This title appears in the Scientific Report : 2016 

Development and Application of a Multiscale Model for the Magnetic Fusion Edge Plasma Region
Hasenbeck, Felix Martin Michael (Corresponding author)
Plasmaphysik; IEK-4
Jülich Forschungszentrum Jülich GmbH Zentralbibliothek, Verlag 2016
188 S.
RWTH Aachen, Diss., 2015
978-3-95806-120-0
Book
Dissertation / PhD Thesis
Plasma-Wall-Interaction
Schriften des Forschungszentrums Jülich Reihe Energie & Umwelt / Energy & Environment 307
OpenAccess
Please use the identifier: http://hdl.handle.net/2128/9912 in citations.
Plasma edge particle and energy transport perpendicular to the magnetic field playsa decisive role for the performance and lifetime of a magnetic fusion reactor. For the particles, classical and neoclassical theories underestimate the associated radial transport by at least an order of magnitude. Drift fluid models, including mesoscale processes on scales down to tenths of millimeters and microseconds, account for the experimentally found level of radial transport; however, numerical simulations for typical reactor scales (of the order of seconds and centimeters) are computationally very expensive. Large scale code simulations are less costly but usually lack an adequate model for the radial transport. The multiscale model presented in this work aims at improving the description of radial particle transport in large scale codes by including the effects of averaged local drift fluid dynamics on the macroscale profiles. The multiscale balances are derived from a generic multiscale model for a fluid, using the Braginskii closure for a collisional, magnetized plasma, and the assumptions of the B2 code model(macroscale balances) and the model of the local version of the drift fluid code ATTEMPT (mesoscale balances). A combined concurrent–sequential coupling procedure is developed for the implementation of the multiscale model within a coupled codesystem. An algorithm for the determination of statistically stationary states and adequate averaging intervals for the mesoscale data is outlined and tested, proving that it works consistently and efficiently.