Gradient Reconstruction in Lattice Boltzmann Methods for Systems of Conservation Laws
Abstract
The automatic derivation turns a declared system of conservation laws into a lattice Boltzmann scheme, giving each conserved physical quantity a set of q populations whose linear equilibrium embeds the physical flux in their first moment. When the flux depends on gradients of the conserved state, those gradients are supplied by tracking them as additional transported fields. Since lattice Boltzmann is commonly memory-bound, these additional degrees of freedom reduce the achievable throughput. To reclaim it, we reconstruct the gradients from the moment structure of the equilibrium instead of transporting them. To leading order, the first moment of a conserved quantity's non-equilibrium part carries its gradient. Because the reconstructed flux enters its own equilibrium reference, that moment is the image of the gradient under a linear operator built from the diffusive-flux Jacobian. The reconstruction is that operator's algebraic inverse, generic across gradient-form constitutive closures and generated automatically from each declared flux. The resulting scheme carries the conserved quantities alone, reconstructing the required gradients from the populations and forming the fluxes locally. It converges at second order in double precision across advection-diffusion-reaction, Allen-Cahn, Navier-Stokes, resistive magnetohydrodynamics and homogenized compressible Navier-Stokes-Fourier systems, matching the accuracy of gradient tracking at equal resolution. On an NVIDIA RTX A5000 it is up to 3.7 times faster in single precision, the memory-bound kernels reaching up to 97% of peak memory bandwidth.
Related
- A Gradient-Reconstruction Lattice Boltzmann Method for Compressible Navier--Stokes--Fourier Equations
- Conservation-Syndrome Quantum Error Correction for Lattice-Boltzmann Quantum Algorithms
- Bridging the Lattice Boltzmann Method and Moment Methods
- Lattice Boltzmann Methods with Anisotropic Equilibrium Distributions
- Rotation Collision based Quantum Lattice Boltzmann Methods
BibTeX
@misc{kummerlander2026gradient,
title = {{Gradient Reconstruction in Lattice Boltzmann Methods for Systems of Conservation Laws}},
author = {Adrian Kummerländer and Fedor Bukreev and Mathias J. Krause},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.06199},
eprint = {2609.06199},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.06199},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Gradient Reconstruction in Lattice Boltzmann Methods for Systems of Conservation Laws
AU - Adrian Kummerländer
AU - Fedor Bukreev
AU - Mathias J. Krause
PY - 2026
DA - 2026/09/05
PB - arXiv
DO - 10.48550/arxiv.2609.06199
UR - https://arxiv.org/abs/2609.06199
AB - The automatic derivation turns a declared system of conservation laws into a lattice Boltzmann scheme, giving each conserved physical quantity a set of q populations whose linear equilibrium embeds the physical flux in their first moment. When the flux depends on gradients of the conserved state, those gradients are supplied by tracking them as additional transported fields. Since lattice Boltzmann is commonly memory-bound, these additional degrees of freedom reduce the achievable throughput. To reclaim it, we reconstruct the gradients from the moment structure of the equilibrium instead of transporting them. To leading order, the first moment of a conserved quantity's non-equilibrium part carries its gradient. Because the reconstructed flux enters its own equilibrium reference, that moment is the image of the gradient under a linear operator built from the diffusive-flux Jacobian. The reconstruction is that operator's algebraic inverse, generic across gradient-form constitutive closures and generated automatically from each declared flux. The resulting scheme carries the conserved quantities alone, reconstructing the required gradients from the populations and forming the fluxes locally. It converges at second order in double precision across advection-diffusion-reaction, Allen-Cahn, Navier-Stokes, resistive magnetohydrodynamics and homogenized compressible Navier-Stokes-Fourier systems, matching the accuracy of gradient tracking at equal resolution. On an NVIDIA RTX A5000 it is up to 3.7 times faster in single precision, the memory-bound kernels reaching up to 97% of peak memory bandwidth.
ER -