Machine learning prediction of plasma behavior from discharge configurations on WEST

TL;DR

This paper introduces a transformer-based machine learning model that predicts key plasma parameters on the WEST tokamak using pre-discharge signals, enabling faster scenario planning and control.

Contribution

The study presents a novel application of transformer models for plasma prediction, trained on WEST data, with high accuracy and rapid inference suitable for real-time use.

Findings

Average MSE loss of 0.026

R^2 of 0.94 indicating high prediction accuracy

Inference time around 0.1 seconds

Abstract

Accurately predicting plasma behavior based on discharge configurations is essential for the safe and efficient operation of tokamak experiments. While physics-based integrated modeling codes provide valuable insights, their high computational cost limits their applicability for fast scenario design and control optimization. In this study, we propose a transformer-based machine learning model to predict key global plasma parameters on the WEST tokamak, including the normalized beta ($\beta_{n}$), toroidal beta ($\beta_{t}$), poloidal beta ($\beta_{p}$), plasma stored energy ($W_{\mathrm{mhd}}$), safety factor at the magnetic axis ($q_{0}$), and safety factor at the 95% flux surface ($q_{95}$). The model uses only signals that can be defined before the discharge, such as magnetic coil currents, auxiliary heating power, plasma current reference, and line-averaged plasma density. Trained on 550 discharges from the WEST campaigns, the model demonstrates an average mean square error (MSE) loss of 0.026, an average coefficient of determination $R^{2}$ of 0.94, and achieves inference times on the order of 0.1 seconds. These results highlight the potential of data-driven surrogate models for assisting in discharge planning, scenario evaluation, and real-time control of tokamak plasmas.