In this work, we address a new feedforward control scheme for the normalized beta (β
N) in tokamak plasmas, using the deep reinforcement learning (RL) technique. The deep RL algorithm optimizes an artificial decision-making agent that adjusts the discharge scenario to obtain a given target β
N from the state–action–reward sets explored by its own trial and error in a virtual tokamak environment. The virtual environment for the RL training is constructed using a long short-term memory (LSTM) network that imitates the plasma responses to external actuator controls, which is trained using five years’ worth of KSTAR experimental data. The RL agent then experiences numerous discharges with different actuator controls in the LSTM simulator, and its internal parameters are optimized in the direction of maximizing the reward. We analyze a series of KSTAR experiments conducted with the RL-determined scenarios to validate the feasibility of the beta control scheme in a real device. We discuss the successes and limitations of feedforward beta control by RL, and suggest a future research path for this area of study.
A tokamak, a torus-shaped nuclear fusion device, needs an electric current in the plasma to produce magnetic field in the poloidal direction for confining fusion plasmas. Plasma current is conventionally generated by electromagnetic induction. However, for a steady-state fusion reactor, minimizing the inductive current is essential to extend the tokamak operating duration. Several non-inductive current drive schemes have been developed for steady-state operations such as radio-frequency waves and neutral beams. However, commercial reactors require minimal use of these external sources to maximize the fusion gain, Q, the ratio of the fusion power to the external power. Apart from these external current drives, a self-generated current, so-called bootstrap current, was predicted theoretically and demonstrated experimentally. Here, we reveal another self-generated current that can exist in a tokamak and this has not yet been discussed by present theories. We report conclusive experimental evidence of this self-generated current observed in the KSTAR tokamak.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.