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

arXiv:2404.17466v1 Announce Type: cross
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 …

abstract algorithms arxiv building cs.lg deep learning deep learning algorithms detection development dynamic embeddings fusion low networks neural networks new paradigm novel paradigm physics physics.comp-ph physics.plasm-ph plasma study systems transfer transfer learning transitions type unique via