Theory Weekly Highlights for June 2014

June 27, 2014

The MARS-Q code has been imported to the GA cluster and will be released publicly via SVN once additional testing is completed. The new version is a major upgrade of the MARS-K code, incorporating new features developed by Y.Q. Liu (Culham) and Z. Wang (PPPL). New capabilities include finite orbit width effects for the kinetic contributions, and a quasilinear option to include the effect of the mode or plasma response on the rotation profile through NTV effects. The new version also includes options for running with MPI.

Ron Waltz, Gary Staebler, and Jeff Candy attended the 41st EPS plasma physics meeting in Berlin. Staebler presented an invited talk on “H-mode Transitions and Limit Cycle Oscillations from Mean Field Transport Equations”. Waltz presented his work with Eric Bass on “Prediction of fusion alpha loss in ITER from local marginal stability to Alfvén eigenmodes,” and Candy presented a paper on “Improved ITER performance modeling via zonal-flow stabilization”.

June 20, 2014

DIII-D has two massive gas jets for disruption mitigation: MEDUSA, located above the midplane at 15º toroidally, and CERBERUS, located below the midplane at 135º. In experiments, the toroidal peaking of radiated energy as measured by two radiated power detectors at 90º and 210º degrees shows no significant variation as a function of whether one or the other or both jets is used. A NIMROD simulation using both jets, when compared with simulations using each individual jet (see Highlights from May 2 and February 7, 2014 at Theory Weekly Highlights for May 2014 and Theory Weekly Highlights for February 2014), reveals that the lack of variation may be an artifact of the limited number of measurement locations. The simulation toroidal peaking factors (TPFs) are calculated in two ways: 1) using information only from the 90 and 210 degree toroidal locations, and 2) using the full toroidal information; these are referred to as a “synthetic TPF” and “real TPF”, respectively. The synthetic TPF is calculated during the pre-thermal quench and thermal quench phases, and falls within the measurement error bars for all three cases during the TQ and two of three cases during the pre-TQ, but over-predicts the MEDUSA-only TPF. Overall, the lack of any significant trend in the DIII-D TPF measurements as the jet number and location is varied is reproduced by the NIMROD synthetic TPFs. However, when the real TPF is calculated, a more intuitive picture emerges, in which the TPF in both the pre-TQ and TQ drops when using two jets, compared with either jet individually. This suggests that there may in reality be a reduction in the TPF with the addition of a second jet in DIII-D, which is missed by the available diagnostics.

June 13, 2014

A new version of OMFIT is ready to be released (v.0.4.10). New versions of the OMFIT integrated modeling framework have been regularly released on an almost-monthly basis. Since the last major release, several updates have been devoted to extending the framework's capabilities, refining the interactive graphical user interface, and expanding and improving the physics modules. The framework has been extended to interface with batch queuing systems, enabling resource allocation and job management on high performance computing (HPC) systems. Presently OMFIT features 30 publicly available modules, covering a broad range of topical areas. Development of an IPS module has demonstrated the capability of managing and executing Integrated Plasma Simulator (IPS) workflows via OMFIT. Recently, two workshops have taken place at GA to illustrate OMFIT capabilities in the area of kinetic equilibrium reconstructions and transport analyses. The latest tutorial has been recorded and remotely broadcasted. Documentation and tutorials about OMFIT and its modules can be found at http://gafusion.github.io/OMFIT-source.

Holger St John completed his final week this week after 35 productive years at GA. Over the course of his career, Holger has been involved intimately in the development of many of the key neutronics, and plasma equilibrium and transport codes used routinely at DIII-D and elsewhere, particularly the ONETWO, GAEQ, and MCGO codes. His expertise in both physics and numerical methods will be sorely missed. Sterling Smith will be taking over stewardship of the highly utilized ONETWO transport code in his stead.

June 06, 2014

Simulations of an ITER hybrid DT scenario were carried out using TGLF and checked with subsequent nonlinear electromagnetic GYRO simulations. Specifically, freezing the TGLF-predicted profiles and simulating the turbulence at various radii with GYRO, complete zonal-flow-generated stabilization was observed at radii inside about r/a=0.6. TGLF was then modified to reduce its fluxes slightly to GYRO levels in the limited region 0.35 ⇐ r/a ⇐ 0.55, improving the total integrated fusion power by about 8%. Methods to add this effect to TGLF are being explored for more accurate simulations of ITER and ITER-like devices. We expect that adding zonal-flow stabilization to TGLF at all simulation radii may yield additional performance increases. Further, the hybrid case studied had a relatively low current and high safety factor. Performance improvements for the baseline ITER case are potentially even greater since the stabilization effect is stronger at low q.



Disclaimer
These highlights are reports of research work in progress and are accordingly subject to change or modification