Table of Contents
Contemporary Issues in Plasma Physics and Nuclear Fusion
Components and Modeling Support for Validation Platforms for Fusion Science
The FES Experimental Plasma Research program has the long term performance measure of demonstrating enhanced fundamental understanding of magnetic confinement and improving the basis for future burning plasma experiments. This can be accomplished through investigations and validations of the linkage between prediction and measurement for scientific leverage in testing the theories and scaling the phenomena that are relevant to future burning plasma systems. This research includes investigations in a variety of concepts such as stellarators, spherical tori, and reversed field pinches. Key program issues include initiation and increase of plasma current; dissipation of plasma exhaust power; symmetric-torus confinement prediction; stability, continuity, and profile control of low-aspect-ratio symmetric tori; quasi-symmetric and three-dimensional shaping benefits to toroidal confinement performance; divertor design for three-dimensional magnetic confinement configurations, and the plasma-materials interface. Presentations in this session will entail scientific and engineering developments, including computational modeling, in support of current experiments in these research activities, in particular for the small-scale concept exploration experiments. The work should have a strong potential for commercialization. Overall, support of research that can best help deepen the scientific foundations of understanding and improve the tokamak concept is an important focus area for this session.
Simulation and Data Analysis Tools for Magnetically Confined Plasmas
The predictive simulation of magnetically confined fusion plasmas is important for the design and evaluation of plasma discharge feedback and control systems; the design, operation, and performance assessment of existing and proposed fusion experiments; the planning of experiments on existing devices; and the interpretation of the experimental data obtained from these experiments. Developing a predictive simulation capability for magnetically confined fusion plasmas is very challenging because of the enormous range of overlapping temporal and spatial scales; the multitude of strongly coupled physical processes governing the behavior of these plasmas; and the extreme anisotropies, high dimensionalities, complex geometries, and magnetic topologies characterizing most magnetic confinement configurations.
Although considerable progress has been made in recent years toward the understanding of these processes in isolation, there remains a critical need to integrate them in order to develop an experimentally validated integrated predictive simulation capability for magnetically confined plasmas. In addition, the increase in the fidelity and level of integration of fusion simulations enabled by advances in high performance computing hardware and associated progress in computational algorithms has been accompanied by orders of magnitude increases in the volume of generated data. In parallel, the volume of experimental data is also expected to increase considerably, as U.S. scientists plan to collaborate on a new generation of overseas long-pulse superconducting fusion experiments. Accordingly, a critical need exists for developing data analysis tools addressing big data challenges associated with computational and experimental research in fusion energy science.
Talks in this session will entail simulation and data analysis tools for magnetic fusion energy science addressing the challenges described above. Areas of interest include, but are not limited to: (1) algorithms incorporating advanced mathematical techniques; (2) algorithms targeting novel computing architectures, including Graphics Processing Unit (GPU), manycore, and heterogeneous computing platforms; (3) verification and validation tools, including efficient methods for facilitating comparison of simulation results with experimental data; (4) data management, visualization, and analysis tools for local and remote multi-dimensional time-dependent datasets resulting from large scale simulations or experiments; (5) techniques for coupling simulation codes, including coupling across different computer platforms and through high speed networks; (6) methodologies for building highly configurable and modular scientific codes and flexible user-friendly interfaces; and (7) remote collaboration tools that enhance the ability of geographically distributed groups of scientists to interact and collaborate in real-time.