Machine learning prediction of plasma behavior from discharge configurations on WEST

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 ($β_{n}$), toroidal beta ($β_{t}$), poloidal beta ($β_{p}$), 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.

Figures (8)

• Figure 1: The Granger causality and Pearson correlation coefficient between input and output signals. For Granger causality, a smaller coefficient indicates a stronger causal relationship, whereas for the Pearson correlation coefficient, a larger value reflects a stronger linear correlation. PowerLH* and PhaLH* are the actual powers and corresponding phases of the LHW system, PowerIC* is the powers of the ICRH system, Ip is the reference of plasma current, the remaining signals correspond to various locations of the PF coils.
• Figure 2: The workflow of machine learning model in the present work
• Figure 3: Three types of typical discharge prediction in the test set. (a) The Ohmic heating-only discharge. (b) The discharge with LHW heating. (c) The discharge with ICRH heating. The targets represent experimental measurements, while the predictions correspond to the model estimations.
• Figure 4: Regression plots of the output signals. Both the target and predicted values are averaged over the duration of each discharge. Except for $q_{0}$ and $q_{95}$, the model achieves a coefficient of determination $R^{2}$ greater than 0.890. The performance for $q_{0}$ and $q_{95}$ is lower compared to the other signals.
• Figure 5: Discharge parameter distributions in the whole dataset. All ICRH and LHW discharge with a total power of zero are removed.