Tracking Blobs with Machine Learning in the Turbulent Edge Plasma of a Tokamak

The analysis of turbulence in plasmas is fundamental in fusion research. Experimental studies are challenging due to the diverse processes that drive the high-speed dynamics of turbulent phenomena. This work presents a novel application of motion tracking to identify and track turbulent filaments in fusion plasmas, called blobs, in high-frequency video data obtained from Gas Puff Imaging diagnostics. We compare four baseline methods trained on synthetic data and then test on synthetic and real data obtained from plasmas in the Tokamak à Configuration Variable. From the size of the blobs estimated from each of the baseline methods employed, we identify the regime of the blob dynamics as described in Myra et al. [1] which agrees with state-of-the-art conditional averaging methods. As an on-going analysis, we use our blob-tracking models to compare 1) a low- and a high-density plasma, as well as 2) a negative and a positive triangularity plasma, extending our study in [2], in order to find any differences in underlying characteristics of edge fluctuations.

[1] J. R. Myra et al 2006 Phys. Plasmas 13, 112502

[2] W. Han et al 2021 Nucl. Fusion 61 034003