Related papers: Learning Plasma Dynamics and Robust Rampdown Traje…
Spherical tokamaks (STs) have been shown to possess properties desirable for a fusion power plant such as achieving high plasma ? and having increased vertical stability. To understand the confinement properties that might be expected in…
We present the FreeGSNKE Pulse Design Tool (FPDT), an open-source, Python-based computational framework that enables in silico testing and predictive design of tokamak plasma scenarios and control strategies. The FPDT couples the FreeGSNKE…
We describe tests validating progress made toward acceleration and automation of hydrodynamic codes in the regime of developed turbulence by three Deep Learning (DL) Neural Network (NN) schemes trained on Direct Numerical Simulations of…
The controlled ramp down of the toroidal plasma current in the ITER tokamak is simulated using a simple model that employs cylindrical geometry. The magnetohydrodynamical (MHD) stability of the plasma throughout the whole current ramp is…
In tokamak plasmas, sheared flows perpendicular to the driving temperature gradients can strongly stabilize linear modes. While the system is linearly stable, regimes with persistent nonlinear turbulence may develop, i.e. the system is…
Anti-dynamo modes are usually found in spheromaks plasma devices experiments due to the fact that Cowling anti-dynamo theorem is naturally applied to axisymmetric devices and flows. In this paper full consideration is given to the existence…
Plasma modeling is central to the design of nuclear fusion reactors, yet simulating collisional plasma kinetics from first principles remains a formidable computational challenge: the Vlasov-Maxwell-Landau (VML) system describes…
Spiking Neural Networks (SNNs) offer a promising energy-efficient alternative to Artificial Neural Networks (ANNs) by utilizing sparse and asynchronous processing through discrete spike-based computation. However, the performance of deep…
Designing superconducting coils for a tokamak fusion device is a highly coupled, non-linear design problem. The coils have many disparate engineering requirements from structural to power electronics, as well strict limits placed on the…
In the quest for controlled thermonuclear fusion, tokamaks present complex challenges in understanding burning plasma dynamics. This study introduces a multi-region multi-timescale transport model, employing Neural Ordinary Differential…
The high dimensionality and complex dynamics of turbulent flows remain an obstacle to the discovery and implementation of control strategies. Deep reinforcement learning (RL) is a promising avenue for overcoming these obstacles, but…
The MAST (Mega Ampere Spherical Tokamak) real time plasma position controller is based on an optical linear camera placed on the mid plane of the vessel. This solution has the advantage of being a direct observation of the D…
A new version of GBS (Ricci et al. Plasma Phys. Control. Fusion 54, 124047, 2012; Halpern et al. J. Comput. Phys. 315, 388-408, 2016; Paruta et al. Phys. Plasmas 25, 112301, 2018) is described. GBS is a three-dimensional, flux-driven,…
The integration of event cameras and spiking neural networks (SNNs) promises energy-efficient visual intelligence, yet scarce event data and the sparsity of DVS outputs hinder effective training. Prior knowledge transfers from RGB to DVS…
Modeling thermal states for complex space missions, such as the surface exploration of airless bodies, requires high computation, whether used in ground-based analysis for spacecraft design or during onboard reasoning for autonomous…
A formulation of a multi-input single-output closed-loop subspace method for system identification is employed for the purpose of obtaining control-relevant model of the vacuum-plasma response in the Damavand tokamak. Such a model is…
Spontaneous spin-up are widely observed in tokamak plasmas, which is crucially important for plasma confinement. A kinetic theory is proposed to show that a toroidal rotation of core plasma is induced by the topological transition of orbits…
This work presents a novel methodology for analysis and control of nonlinear fluid systems using neural networks. The approach is demonstrated on four different study cases being the Lorenz system, a modified version of the…
We propose an Entropy-Optimal Manifold Clustering (EOMC) - and show that it mitigates the cost scaling and robustness issues of the existing dimensionality reduction and manifold learning tools in nonstationary and nonlinear situations,…
Machine learning models are exceptionally effective in capturing complex non-linear relationships of high-dimensional datasets and making accurate predictions. However, their intrinsic ``black-box'' nature makes it difficult to interpret…