Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Neural Score-Based Particle Method for the Vlasov-Maxwell-Landau System

Vasily Ilin, Jingwei Hu

Abstract

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 six-dimensional phase-space transport under self-consistent electromagnetic fields together with the nonlinear, nonlocal Landau collision operator. A recent deterministic particle method for the full VML system estimates the velocity score function via the blob method, a kernel-based approximation with $O(n^2)$ cost. In this work, we replace the blob score estimator with score-based transport modeling (SBTM), in which a neural network is trained on-the-fly via implicit score matching at $O(n)$ cost. We prove that the approximated collision operator preserves momentum and kinetic energy, and dissipates an estimated entropy. We also characterize the unique global steady state of the VML system and its electrostatic reduction, providing the ground truth for numerical validation. On three canonical benchmarks -- Landau damping, two-stream instability, and Weibel instability -- SBTM is more accurate than the blob method, achieves correct long-time relaxation to Maxwellian equilibrium where the blob method fails, and delivers $50\%$ faster runtime with $4\times$ lower peak memory.

A Neural Score-Based Particle Method for the Vlasov-Maxwell-Landau System

Abstract

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 six-dimensional phase-space transport under self-consistent electromagnetic fields together with the nonlinear, nonlocal Landau collision operator. A recent deterministic particle method for the full VML system estimates the velocity score function via the blob method, a kernel-based approximation with cost. In this work, we replace the blob score estimator with score-based transport modeling (SBTM), in which a neural network is trained on-the-fly via implicit score matching at cost. We prove that the approximated collision operator preserves momentum and kinetic energy, and dissipates an estimated entropy. We also characterize the unique global steady state of the VML system and its electrostatic reduction, providing the ground truth for numerical validation. On three canonical benchmarks -- Landau damping, two-stream instability, and Weibel instability -- SBTM is more accurate than the blob method, achieves correct long-time relaxation to Maxwellian equilibrium where the blob method fails, and delivers faster runtime with lower peak memory.

Paper Structure

This paper contains 13 sections, 3 theorems, 52 equations, 13 figures, 2 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Background and Method
  3. The Vlasov--Maxwell--Landau System
  4. Regularized Landau Operator and Particle Method
  5. Score Estimation
  6. Numerical Experiments
  7. Setup
  8. Landau Damping
  9. Two-Stream Instability
  10. Weibel Instability
  11. Computational Efficiency
  12. Conclusion
  13. Proofs of Equilibrium Theorems

Key Result

Theorem 2.1

Consider the Vlasov--Maxwell--Landau system eq:vlasov-eq:maxwell_full with collision frequency $\nu > 0$, periodic spatial domain $x \in \mathbb{T}^{3}$, $v \in \mathbb{R}^{3}$, and a uniform neutralizing background ion density $\rho_{\mathrm{ion}} > 0$. Let $f(x,v) > 0$ be a sufficiently smooth ste where $B_\infty$ is the spatial mean of the initial magnetic field (a conserved quantity), and the

Figures (13)

  • Figure 1: Landau damping at $\nu = 0.4$: estimated entropy production and total energy across particle counts. SBTM (red) dissipates estimated entropy consistently and maintains near-constant total energy. The blob method (blue) overestimates estimated entropy production and exhibits energy drift.
  • Figure 2: Landau damping: $L^2$ norm of the electric field over time at $\nu = 0.4$. The linear theory damping rate is shown for reference (at $\nu = 0.4$ it is not expected to be accurate). The blob method's damping rate varies with $n$ (compare (a), (c), and (e)), converging toward the SBTM rate. The SBTM rate is consistent across all particle counts (b), (d), (f).
  • Figure 3: Landau damping: velocity-space distributions at $n = 10^6$. Rows 1--2: $\nu = 0.4$ (heatmaps and $v_2$-marginal). Rows 3--4: $\nu = 1.0$. SBTM maintains smooth Gaussian tails; the blob method shows unphysical sharp cutoffs at $|v_2| \approx 3$.
  • Figure 4: Two-stream instability: energy and estimated entropy diagnostics at $\nu = 0.32$. Blob (blue) vs SBTM (red) at $n = 5 \times 10^5$ (solid), $10^6$ (dashed), $3 \times 10^6$ (dotted).
  • Figure 5: Two-stream instability: $(x, v_1)$ phase space at $\nu = 0.32$ across particle counts. SBTM achieves full vortex dissipation at all particle counts, while the blob method requires $n = 3 \times 10^6$ to approach the same level of dissipation.
  • ...and 8 more figures

Theorems & Definitions (7)

  • Theorem 2.1: Global steady state of the VML system
  • Theorem 2.2: Global steady state of the VPL system
  • Theorem 2.3
  • proof
  • Remark 2.4
  • proof : Proof of Theorem \ref{['thm:equilibrium']}
  • proof : Proof of Theorem \ref{['thm:equilibrium_vpl']}