Plasma Surrogate Modelling using Fourier Neural Operators
Authors:
Vignesh Gopakumar,
Stanislas Pamela,
Lorenzo Zanisi,
Zongyi Li,
Ander Gray,
Daniel Brennand,
Nitesh Bhatia,
Gregory Stathopoulos,
Matt Kusner,
Marc Peter Deisenroth,
Anima Anandkumar,
JOREK Team,
MAST Team
Abstract:
Predicting plasma evolution within a Tokamak reactor is crucial to realizing the goal of sustainable fusion. Capabilities in forecasting the spatio-temporal evolution of plasma rapidly and accurately allow us to quickly iterate over design and control strategies on current Tokamak devices and future reactors. Modelling plasma evolution using numerical solvers is often expensive, consuming many hou…
▽ More
Predicting plasma evolution within a Tokamak reactor is crucial to realizing the goal of sustainable fusion. Capabilities in forecasting the spatio-temporal evolution of plasma rapidly and accurately allow us to quickly iterate over design and control strategies on current Tokamak devices and future reactors. Modelling plasma evolution using numerical solvers is often expensive, consuming many hours on supercomputers, and hence, we need alternative inexpensive surrogate models. We demonstrate accurate predictions of plasma evolution both in simulation and experimental domains using deep learning-based surrogate modelling tools, viz., Fourier Neural Operators (FNO). We show that FNO has a speedup of six orders of magnitude over traditional solvers in predicting the plasma dynamics simulated from magnetohydrodynamic models, while maintaining a high accuracy (MSE in the normalised domain $\approx$ $10^{-5}$). Our modified version of the FNO is capable of solving multi-variable Partial Differential Equations (PDE), and can capture the dependence among the different variables in a single model. FNOs can also predict plasma evolution on real-world experimental data observed by the cameras positioned within the MAST Tokamak, i.e., cameras looking across the central solenoid and the divertor in the Tokamak. We show that FNOs are able to accurately forecast the evolution of plasma and have the potential to be deployed for real-time monitoring. We also illustrate their capability in forecasting the plasma shape, the locations of interactions of the plasma with the central solenoid and the divertor for the full (available) duration of the plasma shot within MAST. The FNO offers a viable alternative for surrogate modelling as it is quick to train and infer, and requires fewer data points, while being able to do zero-shot super-resolution and getting high-fidelity solutions.
△ Less
Submitted 18 June, 2024; v1 submitted 10 November, 2023;
originally announced November 2023.
Rydberg States of H$_3$ and HeH as Potential Coolants for Primordial Star Formation
Authors:
Gokul Kannan,
Jeremy R. Chien,
Anthony J. Benjamin,
Niranjan Bhatia,
Richard J. Saykally
Abstract:
Current theory and measurements establish the age of the universe as ca. 13.8 billion years. For the first several hundred million years of its existence, it was a dark, opaque void. After that, the hydrogen atoms comprising most of the "ordinary" matter began to condense and ionize, eventually forming the first stars that would illuminate the sky. Details of how these "primordial" stars formed ha…
▽ More
Current theory and measurements establish the age of the universe as ca. 13.8 billion years. For the first several hundred million years of its existence, it was a dark, opaque void. After that, the hydrogen atoms comprising most of the "ordinary" matter began to condense and ionize, eventually forming the first stars that would illuminate the sky. Details of how these "primordial" stars formed have been widely debated, but remain elusive. A central issue in this process is the mechanism by which the primordial gas (mainly hydrogen and helium atoms) collected via the action of dark matter cools and further accretes to fusion densities. Current models invoke collisional excitation of H$_2$ molecular rotations and subsequent radiative rotational transitions allowed by the weak molecular quadrupole moment. In this article, we review the salient considerations, and present some new ideas, bases on recent spectroscopic observations of neutral H$_3$ Rydberg electronic state emission in the mid-infrared.
△ Less
Submitted 3 November, 2020;
originally announced November 2020.
