Classification of tokamak plasma confinement states with convolutional recurrent neural networks

TL;DR

This paper develops and compares CNN and Conv-LSTM models for automatic detection of plasma confinement states and ELMs in tokamak data, improving real-time plasma monitoring.

Contribution

It introduces two deep learning methods for classifying confinement states and ELMs in tokamak plasma data, demonstrating their effectiveness.

Findings

Conv-LSTM outperforms CNN in detection accuracy.

Models achieve high ROC and Kappa scores.

Deep learning enables real-time plasma state monitoring.

Abstract

During a tokamak discharge, the plasma can vary between different confinement regimes: Low (L), High (H) and, in some cases, a temporary (intermediate state), called Dithering (D). In addition, while the plasma is in H mode, Edge Localized Modes (ELMs) can occur. The automatic detection of changes between these states, and of ELMs, is important for tokamak operation. Motivated by this, and by recent developments in Deep Learning (DL), we developed and compared two methods for automatic detection of the occurrence of L-D-H transitions and ELMs, applied on data from the TCV tokamak. These methods consist in a Convolutional Neural Network (CNN) and a Convolutional Long Short Term Memory Neural Network (Conv-LSTM). We measured our results with regards to ELMs using ROC curves and Youden's score index, and regarding state detection using Cohen's Kappa Index.