Tomographic reconstruction of tokamak edge turbulence from single visible camera data and automatic turbulence structure tracking

摘要

Following the lead of Nguyen van Yen (2012 Nucl. Fusion 52 013005), this article tackles the problem of tomographic inversion with one camera assuming a constant emissivity of light along the magnetic field lines. In this way, the 3D problem reduces to 2D by helical symmetry, allowing for the reconstruction of any poloidal plane in the field of view of the camera. It is shown in this article that the complexity of using a wavelet basis for the reconstruction, as presented by Nguyen van Yen, is not necessary. The method is also validated by confronting it to 3D numerical data coming from the TOKAM3X code, for which the emissivity is slowly varying along the field lines (up to 20%), showing the robustness of the reconstruction. The technique is then applied to real camera data recorded during a D-shaped ohmic plasma shot realized in the COMPASS tokamak. The method is experimentally validated by comparing reconstructed data from camera and ion saturation currents measured by Langmuir probes in the divertor region. Finally, it is shown that automatic detection and tracking of structures visible in the reconstructed poloidal plane enables unique investigations of edge plasma physics and opens wide perspectives for this method.