Pith. sign in

REVIEW 3 major objections 4 minor 1 cited by

Unconditionally Long-Time Stable Variable-Step Second-Order ETD Schemes for the 2D Periodic Incompressible NSE

T0 review · 3 major / 4 minor · reviewed 2026-08-03 · deepseek-v4-flash

Pith's one-line read This paper proposes a second-order exponential time-differencing scheme for the periodic incompressible Navier–Stokes equations that it claims is unconditionally uniform-in-time stable: the discrete L2 energy stays bounded for all time, for

desk verdict The new scheme and experiments are worth a look, but the central 'arbitrary step size' stability theorem is not proved: (4.19) compares a left-endpoint Riemann sum to an integral in the wrong direction. read the letter →

arxiv 2602.10268 v3 pith:PGDTVODT submitted 2026-02-10 math.NA cs.NA

classification math.NAcs.NA MSC 65M1265L0465L0576D05
keywords Navier–Stokesequationsexponentialtimedifferencingmean-revertingscalarauxiliaryvariableunconditionalstabilitylong-timesteppingadaptiveuniform-in-timebound
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 aims to show that a second-order exponential time-differencing scheme, augmented with a mean-reverting scalar auxiliary variable and a dynamic second-order correction, preserves the uniform-in-time energy boundedness of the continuous Navier–Stokes equations even when the time step is changed arbitrarily. If true, this would be the first higher-order variable-step method for the incompressible NSE with a provable L∞(0,∞;L2) bound under general L2 forcing, opening the way to reliable computation of long-time statistics. The scheme is fully linear in the fluid variables, requiring two Stokes solves and one cubic root per step, and comes with an embedded adaptive first-order companion. The central stability theorem is Theorem 4.3, whose proof uses exponential decay factors from the viscous and mean-reverting terms and a purported step-size-independent Riemann-sum estimate.

What carries the argument

The mean-reverting scalar auxiliary variable r(t), driven by dr/dt + γr = −(r−1)^{k-1}⟨B(u,u),u⟩ with γ>0, is the load-bearing mechanism: it damps any deviation of r from 0, and the r+1 shift in the stability estimate converts the damping into exponential decay. The ETD filtering functions φ0(z)=e^{-z} and φ1(z)=(1-e^{-z})/z treat the Stokes operator exactly and yield the identity that turns the viscous dissipation into a contraction e^{-θτ} per step. The dynamic second-order correction, using (1-r²) in the nonlinear term, makes the scheme second-order accurate even though r itself is only first-order accurate.

What would settle it

For a partition consisting of a single step of size τ, inequality (4.19) would assert τ ≤ ∫_0^τ e^{-θx}dx = (1−e^{-θτ})/θ, which fails for every τ>0; computing these two quantities for any θ>0, τ>0 settles that the proof's key estimate is invalid as written.

Watch

Extended reading notes

Core claim

The central claim is that the ETD-mr-SAV-MS2o scheme is unconditionally long-time stable: for any variable time-step sequence, the quantity ||u^{n+1}||^2 + |r^{n+1}+1|^2 decays like e^{-θ∑τ} toward a data-dependent constant, where θ = min{νλ1, γ} and the constant involves only the forcing size and γ. This bound is independent of the Reynolds number and of all step sizes, meaning the numerical solution cannot blow up even with arbitrarily large or wildly varying time steps. The proof cancels the nonlinear advection exactly via the skew-symmetry relation ⟨B(u,u),u⟩=0, leaving only viscous dissipation and the mean-reverting damping of the auxiliary variable r.

Load-bearing premise

The theorem's step-size-free constant depends on inequality (4.19), which treats a left-endpoint Riemann sum of a decreasing exponential as if it were bounded above by the integral; that inequality is generally false, so the advertised independence from time-step sizes is not established by the given proof.

Editorial extensions

If this is right

  • If the central claim holds, the method can integrate the 2D/3D periodic NSE over arbitrarily long intervals with a guaranteed uniform L2 bound, regardless of Reynolds number or step-size choices (subject to bounded forcing).
  • Second-order accuracy under variable steps is achieved at a cost of only two Stokes solves and one scalar cubic root per time step.
  • The embedded first-order companion enables automatic step-size control with a preserved long-time stability certificate for the accepted steps.
  • For unforced flows (F=0), the bound predicts exponential decay of the discrete energy with rate θ, matching the continuous dissipation structure.
  • The framework is positioned as a foundation for approximating long-time statistical quantities and rare-event statistics in turbulent regimes without step-size restrictions.

Reading between the lines

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

  • The proof's key step-size-independent constant rests on inequality (4.19), which mislabels a left-endpoint Riemann sum as a right-endpoint one; for a single step the claimed inequality requires τ ≤ (1–e^{–θτ})/θ, which is false. A corrected bound would likely replace the constant by one depending on max τ, so the advertised 'unconditional' step-size independence may not follow from this argument,
  • The mean-reverting parameter γ is central: as γ→0 the scheme reduces to a standard SAV-ZEC formulation and the uniform bound degrades; testing γ across orders of magnitude would reveal how the long-time constant actually scales.
  • Because the cancellation of the nonlinear term uses only the skew-symmetry of B, the same proof strategy may extend to spatial discretizations that preserve ⟨B(u,v),v⟩=0, making the variable-step stability robust beyond Fourier spectral methods.
  • The adaptive controller's step acceptance/rejection rule could be supplemented by monitoring the actual one-step contraction factor e^{-θτ}, giving a computable certificate of the uniform bound during simulation; the paper does not propose such a certificate.
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

3 major / 4 minor

Summary. The paper proposes a variable-step, second-order exponential time-differencing (ETD) scheme for the periodic incompressible Navier–Stokes equations, combining a mean-reverting scalar auxiliary variable (mr-SAV) formulation with a dynamic second-order SAV correction. An embedded adaptive variant is also given. The central theoretical claim (Theorem 4.3) is an unconditional uniform-in-time L2 bound for arbitrary variable step sizes, holding for all Reynolds numbers under bounded forcing. Numerical experiments in 2D demonstrate second-order convergence, long-time boundedness, and effective adaptive error control.

Significance. The construction is attractive and, if Theorem 4.3 were proved, would constitute a notable advance: to my knowledge no earlier higher-order variable-step scheme for the NSE has a rigorous uniform-in-time energy bound under general forcing. The per-step cost is low (two Stokes solves plus one scalar cubic equation), the stability analysis is self-contained, and the numerical study is fairly thorough. However, the proof of the main theorem contains a false Riemann-sum inequality and an insufficiently sharp one-step forcing estimate; the advertised step-size-independent conclusion is therefore not established.

major comments (3)
  1. [Section 4, Eq. (4.19)] The claimed inequality is reversed. On the partition 0=S^n_{n+1}<...<S^n_1, the sum Σ_i e^{-θS^n_{i+1}}(S^n_i−S^n_{i+1}) is a left-endpoint Riemann sum for the decreasing function x↦e^{-θx}, hence it is ≥ the integral, not ≤. For n=1 it asserts τ_2≤(1−e^{-θτ_2})/θ, which is false for large θτ_2 (e.g. θ=1, τ_2=10 gives 10≤0.99995). Consequently (4.8) does not follow from (4.18); the step-independent bound in Theorem 4.3 and in the abstract is unsupported.
  2. [Section 4, Eqs. (4.17) and (4.12)] Independently of (4.19), the one-step recursion has a forcing increment linear in τ_{n+1}: X_{n+1}≤e^{-θτ_{n+1}}X_n+τ_{n+1}C. Even with a correct discrete Gronwall inequality, this recursion cannot yield a constant independent of the step sizes; for one large step the bound gives τ_2C, whereas the scheme's forcing term τφ_1(τνL)F in (3.3a) tends to (νL)^{-1}F and the exact scalar model gives (1−e^{-θτ})/θ·F. The estimate (4.12) is therefore too crude to support the advertised step-independent constant. A repair should sharpen (4.12) to a factor of (1−e^{-θτ_{n+1}}), or the theorem's claim must be weakened.
  3. [Section 5, Examples 5.2–5.3] The experiments use uniform steps τ≤0.01 or adaptive steps with τ_max=1e-2 and tolerances that keep steps small; for the reported parameters θτ is small (e.g. ν=1/40, λ_1=1, τ=0.01 gives θτ≈2.5e-4 when θ=νλ_1). The tests therefore do not exercise the large-θτ regime in which (4.19) fails and cannot be invoked as numerical confirmation of the arbitrary-step claim. A test with a single very large step (or with θτ≫1) would be informative.
minor comments (4)
  1. [Algorithm 1, Step 4] The indicator e_q=|r^{n+1}-1| appears to measure deviation from 1, while the construction drives r toward 0; Figure 5(c) plots |r|. This is inconsistent and should be corrected or clarified.
  2. [Equation (3.9b)] For the k=1 variant of (2.4b), the r-update appears to lack the factor (1−¯r^{n+1}) that would make the nonlinear-term cancellation analogous to that in Theorem 4.3; please verify the displayed formula.
  3. [Notation around (4.18)–(4.19)] The symbols S_i^n and S_n are used with overlapping meanings, and the phrase 'right-end Riemann sum' in (4.19) should be corrected to 'left-endpoint' (or the inequality reversed).
  4. [Remark 3.2] The claim that the 'explicit Adams–Bashforth ETD multistep scheme is second order; see [30]' is stated without detail; please provide a precise statement or derivation, since this is a key part of the heuristic accuracy argument.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the stability proof is self-contained and does not reduce to fitted inputs or load-bearing self-citations.

full rationale

I walked the derivation chain of Theorem 4.3. The one-step inequality (4.17) is obtained by direct energy estimates from the scheme (3.3), using only the definitions of φ0 and φ1, the Poincaré inequality, Cauchy–Schwarz, and elementary algebraic identities. Iterating (4.17) and bounding the resulting sum by (1−e^{−θS})/θ is an elementary calculation; no parameter is fitted to data, and no prior result is invoked to establish the stability bound. The paper cites the same authors' earlier mr-SAV works [47,48] and the dynamic SAV correction [49] as motivation for the formulation, but Theorem 4.3 does not rely on those citations—the stability statement is proved directly in the paper. The self-citations are therefore contextual and not load-bearing. The apparent issue in inequality (4.19) (the sum is a left-endpoint Riemann sum of a decreasing function and is not bounded by the integral, as noted in the skeptical review) is a mathematical correctness concern, not a circularity: the theorem may be unsupported or false, but it is not equivalent to its own inputs by construction. No circular step was found.

Assumptions & free parameters 2 free parameters · 4 assumptions · 1 invented entities

The scheme introduces one non-physical SAV variable and user-chosen algorithmic parameters. The central stability proof is self-contained given standard functional analysis of the periodic Stokes operator, but it rests on the false Riemann-sum inequality (4.19).

free parameters (2)
  • gamma (mean-reverting parameter) = 100, 1000 in experiments
    User-specified, not fitted; the stability proof holds for any gamma>0, but the experiments choose specific values.
  • Adaptive controller parameters (rho, tol_u, tol_q, tau_min, tau_max) = rho=0.95, tol_u=tol_q=1e-4, tau_min=1e-5, tau_max=1e-2
    Hand-chosen in the long-time experiments; not fitted to data but affect the adaptive path.
assumptions (4)
  • standard math Periodic zero-mean boundary conditions and Leray-Hopf projection with the skew-symmetry <B(u,u),u>=0.
    Invoked in Section 2.1 and used in the energy cancellation in Theorem 4.3.
  • standard math Poincare inequality and spectral representation of the Stokes operator L with eigenvalues lambda_k.
    Used in Lemma 4.2 and the energy estimate (4.15)-(4.16).
  • domain assumption The extended mr-SAV system (2.4) is a faithful reformulation of the NSE, with rigorous convergence of the modified system to NSE assumed heuristically.
    The scheme is built for (2.4), and the paper only gives a formal accuracy argument (Remark 3.2) that the SAV correction does not degrade second-order accuracy.
  • domain assumption Regularity assumption (u_i, r_i) in (H^alpha)^d x R with alpha>=3/4 to ensure B(..) in V'.
    Stated in Theorem 4.3; needed for the scheme to be well-defined, per Remark 4.1.
invented entities (1)
  • r(t) — mean-reverting scalar auxiliary variable
    purpose: Enables explicit treatment of the nonlinear advection term while maintaining long-time stability via mean reversion.
    r(t) has no physical counterpart; its behavior is controlled by the scheme. The paper's own experiments show it stays small for small steps, but no falsifiable physical prediction is attached.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Unconditionally Long-Time Stable Variable-Step Second-Order ETD Schemes for the 2D Periodic Incompressible NSE." pith.science (2026). https://pith.science/paper/PGDTVODT

@misc{pith2026260210268,
  author       = {Pith},
  title        = {Pith review of: Unconditionally Long-Time Stable Variable-Step Second-Order ETD Schemes for the 2D Periodic Incompressible NSE},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/PGDTVODT}},
  note         = {Machine review of arXiv:2602.10268}
}
read the original abstract

We develop an efficient, unconditionally stable, variable step second order exponential time differencing scheme for the incompressible Navier Stokes equations in two and three spatial dimensions under periodic boundary conditions, together with an embedded adaptive time stepping variant. The scheme is unconditionally uniform in time stable in the sense that the numerical solution admits a time uniform bound in Linfinity over time with values in L2 to the power d whenever the external forcing term is uniformly bounded in time in L2, for all Reynolds numbers and for arbitrary choices of time step sizes. At each time step, the method requires the solution of two time dependent Stokes problems, which can be evaluated explicitly in the periodic setting using Fourier techniques, along with the solution of a single scalar cubic algebraic equation. Beyond the standard exponential time differencing framework, the proposed scheme incorporates two recently developed ingredients. The first is a dynamic second order scalar auxiliary variable correction, which is essential for achieving second order temporal accuracy. The second is a mean reverting scalar auxiliary variable multistep formulation, which plays a central role in ensuring long time stability. The proposed methods overcome key limitations of existing approaches for the Navier Stokes equations. Classical Runge Kutta schemes generally lack provable long time stability, while IMEX and scalar auxiliary variable based BDF methods typically do not admit unconditional stability guarantees in the variable step setting. Numerical experiments in two spatial dimensions confirm second order temporal accuracy, uniform long time stability, and effective error control provided by the adaptive strategy. Rigorous convergence analysis and a systematic investigation of long time statistical properties will be pursued in future work.

Figures

Figures reproduced from arXiv: 2602.10268 by the authors.

Figure 1
Figure 1. The L2 error of velocity (left), vorticity (middle), and absolute error of the auxiliary variable (right), computed by ETD-mr-SAV-MS1o and ETD-mr-SAV-MS2o at T = 1, plotted against the time￾step size τ = 0.1 × 2−k, for k = 0, 1, · · · , 6. Next, we examine accuracy under variable time stepping. Following [58], we generate a variable-step sequence {τn} by applying a 10% perturbation to the uniform step size 1/N. Erro… view at source ↗
Figure 2
Figure 2. The figure indicates that the proposed schemes remain robust under variable [PITH_FULL_IMAGE:figures/full_fig_p019_2.png] view at source ↗
Figure 2
Figure 2. The L2 error of velocity (left), vorticity (middle), and absolute error of the auxiliary variable (right), computed by ETD-mr-SAV-MS1o and ETD-mr-SAV-MS2o at T = 1, plotted against the number of time steps N = 2k under a 10% perturbed variable-step sequence, for k = 5, . . . , 10. periodic NSE on (0, 2π) 2 in vorticity–streamfunction form (1.2), with external forcing f(x, y;t) = m cos(my), (5.1) following [60]. The … view at source ↗
Figures from the paper (5 more)
Figure 3
Figure 3. Figure 3: shows the evolution of the L 2 norm of vorticity (enstrophy) over the entire interval. We observe that the enstrophy remains uniformly bounded for both fixed-step and adaptive simulations, consistent with the long-time stability theory established in Section 4. For the…
Figure 4
Figure 4. Figure 4: Enstrophy (L2 norm of vorticity) versus time computed by ETD-mr-SAV-MS2o and ETD-mr￾SAV-MS12 for ν = 1/40 and m = 4. The red dotted line denotes the mean value µ; the orange dashed and green dash-dotted lines correspond to the thresholds µ + 2σ and µ + 3σ, respectively…
Figure 5
Figure 5. Figure 5: (a) Adaptive time step sizes as a function of time. (b) Number of steps over time for constant [PITH_FULL_IMAGE:figures/full_fig_p023_5.png]
Figure 6
Figure 6. Figure 6: (a) Adaptive time step sizes as a function of time. (b) Enstrophy ( [PITH_FULL_IMAGE:figures/full_fig_p025_6.png]
Figure 7
Figure 7. Figure 7: Probability density functions of enstrophy obtained by ETD-mr-SAV-MS2o and ETD-mr [PITH_FULL_IMAGE:figures/full_fig_p026_7.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. A Linear Variable-Step Embedded ETD Scheme with Uniform-in-Time Stability for the 2D Navier--Stokes Equations

    math.NA 2026-07 reject novelty 6.0 of 10

    A fully linear ETD-mr-SAV scheme for the 2D NSE is proposed with an advertised uniform-in-time stability theorem, but the theorem's key energy identity is algebraically wrong, so the main claim is unproven.

Reference graph

Works this paper leans on

60 extracted references · 19 canonical work pages · cited by 1 Pith paper

  1. [1]

    Frisch, A

    U. Frisch, A. N. Kolmogorov, Turbulence: the legacy of A. N. Kolmogorov, Cam- bridge University Press, 1995

  2. [2]

    Constantin, C

    P. Constantin, C. Foiaş, Navier-Stokes equations, University of Chicago Press, 1988

  3. [3]

    Foias, O

    C. Foias, O. Manley, R. Rosa, R. Temam, Navier-Stokes equations and turbulence, Vol. 83, Cambridge University Press, 2001

  4. [4]

    A.Majda, X.Wang, Nonlineardynamicsandstatisticaltheoriesforbasicgeophysical flows, Cambridge University Press, 2006. 28

  5. [5]

    Temam, Infinite-dimensional dynamical systems in mechanics and physics, Vol

    R. Temam, Infinite-dimensional dynamical systems in mechanics and physics, Vol. 68, Springer-Verlag, New York, 1988

  6. [6]

    J. G. Heywood, R. Rannacher, Finite-element approximation of the nonstationary Navier–Stokes problem. Part IV: error analysis for second-order time discretization, SIAMJournalonNumericalAnalysis27(2)(1990)353–384.doi:10.1137/0727022

  7. [7]

    Lubich, A

    C. Lubich, A. Ostermann, Runge-Kutta time discretization of reaction-diffusion and Navier-Stokes equations: nonsmooth-data error estimates and applications to long- time behaviour, Applied Numerical Mathematics 22 (1-3) (1996) 279–292.doi: 10.1016/S0168-9274(96)00038-4

  8. [8]

    F. Tone, D. Wirosoetisno, On the long-time stability of the implicit Euler scheme for the two-dimensional Navier–Stokes equations, SIAM Journal on Numerical Analysis 44 (1) (2006) 29–40.doi:10.1137/040618527

Show all 60 references
  1. [9]

    F. Tone, On the long-time stability of the Crank–Nicolson scheme for the 2D Navier– Stokes equations, Numerical Methods for Partial Differential Equations: An Inter- national Journal 23 (5) (2007) 1235–1248.doi:10.1002/num.20219

  2. [10]

    Geveci, On the convergence of a time discretization scheme for the Navier- Stokes equations, Mathematics of Computation 53 (187) (1989) 43–53.doi: 10.1090/S0025-5718-1989-0969488-5

    T. Geveci, On the convergence of a time discretization scheme for the Navier- Stokes equations, Mathematics of Computation 53 (187) (1989) 43–53.doi: 10.1090/S0025-5718-1989-0969488-5

  3. [11]

    J. Simo, F. Armero, Unconditional stability and long-term behavior of transient algorithms for the incompressible Navier-Stokes and Euler equations, Computer Methods in Applied Mechanics and Engineering 111 (1-2) (1994) 111–154.doi: 10.1016/0045-7825(94)90042-6

  4. [12]

    N. Ju, On the global stability of a temporal discretization scheme for the Navier– Stokes equations, IMA Journal of Numerical Analysis 22 (4) (2002) 577–597.doi: 10.1093/imanum/22.4.577

  5. [13]

    Heister, M

    T. Heister, M. A. Olshanskii, L. G. Rebholz, Unconditional long-time stability of a velocity–vorticity method for the 2D Navier–Stokes equations, Numerische Mathe- matik 135 (1) (2017) 143–167.doi:10.1007/s00211-016-0794-1. 29

  6. [14]

    A. T. Hill, E. Süli, Approximation of the global attractor for the incompressible Navier–Stokes equations, IMA Journal of Numerical Analysis 20 (4) (2000) 633– 667.doi:10.1093/imanum/20.4.633

  7. [15]

    Rebholz, F

    L. Rebholz, F. Tone, Long-timeH1-stability of BDF2 time stepping for 2D Navier– Stokes equations, Applied Mathematics Letters 141 (2023) 108624.doi:10.1016/ j.aml.2023.108624

  8. [16]

    Contri, B

    A. Contri, B. Kovács, A. Massing, Error analysis of BDF 1–6 time-stepping methods for the transient Stokes problem: velocity and pressure estimates, SIAM Journal on Numerical Analysis 63 (4) (2025) 1586–1616.doi:10.1137/23M1606800

  9. [17]

    García-Archilla, V

    B. García-Archilla, V. John, J. Novo, Error analysis of BDF schemes for the evo- lutionary incompressible Navier–Stokes equations, arXiv preprint arXiv:2506.16917 (2025)

  10. [18]

    Gottlieb, F

    S. Gottlieb, F. Tone, C. Wang, X. Wang, D. Wirosoetisno, Long time stability of a classical efficient scheme for two-dimensional Navier–Stokes equations, SIAM Journal on Numerical Analysis 50 (1) (2012) 126–150.doi:10.1137/110834901

  11. [19]

    X. Wang, An efficient second order in time scheme for approximating long time statistical properties of the two dimensional Navier–Stokes equations, Numerische Mathematik 121 (4) (2012) 753–779.doi:10.1007/s00211-012-0450-3

  12. [20]

    Cheng, C

    K. Cheng, C. Wang, Long time stability of high order multistep numerical schemes for two-dimensional incompressible Navier–Stokes equations, SIAM Journal on Nu- merical Analysis 54 (5) (2016) 3123–3144.doi:10.1137/16M1061588

  13. [21]

    Du, W.-x

    Q. Du, W.-x. Zhu, Stability analysis and application of the exponential time differ- encing schemes, Journal of Computational Mathematics (2004) 200–209

  14. [22]

    Q. Du, W. Zhu, Analysis and applications of the exponential time differencing schemes and their contour integration modifications, BIT Numerical Mathematics 45 (2) (2005) 307–328. 30

  15. [23]

    L. Ju, X. Li, Z. Qiao, H. Zhang, Energy stability and error estimates of exponential time differencing schemes for the epitaxial growth model without slope selection, Mathematics of Computation 87 (312) (2018) 1859–1885

  16. [24]

    Q. Du, L. Ju, X. Li, Z. Qiao, Maximum bound principles for a class of semilinear parabolic equations and exponential time-differencing schemes, SIAM Review 63 (2) (2021) 317–359

  17. [25]

    H.Wang, J.Sun, H.Zhang, X.Qian, Anoveluptofourth-orderequilibria-preserving and energy-stable exponential Runge-Kutta framework for gradient flows, CSIAM Transactions on Applied Mathematics 6 (1) (2025) 106–147

  18. [26]

    W. Chen, S. Wang, X. Wang, Energy Stable Arbitrary Order ETD-MS Method for Gradient Flows with Lipschitz Nonlinearity, CSIAM Transactions on Applied Mathematics 2 (3) (2021) 460–483

  19. [27]

    Z. Fu, J. Yang, Energy-decreasing exponential time differencing Runge–Kutta meth- ods for phase-field models, Journal of Computational Physics 454 (2022) 110943

  20. [28]

    Z.Fu, J.Shen, J.Yang, Higher-orderenergy-decreasingexponentialtimedifferencing Runge-Kutta methods for gradient flows, Science China Mathematics 68 (7) (2025) 1727–1746

  21. [29]

    W. Chen, Z. F. Fu, S. Wang, X. Wang, Convergence of long-time stable variable- step arbitrary order ETD-MS scheme for gradient flows with Lipschitz nonlinearity, arxiv preprint arXiv.2512.01601 (2025).doi:https://doi.org/10.48550/arXiv. 2512.01601

  22. [30]

    L. Zhang, Un schéma de semi-discrétisation en temps pour des systèmes différentiels discrétisés en espace par la méthode de fourier: résolution numérique des équations de navier-stokes stationnaires par la méthode multigrille, Ph.D. thesis, Paris 11 (1987)

  23. [31]

    R. S. Rogallo, An ILLIAC program for the numerical simulation of homogeneous incompressible turbulence, Tech. rep., NASA TM-73203 (1977). 31

  24. [32]

    G. L. Kooij, M. A. Botchev, B. J. Geurts, An exponential time integrator for the in- compressible Navier–Stokes equation, SIAM Journal on Scientific Computing 40 (3) (2018) B684–B705.doi:10.1137/17M1121950

  25. [33]

    S. Li, Z. J. Wang, L. Ju, L. Luo, Fast time integration of Navier-Stokes equations with an exponential-integrator scheme, in: 2018 AIAA Aerospace Sciences Meeting, 2018, p. 0369.doi:10.2514/6.2018-0369

  26. [34]

    S. Li, L. Ju, H. Si, Adaptive exponential time integration of the Navier-Stokes equa- tions, in: AIAA Scitech 2020 Forum, 2020, p. 2033.doi:10.2514/6.2020-2033

  27. [35]

    J. Shen, J. Xu, J. Yang, The scalar auxiliary variable (SAV) approach for gradient flows, Journal of Computational Physics 353 (2018) 407–416.doi:10.1016/j.jcp. 2017.10.021

  28. [36]

    J. Shen, J. Xu, J. Yang, A new class of efficient and robust energy stable schemes for gradient flows, SIAM Review 61 (3) (2019) 474–506.doi:10.1137/17M1150153

  29. [37]

    Akrivis, B

    G. Akrivis, B. Li, D. Li, Energy-decaying extrapolated RK–SAV methods for the Allen–Cahn and Cahn–Hilliard equations, SIAM Journal on Scientific Computing 41 (6) (2019) A3703–A3727.doi:10.1137/19M1264412

  30. [38]

    L. Lin, Z. Yang, S. Dong, Numerical approximation of incompressible Navier-Stokes equations based on an auxiliary energy variable, Journal of Computational Physics 388 (2019) 1–22.doi:10.1016/j.jcp.2019.03.012

  31. [39]

    L. Lin, N. Ni, Z. Yang, S. Dong, An energy-stable scheme for incompressible Navier- Stokes equations with periodically updated coefficient matrix, Journal of Computa- tional Physics 418 (2020) 109624

  32. [40]

    Huang, J

    F. Huang, J. Shen, Stability and error analysis of a second-order consistent splitting scheme for the Navier–Stokes equations, SIAM Journal on Numerical Analysis 61 (5) (2023) 2408–2433.doi:10.1137/23M1556022

  33. [41]

    Huang, J

    F. Huang, J. Shen, Stability and error analysis of a new class of higher-order consis- tentsplittingschemesfortheNavier–Stokesequations, MathematicsofComputation (2025).doi:10.1090/mcom/4132. 32

  34. [42]

    Y. Di, Y. Ma, J. Shen, J. Zhang, A variable time-step IMEX-BDF2 SAV scheme and its sharp error estimate for the Navier–Stokes equations, ESAIM: Mathematical Modelling and Numerical Analysis 57 (3) (2023) 1143–1170

  35. [43]

    B. Ji, H. Liao, A unifiedL2 norm error analysis of SAV-BDF schemes for the incom- pressible Navier–Stokes equations, Journal of Scientific Computing 100 (1) (2024) 5

  36. [44]

    S. Zeng, Z. Xie, X. Yang, J. Wang, Fully discrete, decoupled and energy-stable Fourier-Spectral numerical scheme for the nonlocal Cahn–Hilliard equation coupled with Navier–Stokes/Darcy flow regime of two-phase incompressible flows, Computer Methods in Applied Mechanics and En...

  37. [45]

    G. D. Zhang, X. He, X. Yang, A unified framework of the SAV-ZEC method for a mass-conserved Allen–Cahn type two-phase ferrofluid flow model, SIAM Journal on Scientific Computing 46 (2) (2024) B77–B106.doi:10.1137/23M1569125

  38. [46]

    Zhang, S

    G.-D. Zhang, S. Zhou, Y. Huang, X. He, X. Yang, A diffuse interface model and fully decoupled, energy-stable scheme for the two-phase ferrofluid flows in porous media, Journal of Computational Physics (2025) 114561

  39. [47]

    D. Han, X. Wang, A highly efficient second-order accurate long-time dynamics pre- serving scheme for some geophysical fluid models, arXiv preprint arXiv:2510.05360 (2025)

  40. [48]

    Coleman, D

    J. Coleman, D. Han, X. Wang, An efficient scheme for approximating long-time dy- namics of a class of non-linear models, Communications in Information and Systems 25 (2) (2025) 273–299

  41. [49]

    D. Hou, Z. Qiao, An implicit–explicit second-order BDF numerical scheme with variable steps for gradient flows, Journal of Scientific Computing 94 (2) (2023) 39

  42. [50]

    Huang, J

    F. Huang, J. Shen, A new class of implicit–explicit BDFk SAV schemes for general dissipative systems and their error analysis, Computer Methods in Applied Mechan- ics and Engineering 392 (2022) 114718.doi:10.1016/j.cma.2022.114718. 33

  43. [51]

    X. Yang, A new efficient fully-decoupled and second-order time-accurate scheme for Cahn–Hilliard phase-field model of three-phase incompressible flow, Computer Methods in Applied Mechanics and Engineering 376 (2021) 113589.doi:10.1016/ j.cma.2020.113589

  44. [52]

    X. Yang, A novel fully-decoupled, second-order and energy stable numerical scheme of the conserved Allen–Cahn type flow-coupled binary surfactant model, Computer Methods in Applied Mechanics and Engineering 373 (2021) 113502.doi:10.1016/ j.cma.2020.113502

  45. [53]

    X. Li, J. Shen, Z. Liu, New SAV-pressure correction methods for the Navier-Stokes equations: stability and error analysis, Mathematics of Computation 91 (333) (2022) 141–167.doi:10.1090/mcom/3651

  46. [54]

    E. Fehlberg, Low-order classical Runge–Kutta formulas with stepsize control and their application to some heat transfer problems, NASA Technical Report TR R- 315, NASA, nASA TR R-315 (1969). URLhttps://ntrs.nasa.gov/citations/19690021375

  47. [55]

    J. R. Dormand, P. J. Prince, A family of embedded Runge–Kutta formulae, Journal of Computational and Applied Mathematics 6 (1) (1980) 19–26.doi:10.1016/ 0771-050X(80)90013-3. URLhttps://doi.org/10.1016/0771-050X(80)90013-3

  48. [56]

    DeCaria, M

    V. DeCaria, M. Schneier, An embedded variable step IMEX scheme for the incom- pressible Navier–Stokes equations, Computer Methods in Applied Mechanics and Engineering 376 (2021) 113661.doi:10.1016/j.cma.2020.113661. URLhttps://doi.org/10.1016/j.cma.2020.113661

  49. [57]

    Hairer, G

    E. Hairer, G. Wanner, S. P. Nørsett, Solving ordinary differential equations I: Non- stiff problems, Springer, 1993

  50. [58]

    W. Chen, X. Wang, Y. Yan, Z. Zhang, A second order BDF numerical scheme with variable steps for the Cahn–Hilliard equation, SIAM Journal on Numerical Analysis 57 (1) (2019) 495–525.doi:10.1137/18M1206084. 34

  51. [59]

    A. K. Kassam, L. N. Trefethen, Fourth-order time-stepping for stiff PDEs, SIAM Journal on Scientific Computing 26 (4) (2005) 1214–1233.doi:10.1137/ S1064827502410633

  52. [60]

    Armbruster, B

    D. Armbruster, B. Nicolaenko, N. Smaoui, P. Chossat, Symmetries and dynamics for 2-D Navier-Stokes flow, Physica D: Nonlinear Phenomena 95 (1) (1996) 81–93. doi:10.1016/0167-2789(96)00006-1. 35

Pith tools

Reviewed August 3, 2026 · model on record in the stance chip above.