A runaway electron avalanche surrogate for partially ionized plasmas

A physics-constrained deep learning surrogate that predicts the exponential ‘avalanche’ growth rate of runaway electrons (REs) for a plasma containing partially ionized impurities is developed. Specifically, a physics-informed neural network (PINN) that learns the adjoint of the relativistic Fokker–...

Full description

Bibliographic Details

Published in:	Nuclear Fusion
Main Authors:	Jonathan S. Arnaud, Xian-Zhu Tang, Christopher J. McDevitt
Format:	Article
Language:	English
Published:	IOP Publishing 2025-01-01
Subjects:	tokamak disruptions runaway electrons physics informed neural networks physics informed machine learning
Online Access:	https://doi.org/10.1088/1741-4326/ae00db