pith. sign in

arxiv: 2604.08917 · v1 · submitted 2026-04-10 · 🧮 math.NA · cs.NA· math.OC

An unfitted finite element method for PDE-constrained shape optimization via shape gradient flow

Wei Gong , Chuwen Ma , Ziyi Zhang This is my paper

Pith reviewed 2026-05-10 17:48 UTC · model grok-4.3

classification 🧮 math.NA cs.NAmath.OC
keywords PDE-constrained shape optimizationunfitted finite element methodshape gradient flowcut finite elementsghost penalizationcubic splinesconvergence rates
0
0 comments X

The pith

An unfitted finite element method solves PDE-constrained shape optimization by evolving the boundary along a shape gradient flow.

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

The paper develops a numerical technique for finding optimal shapes subject to PDE constraints by modeling the process as a continuous flow in which the boundary moves according to a derived velocity field. The state and adjoint equations are discretized on a fixed background mesh that cuts through the moving boundary, with cubic splines used to represent and advance the interface independently of the mesh. Ghost penalization stabilizes the cut-element discretizations of the PDEs. Under reasonable assumptions on the data and geometry, the discretization errors converge at optimal rates with respect to mesh size, and this is confirmed by numerical tests. The approach matters because many design problems in engineering require repeated solution of physics equations on changing domains, and avoiding repeated remeshing can simplify the computation.

Core claim

The central claim is that the shape gradient flow system, consisting of the state equation, the adjoint equation, the velocity equation, and the flow map that generates boundary evolution, can be discretized by an unfitted finite element method in which the boundary is tracked by cubic splines and the PDEs are solved on cut elements with ghost penalization, yielding optimal convergence rates under reasonable assumptions.

What carries the argument

The unfitted cut finite element discretization with ghost penalization for the coupled state-adjoint-velocity system, paired with independent cubic-spline representation of the moving boundary that decouples mesh and geometry evolution.

If this is right

  • The method computes optimal shapes without remeshing the domain at each step of the flow.
  • Optimal convergence rates hold for the discretization error in the state, adjoint, and shape variables as the mesh size tends to zero.
  • Numerical experiments confirm the theoretical rates for the proposed unfitted scheme.
  • The unfitted approach offers an alternative to evolving fitted finite element methods for the same shape gradient flow system.

Where Pith is reading between the lines

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

  • The technique may reduce the computational overhead associated with mesh quality degradation during large shape deformations.
  • Adaptive refinement concentrated near the cut boundary could further improve efficiency while preserving the unfitted character.
  • Similar cut-element techniques with spline boundary tracking might extend to other interface evolution problems such as free-boundary flows.

Load-bearing premise

The convergence analysis depends on reasonable assumptions about the smoothness of solutions and the geometry of the evolving domain whose precise verification is not detailed for general cases.

What would settle it

A sequence of numerical experiments on a benchmark problem with successively refined meshes that fails to exhibit the predicted optimal convergence orders in the error measures would falsify the claim.

read the original abstract

In this paper, we propose an unfitted finite element method to solve PDE-constrained shape optimization problems via shape gradient flow. The shape gradient flow system consists of the state equation, the adjoint equation, the velocity equation, as well as the flow map that generates the evolution of the boundary driven by the velocity field, which can be viewed as a limit system of the classical shape gradient descent algorithm. In \cite{GongLiRao} the authors proposed an evolving finite element method to solve the shape gradient flow system. Instead, in this paper, we propose an unfitted finite element method in which the evolution of the boundary is realized by cubic splines and the equations are solved by cut finite element methods with ghost penalization. Under reasonable assumptions, we are able to prove some optimal convergence rates that are further validated by numerical experiments.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 2 minor

Summary. The paper proposes an unfitted cut finite element method with ghost penalization and cubic-spline boundary tracking to discretize the shape gradient flow system (state equation, adjoint equation, velocity equation, and flow map) for PDE-constrained shape optimization. It replaces the evolving fitted meshes of prior work with a cut-FEM approach and claims to prove optimal convergence rates under reasonable assumptions, which are then validated numerically.

Significance. If the convergence analysis is complete, the method supplies a remeshing-free discretization for shape optimization that leverages standard cut-FEM and spline tools, offering practical advantages for complex geometries. The numerical confirmation of rates adds credibility, and the overall framework extends existing unfitted techniques in a coherent way.

major comments (2)
  1. [Abstract and theoretical sections] The convergence analysis (referenced in the abstract and presumably detailed in the theoretical sections) invokes 'reasonable assumptions' for the optimal rates without an explicit enumerated list or verification that these hold for the coupled state-adjoint-velocity system under cubic-spline boundary evolution and ghost penalization; this is load-bearing for the central claim of optimality.
  2. [Convergence analysis] The error estimates for the unfitted discretization of the velocity equation and its coupling to the flow map are not shown to be independent of the spline approximation order in a way that preserves the overall optimal rate; a concrete bound linking the spline error to the cut-FEM error would be needed to support the claim.
minor comments (2)
  1. Notation for the cut elements, ghost penalty terms, and the discrete flow map could be made more uniform across the state, adjoint, and velocity equations to improve readability.
  2. [Numerical experiments] The numerical experiments section would benefit from an explicit table or plot showing the observed rates against the theoretical prediction for at least two different mesh families.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the positive evaluation of our work and for the detailed, constructive comments. We address each major comment below and will revise the manuscript accordingly to improve clarity and completeness of the analysis.

read point-by-point responses

  1. Referee: [Abstract and theoretical sections] The convergence analysis (referenced in the abstract and presumably detailed in the theoretical sections) invokes 'reasonable assumptions' for the optimal rates without an explicit enumerated list or verification that these hold for the coupled state-adjoint-velocity system under cubic-spline boundary evolution and ghost penalization; this is load-bearing for the central claim of optimality.

    Authors: We agree that an explicit enumeration of the assumptions, together with a verification of their validity for the coupled system, would strengthen the manuscript. In the revised version we will add a dedicated subsection that lists all assumptions used in the convergence analysis. We will also include a short discussion confirming that these assumptions are compatible with the cubic-spline boundary evolution, ghost penalization, and the coupling among the state, adjoint, velocity, and flow-map equations. revision: yes

  2. Referee: [Convergence analysis] The error estimates for the unfitted discretization of the velocity equation and its coupling to the flow map are not shown to be independent of the spline approximation order in a way that preserves the overall optimal rate; a concrete bound linking the spline error to the cut-FEM error would be needed to support the claim.

    Authors: We thank the referee for this observation. We will derive and insert a concrete a-priori bound that relates the cubic-spline approximation error to the cut-FEM error for the velocity equation and its coupling to the flow map. The bound will be stated in terms of the spline degree, the mesh size, and the finite-element polynomial degree, and will be used to show that the spline contribution does not degrade the overall optimal rate when the spline order is chosen consistently with the discretization parameters. This estimate will be added to the convergence-analysis section. revision: yes

Circularity Check

0 steps flagged

No significant circularity in derivation chain

full rationale

The paper introduces an unfitted cut-FEM discretization (with ghost penalization and cubic-spline boundary tracking) for the shape-gradient-flow system (state, adjoint, velocity, and flow-map equations). Convergence rates are proved under explicitly stated reasonable assumptions and validated by separate numerical experiments. No load-bearing step equates a claimed prediction or first-principles result to its own inputs by construction, nor reduces the central claim to a self-citation chain, fitted parameter renamed as prediction, or smuggled ansatz. The replacement of evolving fitted meshes by cut-FEM is a standard technical variation whose error analysis draws on independent ghost-penalty and spline-approximation theory, rendering the derivation self-contained.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on standard assumptions from shape calculus and cut finite element theory plus the 'reasonable assumptions' invoked for the convergence proof. No new physical entities or free parameters are introduced in the abstract.

axioms (1)
  • domain assumption Standard assumptions of shape calculus and cut finite element analysis hold for the state, adjoint, and velocity problems.
    Invoked to obtain optimal convergence rates.

pith-pipeline@v0.9.0 · 5447 in / 1259 out tokens · 36775 ms · 2026-05-10T17:48:31.678326+00:00 · methodology

discussion (0)

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

Reference graph

Works this paper leans on

32 extracted references · 32 canonical work pages

  1. [1]

    Allaire, C

    G. Allaire, C. Dapogny and F. Jouve.Shape and Topology Optimization, in Geometric Partial Differential Equations, Part II.A. Bonito and R. Nochetto, eds., Handb. Numer. Anal. 22, Elsevier, Amsterdam, 2021, pp. 1–132

    work page 2021

  2. [2]

    Belytschko and T

    T. Belytschko and T. Black, Elastic crack growth in finite elements with minimal remeshing.Inter. J. Numer. Methods Engrg., 45(1999), No. 5, pp. 601-620

    work page 1999

  3. [3]

    Bernland, E

    A. Bernland, E. Wadbro and M. Berggren, Acoustic shape optimization using cut finite elements.Int. J. Numer. Methods Engrg., 113(2018), pp. 432-449

    work page 2018

  4. [4]

    Burman, D

    E. Burman, D. Elfverson, P. Hansbo, M. G. Larson and K. Larsson, A cut finite element method for the Bernoulli free boundary value problem.Comput. Methods Appl. Mech. Engrg., 317(2017), pp. 598–618

    work page 2017

  5. [5]

    Burman, D

    E. Burman, D. Elfverson, P. Hansbo, M. G. Larson and K. Larsson, Shape optimization using the cut finite element method.Comput. Methods Appl. Mech. Engrg., 328(2018), pp. 242–261

    work page 2018

  6. [6]

    Burman, P

    E. Burman, P. Hansbo and M. G. Larson, Cutfem based on extended finite element spaces.Numer. Math., 152(2022), No. 2, pp. 331–369

    work page 2022

  7. [7]

    Burman, C

    E. Burman, C. He and M. G. Larson, Comparison of shape derivatives using CutFEM for ill-posed Bernoulli free boundary problem.J. Sci. Comput., 88(2021), No. 2, Paper No. 35, 28 pp

    work page 2021

  8. [8]

    Chenais and E

    D. Chenais and E. Zuazua. Finite-element approximation of 2D elliptic optimal design.J. Math. Pures Appl., 85(2006), pp. 225–249

    work page 2006

  9. [9]

    Deckelnick, P

    K. Deckelnick, P. J. Herbert and M. Hinze. Convergence of a steepest descent algorithm in shape optimisation using W1,∞ functions.ESAIM Math. Model. Numer. Anal., 59(2025), 1505-1529

    work page 2025

  10. [10]

    M. C. Delfour and J. P. Zol´esio.Shapes and Geometries: Metrics, Analysis, Differential Calculus, and Optimization, 2nd edn., SIAM, Philadelphia, 2011. 24

    work page 2011

  11. [11]

    Dziuk and C

    G. Dziuk and C. M. Elliott. Finite elements on evolving surfaces.IMA J. Numer. Anal., 27(2007), pp. 262–292

    work page 2007

  12. [12]

    Edelmann

    D. Edelmann. Finite element analysis for a diffusion equation on a harmonically evolving domain.IMA J. Numer. Anal., 42(2022), pp. 1866–1901

    work page 2022

  13. [13]

    C. M. Elliott and T. Ranner. A unified theory for continuous-in-time evolving finite element space approximations to partial differential equations in evolving domains.IMA J. Numer. Anal., 41(2021), No. 3, pp. 1696–1845

    work page 2021

  14. [14]

    C. M. Elliott and V . Styles. An arbitrary Lagrangian–Eulerian ESFEM for solving PDEs on evolving surfaces.Milan J. Math., 80(2012), pp. 469–501

    work page 2012

  15. [15]

    Eppler, H

    K. Eppler, H. Harbrecht and R. Schneider. On convergence in elliptic shape optimization.SIAM J. Control Optim., 46(2007), No. 1, pp. 61–83

    work page 2007

  16. [16]

    Fumagalli, N

    I. Fumagalli, N. Parolini and M. Verani. Shape optimization for Stokes flows: a finite element convergence analysis. ESAIM Math. Model. Numer. Anal., 49(2015), No. 4, pp. 921–951

    work page 2015

  17. [17]

    W. Gong, B. Li and Q. Rao. Convergent evolving finite element approximations of boundary evolution under shape gradient flow.IMA J. Numer. Anal., 44(2024), No. 5, 2667–2697

    work page 2024

  18. [18]

    W. Gong, J. Li and S. Zhu. Improved discrete boundary type shape gradients for PDE-constrained shape optimiza- tion.SIAM J. Sci. Comput., 44(2022), pp. A2464–A2505

    work page 2022

  19. [19]

    Hansbo and P

    A. Hansbo and P. Hansbo. An unfitted finite element method, based on Nitsche’s method, for elliptic interface problems.Comput. Methods Appl. Mech. Engrg., 191(2002), No. 47-48, pp. 5537–5552

    work page 2002

  20. [20]

    Hansbo, M.G

    P. Hansbo, M.G. Larson, K. Larsson. Cut finite element methods for linear elasticity problems. Geometrically Un- fitted Finite Element Methods and Applications, 25-63, Lect. Notes Comput. Sci. Eng., 121, Springer, Cham, 2017

    work page 2017

  21. [21]

    Hiptmair, A

    R. Hiptmair, A. Paganini and S. Sargheini. Comparison of approximate shape gradients.BIT Numer. Math., 55(2015), pp. 459–485

    work page 2015

  22. [22]

    Huang, H

    P. Huang, H. Wu and Y . Xiao. An unfitted interface penalty finite element method for elliptic interface problems. Comput. Methods Appl. Mech. Engrg.,323(2017), pp. 439-460

    work page 2017

  23. [23]

    Kiniger and B

    B. Kiniger and B. Vexler. A priori error estimates for finite element discretizations of a shape optimization problem. ESAIM Math. Model. Numer. Anal., 47(2013), pp. 1733–1763

    work page 2013

  24. [24]

    Kov ´acs, B

    B. Kov ´acs, B. Li and C. Lubich. Convergence of finite elements on an evolving surface driven by diffusion on the surface.Numer. Math.,137(2017), pp. 643–689

    work page 2017

  25. [25]

    C. Ma, Q. Zhang and W. Zheng. A fourth-order unfitted characteristic FEM for solving the advection-diffusion equation on time-varying domains.SIAM J. Numer. Anal., 60(2022), pp. 2203–2224

    work page 2022

  26. [26]

    Mo ¨es, J

    N. Mo ¨es, J. Dolbow, and T. Belytschko. A finite element method for crack growth without remeshing.Inter. J. Numer. Methods Engrg., 46(1999), No. 1, pp. 131–150

    work page 1999

  27. [27]

    Nicaise and S

    S. Nicaise and S. Sauter. Efficient numerical solution of Neumann problems on complicated domains.Calcolo, 43(2006), No. 2, pp. 95–120

    work page 2006

  28. [28]

    Osher and J

    S. Osher and J. A. Sethian. Fronts propagating with curvature-dependent speed: Algorithms based on Hamilton- Jacobi formulations.J. Comput. Phys., 79(1988), No. 1, pp. 12-49

    work page 1988

  29. [29]

    C. S. Peskin. Flow patterns around heart valves: A numerical method.J. Comput. Phys., 10(1972), No. 2, pp. 252– 271

    work page 1972

  30. [30]

    Schulz and M

    V . Schulz and M. Siebenborn. Computational comparison of surface metrics for PDE constrained shape optimization. Comput. Methods Appl. Math.,16(2016), pp. 485–496

    work page 2016

  31. [31]

    Schulz, M

    V . Schulz, M. Siebenborn and K. Welker. Efficient PDE constrained shape optimization based on Steklov-Poincar´e- type metrics.SIAM J. Optim.,26(2016), pp. 2800–2819

    work page 2016

  32. [32]

    Sokolowski and J.P

    J. Sokolowski and J.P. Zol ´esio. Introduction to Shape Optimization: Shape Sensitivity Analysis.Springer, Heidel- berg, 1992. 25

    work page 1992