Speaker
Tom Farley
Description
See the full Abstract at http://ocs.ciemat.es/EPS2018ABS/pdf/P5.1069.pdf
Filament statistics with imaging techniques and comparison with langmuir
probes
T. Farley1,2, F. Militello1, N.R. Walkden1, J. Harrison1, S.S. Silburn1, J.W. Bradley2
1)
CCFE, Culham Science Centre, Abingdon, Oxon, OX14 3DB, UK
2)
Department of Electrical Engineering and Electronics, University of Liverpool,
L69 3GJ, UK
The dynamics of plasma filaments govern the transport of particles in the scrape-off layer
(SOL) of tokamak plasmas [D’Ippolito et. al., PoP, 2011] and thus scrape off layer density
profiles. A technique has been developed which maps the light emission in fast visible camera
images to a midplane grid of field line emission, using camera registration and magnetic
equilibrium information. Emission along field lines is used as a proxy for plasma density,
facilitating the identification and tracking of SOL plasma filaments. This pseudo-inversion
filament analysis technique is described in detail and benchmarked against synthetic camera
data in new paper [T Farley et. al., RSI, 2018 (in preparation)].
Here the technique is applied to L-mode fast camera data from the MAST tokamak.
Considering individual field lines from within 2D field line emission maps, we construct
pseudo-Langmuir probes. These probes are used to calculate probability density functions
(PDFs) and time averaged radial profiles and perform conditional averaging and other related
statistical techniques. The statistics of the filaments’ properties are interpreted using the
theoretical ergodic framework described in [F. Militello & J.T. Omotani, PPCF, 2016] in order
to better understand how time averaged filament dynamics produce SOL density profiles.
Toroidal filament separation is found to be exponentially distributed, indicating toroidally
uniform, independent generation of filaments with no toroidal mode number. Filament waiting
times are also seen to be exponentially distributed, indicating filament generation is a Poisson
process in agreement with Langmuir probe time series measurements. These findings provide
important validation of assumptions in the ergodic framework and give insight into the
physics of the filaments.
Finally, the statistics measured with the pseudo-Langmuir probe technique are compared
directly with Langmuir probe measurements made in similar discharges and with results from
the full 2D pseudo-inversion technique.
