Analysis of a set of TGLF transport calculations for DIII-D L-Mode dimensionless scaling experiments shows clearly the energy transport shortfall in the near edge region. The study includes scans in gyro-radius, collisionality, beta, and safety factor. Comparing the model predicted power flows from TGLF using the experimental profiles from the gyro-radius scan to the experimental power flows shows the energy transport shortfall in the near edge region. While the location of the transport shortfall remains near ρ = 0.7, the magnitude of the shortfall is clearly larger for low ρ* discharges. For the safety factor scan, the location of the transport shortfall appears to be sensitive to the q-profile. In addition, TGLF modeling of 16 TFTR L-mode shows that the transport shortfall in DIII-D cannot simply be attributed to cold edge temperatures in DIII-D. The temperatures in the TFTR L-modes at ρ = 0.90 were typically three to four times higher than the DIII-D discharges and yet the transport shortfall remains.
The most recent version of the GATO mapping and ideal stability code was placed in the SVN subversion control system and is available publicly. In addition to some minor bug fixes, the newest version includes a few enhancements to make the mapping more generally useful as a stand-alone interface to other stability codes. An option is now included to read an input electron number density profile expressed as a function of an arbitrary radial variable; typically, for example, the profiles from GAPROFILES are given in terms of the square root of the toroidal flux. The coding for including the fast ion pressure profile was also rewritten to allow the profile to be expressed in terms of any radial variable. Calculations for the pedestal and total stored energy, the local field line pitch at each mesh point, were added, and an option is included to set the single point axis q to a specific value to circumvent numerical issues on axis. A module to run GATO under the new OMFIT integrated modeling framework has also been put together.
GYRO simulations exploring the effects of rotation on reverse shear Alfvén eigenmode (RSAE) eigenmode structure and stability in DIII-D have shown that the previously observed twisting of the eigenmode results from the shear in the local mode frequency. The twisting pattern was previously observed in rotation-free simulations of the benchmark discharge #142111 at t=725 ms and was attributed to shear in the local diamagnetic drift frequency of the driving energetic particle (EP) population. Scanning of the rotation rate (using scalar multiples of the carbon rotation profile) shows that the twist pattern can be generally attributed to shear in the local mode frequency. This includes rotation contributions in addition to the diamagnetic drive. The pattern can be “unwound” or “wound tighter”, depending on the relative signs of the two contributions, with corresponding increases or decreases in the linear growth rate. For a typical case with monotonic profiles, the growth rate increases with counter-current rotation and decreases with co-current rotation, suggesting a possible mechanism to reduce AE-induced EP transport.
Valerie Izzo and Nate Ferraro attended the Center for Extended-Magnetohydrodynamics Modeling (CEMM) SciDAC project meeting in Madison, WI. The main topics of the meeting included recent progress in fluid modeling of disruptions, sawtooth cycles, ELMs, and plasma response. Methods for self-consistently including energetic particle species and kinetic effects in NIMROD, M3D, and M3D-C1 were also discussed.
For quantitative comparison of computed fast-ion distributions with measurements, a series of self-consistent simulations using the 5D finite-orbit Monte-Carlo code ORBIT-RF coupled with the full-wave code AORSA has been performed by scanning the assumed fraction of core fast-wave (FW) power absorption. The fast-ion distributions that yield the neutron enhancement factor comparable to the experimental ones are passed to the fast-ion diagnostic simulation code FIDASIM to compute the synthetic fast-ion D-alpha (FIDA) signals. ORBIT-RF/AORSA reproduces the synergistic effect measured in neutron reaction rates and vertical FIDA signals in the two-frequency FW heating (4th harmonic 60 MHz and 6th harmonic 90 MHz) compared to those obtained in a single-frequency FW heating (4th harmonic 60 MHz) . For the tangential FIDA signals, ORBIT-RF/AOSA also reproduces the trend of the measured FIDA signals, which indicate that the tangential components of the fast ions are hardly accelerated by either a single-frequency FW or two-frequency FW. This synergy arises from finite Larmor radius effects that occur since a substantial fraction of fast ions above the neutral-beam injection energy is present due to preheating by the 60 MHz FW. Therefore, the additional 90 MHz FW damps significantly on the beam-ion tails and produces a synergy.
Sam Lazerson from PPPL visited GA last week to collaborate on 3D equilibrium reconstructions and issues of 3D plasma response calculations.
A new version of the multimode transport model developed by Lehigh University together with the necessary drivers has recently been incorporated into simulations of AT-type tokamak configurations using the parallel version of the ONETWO/GCNMP transport package. Currently this confinement model (that includes ETG, ITG, TEM, and drift resistive-ballooning modes) is being applied to simulate high normalized beta (~ 4.8) type ARIES AT configurations. Part of the investigation includes a flux solver approach to the transport equations that requires only two calls per time step and hence may lead to improved computational efficiency. To date we have observed that the predicted transport (particularly due to TEM modes) is significantly less than that computed with the GLF23 model. Detailed comparisons against other transport models including TGLF are underway. Our calculations currently include 40MW of fast wave power for current profile control and indicate that a low fraction of ohmic current is required. Refinement of these results should decrease the ohmic current fraction even further.
Disclaimer
These highlights are reports of research work in progress and are accordingly subject to change or modification