Long-time behavior of solutions to a fluid dynamic shape optimization problem via phase-field method

Christian Kahle; John Sebastian H. Simon; Michael Hinze

arxiv: 2601.13293 · v2 · submitted 2026-01-19 · 🧮 math.OC · math.AP

Long-time behavior of solutions to a fluid dynamic shape optimization problem via phase-field method

Michael Hinze , Christian Kahle , John Sebastian H. Simon This is my paper

Pith reviewed 2026-05-16 12:57 UTC · model grok-4.3

classification 🧮 math.OC math.AP

keywords shape optimizationphase-field methodNavier-Stokes equationsporous media approximationlong-time behaviortime-dependent optimizationtopology optimizationstationary limit

0 comments

The pith

As the time horizon tends to infinity, minima of the time-dependent fluid shape optimization problem converge to minima of the corresponding stationary problem.

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

The paper establishes that for a shape and topology optimization problem governed by the time-dependent Navier-Stokes equations, the optimal phase-field representations and their objective values approach those of the stationary version when the simulation time is taken to infinity. This convergence holds under a porous-media approximation of the fluid flow and a stationary phase-field that acts as a smooth indicator for the domain. A convergence rate for the objective functional is derived to control the limit process. The result justifies using long but finite time horizons in computations to recover stationary optimal shapes without directly solving the infinite-horizon problem.

Core claim

If the time horizon tends to infinity, minima of the time-dependent problem converge towards minima of the corresponding stationary problem. A convergence rate with respect to the time horizon of the values of the objective functional is derived. This allows proving that the solution to the time-dependent problem converges to a phase-field that minimizes the stationary problem.

What carries the argument

Stationary phase-field representation of the topology as a smooth indicator function, combined with porous-media approximation of the time-dependent Navier-Stokes equations, used to analyze the long-time limit of the optimization problem.

If this is right

Finite but sufficiently long time-dependent simulations can approximate stationary optimal shapes with an error controlled by the derived convergence rate.
The phase-field obtained from the long-time limit is guaranteed to be a stationary minimizer.
The same limit argument applies to other objective functionals that satisfy the same growth and continuity properties used in the proof.
Numerical validation on test cases confirms that the observed convergence matches the analytical rate.

Where Pith is reading between the lines

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

Practical shape-optimization codes could stop a time-dependent run once the objective change falls below a threshold tied to the proven rate, rather than guessing the horizon length.
The porous-media approximation may introduce a small bias in the recovered shape that persists even in the infinite-time limit.
The result suggests that similar long-time convergence statements could be proved for other parabolic or hyperbolic state equations in shape optimization.

Load-bearing premise

The time-dependent Navier-Stokes system under the porous-media approximation admits suitable convergence properties as the time horizon tends to infinity and that minimizers exist for both problems.

What would settle it

A numerical experiment on a specific domain and Reynolds number where the objective value of the time-dependent problem stays bounded away from the stationary minimum by a fixed positive amount no matter how large the time horizon is made.

read the original abstract

We investigate the long time behavior of solutions to a shape and topology optimization problem with respect to the time-dependent Navier--Stokes equations. The sought topology is represented by a stationary phase-field that represents a smooth indicator function. The fluid equations are approximated by a porous media approach and are time-dependent. In the latter aspect, the considered problem formulation extends earlier work. We prove that if the time horizon tends to infinity, minima of the time-dependent problem converge towards minima of the corresponding stationary problem. To do so, a convergence rate with respect to the time horizon, of the values of the objective functional, is analytically derived. This allowed us to prove that the solution to the time-dependent problem converges to a phase-field, as the time horizon goes to infinity, which is proven to be a minimizer for the stationary problem. We validate our results by numerical investigation.

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 proves that time-dependent phase-field Navier-Stokes shape optimization converges to the stationary problem as T goes to infinity, with an explicit rate on the objective values.

read the letter

The main result is that minimizers of the time-dependent problem approach minimizers of the stationary one as the time horizon tends to infinity, and they obtain a rate on the objective functional that makes the limit argument work. This extends their prior stationary phase-field work by making the Navier-Stokes equations time-dependent under the same porous-media approximation, while keeping the phase-field itself stationary. They derive the rate from long-time decay estimates on the fluid system and then pass to the limit on the objective values before showing the limiting phase-field is a minimizer. Numerical examples illustrate the convergence in practice. The quantitative rate is the part that adds real value beyond qualitative existence results common in this area. It gives a concrete handle on how large T needs to be for the time-dependent run to approximate the stationary optimum. The potential weak point is uniformity of the decay constant with respect to the phase-field. The damping term alpha(phi) changes with phi, so the energy estimates for the porous-media Navier-Stokes system can have phi-dependent constants. If those constants are not bounded uniformly over the admissible set, the 1/T rate may not hold uniformly on minimizing sequences, and the liminf step for the stationary objective could fail to close. The abstract states they prove the result, so the estimates presumably control this, but it is the step that requires the closest check in the full derivation. This is useful reading for anyone doing numerical shape optimization with time-dependent incompressible flows and phase-field representations. It supplies a theoretical justification for running simulations to large T rather than solving the stationary problem directly. The work is a solid, incremental advance in a specialized corner of PDE-constrained optimization, with enough concrete claims to merit referee attention.

Referee Report

2 major / 2 minor

Summary. The manuscript proves that, as the time horizon T tends to infinity, minimizers of a time-dependent fluid-dynamic shape optimization problem (with phase-field representation and porous-media approximation of the Navier-Stokes equations) converge to minimizers of the corresponding stationary problem. An explicit convergence rate for the objective functional values is derived analytically and used to pass to the limit, showing that the limiting phase-field is a stationary minimizer; the claims are supported by numerical experiments.

Significance. If the uniformity of the derived rate can be established, the result supplies a rigorous justification for approximating stationary shape optimization problems by long-time integration of the time-dependent system. It extends earlier stationary analyses, provides an explicit rate, and combines analysis with numerics, which is useful for PDE-constrained optimization in fluids.

major comments (2)

[main theorem / rate derivation] Proof of the main convergence theorem (rate derivation for |J_T(φ) − J_∞(φ)|): the energy decay estimate for the porous-media Navier-Stokes system yields a constant whose dependence on inf α(φ) and the Reynolds number is not shown to be uniform over the admissible set of phase-fields. Without uniformity the liminf argument for J_∞(φ_T) → inf J_∞ along minimizing sequences φ_T does not close.
[statement of main result] Existence of minimizers for the time-dependent and stationary problems is assumed without explicit reference or proof sketch; this assumption is load-bearing for the convergence statement and should be justified or cited.

minor comments (2)

[introduction] Notation for the objective functionals J_T and J_∞ should be introduced with a single consistent definition early in the paper rather than piecemeal.
[numerics] In the numerical section, state the precise values of the regularization parameters, mesh size, and time-step size used to generate the reported convergence plots.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and constructive comments on our manuscript. The two major points raised can be addressed by adding clarifications and a brief justification in a revised version, which we outline below.

read point-by-point responses

Referee: Proof of the main convergence theorem (rate derivation for |J_T(φ) − J_∞(φ)|): the energy decay estimate for the porous-media Navier-Stokes system yields a constant whose dependence on inf α(φ) and the Reynolds number is not shown to be uniform over the admissible set of phase-fields. Without uniformity the liminf argument for J_∞(φ_T) → inf J_∞ along minimizing sequences φ_T does not close.

Authors: We agree that uniformity of the constant is essential for the argument to close. In the energy decay estimate, the constant depends on a lower bound for α(φ) and on the (fixed) Reynolds number. Because the admissible phase-fields satisfy 0 ≤ φ ≤ 1 and α(φ) is constructed to be bounded below by a positive constant δ > 0 independent of φ (via the standard regularization α(φ) = α_min + (α_max - α_min)φ^2 with α_min > 0), the lower bound is uniform over the entire admissible set. The Reynolds number is a fixed problem parameter and does not vary. We will insert an explicit remark (or short lemma) immediately after the energy decay estimate in the revised manuscript to record this uniformity, thereby completing the liminf passage. revision: yes
Referee: Existence of minimizers for the time-dependent and stationary problems is assumed without explicit reference or proof sketch; this assumption is load-bearing for the convergence statement and should be justified or cited.

Authors: We acknowledge that the existence statements are used crucially and should be made explicit. In the revised manuscript we will add a short paragraph (or appendix subsection) that sketches the existence proof for both the time-dependent and stationary problems, relying on standard arguments: weak compactness in the phase-field space, lower semicontinuity of the objective, and the fact that the porous-media Navier–Stokes system admits unique weak solutions for each fixed phase-field. We will also cite the relevant literature (e.g., the stationary analysis in the authors’ earlier work and standard results on phase-field regularization of fluid optimization) to keep the manuscript self-contained. revision: yes

Circularity Check

0 steps flagged

No significant circularity; analytic derivation of convergence rate stands independently

full rationale

The paper derives the long-time convergence of time-dependent minimizers to stationary ones by establishing an explicit rate on the objective functional values from the decay properties of the porous-media Navier-Stokes system. No step reduces a claimed prediction to a fitted input, self-definition, or load-bearing self-citation chain; the rate estimate is obtained directly from energy estimates on the model equations under the stated assumptions of existence and suitable convergence. The extension of prior work is noted but does not carry the central proof. The argument remains self-contained against the model equations and does not exhibit any of the enumerated circular patterns.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central claim rests on standard existence results for minimizers in PDE-constrained optimization and on approximation properties of the porous-media model for Navier-Stokes; no new free parameters or invented entities are introduced.

axioms (2)

domain assumption Existence of minimizers for both the time-dependent and stationary optimization problems
Invoked to state that minima exist and can be compared in the limit.
domain assumption Suitable regularity and convergence properties of the porous-media approximation to the time-dependent Navier-Stokes equations
Required for the phase-field to remain well-defined and for the objective functional to behave continuously as time horizon grows.

pith-pipeline@v0.9.0 · 5451 in / 1327 out tokens · 29132 ms · 2026-05-16T12:57:10.401226+00:00 · methodology

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 unclear

?

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

Theorem 6.1: |JT(φT)−Js(φs)|≤C(1/√T+1/T) obtained from Gronwall on ws=us−vs with rate γ from Assumption 3.3
IndisputableMonolith/Foundation/AbsoluteFloorClosure.lean absolute_floor_iff_bare_distinguishability unclear

?

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

Existence of minimizers via weak-* L∞ + weak H1 compactness and lower-semicontinuity of Eε

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.