Magnetic control of tokamak plasmas through deep reinforcement learning.

Nuclear fusion using magnetic confinement, in particular in the tokamak configuration, is a promising path towards sustainable energy. A core challenge is to shape and maintain a high-temperature plasma within the tokamak vessel. This requires high-dimensional, high-frequency, closed-loop control using magnetic actuator coils, further complicated by the diverse requirements across a wide range of plasma configurations. In this work, we introduce a previously undescribed architecture for tokamak magnetic controller design that autonomously learns to command the full set of control coils. This architecture meets control objectives specified at a high level, at the same time satisfying physical and operational constraints. This approach has unprecedented flexibility and generality in problem specification and yields a notable reduction in design effort to produce new plasma configurations. We successfully produce and control a diverse set of plasma configurations on the Tokamak à Configuration Variable1,2, including elongated, conventional shapes, as well as advanced configurations, such as negative triangularity and 'snowflake' configurations. Our approach achieves accurate tracking of the location, current and shape for these configurations. We also demonstrate sustained 'droplets' on TCV, in which two separate plasmas are maintained simultaneously within the vessel. This represents a notable advance for tokamak feedback control, showing the potential of reinforcement learning to accelerate research in the fusion domain, and is one of the most challenging real-world systems to which reinforcement learning has been applied.

Essay: Overcoming the Obstacles to a Magnetic Fusion Power Plant.

Fusion occurs when light nuclei combine to form heavier nuclei. The energy released in this process powers the stars and can provide humankind with a safe, sustainable, and clean source of baseload electricity, a valuable tool in the fight against climate change. To overcome the Coulomb repulsion of like-charged nuclei, fusion reactions necessitate temperatures of tens of millions of degrees or thermal energies of tens of keV, at which matter exists only in the form of plasma. Plasma is an ionized state of matter that is rare on Earth but characterizes most of the visible universe. The quest for fusion energy is thus intrinsically associated with plasma physics. In this Essay, I lay out my view of the challenges on the path to fusion power plants. As these need to be sizable and inevitably complex, large-scale collaborative enterprises are required, involving not only international cooperation but also private-public industrial partnerships. We focus on magnetic fusion, in particular on the tokamak configuration, relevant to the International Thermonuclear Experimental Reactor (ITER), the largest fusion device to be built in the world. Part of a series of Essays which concisely present author visions for the future of their field.