Electron neural closure for turbulent magnetosheath simulations: energy channels

2025-09-30

AI Summary

Conventional local closure models—such as the double-adiabatic approximation or the magnetic-loop-pressure (MLP) model—fail to accurately capture the nonlocal spatial correlations underlying the pressure–strain interaction energy channel in magnetosheath turbulence. Method: We propose a nonlocal five-moment electron pressure tensor closure model, implemented via an energy-conserving fully convolutional neural network (FCNN) surrogate that directly learns the nonlocal response kernel from high-fidelity kinetic particle-in-cell simulation data. Contribution/Results: The model faithfully reconstructs both the spatial distribution and conditional statistics of the pressure–strain tensor; small-scale structural fidelity improves markedly with increased training data, and its generalization capability surpasses existing local closures. By enabling efficient, physics-informed closure of the kinetic equation, this approach establishes a new paradigm for multiscale energy transport modeling in collisionless plasmas.

Technology Category

• Machine Learning: Kernel Methods
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems

Abstract

In this work, we introduce a non-local five-moment electron pressure tensor closure parametrized by a Fully Convolutional Neural Network (FCNN). Electron pressure plays an important role in generalized Ohm's law, competing with electron inertia. This model is used in the development of a surrogate model for a fully kinetic energy-conserving semi-implicit Particle-in-Cell simulation of decaying magnetosheath turbulence. We achieve this by training FCNN on a representative set of simulations with a smaller number of particles per cell and showing that our results generalise to a simulation with a large number of particles per cell. We evaluate the statistical properties of the learned equation of state, with a focus on pressure-strain interaction, which is crucial for understanding energy channels in turbulent plasmas. The resulting equation of state learned via FCNN significantly outperforms local closures, such as those learned by Multi-Layer Perceptron (MLP) or double adiabatic expressions. We report that the overall spatial distribution of pressure-strain and its conditional averages are reconstructed well. However, some small-scale features are missed, especially for the off-diagonal components of the pressure tensor. Nevertheless, the results are substantially improved with more training data, indicating favorable scaling and potential for improvement, which will be addressed in future work.

Problem

• Developing neural network closure for electron pressure in plasma simulations
• Improving turbulent magnetosheath modeling with non-local pressure tensor
• Enhancing energy channel analysis in decaying plasma turbulence

Innovation

• Fully Convolutional Neural Network parametrizes electron pressure tensor closure
• Surrogate model developed for kinetic Particle-in-Cell turbulence simulation
• Learned equation of state outperforms local closures and adiabatic expressions