Full Shot Predictions for the DIII-D Tokamak via Deep Recurrent Networks

TL;DR

This paper develops a data-driven deep recurrent network model to predict the full temporal evolution of plasma discharges in the DIII-D tokamak, aiding understanding and control of plasma dynamics for nuclear fusion.

Contribution

It introduces a novel application of deep recurrent networks to model entire plasma shot sequences in tokamaks, improving predictive capabilities.

Findings

Deep recurrent networks accurately predict plasma shot evolution.

Training and inference procedures significantly impact prediction quality.

Model calibration enhances reliability of plasma behavior forecasts.

Abstract

Although tokamaks are one of the most promising devices for realizing nuclear fusion as an energy source, there are still key obstacles when it comes to understanding the dynamics of the plasma and controlling it. As such, it is crucial that high quality models are developed to assist in overcoming these obstacles. In this work, we take an entirely data driven approach to learn such a model. In particular, we use historical data from the DIII-D tokamak to train a deep recurrent network that is able to predict the full time evolution of plasma discharges (or "shots"). Following this, we investigate how different training and inference procedures affect the quality and calibration of the shot predictions.