Observation of a new type of self-generated current in magnetized plasmas

Nonperturbative theory of the low-to-high confinement transition through stochastic simulations and information geometry: Correlation and causal analyses

The low-to-high confinement (L-H) transition signifies one of the important plasma bifurcations occurring in magnetic confinement plasmas, with vital implications for exploring high-performance regimes in future fusion reactors. In particular, the accurate turbulence statistical description of self-regulation and causal relation among turbulence and shear flows is essential for accessing enhanced plasma performance and advanced operation scenarios. To address this, we provide a nonperturbative theory of the L-H transition by stochastic simulations of a reduced L-H transition model and detailed statistical analysis. By calculating time-dependent probability density functions (PDFs) of turbulence, zonal flows, and the mean pressure gradient, we elucidate how statistical properties change over time with the help of the information geometry theory (information rate, causal information rate), highlighting its utility in capturing self-regulation and causal relation among turbulence, zonal flow shears, and the mean flow shears. Furthermore, stochastic noises in turbulence, zonal flows, and/or input power are shown to induce uncertainty in the power threshold Q_{c} above which the L-H transition occurs while leading to a rather gradual L-H transition. A time-dependent PDF of power loss over the L-H transition is presented.

Past rewinding of fluid dynamics from noisy observation via physics-informed neural computing

Reconstructing the past of observed fluids has been known as an ill-posed problem due to both numerical and physical challenges, especially when observations are distorted by inevitable noise, resolution limits, or unknown factors. When employing traditional differencing schemes to reconstruct the past, the computation often becomes highly unstable or diverges within a few backward time steps from the distorted and noisy observation. Although several techniques have been recently developed for inverse problems, such as adjoint solvers and supervised learning, they are also unrobust against errors in observation when there is time-reversed simulation. Here we present that by using physics-informed neural computing, robust time-reversed fluid simulation is possible. By seeking a solution that closely satisfies the given physics and observations while allowing for errors, it reconstructs the most probable past from noisy observations. Our work showcases time rewinding in extreme fluid scenarios such as shock, instability, blast, and magnetohydrodynamic vortex. Potentially, this can be applied to trace back the interstellar evolution and determining the origin of fusion plasma instabilities.

Avoiding fusion plasma tearing instability with deep reinforcement learning

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 performance1-4. 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.

Solving real-world optimization tasks using physics-informed neural computing

Optimization tasks are essential in modern engineering fields such as chip design, spacecraft trajectory determination, and reactor scenario development. Recently, machine learning applications, including deep reinforcement learning (RL) and genetic algorithms (GA), have emerged in these real-world optimization tasks. We introduce a new machine learning-based optimization scheme that incorporates physics with the operational objectives. This physics-informed neural network (PINN) could find the optimal path in well-defined systems with less exploration and also was capable of obtaining narrow and unstable solutions that have been challenging with bottom-up approaches like RL or GA. Through an objective function that integrates governing laws, constraints, and goals, PINN enables top-down searches for optimal solutions. In this study, we showcase the PINN applications to various optimization tasks, ranging from inverting a pendulum, determining the shortest-time path, to finding the swingby trajectory. Through this, we discuss how PINN can be applied in the tasks with different characteristics.