arxiv: 2604.10616 · v3 · submitted 2026-04-12 · 🧮 math.AP · cs.NA· math.NA

Local Well-Posedness of a Modified NSCH-Oldroyd System: PINN-Based Numerical Computation

Woojeong Kim This is my paper

Pith reviewed 2026-05-12 01:45 UTC · model grok-4.3

classification 🧮 math.AP cs.NAmath.NA

keywords Navier-Stokes-Cahn-HilliardOldroyd systemlocal well-posednessthrombus modelingphysics-informed neural networksenergy decayMetropolis-Hastings sampling

0 comments p. Extension

The pith

Added diffusion to the deformation variable makes a modified NSCH-Oldroyd system locally well-posed while preserving its energy dissipation.

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

The paper modifies the Navier-Stokes-Cahn-Hilliard-Oldroyd system for thrombus modeling by adding a diffusion term to the deformation variable. This change is constructed to keep the original dissipative energy structure intact. The central result is a proof that the modified system is locally well-posed, so short-time solutions exist and remain unique. PINN-based computations then illustrate the system on representative thrombus cases, with residual losses and benchmark errors evaluated through Metropolis-Hastings sampling driven by the energy decay. The approach therefore supplies both an analytical foundation and a practical numerical tool for the modified model.

Core claim

By introducing an additional diffusion term in the deformation variable while preserving the dissipative energy structure, the modified NSCH-Oldroyd system admits local well-posedness. Unique local-in-time solutions therefore exist. The same energy structure supports PINN numerical illustrations for thrombus cases, where Metropolis-Hastings sampling based on energy decay produces low residual losses and benchmark errors.

What carries the argument

The diffusion-enhanced equation for the deformation variable that preserves the dissipative energy structure of the original system

If this is right

Unique local-in-time solutions exist for the modified system.
The energy dissipation property remains available to guide numerical stability.
PINN approximations for thrombus cases can be validated against energy-decay benchmarks.
The physical interpretation for thrombus modeling is unchanged by the modification.

Where Pith is reading between the lines

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

Similar diffusion enhancements could regularize other viscoelastic or phase-field fluid models while retaining energy structure.
Hybrid analytical-numerical pipelines pairing local well-posedness with PINN solvers may extend to more complex biological flow geometries.
Tuning the diffusion coefficient could be tested for optimal numerical performance without shifting the physical regime.

Load-bearing premise

The added diffusion term for the deformation variable preserves the associated dissipative energy structure of the original system without altering its physical meaning for thrombus modeling.

What would settle it

A finite-time blow-up solution or an observed violation of the energy dissipation inequality in the modified system would falsify the local well-posedness claim.

Figures

Figures reproduced from arXiv: 2604.10616 by Woojeong Kim.

**Figure 1.** Figure 1: On time domain [0, .2], phase field variable ϕ (=0 on clot and =1 on blood) on (a)original thrombus system with λe = 1 on thrombus and λe = 0 on blood in [6], (b)modified new model with generalized viscoelasticity ν(ϕ) = (1 − ϕ) + 10−5ϕ. (a) displays mixed and ambiguous phase field shape change, not describing interplay of dynamics on shock region. However, (b) shows more stable formation for the interface… view at source ↗

**Figure 2.** Figure 2: Change of tr(F F −I) (a)before and (b)after adding the diffusive F term in new model. We observe more stable behavior of F near the interfacial region. Since diffusion slows the variation of F toward the equilibrium state F = I, the solver becomes more stable, as shown in [PITH_FULL_IMAGE:figures/full_fig_p043_2.png] view at source ↗

**Figure 4.** Figure 4: (a) Before applying AA-PINN method and (b) after applying this on same experiment setting as [PITH_FULL_IMAGE:figures/full_fig_p048_4.png] view at source ↗

**Figure 5.** Figure 5: Axial section figures of evolution for each unit train time interval ∆unitt = 0.05(a) and ∆unitt = 1.0(b). These have parameter as in the Case A in [PITH_FULL_IMAGE:figures/full_fig_p050_5.png] view at source ↗

**Figure 6.** Figure 6: Mass conservation error and divergence free condition error for each of unit train time intervals 0.5(a) and 1.0(b). The parameter setting comes from the Case A in [PITH_FULL_IMAGE:figures/full_fig_p051_6.png] view at source ↗

**Figure 7.** Figure 7: Energy dissipation for each of unit train time intervals 0.05(a) and 0.1(b). The parameter setting comes from the Case A in [PITH_FULL_IMAGE:figures/full_fig_p051_7.png] view at source ↗

**Figure 8.** Figure 8: Axial section figures of evolutionas (a) and (b). These (a) and (b) have parameter as in the Case B and Case B′ in [PITH_FULL_IMAGE:figures/full_fig_p052_8.png] view at source ↗

**Figure 9.** Figure 9: Case C ′ in (a) and (b) and Case C in (c) and (d) in [PITH_FULL_IMAGE:figures/full_fig_p053_9.png] view at source ↗

**Figure 10.** Figure 10: Case C ′ in (a) and (b) of [PITH_FULL_IMAGE:figures/full_fig_p054_10.png] view at source ↗

**Figure 11.** Figure 11: Axial section figures of Case D in [PITH_FULL_IMAGE:figures/full_fig_p054_11.png] view at source ↗

**Figure 12.** Figure 12: h = 0.05 of Case D′ in [PITH_FULL_IMAGE:figures/full_fig_p055_12.png] view at source ↗

**Figure 13.** Figure 13: Energy dissipation figures. These have parameter as in the Case A in [PITH_FULL_IMAGE:figures/full_fig_p056_13.png] view at source ↗

**Figure 14.** Figure 14: Energy dissipation figures. These have parameter as in the Case D′ in [PITH_FULL_IMAGE:figures/full_fig_p056_14.png] view at source ↗

read the original abstract

Motivated by thrombus modeling, we study a modified Navier-Stokes-Cahn-Hilliard system and consider PINN-based numerical illustrations for the modified system. To enable the analysis, we introduce a diffusion-enhanced system for the deformation variable while preserving the associated dissipative energy structure. We prove local well-posedness for this new system. We also present PINN-based numerical illustrations for representative thrombus cases and report residual losses and benchmark errors obtained with Metropolis-Hastings sampling based on the energy decay.

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.

Desk Editor's Note private letter to a colleague

The paper adds diffusion to the deformation variable to get local well-posedness for a modified NSCH-Oldroyd system and shows PINN runs for thrombus cases, but the details stay thin.

read the letter

The main new thing is the diffusion term added to the deformation variable so that local well-posedness can be proved while the dissipative energy structure is kept. The authors treat this as a technical fix that does not change the underlying thrombus model much, then run PINN simulations on representative cases and report residual losses plus benchmark errors from Metropolis-Hastings sampling driven by the energy decay. That combination is the concrete output. The well-posedness claim rests on standard energy methods once the extra diffusion is in place, and the numerics are at least tied back to the same energy law instead of floating free. Both pieces are checkable in principle. The soft spots are the usual ones for this style of work. The abstract gives no function spaces, no precise a priori estimates, and no indication of how the diffusion coefficient is chosen or whether it affects long-time behavior. On the numerical side, the sampling is reasonable but the error analysis looks light; residual losses alone do not tell you much about approximation quality or sensitivity to the Metropolis-Hastings parameters. The physical justification for the added term is also brief. Readers who already work with NSCH-Oldroyd or similar phase-field fluid models will see the incremental step clearly and can judge the estimates themselves. The paper is narrow enough that it will not move the broader field, but the claims are specific and the methods are reproducible enough to warrant referee time. I would send it out for review rather than desk-reject.

Referee Report

1 major / 2 minor

Summary. The manuscript introduces a diffusion-enhanced modification to the Navier-Stokes-Cahn-Hilliard-Oldroyd system motivated by thrombus modeling. By adding a diffusion term to the deformation variable while preserving the dissipative energy structure, the authors prove local well-posedness. They further present PINN-based numerical illustrations for representative thrombus cases, employing Metropolis-Hastings sampling guided by the energy decay law and reporting residual losses together with benchmark errors.

Significance. If the local well-posedness result holds via the preserved energy law and the PINN computations accurately reproduce the expected decay, the work supplies both an analytically tractable regularization and a physics-informed numerical tool for a biologically relevant non-Newtonian fluid model. The energy-structure preservation is a clear technical strength that permits standard PDE techniques, and the Metropolis-Hastings sampling tied to the energy functional is a constructive validation feature.

major comments (1)

[modified system definition and energy estimates] The central well-posedness argument relies on the claim that the added diffusion term leaves the dissipative energy structure intact (abstract and the modified-system section). A detailed verification that the new term does not introduce uncontrolled growth in the energy estimates or alter the physical interpretation for thrombus dynamics would strengthen the result; without an explicit computation of the modified energy law, it is difficult to confirm the estimates close.

minor comments (2)

[abstract and numerical-results section] The abstract states that residual losses and benchmark errors are reported, yet the precise norms, sampling parameters, and comparison baselines (e.g., against finite-element or spectral methods) are not summarized; adding a short table or sentence would improve readability.
[preliminaries] Notation for the deformation variable and the precise function spaces in which local well-posedness is obtained should be stated explicitly at the outset rather than introduced only in the proof.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the positive evaluation and the constructive comment on the energy estimates. We will incorporate an explicit verification of the modified energy law to strengthen the well-posedness argument.

read point-by-point responses

Referee: The central well-posedness argument relies on the claim that the added diffusion term leaves the dissipative energy structure intact (abstract and the modified-system section). A detailed verification that the new term does not introduce uncontrolled growth in the energy estimates or alter the physical interpretation for thrombus dynamics would strengthen the result; without an explicit computation of the modified energy law, it is difficult to confirm the estimates close.

Authors: We agree that an explicit computation is needed to confirm the estimates. In the revised manuscript we will add a dedicated subsection deriving the energy balance for the modified system. The calculation shows that the new diffusion term on the deformation variable produces an additional non-positive contribution to the dissipation rate, so the a priori estimates close exactly as in the original structure and no uncontrolled growth appears. On the physical side, the term is introduced as a regularization that improves numerical stability for thrombus simulations without changing the viscoelastic constitutive law; we will add a short paragraph in the model section clarifying this interpretation. revision: yes

Circularity Check

0 steps flagged

No significant circularity identified

full rationale

The paper's core claim is local well-posedness for a deliberately modified NSCH-Oldroyd system obtained by adding a diffusion term to the deformation variable. This modification is introduced explicitly to keep the dissipative energy structure intact, after which standard PDE estimates are applied; the well-posedness result therefore does not reduce to the modification itself by definition. The subsequent PINN computations report residuals and benchmark errors against the model's built-in energy decay, but these quantities serve as numerical validation rather than as a 'prediction' that is forced by the fitting procedure. No self-citations, uniqueness theorems, or ansatzes are invoked in a load-bearing way, and the derivation chain remains independent of its own inputs.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central claim rests on standard mathematical assumptions for PDE local existence and the preservation of the dissipative energy structure after modification; no new free parameters or invented entities are introduced in the abstract.

axioms (2)

standard math Appropriate Sobolev or Banach function spaces and initial data regularity suffice for local existence and uniqueness in the modified system.
Invoked implicitly to establish local well-posedness after adding the diffusion term.
domain assumption The added diffusion term preserves the dissipative energy structure of the original NSCH-Oldroyd system.
Stated as the key property that enables both the analysis and the numerical validation via energy decay.

pith-pipeline@v0.9.0 · 5374 in / 1419 out tokens · 50221 ms · 2026-05-12T01:45:04.819350+00:00 · methodology

Review history (2 revisions) →

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel (J-cost uniqueness) echoes

?

echoes
ECHOES: this paper passage has the same mathematical shape or conceptual pattern as the Recognition theorem, but is not a direct formal dependency.

We modify the system in [6] by adding a small diffusion term to the deformation equation while preserving the underlying physical and analytical stability of the model... the energy-dissipation identity holds: (2.37)–(2.39)
IndisputableMonolith/Foundation/RealityFromDistinction.lean reality_from_one_distinction unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

Theorem 1.1 (local well-posedness in D(A)×H³×H² for d=2,3)

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

33 extracted references · 33 canonical work pages · 1 internal anchor

[1]

Azouani, E

A. Azouani, E. Olson, and E. S. Titi, Continuous data assimilation using general interpolant observables,Journal of Nonlinear Science24(2) (2014), 277–304

work page 2014
[2]

Foias, O

C. Foias, O. P. Manley, R. Temam, and Y. M. Tr` eve, Asymptotic analysis of the Navier–Stokes equations,Physica D: Nonlinear Phenomena9(1–2) (1983), 157–188

work page 1983
[3]

Ern and J.-L

A. Ern and J.-L. Guermond,Theory and Practice of Finite Elements, Applied Mathematical Sciences, Vol. 159, Springer, New York, 2004

work page 2004
[4]

Marschall

J. Marschall. The trace of Sobolev-Slobodeckij spaces on Lipschitz domains.Manuscripta Mathematica, 58(1- 2):47–65, 1987

work page 1987
[5]

Brezis and P

H. Brezis and P. Mironescu. Gagliardo–Nirenberg inequalities and non-inequalities: The full story.Annales de l’Institut Henri Poincar´ e (C) Analyse Non Lin´ eaire, 35(5):1355–1376, 2018. 56

work page 2018
[6]

W. Kim, K. Tawri, and R. Temam. Local well-posedness of a three-dimensional phase-field model for thrombus and blood flow.Rev. Real Acad. Cienc. Exactas F´ ıs. Nat. Ser. A Mat., 116(4):149, 2022

work page 2022
[7]

Cattabriga

L. Cattabriga. Su un problema al contorno relativo al sistema di equazioni di Stokes.Rendiconti del Seminario Matematico della Universit` a di Padova, 31:308–340, 1961

work page 1961
[8]

L. E. Payne and H. F. Weinberger. An optimal Poincar´ e inequality for convex domains.Archive for Rational Mechanics and Analysis, 5(1):286–292, 1960

work page 1960
[9]

Agmon, A

S. Agmon, A. Douglis, and L. Nirenberg. Estimates near the boundary for solutions of elliptic partial differential equations satisfying general boundary conditions.Comm. Pure Appl. Math., 12:623–727, 1959

work page 1959
[10]

D. S. Grebenkov and B.-T. Nguyen, Geometrical structure of Laplacian eigenfunctions,SIAM Review, vol. 55, no. 4, pp. 601–667, 2013

work page 2013
[12]

S. Xu, Z. Xu, O. V. Kim, R. I. Litvinov, J. W. Weisel, and M. Alber. Model predictions of deformation, embolization and permeability of partially obstructive blood clots under variable shear flow.J. R. Soc. Interface, 14(136):20170441, 2017

work page 2017
[13]

R. Qiu, R. Huang, Y. Xiao, J. Wang, Z. Zhang, J. Yue, Z. Zeng, and Y. Wang. Physics-informed neural networks for phase-field method in two-phase flow.Phys. Fluids, 34(5):052109, 2022

work page 2022
[14]

C. L. Wight and J. Zhao. Solving Allen-Cahn and Cahn-Hilliard Equations using the Adaptive Physics Informed Neural Networks.Commun. Comput. Phys., 29(3):930–954, 2021

work page 2021
[15]

L. D. McClenny and U. Braga-Neto. Self-Adaptive Physics-Informed Neural Networks using a Soft Attention Mechanism.arXiv preprint arXiv:2009.04544, 2020

work page arXiv 2009
[16]

Penwarden, A

M. Penwarden, A. D. Jagtap, S. Zhe, G. E. Karniadakis, and R. M. Kirby. A unified scalable framework for causal sweeping strategies for Physics-Informed Neural Networks (PINNs) and their temporal decompositions. J. Comput. Phys., 493:112464, 2023

work page 2023
[17]

J. Hua, P. Lin, C. Liu, and Q. Wang. Energy law preservingC 0 finite element schemes for phase field models in two-phase flow computations.J. Comput. Phys., 230(19):7115–7131, 2011

work page 2011
[18]

Tierra, J

G. Tierra, J. P. Pavissich, R. Nerenberg, Z. Xu, and M. S. Alber. Multicomponent model of deformation and detachment of a biofilm under fluid flow.J. R. Soc. Interface, 12(106):20150045, 2015

work page 2015
[19]

Kalesh, T

D. Kalesh, T. Merembayev, S. Omirbekov, and Y. Amanbek. Application of physics-informed neural networks for two-phase flow model with variable diffusion and experimental diagnostic framework.Results Eng., 26:105439, 2025

work page 2025
[20]

E. J. R. Coutinho, M. Dall’Aqua, L. McClenny, M. Zhong, U. Braga-Neto, and E. Gildin. Physics-informed neural networks with adaptive localized artificial viscosity.J. Comput. Phys., 489:112265, 2023

work page 2023
[21]

Abbasi, A

J. Abbasi, A. D. Jagtap, B. Moseley, A. Hiorth, and P. Østebø Andersen. Challenges and advancements in mod- eling shock fronts with physics-informed neural networks: A review and benchmarking study.Neurocomputing, 657:131440, 2025

work page 2025
[22]

A computational study of a data assimilation algorithm for the two- dimensional Navier–Stokes equations,

M. Gesho, E. Olson, and E. S. Titi, “A computational study of a data assimilation algorithm for the two- dimensional Navier–Stokes equations,”Communications in Computational Physics, vol. 19, no. 4, pp. 1094–1110, 2016, doi:10.4208/cicp.060515.161115a

work page doi:10.4208/cicp.060515.161115a 2016
[23]

Auto-Adaptive PINNs with Applications to Phase Transitions

K. Buck and W. Kim, Auto-Adaptive PINNs with Applications to Phase Transitions,arXiv preprint arXiv:2510.23999, 2025

work page internal anchor Pith review Pith/arXiv arXiv 2025
[24]

Mokbel, H

D. Mokbel, H. Abels, and S. Aland. A phase-field model for fluid–structure interaction.Journal of Computational Physics, 2018

work page 2018
[25]

Model predictions of deformation, embolization and permeability of partially obstructive blood clots under variable shear flow,

S. Xu, Z. Xu, O. V. Kim, R. I. Litvinov, J. W. Weisel, and M. Alber, “Model predictions of deformation, embolization and permeability of partially obstructive blood clots under variable shear flow,”Journal of The Royal Society Interface, 14(136):20170441, 2017

work page 2017
[26]

Zheng, A

X. Zheng, A. Yazdani, H. Li, J. D. Humphrey, and G. E. Karniadakis. A three-dimensional phase-field model for multiscale modeling of thrombus biomechanics in blood vessels.PLoS Comput. Biol., 16(4):e1007709, 2020

work page 2020
[27]

T. H. B. Demont, S. K. F. Stoter, and E. H. van Brummelen. Numerical investigation of the sharp-interface limit of the Navier–Stokes–Cahn–Hilliard equations.Journal of Fluid Mechanics, 2023

work page 2023
[28]

Temam,Infinite-Dimensional Dynamical Systems in Mechanics and Physics, 2nd ed., Applied Mathematical Sciences, vol

R. Temam,Infinite-Dimensional Dynamical Systems in Mechanics and Physics, 2nd ed., Applied Mathematical Sciences, vol. 68, Springer-Verlag, New York, 1997

work page 1997
[29]

Lions and E

J.-L. Lions and E. Magenes,Non-homogeneous Boundary Value Problems and Applications, Vol. I, Springer- Verlag, New York–Heidelberg, 1972. Translated from the French by P. Kenneth.Die Grundlehren der mathema- tischen Wissenschaften, Band 181

work page 1972
[30]

R. Qiu, R. Huang, Y. Xiao, J. Wang, Z. Zhang, J. Yue, Z. Zeng, and Y. Wang,Physics-informed neural networks for phase-field method in two-phase flow, Phys. Fluids34(5), 052109 (2022). doi:10.1063/5.0091063. 57

work page doi:10.1063/5.0091063 2022
[31]

Abbasi, A

J. Abbasi, A. D. Jagtap, B. Moseley, A. Hiorth, and P. Ø. Andersen,Challenges and Advancements in Modeling Shock Fronts with Physics-Informed Neural Networks: A Review and Benchmarking Study, Neurocomputing657 (2025), 131440

work page 2025
[32]

P. L. Lions and N. Masmoudi. Global solutions for some Oldroyd models of non-Newtonian flows.Chinese Ann. Math. Ser. B, 21(2):131–146, 2000

work page 2000
[33]

Temam.Navier-Stokes equations: theory and numerical analysis

R. Temam.Navier-Stokes equations: theory and numerical analysis. AMS Chelsea Publishing, Providence, RI,

work page
[34]

Reprint of the 1984 edition. 58

work page 1984