CD)
ICRF Heating & CD Simulation
- Various ICRF heating scenarios for KSTAR, which include the minority-ion heating, mode-
conversion heating, 2nd harmonic heating etc.
- Flow generation from ICRF heating also being studied
Plan for Integrated Scenario Modeling
- trying to utilize the global gyrofluid code for a more self-consistent modeling of integrated
operation scenario, particularly considering turbulence evolution & pedestal formation
(cf. talk by Prof. Y.S. Na)
17
Modeling of NBI Heating & CD• A detailed simulation performed of NBI heating & CD using NUBEAM code
- to support the optimization study of 2nd KSTAR NBI configuration
- comparison made of three possible NBI configurations
<For type A> <For type B2>
<For type B1>
Modeling of ICRF Heating & Flow Generation
18
0 5 10 15 20 25 30-0.25
-0.2
-0.15
-0.1
-0.05
0
0.05
0.1
0.15
0.2
0.25
Hydrogen concentration [%]
toro
idal
forc
e at
(=1
) [N
]
/2-/2
0 5 10 15 20 25 30-0.01
-0.005
0
0.005
0.01
0.015
0.02
Hydrogen concentration [%]
toro
idal
forc
e at
(=1
) [N
]
/2-/2
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 10
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
0.2
nornalized minor radius
F(
) [N
]
0 = /2
1% H2% H5% H10% H20% H30% H
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1-0.14
-0.12
-0.1
-0.08
-0.06
-0.04
-0.02
0
0.02
0.04
0.06
nornalized minor radius
F(
) [N
]
0 = /2
1% H2% H5% H10% H20% H30% H
Toroidal force Poloidal force
on last flux surface
• Momentum transfer from RFs, calculated using TORIC for minority ion heating case
ne=5×1019 m-3
(cf. talk by Dr. B.H. Park)
19
Simulation Study of KSTAR Divertor/SOL Transport
- Initial study performed using UEDGE-DEGAS for KSTAR divertor design
- Also, a benchmark study performed using KTRAN(SNU-developed0 code in
collaboration with SNU group
- Recently, a more detailed study started using B2-EIRENE or SOLPS
Simulation Study of PWI
- Recently, ERO code imported from Julich group in Germany
- Simulation study to be started soon for the modeling of erosion, retention phenomena
Divertor Transport & PWI
Modeling of Charging and Transport of Dust Particle
- in collaboration with Prof. N.S. Yoon (Cung-Buk National Univ.) (cf. talk by Prof. N.S. Yoon)
20
Simulation of KSTAR Divertor Transport
• Simulation results using KTRAN for KSTAR divertor model
Decrease of Impurity density, Increase of radiation
Max :1.5e18Max :8.69e18[m-3] [m-3]
[W/m2] [W/m2]
Impurity density
Power Radiation
TungstenCarbon
(cf. talk by Mr. S.B. Shim)
21
Summary of Research Activities
Research area
Major research subjects Simulation codes Man power
Turbulent transport(WCI-project)
- Intrinsic rotation and momentum trans-port
- Transport barrier formation (ITB, ETB)- Non-local transport- Particle & impurity transport- Turbulence & MHD interaction etc.
- Global gyrofluid (TRB, BOUT++)- Global gyroki-netic (gKPSP, a new full-f)
5(+2+6)
*2: domestic collabora-tor*6 : for-eigner
MHD insta-bility
- Equilibrium & 3D perturbed field- Sawtooth, LM, NTM- Disruption- Energetic particle mode
- EFIT, MAFLO- NIMROD, 4-field- TSC- NOVA-K
2(+2)
Pedestal & ELM
- Pedestal stability (linear, non-linear)- ELM control by RMP- ELM control by PPM- NTV & neoclassical transport
- BOUT++- XGC0+M3D- M3D- XGC0
3(+1)
Integrated Scenario
- KSTAR operation scenarios- Heating & CD modeling (NBI, ICRF,
LHCD, ECH/ECCD)
- ASTRA- NUBEAM, TORIC, LSC, TORAY
2(+1)
Divertor & PWI
- KSTAR divertor transport- PWI (erosion, retention)- Dust*
- B2-EIRENE(SOLPS)
- ERO- (Prof. N.S. Yoon)
1(+1)