Full-pulse Tomographic Reconstruction with Deep Neural Networks

TL;DR

This paper presents a deep neural network approach for rapid 2D plasma tomography reconstruction in fusion devices, enabling real-time visualization of plasma phenomena during discharges.

Contribution

The authors develop and train a neural network that can quickly produce accurate tomographic reconstructions, significantly reducing computation time compared to traditional methods.

Findings

High-accuracy reconstructions comparable to traditional methods

Reconstruction time reduced to a few seconds per pulse

Enables real-time plasma phenomena visualization

Abstract

Plasma tomography consists in reconstructing the 2D radiation profile in a poloidal cross-section of a fusion device, based on line-integrated measurements along several lines of sight. The reconstruction process is computationally intensive and, in practice, only a few reconstructions are usually computed per pulse. In this work, we trained a deep neural network based on a large collection of sample tomograms that have been produced at JET over several years. Once trained, the network is able to reproduce those results with high accuracy. More importantly, it can compute all the tomographic reconstructions for a given pulse in just a few seconds. This makes it possible to visualize several phenomena -- such as plasma heating, disruptions and impurity transport -- over the course of a discharge.