5 papers
A Physics-Informed Neural Network for Solving the Quasi-static Magnetohydrodynamic Equations
Jonathan S. Arnaud, Christopher J. McDevitt, Golo Wimmer +1
A physics-informed neural network (PINN) is developed, for the first time, to learn the time-dependent quasi-static magnetohydrodynamic (MHD) equations in axisymmetric tokamak geom…
An Adjoint Formulation of Energetic Particle Confinement
Christopher J. McDevitt, Jonathan S. Arnaud
An adjoint formulation of energetic particle confinement in axisymmetric tokamak geometry is derived and evaluated using a physics-informed neural network (PINN). The PINN estimate…
A Runaway Electron Avalanche Surrogate for Partially Ionized Plasmas
Jonathan S. Arnaud, Xian-Zhu Tang, Christopher J. McDevitt
A physics-constrained deep learning surrogate that predicts the exponential ``avalanche'' growth rate of runaway electrons (REs) for a plasma containing partially ionized impuritie…
An Efficient Surrogate Model of Secondary Electron Formation and Evolution
Christopher J. McDevitt, Jonathan Arnaud, Xian-Zhu Tang
This work extends the adjoint-deep learning framework for runaway electron (RE) evolution developed in Ref. [C. McDevitt et al., A physics-constrained deep learning treatment of ru…
A Physics-Constrained Deep Learning Treatment of Runaway Electron Dynamics
Christopher J. McDevitt, Jonathan Arnaud, Xian-Zhu Tang
An adjoint formulation leveraging a physics-informed neural network (PINN) is employed to advance the density moment of a runaway electron (RE) distribution forward in time. A dist…