Bulletin of the American Physical Society
56th Annual Meeting of the APS Division of Plasma Physics
Volume 59, Number 15
Monday–Friday, October 27–31, 2014; New Orleans, Louisiana
Session JO3: Turbulence and Transport |
Hide Abstracts |
Chair: Emily Belli, General Atomics Room: Salon D |
Tuesday, October 28, 2014 2:00PM - 2:12PM |
JO3.00001: Wavenumber-resolved core turbulence studies in the ASDEX Upgrade tokamak and comparison with non-linear gyrokinetic simulations with the GENE code Tim Happel, Alejandro Ba{\~n}{\'o}n Navarro, Garrard Conway, Tobias G{\"o}rler, Frank Jenko, Francois Ryter, Ulrich Stroth Core plasma turbulence determines transport properties and impacts on the efficiency of a fusion reactor. Gyrokinetic codes are developed to predict dominant instabilities and the turbulence level, which causes the observed particle and heat losses. A careful validation of these codes is mandatory to improve the reliability of predictions. To this end, core turbulence is investigated in ASDEX Upgrade by means of Doppler reflectometry, which provides the perpendicular velocity of turbulent structures and their fluctuation level. H-mode discharges have been performed in which ECRH is used to drive the turbulence from the ITG turbulence regime towards the TEM regime. In general, the turbulence level increases from core towards the edge. With increasing $R/L_{T_e}$, core large scale structures show larger fluctuation amplitudes while their phase velocity is altered with respect to that of small structures. Results are compared with gyrokinetic simulations with the GENE code. Linear results show a transition from ITG towards TEM turbulence close to the radial ECRH deposition location. After matching of heat fluxes to results from power balance analysis, the radial trend in the turbulence level is reproduced. The response to additional heating is opposite to the experimental findings. [Preview Abstract] |
Tuesday, October 28, 2014 2:12PM - 2:24PM |
JO3.00002: The effect of electron-ion collisionality on ETG turbulence Greg Colyer, Alex Schekochihin, Colin Roach, Michael Barnes, Young-chul Ghim, Bill Dorland, Felix Parra In electrostatic simulations of MAST plasma at electron-gyroradius scales with adiabatic ions, using the local flux-tube gyrokinetic code GS2, we find that the saturated electron heat flux decreases as the electron collisionality decreases. At early simulation times the heat flux quasi-saturates at a level independent of electron collisionality; however the zonal fluctuation component continues to grow slowly until much later simulation times, eventually reducing the heat flux at low collisionality. The heat flux at the longest simulation times is the saturated level relevant to energy transport, in the gyrokinetic expansion. We outline an explanation based on zonal-nonzonal interactions and the scaling of the zonal damping rate with electron-ion collisionality, and we discuss the correlation times of the zonal and nonzonal components of the microturbulence. Improved energy confinement with decreasing collisionality has previously been observed on NSTX and MAST, and is favourable towards the performance of future, hotter devices. [Preview Abstract] |
Tuesday, October 28, 2014 2:24PM - 2:36PM |
JO3.00003: The measurement of geodesic acoustic mode magnetic field oscillations in J-TEXT tokamak T. Lan, J. Wu, H.G. Shen, T.J. Deng, A.D. Liu, J.L. Xie, H. Li, W.D. Liu, C.X. Yu, Y. Sun, H. Liu, Z.P. Chen, G. Zhuang Geodesic acoustic mode (GAM) magnetic field oscillations have been investigated using three-dimension magnetic probe and Langmuir probe arrays in the edge of J-TEXT tokamak. The probe arrays are placed on the two top windows of tokamak, separated toroidally. Inside the LCFS, GAM shows apparent oscillations in floating potential. In contrast, GAM magnetic field oscillations are not significant in raw magnetic fields signals. Using toroidal correlation technique, the GAM magnetic field oscillations are distinguished from ambient magnetic field. The amplitudes of three dimension GAM magnetic field fluctuations, as well as the dependence with local plasma parameters such as safety factor and plasma beta, are coincident with theoretical predictions. And its toroidal symmetry mode structure, i.e. n=0, is identified. Furthermore, the GAM current sheet, in which GAM oscillates, is firstly verified with magnetic probes arrays in different radial positions, which may help us to understand the radial structure of GAM. [Preview Abstract] |
Tuesday, October 28, 2014 2:36PM - 2:48PM |
JO3.00004: Toroidal drift modes in tokamaks: a new model for small ELMs Arkaprava Bokshi, David Dickinson, Howard Wilson Toroidal drift instabilities, such as the ion-temperature gradient (ITG) mode, are likely drivers of turbulent transport in tokamaks. Depending on the radial drive profile, two distinct mode structures can emerge: for a peaked profile, the violent Isolated Mode (IM) exists on the outboard-midplane, whereas for a linear profile, the more benign General Mode (GM) sits at the top/bottom of the plasma. The IM only exists in special conditions, so we expect the GM to usually drive turbulence. A new global code, based on an electrostatic gyrokinetic toroidal ITG model, has been developed and benchmarked to study the time-evolution of these linear modes. While we consider the ITG mode, the results are expected to be valid for most microinstabilities. A key result is that as the flow-shear evolves through a critical value, the GM evolves into the IM and then back to the GM. Curiously, the mode structure transiently passes through the violent IM phase independent of how fast the equilibrium evolves! For a pedestal evolving between ELMs, if a GM-IM-GM transition occurs, the burst in linear growth during the IM phase could drive a small ELM. The associated transport would maintain the pedestal pressure gradient below the peeling-ballooning limit, avoiding Type I ELMs. [Preview Abstract] |
Tuesday, October 28, 2014 2:48PM - 3:00PM |
JO3.00005: TEM-turbulence in stellarators and its optimization Josefine H.E. Proll, Per Helander, Samuel Lazerson, Harry Mynick, Pavlos Xanthopoulos Quasi-isodynamic stellarators, which are especially optimized for neoclassical transport, have been shown to be resilient towards trapped-electrons modes (TEMs) in large regions of parameter space. In these configurations, all particles have average ``good curvature.'' It was shown analytically that, thanks to this property, particles that bounce faster than the mode in question draw energy from it near marginal stability, so that the ordinary density-gradient-driven TEM has to be stable in the electrostatic and collisionless limit.This has been confirmed in linear flux-tube simulations that were performed with the GENE code. Several magnetic field configurations were compared and it was found that the growth rates of the TEMs drop with increasing degree of quasi-isodynamicity. These findings can be used to optimize stellarators with respect to TEM turbulence by reducing the fraction of trapped particles with bounce averaged ``bad curvature.'' An appropriate proxy function has therefore been designed to be implemented in STELLOPT, a stellarator optimization tool that can now be used to further explore the configuration space of neoclassically optimized stellarators with the aim to extract designs with improved turbulent transport. [Preview Abstract] |
Tuesday, October 28, 2014 3:00PM - 3:12PM |
JO3.00006: Gyrokinetic particle simulations of kinetic ballooning mode in tokamak pedestal Ihor Holod The pedestal height and width in tokamak H-mode operation are widely believed to be constrained by mesoscale peeling-ballooning modes and microscopic kinetic ballooning modes (KBM). However, direct evidences of the KBM turbulence in pedestal are very limited. The role of the drift-Alfvenic microturbulence during the pedestal recovery period is not clear. Here we use gyrokinetic toroidal code (GTC) to study the edge instability of a DIII-D discharge {\#}131997 using realistic geometry and plasma profiles and focusing on the pedestal region with steep pressure gradient. First, electrostatic simulations find a reactive trapped electron mode with an unusual eigenmode structure, which peaks at the poloidal angle $\theta = \pm \pi $/2. The electron collisions decrease the growth rate by about one-half. Next, the plasma pressure is scanned in GTC electromagnetic simulations to identify the boundary for the KBM onset. At the finite electron beta an electromagnetic instability is found with KBM characteristics. The linear growth rate increases with $\beta_{\mathrm{e}}$ and the mode propagation is in the ion diamagnetic direction. Nonlinear simulations of the KBM turbulence will also be presented. [Preview Abstract] |
Tuesday, October 28, 2014 3:12PM - 3:24PM |
JO3.00007: Gyrokinetic study of edge blobs and divertor heat-load footprint C.-S. Chang, S.-H. Ku, M. Churchill, S. Zweben In an attempt to better understand the complicated physics of the inter-related ``intermittent plasma objects (blobs)'' and divertor heat-load footprint, the full-function gyrokinetic PIC code XGC1 has been used in realistic diverted geometry. Neoclassical and turbulence physics are simulated together self-consistently in the presence of Monte Carlo neutral particles. Blobs are modeled here as electrostatic nonlinear turbulence phenomenon. It is found that the ``blobs'' are generated, together with the ``holes,'' around the steep density gradient region. XGC1 reasserts the previous findings [1] that blobs move out convectively into the scrape-off layer, while the holes move inward toward plasma core. The measured radial width of the divertor heat load, mapped to the outer midplane, is found to be much less than the median radial size of the intermittent plasma objects, but is rather closer to the width of neoclassical orbit excursion from pedestal to divertor, yielding approximately the 1/Ip-type scaling found from our previous pure neoclassical simulation [3] or a heuristic neoclassical argument by Goldston [4]. However, it also shows some spreading by the intermittent turbulence. In ITER plasma edge, where the ion banana width at separatrix becomes negligibly small compared to the meso-scale blob size, blobs may saturate the 1/Ip scaling. \\[4pt] [1] D. D'Ippolito et al., Phys. Plasmas \underline {18} (2011) 060501 \\[0pt] [2] J. Boedo et al., Phys. Plasmas \underline {10} (2003) 1670 \\[0pt] [3] Report on DOE FES Joint Facilities Research Milestone 2010 on Heat-Load Width, Appendix H \\[0pt] [4] R.J. Goldston, Nucl. Fusion 52 (2012) 013009 [Preview Abstract] |
Tuesday, October 28, 2014 3:24PM - 3:36PM |
JO3.00008: Intrinsic momentum generation in diverted tokamak edge by interaction between turbulence and neoclassical particle dynamics Janghoon Seo, C.-S. Chang, S.-H. Ku, J.-M. Kwon Fluid Reynolds stress from turbulence has usually been considered to be responsible for the anomalous toroidal momentum transport in tokamak plasma. Experiment by S. H. M\"{u}ller et al. [Phys. Rev. Lett. \textbf{106}, 115001 (2011)], however, reported that neither the observed edge rotation profile nor the inward momentum transport phenomenon at the edge region of an H-mode plasma could be explained by the fluid Reynolds stress measured with reciprocating Langmuir-probe. The full-function gyrokinetic code XGC1 is used to explain, for the first time, M\"{u}ller et al's experimental observation. It is discovered that, unlike in the plasma core, the fluid Reynolds stress from turbulence is not sufficient for momentum transport physics in plasma edge. The ``turbulent neoclassical'' physics arising from the interaction between kinetic neoclassical orbit dynamics and plasma turbulence is key in the tokamak edge region across the plasma pedestal into core. [Preview Abstract] |
Tuesday, October 28, 2014 3:36PM - 3:48PM |
JO3.00009: Use of Uncertainty Quantification Techniques for Interpretive and Predictive Transport Analysis of Burning Plasmas Alexei Pankin, Masayuki Yokoyama, Ryohsuke Seki, Chihiro Suzuki, Arnold Kritz, Tariq Rafiq Development of the uncertainty quantification (UQ) and sensitivity analysis (SA) techniques in the applied mathematics community brings new opportunities in the analysis, interpretation and validation of experimental data as well as in the development of new discharge scenarios in predictive transport modeling. The UQ techniques have been recently used to develop a new validation method for predictive transport codes [A.Y. Pankin {\it et al.} Phys. Plasmas 20 (2013) 102501]. In this research, the use of UQ and SA techniques is extended to the interpretive analysis of experimental data. The progress achieved in implementing UQ methods in the TASK3D-a1 code is described. TASK3D-a1 is a suite of codes for the interpretive transport analysis of LHD experimental data. The DAKOTA toolkit for calculating UQ is implemented in TASK3D-a1, and it is used to investigate the effects related to the instrumental errors and numerical errors resulting from the interpolation of experimental data. The uncertainties in the computation of effective diffusivities and in the verification of the energy and momentum balances associated with these two types of errors are evaluated. The possible application of these techniques for other interpretive modeling codes such as TRANSP and ONETWO is discussed. [Preview Abstract] |
Tuesday, October 28, 2014 3:48PM - 4:00PM |
JO3.00010: Predicting Heat Transport across Multiple Devices with Neural Networks C.J. Luna, R.V. Budny, O. Meneghini, S.P. Smith, J. Penna Three multi-layer, feed-forward, back-propagation neural networks have been built and trained on heat transport data from DIII-D, TFTR, and JET, respectively. A comparative analysis shows that previous success of neu- ral networks in predicting heat transport in DIII-D [1] is reproduced for both TFTR and JET. The effect of using different neural network topolo- gies has been investigated across all of the devices. It is found that the neural networks can consistently predict the total species' heat fluxes for all of the devices, however they have difficulty in predicting the indi- vidual components of the heat fluxes in presence of significant transient variations in stored energy (i.e. non steady-state conditions). Such lim- itation has been addressed by providing the time-derivative information of the plasma parameters that are input to the neural network. Finally, an attempt is made to draw a connection between the most consistently successful neural network topologies and their relevance to the physics of heat transport in tokamak plasmas.\\[4pt] [1] O. Meneghini, et al., Phys. Plasmas 21 (2014) 060702 [Preview Abstract] |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2024 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
100 Motor Pkwy, Suite 110, Hauppauge, NY 11788
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700