FTL: Transfer Learning Nonlinear Plasma Dynamic Transitions in Low Dimensional Embeddings via Deep Neural Networks

TL;DR

This paper introduces FTL, a transfer learning approach using deep neural networks to model nonlinear plasma dynamics in low-dimensional embeddings, enabling efficient detection of plasma instabilities from limited data.

Contribution

The study presents a novel transfer learning framework that leverages pre-trained neural networks to accurately reconstruct nonlinear plasma behaviors from minimal simulation data.

Findings

FTL effectively captures nonlinear kink mode structures.

The model generalizes to various MHD modes.

It demonstrates robustness in identifying plasma instabilities.

Abstract

Deep learning algorithms provide a new paradigm to study high-dimensional dynamical behaviors, such as those in fusion plasma systems. Development of novel model reduction methods, coupled with detection of abnormal modes with plasma physics, opens a unique opportunity for building efficient models to identify plasma instabilities for real-time control. Our Fusion Transfer Learning (FTL) model demonstrates success in reconstructing nonlinear kink mode structures by learning from a limited amount of nonlinear simulation data. The knowledge transfer process leverages a pre-trained neural encoder-decoder network, initially trained on linear simulations, to effectively capture nonlinear dynamics. The low-dimensional embeddings extract the coherent structures of interest, while preserving the inherent dynamics of the complex system. Experimental results highlight FTL's capacity to capture transitional behaviors and dynamical features in plasma dynamics -- a task often challenging for conventional methods. The model developed in this study is generalizable and can be extended broadly through transfer learning to address various magnetohydrodynamics (MHD) modes.