Date/Time
Date(s) - 01/25/2024
1:55 pm - 2:55 pm
Location
Rhines Hall 125
Categories
Abstract
Deep learning methods offer the promise of drastically reducing the computational cost of evaluating a diverse range of plasma physics models. The application of deep learning methods to several plasma applications is, however, hindered by the often sparse experimental and computational data sets available.
Physics-informed machine learning methods, whereby physical constraints are embedded in the training of a neural network, offer a path through which the quantity of data required to train a neural network can be drastically reduced. The present work employs a physics-informed neural network (PINN) to predict relativistic electron formation in a magnetic fusion plasma in the absence of any experimental or simulation data. Such electrons, which are often observed to achieve energies of several mega electron volts, pose an immediate threat to tokamak devices due to their high energy and often localized impact on plasma-facing components.
In this seminar, a PINN trained on the adjoint to the relativistic Fokker-Planck equation will be shown to accurately predict the rate at which such relativistic electrons are generated across a broad range of plasma conditions, thus providing an efficient surrogate for identifying tokamak regimes where such relativistic electrons can be expected to emerge.
Bio
Chris McDevitt, Ph.D.
Associate Professor, Nuclear Engineering
University of Florida
Dr. Chris McDevitt is an associate professor in the Nuclear Engineering Program at the University of Florida where his research is focused on the theory and simulation of fusion plasmas. Prior to joining UF in Fall 2019, he completed his B.S. in physics at the University of California at Santa Cruz and subsequently completed his Ph.D. in physics at the University of California at San Diego, where he focused on the description of turbulence in magnetic fusion plasmas. After a short stint as a visiting scientist at Ecole Polytechnique, he moved to Los Alamos National Laboratory where he worked as a staff scientist.