pith. sign in

arxiv: 2512.11598 · v2 · submitted 2025-12-12 · 🧮 math.NA · cs.NA

A meshless MUSCL method for the BGK-Boltzmann equation

Klaas Willems , Axel Klar , Giovanni Russo , Giovanni Samaey , Sudarshan Tiwari This is my paper

Pith reviewed 2026-05-16 22:32 UTC · model grok-4.3

classification 🧮 math.NA cs.NA
keywords meshless MUSCLBGK-Boltzmannmoving boundariesMOOD limiterALE methodrarefied gas simulationhigh-order accuracynumerical method
0
0 comments X

The pith

A meshless MUSCL scheme on moving grids solves the BGK equation to fourth order in one dimension for flows with moving boundaries.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper develops a meshless method to simulate rarefied gases interacting with moving boundaries using the BGK-Boltzmann equation. It combines an Arbitrary Lagrangian-Eulerian formulation, where points move with the gas mean velocity, with a Moving Least Squares MUSCL discretization and high-order IMEX time stepping. An adapted MOOD limiter prevents oscillations at shocks while preserving accuracy. The approach avoids iterative boundary condition procedures. Tests on cavity flows, shear layers, and shock tubes demonstrate the claimed convergence rates.

Core claim

The method uses a Lagrangian BGK equation discretized on a moving point cloud via meshless MUSCL reconstruction based on Moving Least Squares. Time integration employs an Implicit-Explicit Runge-Kutta scheme. An adapted MOOD criterion relaxes the discrete maximum principle at discontinuities. Boundary conditions are applied directly. This combination achieves fourth-order accuracy in one-dimensional tests and second-order accuracy in two-dimensional simulations involving moving boundaries and rigid bodies.

What carries the argument

Meshless Moving Least Squares MUSCL reconstruction on an ALE moving grid with adapted MOOD limiting for the BGK equation.

If this is right

  • The algorithm handles time-dependent domains and rigid objects without remeshing or special treatments.
  • High-order accuracy is preserved across discontinuities thanks to the MOOD detector.
  • Boundary conditions for moving walls require no iteration or extrapolation.
  • Classical test cases like driven cavities and shock tubes confirm the method's performance.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The Lagrangian movement of grid points may improve resolution in regions of high velocity gradients.
  • Similar techniques could extend to other Boltzmann-type equations or multi-species gases.
  • Computational savings might arise in problems where the flow aligns with boundary motion, reducing the need for fixed fine grids.
  • Further analysis could explore stability limits when the MOOD criterion is relaxed in higher dimensions.

Load-bearing premise

The adapted MOOD criterion relaxes the discrete maximum property at discontinuities without degrading the formal order of accuracy or introducing excessive numerical diffusion on the chosen test problems.

What would settle it

Running the shock tube or driven cavity test cases and measuring convergence rates below fourth order in 1D or second order in 2D, or observing persistent oscillations near discontinuities, would disprove the accuracy claims.

Figures

Figures reproduced from arXiv: 2512.11598 by Axel Klar, Giovanni Russo, Giovanni Samaey, Klaas Willems, Sudarshan Tiwari.

Figure 1
Figure 1. Figure 1: Illustration of the variables used in the first-order spatial discretisation method. [PITH_FULL_IMAGE:figures/full_fig_p008_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: To find the neighbours of the red grid point, the distance to all points in the green highlighted [PITH_FULL_IMAGE:figures/full_fig_p013_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: The relative L 1 error of the density (left) and mean velocity (right) for several schemes for a smooth initial condition in one spatial dimension. Remark 8. Since we fix the number of grid points in velocity space, we are reporting on the convergence of a discrete velocity model, since there is some error related to the velocity discretisation that we are not measuring here. For small Knudsen numbers, as … view at source ↗
Figure 4
Figure 4. Figure 4: The relative L1 error of the density (left) and temperature (right) for several schemes for a smooth initial condition in two spatial dimension. Because the convergence of the mean velocity is the same as the convergence for the density for all methods, the plots are omitted. 4.3 Shock tube In this subsection, and the next two, we shall make use of dimensional quantities and adopt the SI system, so that le… view at source ↗
Figure 5
Figure 5. Figure 5: The numerical solution and the exact solution to Sod’s shock tube at time 0 [PITH_FULL_IMAGE:figures/full_fig_p016_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: The mass error for several numerical methods for Sod’s shock tube. [PITH_FULL_IMAGE:figures/full_fig_p016_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Schematic of the moving plate problem with [PITH_FULL_IMAGE:figures/full_fig_p017_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: The position (left) and velocity (right) of the moving plate up to time [PITH_FULL_IMAGE:figures/full_fig_p018_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: The velocity in the x-direction along the vertical centre line of the square cavity. The gas is in slip flow regime, see the first row in table 2. ρ 0 = 15 π , mean velocity U 0 x = ( tanh 15 π [PITH_FULL_IMAGE:figures/full_fig_p019_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Velocity field of the driven square cavity simulation plotted at several times for various Knudsen [PITH_FULL_IMAGE:figures/full_fig_p020_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Plots of the vorticity for the shear flow problem. [PITH_FULL_IMAGE:figures/full_fig_p021_11.png] view at source ↗
read the original abstract

We present a numerical method for simulating rarefied gases that interact with moving boundaries and rigid bodies. The gas is described by the BGK equation in Lagrangian form and solved using an Arbitrary Lagrangian-Eulerian method, in which grid points move with the local mean velocity of the gas. The main advantage of the moving grid is that the algorithm can deal well with cases where the domain boundaries are time-dependent and the simulation domain contains rigid objects. Due to the irregular nature of the grid, we use a novel meshless MUSCL-like Moving Least Squares Method (MLS) for spatial discretisation coupled with a higher-order Implicit-Explicit Runge-Kutta method. To avoid spurious oscillations at discontinuities, we use the so-called Multi-dimensional Optimal Order Detection (MOOD) method with an adapted criterion to relax the discrete maximum property. Finally, we employ a new implementation of the boundary conditions that requires no iterative or extrapolation procedure. The method achieves fourth-order in 1D and second-order in 2D for simulations with moving boundaries. We demonstrate the method's effectiveness on classical test cases such as the driven square cavity, shear layer, and shock tube.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The paper presents a meshless Arbitrary Lagrangian-Eulerian (ALE) scheme for the BGK-Boltzmann equation on moving point clouds. Spatial discretization uses a MUSCL-like Moving Least Squares (MLS) reconstruction, time integration employs a higher-order IMEX Runge-Kutta method, and oscillations are controlled by an adapted Multi-dimensional Optimal Order Detection (MOOD) limiter that relaxes the discrete maximum property. A non-iterative boundary-condition treatment is introduced for rigid moving bodies. The central claims are fourth-order accuracy in one dimension and second-order accuracy in two dimensions on classical test problems (driven cavity, shear layer, shock tube) with time-dependent domains.

Significance. If the reported convergence rates hold under the combination of moving irregular stencils, MLS fitting, and the adapted MOOD detector, the method would supply a practical meshless alternative for rarefied-gas problems involving rigid-body motion without remeshing. The absence of quantitative error tables or grid-refinement studies in the current manuscript, however, leaves the accuracy claims unsubstantiated.

major comments (2)
  1. [Abstract] Abstract: the statements that the method 'achieves fourth-order in 1D and second-order in 2D' are presented without any accompanying L1/L2 error norms, grid-refinement tables, or comparison against reference solutions. In the absence of these data the central accuracy claims cannot be verified.
  2. [Method] Method section (MOOD adaptation): the adapted MOOD criterion is asserted to activate only at genuine discontinuities while preserving the formal truncation error of the underlying MLS reconstruction elsewhere. With time-dependent ALE point clouds and velocity-induced stencil changes near rigid boundaries, no quantitative bound is given on false-positive activation frequency in smooth regions or on its effect on the observed convergence rates.
minor comments (1)
  1. [Abstract] The abstract lists test cases but does not indicate which norms or reference solutions are used to measure the reported orders.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback. We will revise the manuscript to include quantitative error tables, grid-refinement studies, and additional analysis of the MOOD limiter to fully substantiate the accuracy claims.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the statements that the method 'achieves fourth-order in 1D and second-order in 2D' are presented without any accompanying L1/L2 error norms, grid-refinement tables, or comparison against reference solutions. In the absence of these data the central accuracy claims cannot be verified.

    Authors: We agree that the abstract claims require supporting data for verification. In the revised manuscript we will add L1 and L2 error norms from systematic grid-refinement studies in both 1D and 2D, together with direct comparisons against reference solutions (analytical or high-resolution reference computations). These tables will be placed in the results section and referenced from the abstract. revision: yes

  2. Referee: [Method] Method section (MOOD adaptation): the adapted MOOD criterion is asserted to activate only at genuine discontinuities while preserving the formal truncation error of the underlying MLS reconstruction elsewhere. With time-dependent ALE point clouds and velocity-induced stencil changes near rigid boundaries, no quantitative bound is given on false-positive activation frequency in smooth regions or on its effect on the observed convergence rates.

    Authors: We acknowledge the need for quantitative bounds on false-positive MOOD activations under moving stencils. In the revision we will add a dedicated subsection with numerical experiments that measure activation frequency in smooth regions across the test suite, including near moving boundaries, and we will show that the reported convergence rates remain unaffected. This will be supported by additional convergence plots with and without the limiter. revision: yes

Circularity Check

0 steps flagged

No significant circularity: standard MLS-MOOD construction with empirical order verification

full rationale

The paper constructs its meshless MUSCL scheme from established ingredients (ALE Lagrangian form of BGK, moving-least-squares reconstruction, IMEX Runge-Kutta time stepping, and an adapted MOOD limiter that relaxes the discrete-maximum property only at detected discontinuities). The claimed orders (fourth in 1D, second in 2D) are obtained from numerical convergence studies on standard test problems rather than being fitted parameters or self-defined quantities. No equation or step reduces the central accuracy statements to tautological inputs, and the boundary-condition implementation is presented as a separate technical contribution without serving as a load-bearing premise for the order claims. The derivation chain therefore remains self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The scheme rests on the standard BGK collision model and on the numerical properties of moving-least-squares approximation and MOOD detection; no new physical constants or entities are introduced.

axioms (2)
  • domain assumption The BGK operator provides a sufficient relaxation model for the Boltzmann collision integral in the regimes considered
    Invoked in the problem statement and test-case selection
  • standard math Moving-least-squares reconstruction reproduces polynomials up to the chosen degree on irregular point sets
    Underlying the MUSCL-like spatial discretization

pith-pipeline@v0.9.0 · 5512 in / 1373 out tokens · 45776 ms · 2026-05-16T22:32:04.298448+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

11 extracted references · 11 canonical work pages · 2 internal anchors

  1. [1]

    2000), 813–830

    [1]Andries, P., Le Tallec, P., Perlat, J.-P., and Perthame, B.The Gaussian-BGK model of Boltzmann equation with small Prandtl number.European Journal of Mechanics - B/Fluids 19, 6 (Nov. 2000), 813–830. [2]Ascher, U. M., Ruuth, S. J., and Spiteri, R. J.Implicit-explicit Runge-Kutta methods for time-dependent partial differential equations.Applied Numerical...

    work page 2000

  2. [2]

    [10]Bressan, A.Hyperbolic systems of conservation laws: the one dimensional Cauchy problem, reprint ed

    arXiv:2502.07548 [math]. [10]Bressan, A.Hyperbolic systems of conservation laws: the one dimensional Cauchy problem, reprint ed. No. 20 in Oxford lecture series in mathematics and its applications. Oxford University Press, Oxford,

    work page arXiv

  3. [3]

    2012), 6643–6664

    [14]Chen, S., Xu, K., Lee, C., and Cai, Q.A unified gas kinetic scheme with moving mesh and velocity space adaptation.Journal of Computational Physics 231, 20 (Aug. 2012), 6643–6664. [15]Cho, S. Y., Boscarino, S., Groppi, M., and Russo, G.Conservative semi-Lagrangian schemes for a general consistent BGK model for inert gas mixtures.Communications in Mathe...

    work page arXiv 2012

  4. [4]

    High order semi-Lagrangian methods for the BGK equation

    arXiv:1411.7929 [math]. [28]Groppi, M., Russo, G., and Stracquadanio, G.Boundary conditions for semi-Lagrangian methods for the BGK model.Communications in Applied and Industrial Mathematics 7, 3 (Sept. 2016), 138–164. [29]Holway, Jr., L. H.Kinetic Theory of Shock Structure Using an Ellipsoidal Distribution Function. InProceedings of the fourth internatio...

    work page internal anchor Pith review Pith/arXiv arXiv 2016

  5. [5]

    2000), 1461–1488

    [31]Kurganov, A., and Levy, D.A Third-Order Semidiscrete Central Scheme for Conservation Laws and Convection-Diffusion Equations.SIAM Journal on Scientific Computing 22, 4 (Jan. 2000), 1461–1488. [32]Lancaster, P., and Salkauskas, K.Surfaces generated by moving least squares methods.Mathematics of Computation 37, 155 (1981), 141–158. [33]LeVeque, R. J.Num...

    work page 2000

  6. [6]

    [35]Mieussens, L.Discrete velocity model and implicit scheme for the BGK equation of rarefied gas dynamics

    [34]Levin, D.The approximation power of moving least-squares.Mathematics of Computation 67, 224 (1998), 1517–1531. [35]Mieussens, L.Discrete velocity model and implicit scheme for the BGK equation of rarefied gas dynamics. Mathematical Models and Methods in Applied Sciences 10, 08 (Nov. 2000), 1121–1149. [36]Pareschi, L., and Perthame, B.A Fourier spectra...

    work page 1998

  7. [7]

    A New Class of Conservative Large Time Step Methods for the BGK Models of the Boltzmann Equation

    arXiv:1103.5247 [math]. [42]Seibold, B.M-matrices in meshless finite difference methods. Berichte aus der Mathematik. Shaker, Aachen,

    work page internal anchor Pith review Pith/arXiv arXiv

  8. [8]

    InProceedings of the 1968 23rd ACM national conference(1968), ACM Press, pp

    [43]Shepard, D.A two-dimensional interpolation function for irregularly-spaced data. InProceedings of the 1968 23rd ACM national conference(1968), ACM Press, pp. 517–524. [44]Shrestha, S., Tiwari, S., Klar, A., and Hardt, S.Numerical simulation of a moving rigid body in a rarefied gas.Journal of Computational Physics 292(July 2015), 239–252. [45]Suchde, P...

    work page 1968

  9. [9]

    Y., Che Sidik, N

    [46]Tey, W. Y., Che Sidik, N. A., Asako, Y., W. Muhieldeen, M., and Afshar, O.Moving Least Squares Method and its Improvement: A Concise Review.Journal of Applied and Computational Mechanics 7, 2 (Apr. 2021). 23 [47]Tiwari, S., Klar, A., and Russo, G.A meshfree arbitrary Lagrangian-Eulerian method for the BGK model of the Boltzmann equation with moving bo...

    work page 2021

  10. [10]

    N., and Weideman, J

    [49]Trefethen, L. N., and Weideman, J. A. C.The Exponentially Convergent Trapezoidal Rule.SIAM Review 56, 3 (Jan. 2014), 385–458. [50]V. Bobylev, A., Bisi, M., Groppi, M., Spiga, G., and F. Potapenko, I.A general consistent BGK model for gas mixtures.Kinetic & Related Models 11, 6 (2018), 1377–1393. [51]Willems, K.meshfree4bgk2d.https://github.com/KlaasWi...

    work page 2014

  11. [11]

    2010), 7747–7764

    [54]Xu, K., and Huang, J.-C.A unified gas-kinetic scheme for continuum and rarefied flows.Journal of Compu- tational Physics 229, 20 (Oct. 2010), 7747–7764. [55]Zabelok, S., Arslanbekov, R., and Kolobov, V.Adaptive kinetic-fluid solvers for heterogeneous computing architectures.Journal of Computational Physics 303(Dec. 2015), 455–469. 24

    work page 2010