Full-pulse Tomographic Reconstruction with Deep Neural Networks
This work addresses the slow reconstruction process in fusion plasma diagnostics, making it practical to analyze entire pulses for researchers in nuclear fusion, though it is incremental as it applies existing deep learning methods to this domain.
The paper tackled the computationally intensive problem of plasma tomography in fusion devices by training a deep neural network on historical data, enabling full-pulse reconstructions in seconds with high accuracy, which allows visualization of phenomena like plasma heating and disruptions.
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.