Update on Self-Consistent Plasma Modeling of ARIES Baseline Design Points A.D. Turnbull, R. Buttery, M. Choi, L.L Lao, S. Smith, H. St John General Atomics ARIES Team Meeting Bethesda Md April 04 2011

Approach, Optimization, and Recent Progress Proposed approach: Outline New features Optimization procedure: Physics optimization Code linkages Progress: Reconstructed ARIES-AT Base configuration with added H-mode pedestal: EFIT calculation reproduced 2000 equilibrium Run ONETWO calculation to obtain profiles of ne, ni, Te, Ti, current density, and bootstrap current: Using density and temperature profiles from Jardin and Kessel 2006 Simulated LHCD and ICRF to make up the shortfall in current density Main Results: LHCD can provide up to 0.63 MA for ne(0) ~ 3 x 1020 m-3 LHCD efficiency is poor at higher densities Future plans

Proposal to Perform a Self-Consistent Analysis of ARIES Design Points Similar analysis was done in the past for ARIES designs but: Design points have been updated with improved Systems Code modeling Understanding of physics issues has also evolved considerably: Particularly pedestal modeling and interaction with the core Analysis will involve: Coupled equilibrium, transport, current drive, fuelling, and stability calculations to obtain a steady state solution In a self consistent simulation using: Latest core transport models Coupled to edge pedestal models The simulation would use the same tools as used to model DIII-D and FSNF, providing a consistent up-to-date set of models and tools across the spectrum of current and planned facilities

Proposal is Intended to Repeat Previous 2000 Effort For ARIES-AT With Improved Tools and Understanding New optimized target configuration Self consistent steady state solution Transport model TGLF in place of GLF23: Real shaped geometry instead of GLF23 shifted circle model Self consistent H-mode pedestal: Realistic pedestal height and width predictions with EPED1 Full-wave Lower hybrid modeling Improved resistive wall stability understanding: Stabilization at low rotation Role of error fields in braking at low rotation New angular momentum transport models Key technical issue is that core transport is stiff: Largest leverage to improving confinement is from increasing pedestal height Conversely, H-mode pedestal and ELM physics depends crucially on the heat and particle fluxes coming from the core

Optimization Procedure Aim to use IMFIT to start with for the core loop of transport self consistently optimized with heating and current drive: Re compute EFIT equilibria as needed Incorporate a pedestal optimization: EPED1 model to predict pedestal height and width given pedestal b b optimization: Initialize with bN = 5.7 Check ideal stability with DCON or GATO? Optimize to converge to maximum stable bN q profile optimization: Adjustment of q profile to improve stability and current drive potential Step will need judgement rather than automation Shape and size optimization: Elongation, triangularity and aspect ratio Size adjustments as needed

Physics Overview of Optimization Logic First guess: previous ARIES + pedestal, pass through H&CD & force balance Heat & current drive  Transport solution ONETWO & GENRAY to align profiles Re-EPED when b changes Re-EFIT at every stage b optimization: vary pressure to optimize stability DCON or GATO codes q profile optimization: to improve transport & stability changes in H&CD deposition Shape optimization: vary elongation, triangularity & aspect ratio. Ultimately change size to ensure target performance is reached

Code Linkage Overview of Optimization Procedure IMFIT code structure

Base Line ARIES H-mode Case From 2000 Study ARIES H-mode cross section: q profile:

Previously Optimized ARIES Lower Hybrid Heating and Current Drive Scenarios Using CURRAY Aimed to drive non-inductive off-axis current in 0.8 <ρ< 1.0 (S.C. Jardin, Fusion Engineering and design (2006)) Two scenarios studied with different density: 4 or 5 wave spectra of LH grills, each centered at a different N//, were determined to have relatively broad profile Scenario 1 Scenario 2 ne(0)=2.93×1020 m-3, Te(0)=26.3keV, Zeff=1.9 ne(0)=2.74×1020 m-3, Te(0)=28.2 keV, Zeff=2.0 Freq (GHz) N// θ Power (37 MW) I/P (A/W) 3.6 1.65 -90 3.06 0.053 2.0 4.40 0.049 2.5 8.22 0.039 3.5 8.87 0.024 5.0 12.39 0.013 Freq (GHz) N// θ Power (20 MW) I/P (A/W) 4 1.7 -90 2.94 0.057 2.0 3.24 0.052 2.5 5.43 0.041 4.0 8.17 0.021 133-09/MC/

Scenario 1 Studied Using Initial Density and Temperature Profiles From Jardin 2006 Fusion Engineering and Design Modified to include H-mode edge density pedestal but Te set to be consistent with original pressure from 2006 published result ne(axis) = 5.0x 1020 m-3 T (keV) n (1013 cm-3) Te=TD=TT Electron Zeff=1.69 D and T Normalized toroidal flux Normalized toroidal flux

For Scenario 1 GENRAY Predicts 0 For Scenario 1 GENRAY Predicts 0.63 MA Total Driven Current From Lower Hybrid System Grill f GHz N// θ Power (37 MW) I/P (A/W) 1 3.6 1.65 -90 3.06 0.013 2 2.0 4.40 0.020 3 2.5 8.22 0.024 4 3.5 8.87 0.021 5 5.0 12.39 0.010 Total grill1 grill2 grill3 grill4 grill5 LHCD driven current (A/cm2) GENRAY Driven current at higher density is much smaller Normalized toroidal flux

Additional Current Driven by ICCD System in Core GENRAY f = 96 MHz N//=2.0 P=4.7MW θ = -15o LHCD ILHCD=0.63 MA Driven current (A/cm2) ICRF IICCD=0.11MA Normalized toroidal flux

Comparison of Driven Current Profiles from GENRAY with CURRAY Results Show Lower Current Driven Lower current drive found than in 2006 paper probably due to additional density pedestal in new simulation GENRAY CURRAY Driven current (A/cm2) LHCD ILHCD=0.63 MA LHCD ILHCD=1.07 MA ICRF IICCD=0.11MA ICRF IICCD=0.15MA Normalized toroidal flux

Tools for Self Consistent Optimization of ARIES-AT Design Point Are in Place Initial case set up: Equilibrium with H-mode pedestal reproduced Simulated electron and ion densities and temperatures (ONETWO): Edge density pedestal produced consistent with equilibrium pressure pedestal No temperature pedestal Current drive from Lower Hybrid and ICCD systems reproduced using GENRAY: Driven current shortfall compared with 2006 CURRAY results Presumably due to H-mode pedestal

Future Steps Will Complete Self Consistent Optimization of ARIES-AT Design Point Consistent pedestal included: Both Temperature and density modified to reproduce H-mode pedestals Consistent with equilibrium H-mode pressure pedestal Equilibrium pedestal modified for consistency with peeling-ballooning (ELM) stability EPED1 model to provide pedestal height and width for given pedestal bpol Self consistent steady state scenario iteration: Heating and current drive optimization using TGLF transport mode Pedestal optimization for bpol consistent with transport simulation and EPED1 model Iteration on b, q profile, and ultimately cross section: Varied around design point as needed Resistive wall mode stability considered

Backup slides

Case1 : Plasma Density and Temperature Profiles Based on C. E Case1 : Plasma Density and Temperature Profiles Based on C.E. Kessel’s Paper Ref) C.E. Kessel, Fusion Engineering and design (2006) Zeff=1.69 ne(axis) = 5.0x 1020 m-3 Electron Te=TD=TT Density (1013 cm-3) Temperature (keV) D and T Normalized toroidal flux Normalized toroidal flux

Case1 : High Density Profile with N//s and Powers of Scenario 1 Predicts Small CD Efficiency Grill 5 with N//=5 seems to produce too much current near plasma edge GENRAY grill1 Grill f GHz N// θ Power (37 MW) I/P (A/W) 1 3.6 1.65 -90 3.06 0.0008 2 2.0 4.40 0.0088 3 2.5 8.22 0.0084 4 3.5 8.87 0.0015 5 5.0 12.39 0.0027 grill2 grill3 grill4 grill5 LHCD driven current (A/cm2) Total Total driven current by LH is 0.15 MA Normalized toroidal flux

LHCD Driven Current Profiles from GENRAY and CURRAY ILHCD=1.07 MA Case 2 ILHCD=0.63 MA LHCD driven current (A/cm2) Case 1 ILHCD=0.15 MA Normalized toroidal flux ne(0)=2.93×1020 m-3, Te(0)=26.3keV