Full-Pulse Tomographic Reconstruction with Deep Neural Networks

Abstract

Plasma tomography consists of reconstructing a two-dimensional radiation profile of 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 the entire pulse.

Keywords:

• Plasma tomography
• deep learning
• convolutional neural networks

Acknowledgments

This work has been carried out within the framework of the EUROfusion Consortium and has received funding from the Euratom research and training program 2014–2018 under grant agreement No. 633053. Institute for Plasmas and Nuclear Fusion activities also received financial support from Fundação para a Ciência e a Tecnologia through project UID/FIS/50010/2013. The Titan X GPU used in this work was donated by NVIDIA Corporation.

We would also like to thank E. Pawelec at the University of Opole, Poland, for insights into laser ablation experiments and P. Lomas at JET, United Kingdom Atomic Energy Authority, for several helpful discussions regarding disruptions.

Supplemental Material

Supplemental video files for this article can be accessed at https://doi.org/10.1080/15361055.2017.1390386.