DeepMind, EPFL Explore New Ways to Use AI For Controlling Nuclear Fusion


EPFL’s Swiss Plasma Center (SPC)  and DeepMind teamed up to develop a new magnetic control method for plasmas based on deep reinforcement learning that may make fusion a reality, according to a news release.

The team, which published their study in Nature, applied it to a real-world plasma for the first time in the SPC’s tokamak research facility, TCV. Tokamaks, which are donut-shaped devices, use a powerful magnetic field to confine plasma at extremely high temperatures – hundreds of millions of degrees Celsius, even hotter than the sun’s core – so that nuclear fusion can occur between hydrogen atoms. The resulting release of energy could be one day used as a sustainable source of power.

Scientists can use SPC’s tokamak to investigate new approaches for confining and controlling plasmas, according to the release.

“Our simulator is based on more than 20 years of research and is updated continuously,” said Federico Felici, an SPC scientist and co-author of the study. “But even so, lengthy calculations are still needed to determine the right value for each variable in the control system. That’s where our joint research project with DeepMind comes in.”

The DeepMind team developed an AI algorithm that can create and maintain specific plasma configurations. The algorithm first tried multiple control strategies in simulation to gather experience. After this training period, the AI-based system was able to create and maintain a wide range of plasma shapes and advanced configuration.

Plasma gif
Range of different plasma shapes generated with the reinforcement learning controller
Credit: DeepMind & SPC/EPFL

Finally, the research team tested their new system directly on the tokamak to see how it would perform under real-world conditions.

Martin Riedmiller, control team lead at DeepMind and co-author of the study, added that “our team’s mission is to research a new generation of AI systems – closed-loop controllers – that can learn in complex dynamic environments completely from scratch. Controlling a fusion plasma in the real world offers fantastic, albeit extremely challenging and complex, opportunities.”

The SPC’s collaboration with DeepMind dates back to 2018 when Felici first met DeepMind scientists at a hackathon at the company’s London headquarters. There he explained his research group’s tokamak magnetic-control problem.

“DeepMind was immediately interested in the prospect of testing their AI technology in a field such as nuclear fusion, and especially on a real-world system like a tokamak,” said Felici.