Uncovering edge plasma dynamics via deep learning from partial observations

Speaker 1: Abhilash Mathews


Tuesday, November 3, 2020



PSFC Student Seminars

One of the most intensely studied aspects of magnetic confinement fusion is edge plasma behaviour, which is critical to reactor performance and operation. Drift-reduced Braginskii two-fluid theory has for decades been widely used to model edge plasmas with varying success. We demonstrates that physics-informed neural networks can accurately learn turbulent field dynamics consistent with the two-fluid theory from just partial observations of a plasma's electron density and temperature. This novel computational tool is being developed for plasma diagnosis and model validation in challenging thermonuclear environments, and initial results from analysis of a synthetic plasma will be presented.