Avoiding fusion plasma tearing instability with deep reinforcement learning

Jaemin Seo; SangKyeun Kim; Azarakhsh Jalalvand; Rory Conlin; Andrew Rothstein; Joseph Abbate; Keith Erickson; Josiah Wai; Ricardo Shousha; Egemen Kolemen

doi:10.1038/s41586-024-07024-9

Avoiding fusion plasma tearing instability with deep reinforcement learning

Nature. 2024 Feb;626(8000):746-751. doi: 10.1038/s41586-024-07024-9. Epub 2024 Feb 21.

Authors

Jaemin Seo^{1

2}, SangKyeun Kim^{1

3}, Azarakhsh Jalalvand¹, Rory Conlin^{1

3}, Andrew Rothstein¹, Joseph Abbate^{3

4}, Keith Erickson³, Josiah Wai¹, Ricardo Shousha^{1

3}, Egemen Kolemen^{5

6}

Affiliations

¹ Department of Mechanical and Aerospace Engineering, Princeton University, Princeton, NJ, USA.
² Department of Physics, Chung-Ang University, Seoul, South Korea.
³ Princeton Plasma Physics Laboratory, Princeton, NJ, USA.
⁴ Department of Astrophysical Sciences, Princeton University, Princeton, NJ, USA.
⁵ Department of Mechanical and Aerospace Engineering, Princeton University, Princeton, NJ, USA. ekolemen@princeton.edu.
⁶ Princeton Plasma Physics Laboratory, Princeton, NJ, USA. ekolemen@princeton.edu.

Abstract

For stable and efficient fusion energy production using a tokamak reactor, it is essential to maintain a high-pressure hydrogenic plasma without plasma disruption. Therefore, it is necessary to actively control the tokamak based on the observed plasma state, to manoeuvre high-pressure plasma while avoiding tearing instability, the leading cause of disruptions. This presents an obstacle-avoidance problem for which artificial intelligence based on reinforcement learning has recently shown remarkable performance^1-4. However, the obstacle here, the tearing instability, is difficult to forecast and is highly prone to terminating plasma operations, especially in the ITER baseline scenario. Previously, we developed a multimodal dynamic model that estimates the likelihood of future tearing instability based on signals from multiple diagnostics and actuators⁵. Here we harness this dynamic model as a training environment for reinforcement-learning artificial intelligence, facilitating automated instability prevention. We demonstrate artificial intelligence control to lower the possibility of disruptive tearing instabilities in DIII-D⁶, the largest magnetic fusion facility in the United States. The controller maintained the tearing likelihood under a given threshold, even under relatively unfavourable conditions of low safety factor and low torque. In particular, it allowed the plasma to actively track the stable path within the time-varying operational space while maintaining H-mode performance, which was challenging with traditional preprogrammed control. This controller paves the path to developing stable high-performance operational scenarios for future use in ITER.