Pith. sign in

REVIEW 2 major objections 4 minor 46 references

Asymptotic-preserving schemes for the initial-boundary value problem of hyperbolic relaxation systems

T0 review · 2 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read This paper constructs a first-order asymptotic-preserving scheme that resolves the relaxation limit of hyperbolic initial-boundary value problems with boundary layers on meshes far coarser than the layer thickness.

desk verdict Genuinely new boundary AP scheme for relaxation IBVPs, well-constructed and convincingly tested, but the AP claim rests on consistency alone and would be stronger with a stability estimate. read the letter →

arxiv 2505.22656 v1 pith:3WXXTHZO submitted 2025-05-28 math.NA cs.NA

Yizhou Zhou This is my paper
classification math.NAcs.NA MSC 35L5065M0676N20
keywords asymptotic-preservingschemeshyperbolicrelaxationsystemsinitial-boundaryvalueproblemboundarylayersJin-XinmodelinterfaceIMEXupwindsingularperturbations
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper extends asymptotic-preserving (AP) schemes from initial-value problems to initial-boundary value problems for first-order hyperbolic relaxation systems. Its central claim is that a modified upwind scheme, built from eigenvectors of the matrix $M(\eta)=A^{-1}(I-\eta Q)$, captures the equilibrium limit together with its $O(\varepsilon)$ boundary layer on meshes with $h \gg \varepsilon$, and is first-order accurate in the relaxation limit. The same construction yields a unified scheme for interface problems in which the relaxation parameter jumps across a point. The payoff is that stiff relaxation IBVPs can be solved directly on coarse meshes, without resolving the layer or decomposing the domain.

What carries the argument

The central object is the modified characteristic matrix $M(\eta)=A^{-1}(I-\eta Q)$; for the Jin–Xin model it is $M(\eta)=\begin{pmatrix}-\eta a & 1+\eta\\ 1 & 0\end{pmatrix}$ with $a=f'(u)$. Its eigenvectors interpolate between the eigenvectors of the flux matrix $A$ as $\eta\to0$ and the equilibrium projection as $\eta\to\infty$, and the scheme upwinds along these eigenvectors. In the non-stiff regime $\tau\ll\varepsilon$ this reproduces the classical upwind scheme to $O(\tau^p)$, while in the stiff regime $\tau\gg\varepsilon$ it automatically discretizes the equilibrium system with the reduced boundary condition, using the boundary update $L_-^M A^{-1}(U_0^{n+1}-U_0^n)+\lambda L_-^M(U_1^n-U_0^n)=\frac{\tau}{\varepsilon}L_-^M A^{-1}\text{source}$. For general linear systems the paper avoids a continuous eigenvalue perturbation by defining $R_\pm=R_\pm^A$ for $\tau<\varepsilon$ and $R_\pm=R_\pm^\infty$ for $\tau\ge\varepsilon$.

What would settle it

A concrete computation that would settle the claim is the von Neumann or energy analysis of the linear Jin–Xin scheme (2.13)–(2.15) with $f'<0$ and $\varepsilon\ll h$: if any Fourier mode has amplification factor above 1, or if the discrete energy grows as $\varepsilon\to0$ on a fixed mesh, the first-order convergence claim fails. Equivalently, a numerical run of Example 1 over long time should show boundary errors staying $O(\tau)$; if the error accumulates with time, the missing uniform stability is real.

Watch

Extended reading notes

Core claim

The paper proposes boundary AP schemes (2.13)–(2.15) for the Jin–Xin model and (3.13) for general linear relaxation systems with constant coefficients, and claims that each is a first-order scheme for the asymptotic relaxation limit. As $\varepsilon \to 0$ on a fixed mesh, the limiting scheme is consistent with the limit value (2.8) or (3.11) up to $O(\tau)$ at the boundary and $O(\tau^2)$ in the interior, so the boundary layer, whose thickness is $O(\varepsilon)$, appears as a discrete jump at the boundary point. For the Jin–Xin model the discretization uses left eigenvectors of $M(\eta)=A^{-1}(I-\eta Q)$ with $\eta=(\tau/\varepsilon)^p$; for general linear systems, where the eigenvectors may depend non-analytically on $\eta$, the scheme switches between the flux eigenvectors of $A$ and the limiting eigenvectors at $\eta=\infty$ given by Lemma 3.3. The paper also reformulates the same idea for interface problems and reports first-order convergence in numerical examples, including the case where the classical upwind scheme produces an $O(1)$ error at the boundary.

Load-bearing premise

The load-bearing premise is that the AP verification by substitution and truncation error is enough: the scheme is assumed to be stable uniformly in $\varepsilon$, but no such discrete stability estimate is proved.

Editorial extensions

If this is right

  • For the Jin–Xin model with $f'(u)<0$, the boundary-layer correction $\mu(0,t)$ is captured as a one-cell jump at $j=0$, so the layer does not need to be resolved by the mesh.
  • When $f'(u)>0$, no boundary layer forms and the scheme reduces to the usual first-order upwind scheme, so it is AP in both regimes without switching equations.
  • For general linear systems satisfying the structural stability condition and the generalized Kreiss condition, the same discretization solves the IBVP with coarse meshes and gives the reduced boundary condition automatically.
  • For interface problems with a discontinuous relaxation parameter, the unified scheme (4.9)–(4.10) computes both sides with one system and one mesh, and the numerical examples show that the classical upwind scheme's interface spike is removed.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The paper's AP verification substitutes the formal relaxation limit into the limiting scheme and checks truncation error; it never proves a stability estimate uniform in $\varepsilon$. If such an estimate does not hold, the consistency calculation alone would not establish convergence, so the numerical examples are currently the main evidence for the first-order claim.
  • Because the general linear scheme switches discontinuously between $R^A$ and $R^\infty$ when $\tau$ crosses $\varepsilon$, its behavior near $\tau\approx\varepsilon$ is left unexplored; a smooth interpolation between the two eigenbases could make the scheme more robust in the intermediate regime.
  • The same eigenvector-interpolation idea may extend to multidimensional or characteristic-boundary relaxation problems, where boundary layers are not scalar and the reduced boundary condition is more involved; the paper's references to characteristic boundaries suggest this as a next step.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 4 minor

Summary. The paper proposes asymptotic-preserving (AP) finite-difference schemes for initial-boundary value problems of hyperbolic relaxation systems. For the Jin-Xin model, the author modifies the upwind flux using eigenvectors of the matrix M(η)=A^{-1}(I-ηQ) with η=(τ/ε)^p, resulting in the scheme (2.13)-(2.15). The construction is then extended to general linear constant-coefficient systems, where the scheme (3.13) uses the limiting right-stable/unstable matrices R^A_± for τ<ε and R^∞_± for τ≥ε, with the switch given in (3.12). The AP property is verified by substituting the formal relaxation limit (2.8)/(3.11) into the limiting scheme and checking truncation residuals of order O(τ) at the boundary point and O(τ^2) at interior points. The same idea is adapted to interface problems in Section 4, and numerical experiments in Section 5 report first-order convergence and improved boundary-layer capture compared with the classical upwind scheme.

Significance. If fully established, the proposed schemes would fill a genuine gap: standard upwind discretizations for hyperbolic relaxation IBVPs produce O(1) boundary-value errors when an O(ε) boundary layer is present, and this paper offers a principled modification based on the generalized Kreiss condition and the asymptotic analysis of Yong [36]. The construction is not parameter-fitted: the matrix M(η) and the limiting matrices R^∞_± are derived from the structure of the relaxation system, and the limiting schemes match the expected equilibrium system and reduced boundary condition. The paper contains explicit eigenvector computations in Appendix A and a careful formal consistency analysis in Sections 2.3 and 3.2. However, the AP claim is supported only by a consistency calculation; no uniform-in-ε stability estimate or convergence proof is provided. The numerical experiments are encouraging, but they do not test the regime τ≈ε where the general scheme switches between the two extreme discretizations. The contribution is therefore solid as a formal construction with promising numerics, but the theoretical claim is not yet fully established.

major comments (2)
  1. [Sections 2.3 and 3.2, Eqs. (2.13)-(2.15) and (3.13)] The AP property is established only by a consistency check. Substituting the formal limit (2.8)/(3.11) into the limiting scheme and computing residuals of O(τ) at j=0 and O(τ^2) for j≥1 shows that the limiting scheme is consistent with the limit, but it does not show that solutions of the original scheme converge to the relaxation limit as ε→0. This requires a uniform-in-ε stability estimate for the fully discrete operator, including the boundary projection in the second equations of (2.13)-(2.15)/(3.13) and the implicit source term. The issue is load-bearing for the central claim of coarse-mesh capture of O(ε) boundary layers: at j=0 the limit is discontinuous on the mesh scale, so the residual statement is not a classical local truncation error bound. Please provide a stability argument or explicitly state that the AP property is being claimed only in the sense of formal consistency of the limiting scheme.
  2. [Section 3.2, Eq. (3.12)] The general-linear scheme is analyzed only at the two extremes η=0 and η=∞. The switch at τ≈ε between R^A_± and R^∞_± is a discontinuous change of the discrete flux, and no consistency or stability analysis is given for the intermediate regime. Since the AP claim for the general system covers all ε, this is a gap. The numerical experiments in Section 5 use ε=1e-9, 1e-6, and 1e-4 with fixed τ and therefore do not sweep across τ≈ε; please add either an analysis of the switching region or experiments that vary ε across it.
minor comments (4)
  1. [Section 2.2 heading] The heading contains a typo: 'upind' should be 'upwind'.
  2. [Section 3.2, after Eq. (3.18)] The displayed estimate mixing O(τ^2) and O(τ) terms is confusing; it would be clearer to write the two contributions as separate estimates and then combine them.
  3. [Appendix A, proof of Lemma 2.2] The normalized eigenvectors are compared with unnormalized vectors such as (0,1) and (-a,1); a sentence noting that normalization factors are absorbed into the O(η^{-1}) terms would prevent confusion.
  4. [Section 4, Eqs. (4.9)-(4.10)] It would help to state explicitly that the interface condition (4.8) is enforced at j=0 and to spell out the relation between L^{r,j}_-, L^{l,j}_+ and the left/right limits of ε(x) at grid points, since the notation is otherwise easy to lose.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the AP claim is a direct consistency check of an explicitly constructed limiting scheme against an independently derived relaxation limit; self-citations are rigorous published theorems.

full rationale

The derivation is self-contained in the relevant sense. The scheme (2.13)-(2.15)/(3.13) is constructed from the eigenvectors of M(eta)=A^{-1}(I-eta Q), and the stiff limit of those eigenvectors is computed directly (Lemma 2.2, Appendix A; Lemma 3.3 from [36]), not fitted to the target limit. The AP verification consists of taking epsilon to zero in the scheme, obtaining an explicit limiting scheme, and substituting the independently characterized relaxation limit (2.8)/(3.11) to measure truncation residuals; this is exactly the standard consistency check used for AP schemes. The limits themselves come from asymptotic analyses in [32,36] and [38], whose error estimates are cited as theorems with stated assumptions, not as numerical fits, so the self-citations (notably [38] and [15]) are load-bearing but independent published results and do not reduce the claim to its inputs. The construction does not define its target in terms of the scheme, does not fit parameters to data and rename them as predictions, and does not import a uniqueness theorem from the authors' own prior work to forbid alternatives. The only genuinely missing ingredient is a uniform-in-epsilon stability estimate for the fully discrete operators (2.13)-(2.15)/(3.13), especially across the tau approximately epsilon switch in (3.12); without it, the consistency calculations in Sections 2.3 and 3.2 do not by themselves establish convergence of discrete solutions to the relaxation limit. That is an omitted convergence proof, not a circular reduction, and it does not change the circularity score.

Assumptions & free parameters 1 free parameters · 6 assumptions · 0 invented entities

The central claim rests on the structural assumptions (3.3), invertibility of A and A11, the generalized Kreiss condition, and the cited asymptotic error estimates. These are stated, not proven here. The only hand-chosen numerical parameter is p; nothing is fitted to the reported data.

free parameters (1)
  • p (exponent in η=(τ/ε)^p) = p=2 in Examples 1, 2, 3 and 5; p=4 in Example 4
    Positive integer chosen by the author to control how quickly η transitions between small and large regimes. The AP argument works for any p≥2, and the value does not change the ε→0 limit, so it is a mild scheme parameter rather than a fitted constant.
assumptions (6)
  • domain assumption Block structure (3.3): after a change of variables, A=(A11 A12; A21 A22), Q=diag(0,S) with S negative definite.
    Assumed throughout Section 3; invoked when deriving (3.6)-(3.7) and Lemma 3.3. It restricts the class of relaxation systems.
  • domain assumption A and A11 are invertible.
    Used in (3.6)-(3.7) and in Lemma 3.3; stated before (3.6) and inside Lemma 3.3.
  • domain assumption The boundary matrix B satisfies the generalized Kreiss condition (Definition 3.1).
    Needed for well-posedness of the reduced boundary conditions (3.10); quoted from [36].
  • domain assumption There is no initial layer and the boundary is non-characteristic for both the relaxation system and the equilibrium system.
    Stated in the introduction as simplifying assumptions for the whole paper.
  • standard math Error estimates from [32,36,38] justify replacing the exact solution Uε by the asymptotic solution Uε, and then by its limit U0.
    Invoked in Sections 2.1 and 3.1 to justify using U0 in the truncation-error check; not proved in this paper.
  • standard math The boundary-layer ODE (2.5)/(3.7) has a unique bounded solution on the stable subspace.
    Standard ODE theory combined with the generalized Kreiss condition; used to define the limit (2.8)/(3.11).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Asymptotic-preserving schemes for the initial-boundary value problem of hyperbolic relaxation systems." pith.science (2026). https://pith.science/paper/3WXXTHZO

@misc{pith2026250522656,
  author       = {Pith},
  title        = {Pith review of: Asymptotic-preserving schemes for the initial-boundary value problem of hyperbolic relaxation systems},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/3WXXTHZO}},
  note         = {Machine review of arXiv:2505.22656}
}
read the original abstract

In this work, we present a numerical method for the initial-boundary value problem (IBVP) of first-order hyperbolic systems with source terms. The scheme directly solves the relaxation system using a relatively coarse mesh and captures the equilibrium behavior quite well, even in the presence of boundary layers. This method extends the concept of asymptotic-preserving schemes from initial-value problems to IBVPs. Moreover, we apply this idea to design a unified numerical scheme for the interface problem of relaxation systems.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

46 extracted references · 37 canonical work pages

  1. [36]

    M. Piu, M. Herty, and G. Puppo , Derivation and stability analysis of a macroscopic multila ne model for traffic flow , SIAM J. Appl. Math., 83 (2023), pp. 2052–2072, https://doi.org/ 10.1137/22M1543288

    work page doi:10.1137/22m1543288 2023

  2. [1]

    Introduction. This paper is concerned with the numerical schemes for the initial-boundary value problem (IBVP) of the first-order hyperbolic relaxation system        Ut + A(U )Ux = 1 ǫ Q(U ), x > 0, t > 0, BU (0, t) = b(t), U (x, 0) = g(x). Here A(U ) and Q(U ) are smooth functions of unknown U = U (x, t) ∈ Rn, B is a constant matrix, b(t) and g(x) ...

    work page Pith review arXiv 2025

  3. [2]

    An illustrative example: Jin-Xin model. 2.1. Initial-boundary value problem. Consider the following initial-boundary value problem for the Jin-Xin relaxation model: { ut + vx = 0, vt + ux = (f (u) − v)/ǫ, x > 0, t > 0,(2.1) Buu(0, t) + Bvv(0, t) = b(t).(2.2) Here u = u(x, t) and v = v(x, t) are the unknowns, f (u) is a convex function, ǫ is the relaxation...

    work page

  4. [3]

    General relaxation system. 3.1. Relaxation limits for IBVPs of relaxation systems. We consider the general one dimensional linear relaxation system Ut + AUx = 1 ǫ QU,(3.1) BU (0, t) = b(t).(3.2) Here U ∈ Rn is the unknown, A, Q are n × n matrices with constant coefficients, and B is an n+ × n constant matrix with n+ the number of positive eigenvalues for A....

    work page

  5. [4]

    In this section, we show that our scheme can be also used to compute the interface problem: Ut + AUx = 1 ǫ(x) QU, x ∈ R

    Interface problem. In this section, we show that our scheme can be also used to compute the interface problem: Ut + AUx = 1 ǫ(x) QU, x ∈ R. Here A, Q are constant matrices which have the same partition as that in ( 3.3) and (4.1) ǫ(x) = 1 , x < 0, ǫ (x) = ǫ0 ≪ 1, x ≥ 0. Denote U l(x, t) = U (−x, t) for x < 0 and U r(x, t) = U (x, t) for x > 0. According t...

    work page

  6. [5]

    Example 1 (Linear Jin-Xin model): In this numerical experiment, we check the validity of the boundary AP scheme by showing the convergence rate

    Numerical experiment. Example 1 (Linear Jin-Xin model): In this numerical experiment, we check the validity of the boundary AP scheme by showing the convergence rate. We consider the Jin-Xin model ( 2.1)-(2.2) with linear function f (u) = au. Particularly, we consider three cases. Case 1: We first consider the case with a = −0.5 < 0. In this case, there is...

    work page 2000

  7. [6]

    Bal and Y

    G. Bal and Y. Maday , Coupling of transport and diffusion models in linear transpo rt theory , M2AN Math. Model. Numer. Anal., 36 (2002), pp. 69–86, https://doi.org/10.1051/m2an: 2002007

    doi:10.1051/m2an: 2002

  8. [7]

    Bensoussan, J.-L

    A. Bensoussan, J.-L. Lions, and G. C. Papanicolaou , Boundary layers and homogenization of transport processes, Publ. Res. Inst. Math. Sci., 15 (1979), pp. 53–157, https://doi.org/ 10.2977/prims/1195188427

    arXiv 1979

Show all 46 references
  1. [8]

    Benzoni-Gavage and D

    S. Benzoni-Gavage and D. Serre , Multidimensional hyperbolic partial differential equatio ns, Oxford Mathematical Monographs, The Clarendon Press, Oxfo rd University Press, Oxford,

  2. [9]

    Degond and F

    P. Degond and F. Deluzet , Asymptotic-preserving methods and multiscale models for p lasma physics, J. Comput. Phys., 336 (2017), pp. 429–457, https://doi.org/10.1016/j.jcp.2017.02. 009

    2017 doi

  3. [10]

    Borsche and A

    R. Borsche and A. Klar , Kinetic layers and coupling conditions for scalar equation s on net- works, Nonlinearity, 31 (2018), pp. 3512–3541, https://doi.org/10.1088/1361-6544/aabc91

    2018 doi

  4. [11]

    Borsche and A

    R. Borsche and A. Klar , A nonlinear discrete velocity relaxation model for traffic flo w, SIAM J. Appl. Math., 78 (2018), pp. 2891–2917, https://doi.org/10.1137/17M1152681

    2018 doi

  5. [12]

    Bourgat, P

    J.-F. Bourgat, P. Le Tallec, B. Perthame, and Y. Qiu , Coupling Boltzmann and Euler equations without overlapping , in Domain decomposition methods in science and engineer- ing (Como, 1992), vol. 157 of Contemp. Math., Amer. Math. Soc ., Providence, RI, 1994, pp. 377–398, http...

    1992 doi

  6. [13]

    R. E. Caflisch, S. Jin, and G. Russo , Uniformly accurate schemes for hyperbolic systems with relaxation , SIAM J. Numer. Anal., 34 (1997), pp. 246–281, https://doi.org/10.1137/ S0036142994268090

    1997

  7. [14]

    Z. Cai, Y. F an, and R. Li , A framework on moment model reduction for kinetic equation , SIAM J. Appl. Math., 75 (2015), pp. 2001–2023, https://doi.org/10.1137/14100110X

    2015 doi

  8. [15]

    Huang, R

    J. Huang, R. Li, and Y. Zhou , Coupling conditions for linear hyperbolic relaxation syst ems in two-scale problems , Math. Comp., 92 (2023), pp. 2133–2165, https://doi.org/10.1090/ mcom/3845

    2023

  9. [16]

    Filbet and S

    F. Filbet and S. Jin , A class of asymptotic-preserving schemes for kinetic equat ions and related problems with stiff sources , J. Comput. Phys., 229 (2010), pp. 7625–7648, https:// doi.org/10.1016/j.jcp.2010.06.017

    2010 doi

  10. [17]

    Gatignol , Th´ eorie cin´ etique des gaz ` a r´ epartition discr` ete de vi tesses, vol

    R. Gatignol , Th´ eorie cin´ etique des gaz ` a r´ epartition discr` ete de vi tesses, vol. Vol. 36 of Lecture Notes in Physics, Springer-Verlag, Berlin-New Yor k, 1975

    1975

  11. [18]

    Giovangigli and M

    V. Giovangigli and M. Massot , Asymptotic stability of equilibrium states for multicompo nent reactive flows , Math. Models Methods Appl. Sci., 8 (1998), pp. 251–297, https://doi.org/ 10.1142/S0218202598000123

    1998 doi

  12. [19]

    F. c. Golse, S. Jin, and C. D. Levermore , A domain decomposition analysis for a two-scale linear transport problem , M2AN Math. Model. Numer. Anal., 37 (2003), pp. 869–892, https://doi.org/10.1051/m2an:2003059

    2003 doi

  13. [20]

    J. Hu, S. Jin, and Q. Li , Asymptotic-preserving schemes for multiscale hyperbolic and kinetic equations, in Handbook of numerical methods for hyperbolic problems, vol. 18 of Handb. Numer. Anal., Elsevier/North-Holland, Amsterdam, 2017, p p. 103–129

    2017

  14. [21]

    D. Jou, J. Casas-V ´azquez, and G. Lebon , Extended irreversible thermodynamics , Springer, New York, fourth ed., 2010, https://doi.org/10.1007/978-90-481-3074-0

    2010 doi

  15. [22]

    Jin , Runge-Kutta methods for hyperbolic conservation laws with stiff relaxation terms , J

    S. Jin , Runge-Kutta methods for hyperbolic conservation laws with stiff relaxation terms , J. Comput. Phys., 122 (1995), pp. 51–67, https://doi.org/10.1006/jcph.1995.1196

    1995

  16. [23]

    Jin , Asymptotic preserving (AP) schemes for multiscale kinetic and hyperbolic equations: a review, Riv

    S. Jin , Asymptotic preserving (AP) schemes for multiscale kinetic and hyperbolic equations: a review, Riv. Math. Univ. Parma (N.S.), 3 (2012), pp. 177–216. 20 YIZHOU ZHOU

    2012

  17. [24]

    Jin , Asymptotic-preserving schemes for multiscale physical pr oblems, Acta Numer., 31 (2022), pp

    S. Jin , Asymptotic-preserving schemes for multiscale physical pr oblems, Acta Numer., 31 (2022), pp. 415–489, https://doi.org/10.1017/S0962492922000010, https://doi.org/10. 1017/S0962492922000010

    2022 doi

  18. [25]

    Jin, J.-G

    S. Jin, J.-G. Liu, and L. W ang , A domain decomposition method for semilinear hyperbolic systems with two-scale relaxations , Math. Comp., 82 (2013), pp. 749–779, https://doi.org/ 10.1090/S0025-5718-2012-02643-3

    2013 doi

  19. [26]

    S. Jin, L. Pareschi, and G. Toscani , Diffusive relaxation schemes for multiscale discrete- velocity kinetic equations , SIAM J. Numer. Anal., 35 (1998), pp. 2405–2439, https://doi. org/10.1137/S0036142997315962

    1998 doi

  20. [27]

    C. D. Levermore , Moment closure hierarchies for kinetic theories , J. Statist. Phys., 83 (1996), pp. 1021–1065, https://doi.org/10.1007/BF02179552

    1996 doi

  21. [28]

    Kato , A short introduction to perturbation theory for linear oper ators, Springer-Verlag, New York-Berlin, 1982

    T. Kato , A short introduction to perturbation theory for linear oper ators, Springer-Verlag, New York-Berlin, 1982

    1982

  22. [29]

    Klar , Convergence of alternating domain decomposition schemes f or kinetic and aerody- namic equations , Math

    A. Klar , Convergence of alternating domain decomposition schemes f or kinetic and aerody- namic equations , Math. Methods Appl. Sci., 18 (1995), pp. 649–670, https://doi.org/10. 1002/mma.1670180806

    1995

  23. [30]

    Klar , An asymptotic-induced scheme for nonstationary transport equations in the diffu- sive limit , SIAM J

    A. Klar , An asymptotic-induced scheme for nonstationary transport equations in the diffu- sive limit , SIAM J. Numer. Anal., 35 (1998), pp. 1073–1094, https://doi.org/10.1137/ S0036142996305558

    1998

  24. [31]

    E. W. Larsen, J. E. Morel, and W. F. Miller, Jr. , Asymptotic solutions of numerical trans- port problems in optically thick, diffusive regimes , J. Comput. Phys., 69 (1987), pp. 283– 324, https://doi.org/10.1016/0021-9991(87)90170-7

    1987 doi

  25. [32]

    Lemou and L

    M. Lemou and L. Mieussens , A new asymptotic preserving scheme based on micro-macro formulation for linear kinetic equations in the diffusion li mit, SIAM J. Sci. Comput., 31 (2008), pp. 334–368, https://doi.org/10.1137/07069479X

    2008 doi

  26. [33]

    Xin and W.-Q

    Z. Xin and W.-Q. Xu , Initial-boundary value problem to systems of conservation laws with relaxation, Quart. Appl. Math., 60 (2002), pp. 251–281, https://doi.org/10.1090/qam/ 1900493

    2002 doi

  27. [34]

    Mieussens , Discrete-velocity models and numerical schemes for the Bol tzmann-BGK equa- tion in plane and axisymmetric geometries , J

    L. Mieussens , Discrete-velocity models and numerical schemes for the Bol tzmann-BGK equa- tion in plane and axisymmetric geometries , J. Comput. Phys., 162 (2000), pp. 429–466, https://doi.org/10.1006/jcph.2000.6548

    2000

  28. [35]

    M ¨uller, A history of thermodynamics: the doctrine of energy and entr opy, Springer Science & Business Media, 2007

    I. M ¨uller, A history of thermodynamics: the doctrine of energy and entr opy, Springer Science & Business Media, 2007

    2007

  29. [37]

    W ang and Z

    W.-C. W ang and Z. Xin , Asymptotic limit of initial-boundary value problems for co nserva- tion laws with relaxational extensions , Comm. Pure Appl. Math., 51 (1998), pp. 505–535, https://doi.org/10.1002/(SICI)1097-0312(199805)51:5⟨505::AID-CPA3⟩3.0.CO;2-C

    1998 doi

  30. [38]

    Xin and W.-Q

    Z. Xin and W.-Q. Xu , Stiff well-posedness and asymptotic convergence for a class of linear relaxation systems in a quarter plane , J. Differential Equations, 167 (2000), pp. 388–437, https://doi.org/10.1006/jdeq.2000.3806

    2000

  31. [39]

    Zhou and W.-A

    Y. Zhou and W.-A. Yong , Boundary conditions for hyperbolic relaxation systems wit h char- acteristic boundaries of type II , J. Differential Equations, 310 (2022), pp. 198–234, https:// doi.org/10.1016/j.jde.2021.11.020

    2022 doi

  32. [40]

    Xu , Initial-boundary value problem for a class of linear relaxa tion systems in arbitrary space dimensions , J

    W.-Q. Xu , Initial-boundary value problem for a class of linear relaxa tion systems in arbitrary space dimensions , J. Differential Equations, 183 (2002), pp. 462–496, https://doi.org/10. 1006/jdeq.2001.4130

    2002

  33. [41]

    Xu, Boundary conditions and boundary layers for a multi-dimens ional relaxation model, J

    W.-Q. Xu, Boundary conditions and boundary layers for a multi-dimens ional relaxation model, J. Differential Equations, 197 (2004), pp. 85–117, https://doi.org/10.1016/j.jde.2003.08. 007

    2004 doi

  34. [42]

    Yong, Boundary conditions for hyperbolic systems with stiff sourc e terms , Indiana Univ

    W.-A. Yong, Boundary conditions for hyperbolic systems with stiff sourc e terms , Indiana Univ. Math. J., 48 (1999), pp. 115–137, https://doi.org/10.1512/iumj.1999.48.1611

    1999 doi

  35. [43]

    Yong , Singular perturbations of first-order hyperbolic systems w ith stiff source terms , J

    W.-A. Yong , Singular perturbations of first-order hyperbolic systems w ith stiff source terms , J. Differential Equations, 155 (1999), pp. 89–132, https://doi.org/10.1006/jdeq.1998.3584

    1999

  36. [44]

    Zhou and W.-A

    Y. Zhou and W.-A. Yong , Boundary conditions for hyperbolic relaxation systems wit h char- acteristic boundaries of type I , J. Differential Equations, 281 (2021), pp. 289–332, https:// doi.org/10.1016/j.jde.2021.02.008

    2021 doi

  37. [46]

    Y. Zhu, L. Hong, Z. Yang, and W.-A. Yong , Conservation-dissipation formalism of irre- versible thermodynamics , J. Non-Equilib. Thermodyn., 40 (2015), pp. 67–74

    2015

  38. [2007]

    First-order systems and applications

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.