Pith. sign in

REVIEW 4 major objections 5 minor 26 references

A model and a finite element approximation of the mixed-dimensionality diffusion problem

T0 review · 4 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read A saddle-point formulation ties 1D fiber temperatures to a 3D solid through the H1 inner product, yielding well-posedness and stable convergence on independent meshes.

desk verdict Elegant H1(S) coupling idea with a real gap between the analyzed exact-integral problem and the implemented transverse quadrature; worth refereeing after a revision. read the letter →

arxiv 2608.06976 v1 pith:OUSSWPLD submitted 2026-08-07 math.NA cs.NAmath-phmath.MP

Ignacio Romero , David Portillo This is my paper · ORCID
classification math.NAcs.NAmath-phmath.MP MSC 65N3065N1265N1535J05
keywords mixed-dimensionaldiffusion3D-1DcouplingLagrangemultipliersH1normconstraintinf-supconditionfiniteelementmethodnon-conformingmeshesheatconductioninfibers
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

This paper proposes and analyzes a variational model for heat exchange between a three-dimensional diffusive solid and a one-dimensional diffusive fiber embedded in it. The central claim is that the coupled problem is well posed when the fiber temperature is lifted to the cylindrical volume the fiber sweeps and equality with the solid temperature is enforced through the H1 inner product, with the Lagrange multiplier drawn from the fiber temperature space. The paper further claims that any Galerkin discretization using the fiber temperature space as the multiplier space inherits this stability, with discrete inf-sup constant exactly 1, so the solid and fiber meshes can be chosen independently and convergence follows. If correct, the method gives a robust, mesh-independent recipe for mixed-dimensional diffusion in tissues, composites, embedded conduits, and branched networks.

What carries the argument

The central object is the lifting operator L that maps a function on the fiber curve C to a function constant on each cross section of the swept cylinder S, together with the H1(S) inner product used as the coupling bilinear form. The identity ||Lφ||_{H1(S)} = $\sqrt$(A)||φ||_{H1(C)} turns the abstract inf-sup test into the concrete ratio ||Lμ||^2_{H1(S)}/||μ||^2_Λ = 1, which guarantees both continuous and discrete stability. A second mechanism is the choice of norm on the multiplier space Λ := H1(C), equipped with ||Lμ||_{H1(S)} rather than the plain H1(C) norm, which keeps the stability constant finite and dimensionless in the slender limit.

What would settle it

Compute the discrete inf-sup constant gamma_h from Eq. (35) on a curved fiber with curvature kappa such that kappa R is not negligible, say kappa R ≈ 0.5, using a very fine solid mesh and a fiber mesh refined below the fiber radius. If the norm-identity error is material, gamma_h will depart from 1 and the observed H1-seminorm convergence rate of the solid temperature will fall below the predicted rate of 1; if gamma_h remains at 1 and the rate holds, the curvature error is negligible at the tested slenderness.

Watch

Extended reading notes

Core claim

The discovery is that coupling through the H1(S) inner product rather than the L2 inner product is exactly what stabilizes mixed-dimensional diffusion. Lifting the one-dimensional field to the overlap cylinder S and tying it to the solid field via the norm identity ||Lφ||_{H1(S)} = sqrt(A)||φ||_{H1(C)} makes the continuous saddle-point problem satisfy the inf-sup condition with constant 1, and the same test pair (0, μ) remains available in every conforming discrete setting, so the discrete inf-sup constant is also 1 for arbitrary, non-conforming meshes. Consequently, the discretization converges at mixed finite element theory rates without any special mesh compatibility requirement beyond taking the discrete multiplier space equal to the discrete fiber temperature space.

Load-bearing premise

The load-bearing premise is that the lifted norm identity ||Lphi||_{H1(S)} = $\sqrt$(A)||phi||_{H1(C)} is exact; for curved fibers it holds only up to a relative error O($kappa^{2}$ $R^{2}$), and the theorem statements and stability constants do not explicitly account for that error.

Editorial extensions

If this is right

  • Mixed-dimensional diffusion can be discretized with any finite element space on the solid and any on the fiber, using the fiber temperature space as the multiplier space, and will converge without any requirement that the meshes conform.
  • The discrete inf-sup constant equals 1 for every pair of independently constructed meshes, so no stabilization parameter or mesh compatibility condition is needed; the paper's consistency test confirms gamma_h = 1 at every fiber refinement and an absence of checkerboard multiplier modes.
  • The H1 coupling is not just convenient but necessary: dropping the gradient term from the constraint (a pure L2 coupling) makes the discrete inf-sup constant decay as (h_C/R)^2 under fiber refinement, signaling instability.
  • The same formulation supports multiple fibers, branched networks, transient problems, and other diffusion problems governed by Poisson's equation, with the coupling terms unchanged.
  • The convergence estimate of Theorem 2 gives explicit rates for the solid field in L2 and H1 seminorm, the fiber field in L2, and the multiplier, all confirmed numerically.

Reading between the lines

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

  • A consequence the paper leaves implicit: the norm identity used for stability is exact only for straight fibers, so for curved fibers the stated theorem should ideally carry an explicit O(kappa^2 R^2) correction; the helix example suggests the practical effect is small, but the proof as written does not fully cover curved geometry.
  • The implementation deliberately approximates the constraint by sampling coupling points on two diameters and skipping points outside the solid, and the manufactured exact solution in Section 4.2 is exact only for that quadrature rule; replacing it with a full-disk rule would give a testable O(R^2) change in the discrete multiplier and coupling constant.
  • The same energy-inner-product pairing strategy could plausibly stabilize other mixed-dimensional couplings, such as 3D-2D interface problems or beam-shell-solid mechanics, whenever a norm equivalence analogous to Eq. (19) can be established.
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

4 major / 5 minor

Summary. The paper introduces a variational formulation for coupled 3D-1D diffusion, where a fiber temperature field defined on a curve is lifted to a cylindrical volume S and tied to a solid temperature through an H1(S) inner product enforced by a Lagrange multiplier. The authors prove well-posedness of the continuous saddle-point problem (Theorem 1) and stability plus convergence of a Galerkin finite element discretization with multiplier space equal to the discrete fiber temperature space (Theorem 2). The key advertised feature is that the solid and fiber meshes need not be conforming. Numerical experiments include a consistency test, a convergence test against a manufactured exact solution, a helical fiber self-convergence study, and branched tree networks.

Significance. If the results hold as stated, the paper provides an attractive and simple alternative to existing 3D-1D coupling methods: the inf-sup constant is exactly one by construction, the coupling is stable independently of mesh compatibility, and the discrete well-posedness argument is elementary because the multiplier space is chosen as the fiber temperature space. The authors are explicit about the modeling simplifications (constant cross-sectional temperature, overlapping domains) and about the role of the lifted H1 norm in the multiplier space, which is a useful conceptual clarification. However, the main advertised claim — convergence of the implemented finite element method to the solution of the continuous problem — is not supported as written, because the implementation uses a transverse quadrature rule that is not the bilinear form analyzed in Theorem 2, and because the norm identity used in the theorems is exact only for straight fibers. The underlying idea is sound and the gaps are repairable, but the current manuscript overstates what has been rigorously established.

major comments (4)
  1. [Section 3, Theorem 2 and Eq. (36)] The convergence proof in Theorem 2 and the error estimate (36) are stated for the bilinear form b(v,beta;mu) = (Lmu, Lbeta - v)_{H1(S)} with exact integrals over the cylindrical region S. The implementation described in Appendix A (Algorithms 1 and 2, and Eq. (40)) replaces these integrals by a fixed two-diameter transverse quadrature rule, and Algorithm 1 lines 13-14 additionally skip sample points that fall outside the solid mesh. Section 3.1 explicitly states that this rule is exact only for fields linear on the cross section and that the discrete solution depends on the rule through its second moment. For a fixed fiber radius R the transverse quadrature error is O(R^2) and does not vanish as h_Omega and h_C tend to zero, so the discrete bilinear form b_h does not converge to b. Consequently Theorem 2 does not establish convergence of the implemented method to the solution of the continuous problem (16)-(17). The gap is repairable either by using a quadrature rule that becomes exact in the limit (e.g., exact cross-sectional integration) or by including a variational-crime term in the analysis, but as written the central convergence claim applies to a different numerical method than the one implemented.
  2. [Section 4.2, Eq. (43) and Table 2] The manufactured exact solution in Eq. (43) is explicitly chosen to match the two-diameter quadrature rule of Appendix A: the constant 0.97291667 corresponds to the second moment R^2/6 assigned by that rule, while the exact disk-average constant would be 0.9725. The paper itself acknowledges this in the paragraph after Eq. (43), stating that (43) is the exact solution of the coupled problem 'as discretized, coupling operator and its quadrature included.' Therefore the convergence rates in Table 2 (1.90 for L2, 0.98 for H1, 2.11 for theta, 2.23 for the multiplier) are convergence rates to a quadrature-modified problem, not to the solution of Eqs. (16)-(17). This does not confirm the theoretical convergence result for the continuous model; it only verifies the internal consistency of the discrete implementation against a solution constructed from the same quadrature. The 'confirmation of the theoretical results' claimed in the abstract is therefore not supported by this numerical example.
  3. [Section 2.4, Eq. (19) and Theorem 1] The identity ||Lphi||_{H1(S)} = sqrt(A)||phi||_{H1(C)} in Eq. (19) is used as an exact norm equality in the proof of Theorem 1, in particular in the continuity estimate (25), the coercivity bound on the kernel (27)-(29), and the inf-sup argument (30). The remark following Eq. (14) correctly notes that for curved fibers the gradient contribution carries a cross-sectional factor integral over Sigma of (1 - k xi)^{-1} dA = A(1 + O(k^2 R^2)), so Eq. (19) holds only up to a relative error of order k^2 R^2. This curvature error is not tracked through the proof of Theorem 1 nor in the discrete inf-sup statement of Theorem 2. As stated, the theorems therefore prove well-posedness and discrete stability only in the straight-fiber limit or under an unstated smallness assumption on k R. The authors should either restrict the theorems to straight fibers or, preferably, prove a perturbed version of the stability estimates that includes the curvature terms explicitly, so that the O(k^2 R^2) error is controlled in the same norm.
  4. [Section 3.1 and Appendix A, Eq. (40)] The analysis in Section 3 assumes that the coupling bilinear form is evaluated exactly, while Algorithm 2 evaluates the coupling integrand at a single sample point per link element using the shape functions evaluated at that point. No consistency requirement is imposed on the quadrature rule in the statement of Theorem 2 or in the convergence argument. Since the number of transverse sample points is fixed and independent of the mesh size, the discrete constraint is not an exact integral of the lifted fields. This is a variational crime that is known to affect the constants and possibly the order of convergence in mixed methods. At minimum, the paper should state that Theorem 2 concerns the exact-integral formulation and that the implementation is a separate approximate scheme; ideally it should provide a quadrature-consistency analysis showing that the additional error is of the same order as the finite element interpolation error, perhaps under a geometric assumption such as the fiber lying in a region where all quadrature points remain inside the solid.
minor comments (5)
  1. [Section 3.1, after Eq. (38)] The paragraph discussing the two-diameter transverse rule says the rule is 'exact whenever f varies at most linearly across the section — which is all the model itself resolves — but not for its quadratic variation.' This is an important caveat, but it is also a statement about the quadrature consistency; since the model does resolve the quadratic variation of the solid field u (e.g., the exact solution of Section 4.2 is quadratic), the wording 'all the model itself resolves' is too strong and should be clarified.
  2. [Section 4.2, Eq. (43)] The sentence 'No Dirichlet conditions are imposed on the solid' should be accompanied by an explicit reference to Remark 4, since Theorem 1 as stated requires |partial_D Omega| > 0. The example is exactly the complementary situation discussed in that remark, and the convergence study relies on the coupling-induced well-posedness rather than on the Dirichlet condition of Theorem 1.
  3. [Section 4.3, Table 3] The row for max|lambda_h| reports a fitted slope of 0.15, yet the text only notes that the multiplier does not converge to zero. The lack of an explanation for the essentially flat behavior is a missed opportunity; it would be informative to state whether this is expected because the fiber temperature is not manufactured to make the constraint gap vanish, and how the multiplier values scale with mesh size in that case.
  4. [Appendix A, Algorithm 1] The parameter n_h (the number of transverse subdivisions) is introduced in Algorithm 1 but is not defined in the main text, and the pseudocode does not indicate how the transverse quadrature weights are related to n_h. Since the transverse rule is the source of the quadrature discrepancy discussed in the major comments, a precise definition of the sample points and weights would help the reader assess the consistency of the scheme.
  5. [General] The abstract states 'Numerical examples confirm the theoretical results,' but the theoretical results concern the exact-integral formulation while the numerical examples (especially Sections 4.2 and 4.3) exercise the quadrature-based implementation. The abstract should either be more modest or the theoretical analysis must be extended to cover the implemented scheme.

Circularity Check

2 steps flagged · score 2.0 of 10

Two minor self-referential constructions (multiplier-norm normalization and quadrature-tailored exact solution) are transparently acknowledged; the core well-posedness and convergence proofs are otherwise self-contained.

  1. self definitional [Section 2.4, Eq. (15) and Eq. (30)]
    "Λ:=H 1(C),∥µ∥ Λ :=∥Lµ∥ H 1(S) = √ A∥µ∥ H 1(C) ,(15) ... To prove the inf-sup condition (24) it suffices to restrict the supremum to the particular test pair (v, β) = (0, µ) and note that, by Eqs. (20) and (15),∥(0, µ)∥U = √ A∥µ∥ H 1(C) = ∥µ∥Λ, so that ... = 1>0."

    The discrete and continuous inf-sup constant γ=1 is not obtained from an independent estimate: the multiplier norm is defined as the lifted H1 norm, so the test pair (0,µ) gives b(0,µ;µ)/(∥(0,µ)∥_U ∥µ∥_Λ) = ∥Lµ∥²_{H1(S)}/(∥Lµ∥_{H1(S)}·∥Lµ∥_{H1(S)}) = 1 by the definitions in (15) and (20). The celebrated constant 1 is thus a normalization choice, not a discovered mesh-independent stability bound. The paper itself says in Remark 3 that the norm (15) 'simply makes this scaling explicit.' This is transparent and does not undermine the substantive coercivity-on-kernel proof, so it is a minor self-definitional element rather than fatal circularity.

  2. other [Section 4.2, Eq. (43) and surrounding paragraph]
    "For the quadratic field at hand the axial value would give z+0.97375 and a rule integrating exactly over the disk would give z+0.9725, whereas the two-diameter rule of Appendix A, which assigns the second moment R 2/6 to each transverse direction, gives the constant shown in Eq.(43). ... what matters here is that (43) is the exact solution of the coupled problem as discretized, coupling operator and its quadrature included."

    The 'exact solution' used in the convergence test is not the solution of the continuum saddle-point problem (16)-(17) analyzed in Theorem 2. Its constant 0.97291667 is chosen so that the constraint gap vanishes under the implementation's two-diameter transverse quadrature rule, which assigns second moment R²/6 rather than the disk value R²/4; together with Algorithm 1's skipping of sample points outside the solid mesh, this means the implemented discrete bilinear form b_h differs from the analyzed b. Consequently the rates in Table 2 measure convergence to a quadrature-modified problem, not directly to the solution of Eqs. (16)-(17).

full rationale

The paper's central derivation chain is substantially self-contained. Theorem 1 is proved from the definitions: continuity of a and b via Cauchy-Schwarz and the lifting identities (18)-(19), coercivity on ker(b) via the kernel bound (28) and Poincaré's inequality, and inf-sup via the explicit test pair (0,µ). Theorem 2 replicates the argument on the discrete spaces; because W^h is chosen equal to the discrete fiber-temperature space, the same test pair is available and the discrete inf-sup constant is uniform. No parameter is fitted from numerical data, and the well-posedness conclusion does not rest on an external self-citation: prior work [10] supplies the coupling idea, but the analysis here is self-contained. Two minor self-referential elements are present, both openly acknowledged. First, the multiplier norm in (15) is defined as the lifted H1 norm, making the inf-sup quotient for (0,µ) equal to 1 by definition; the constant γ=1 is a normalization rather than an independently discovered stability constant. Second, the exact solution (43) in Section 4.2 is manufactured to satisfy the discrete constraint under the implementation's two-diameter quadrature rule, so the convergence rates in Table 2 are for the quadrature-modified problem, not directly for the continuum bilinear form of Theorem 2. This is a variational-crime gap, not a hidden fit, and the text explicitly calls (43) 'the exact solution of the coupled problem as discretized.' Finally, the lifting norm identity (19) is only approximate for curved fibers (relative error O(k²R²)), as the paper's own remark concedes; that is a correctness caveat for the curved example, not a circularity. Overall, the core proofs are independent and the self-referential elements are transparent and non-load-bearing for the main well-posedness claim.

Assumptions & free parameters 2 free parameters · 5 assumptions · 0 invented entities

The model introduces no new physical entities. The central claim rests on standard Sobolev and mixed-FE results, on geometric assumptions on the fiber (slender, non-self-intersecting, constant cross-section), and on the modeling simplification of constant cross-sectional fiber temperature. The main non-standard assumption is the exact norm equivalence (19), which is only approximate for curved fibers.

free parameters (2)
  • Coupling length l in H1(S) inner product = l = R = sqrt(A/pi)
    Chosen by hand in Eq. (17) and Remark 3; it sets the relative weight of gradient coupling. If l=0 (L2-only), the discrete inf-sup constant decays as (h_C/R)^2, so the stability claim is conditional on this choice.
  • Multiplier norm constant = Factor sqrt(A) in ||mu||_Lambda = ||Lmu||_{H1(S)} = sqrt(A)||mu||_{H1(C)}
    The inf-sup constant gamma=1 follows from this scaling. It is a norm choice rather than a data fit, but it is load-bearing for the stability theorem.
assumptions (5)
  • standard math Poincare inequality on H1_D(Omega) with nonzero Dirichlet boundary, and on H1_D(C) in Remark 4
    Used in the coercivity step of Theorem 1 and in Remark 4 to control the fiber norm through one fixed endpoint.
  • standard math Brezzi-Fortin mixed saddle-point theory for continuity, kernel coercivity, and inf-sup
    Invoked to prove well-posedness of the continuous and discrete problems and the convergence estimate in Eq. (36).
  • domain assumption Fiber is a slender non-self-intersecting tube (kappa R < 1) with constant cross-section area A, generated by sweeping a circle along a smooth curve
    Used in Section 2.1 and in the curvature discussion after Eq. (14); justifies the volume identities and the slenderness expansion.
  • domain assumption Lifted fiber temperature is constant on each cross section, and matrix and fiber volumes overlap
    This is the modeling simplification stated in Remark 5: the coupled solution is not the exact 3D solution of a slender body embedded in a matrix, but an effective model.
  • ad hoc to paper Norm identity (19), ||Lphi||_{H1(S)} = sqrt(A)||phi||_{H1(C)}, treated as exact in Theorems 1 and 2
    For curved fibers the gradient factor is A(1+O(k^2 R^2)), so the identity is only approximate. The paper acknowledges this but does not carry the error through the proofs.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A model and a finite element approximation of the mixed-dimensionality diffusion problem." pith.science (2026). https://pith.science/paper/OUSSWPLD

@misc{pith2026260806976,
  author       = {Pith},
  title        = {Pith review of: A model and a finite element approximation of the mixed-dimensionality diffusion problem},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/OUSSWPLD}},
  note         = {Machine review of arXiv:2608.06976}
}
read the original abstract

We present the formulation of a boundary value problem that models the coupled behavior of a three-dimensional diffusive solid with one-dimensional diffusive fibers embedded inside it. We introduce a variational statement of the problem that identifies the linked diffusive fields as energy minimizers under a coupling constraint. This saddle-point problem is proved to be well posed. Then, we introduce a finite element discretization of the proposed boundary value problem, and we prove the convergence of the finite element solution to the exact one. The most significant feature of this approximation is that the meshes of the bodies need not be conforming. Numerical examples confirm the theoretical results

Figures

Figures reproduced from arXiv: 2608.06976 by the authors.

Figure 1
Figure 1. Solid matrix with an embedded thermal fiber. The fiber occupies a [PITH_FULL_IMAGE:figures/full_fig_p007_1.png] view at source ↗
Figure 2
Figure 2. Geometry (left) and mesh (right, at a representative fiber refinement): [PITH_FULL_IMAGE:figures/full_fig_p014_2.png] view at source ↗
Figure 3
Figure 3. The discrete inf-sup constant γh of Eq. (35) across the fiber-refinement sweep. The proposed formulation gives γh = 1 at every refinement, while the L 2 constrained formulation leads to γh → 0, pointing at an unstable numerical method. discrete inf-sup constant decays quadratically under fiber refinement. Over the last four meshes the computed values follow γh ≃ 1 12  hC R 2 (42) to three significant digits, so th… view at source ↗
Figures from the paper (11 more)
Figure 4
Figure 4. Figure 4: Value of the multiplier λh along the fiber when using 96 elements in the fiber (symmetric logarithmic scale). L 2 formulation (left) and H1 formulation (right). a checkerboard mode. In the advocated formulation, this mode does not appear as a result of the unconditiona…
Figure 5
Figure 5. Figure 5: Left: brick with the off-axis embedded fiber, running its full height. [PITH_FULL_IMAGE:figures/full_fig_p017_5.png]
Figure 6
Figure 6. Figure 6: Convergence of the solid L 2 and H1 -seminorm errors, the fiber L 2 error, and max |λh|, under uniform refinement. Triangles show the theoretical slopes 2 (L 2 ) and 1 (H1 ). −1.00 −0.75 −0.50 −0.25 0.00 0.25 0.50 0.75 1.00 z 0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.0…
Figure 7
Figure 7. Figure 7: Left: computed fiber temperature at the coarsest and finest levels [PITH_FULL_IMAGE:figures/full_fig_p018_7.png]
Figure 8
Figure 8. Figure 8: The three meshes used in the convergence study. Embedded fiber is [PITH_FULL_IMAGE:figures/full_fig_p019_8.png]
Figure 9
Figure 9. Figure 9: Convergence of the solid field u against the solution on the finest reference mesh. 20 [PITH_FULL_IMAGE:figures/full_fig_p020_9.png]
Figure 10
Figure 10. Figure 10: Computed temperature field at mesh levels 1, 2, and 3 (coarsest to [PITH_FULL_IMAGE:figures/full_fig_p021_10.png]
Figure 11
Figure 11. Figure 11: Column network: t = 10, 30, 60, 100 (left to right) [PITH_FULL_IMAGE:figures/full_fig_p023_11.png]
Figure 12
Figure 12. Figure 12: Bush network: t = 10, 30, 60, 100 (left to right) [PITH_FULL_IMAGE:figures/full_fig_p023_12.png]
Figure 13
Figure 13. Figure 13: Canopy network: t = 10, 30, 60, 100 (left to right) [PITH_FULL_IMAGE:figures/full_fig_p023_13.png]
Figure 14
Figure 14. Figure 14: Windswept network: t = 10, 30, 60, 100 (left to right). 23 [PITH_FULL_IMAGE:figures/full_fig_p023_14.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

26 extracted references · 22 canonical work pages

  1. [1]

    van Heerden, D

    A. van Heerden, D. Judt, S. Jafari, C. Lawson, T. Nikolaidis, and D. Bosak. Aircraft thermal management: Practices, technology, system architec- tures, future challenges, and opportunities.Progress in Aerospace Sciences, 128:100767, January 2022

    work page 2022

  2. [2]

    Angelo and A

    C. Angelo and A. Quarteroni. On the coupling of 1d and 3d diffusion- reaction equations: application to tissue perfusion problems.Mathematical Models and Methods in Applied Sciences, 18(08):1481–1504, August 2008

    2008

  3. [3]

    Laurino and P

    F. Laurino and P. Zunino. Derivation and analysis of coupled PDEs on manifolds with high dimensionality gap arising from topological model reduction.ESAIM: Mathematical Modelling and Numerical Analysis, 53(6):2047–2080, Nov 2019

    work page 2019

  4. [4]

    Kuchta, F

    M. Kuchta, F. Laurino, K. Mardal, and P. Zunino. Analysis and approx- imation of mixed-dimensional PDEs on 3d-1d domains coupled with La- grange multipliers.SIAM Journal on Numerical Analysis, 59(1):558–582, Jan 2021. 24

    work page 2021

  5. [5]

    Berrone, D

    S. Berrone, D. Grappein, and S. Scial` o. 3d-1d coupling on non conforming meshes via a three-field optimization based domain decomposition.Journal of Computational Physics, 448:110738, January 2022

    work page 2022

  6. [6]

    Berrone, C

    S. Berrone, C. Giverso, D. Grappein, L. Preziosi, and S. Scial` o. An opti- mization based 3d-1d coupling strategy for tissue perfusion and chemical transport during tumor-induced angiogenesis.Computers & Mathematics with Applications, 151:252–270, December 2023

    work page 2023

  7. [7]

    T. Koch, M. Schneider, R. Helmig, and P. Jenny. Modeling tissue perfu- sion in terms of 1d-3d embedded mixed-dimension coupled problems with distributed sources.Journal of Computational Physics, 410:109370, June 2020

    work page 2020

  8. [8]

    I. G. Gjerde, K. Kumar, and J. M. Nordbotten. A mixed approach to the poisson problem with line sources.SIAM Journal on Numerical Analysis, 59(2):1117–1139, January 2021

    work page 2021

Show all 26 references
  1. [9]

    T. Koch, H. Wu, and M. Schneider. Nonlinear mixed-dimension model for embedded tubular networks with application to root water uptake.Journal of Computational Physics, 450:110823, February 2022

    2022

  2. [10]

    Portillo and I

    D. Portillo and I. Romero. Embedding structures in continua: linear models and finite element discretizations.Computer Methods in Applied Mechanics and Engineering, 451:118683, 2026

    2026

  3. [11]

    Firmbach, I

    M. Firmbach, I. Steinbrecher, A. Popp, and M. Mayr. Computational challenges in mixed-dimensional beam/solid coupling.PAMM, 23(1), May 2023

    2023

  4. [12]

    Steinbrecher, A

    I. Steinbrecher, A. Popp, and C. Meier. Consistent coupling of positions and rotations for embedding 1d cosserat beams into 3d solid volumes.Com- putational Mechanics, 69(3):701–732, Nov 2022

    2022

  5. [13]

    Steinbrecher, M

    I. Steinbrecher, M. Mayr, M. J. Grill, J. Kremheller, C. Meier, and A. Popp. A mortar-type finite element approach for embedding 1d beams into 3d solid volumes.Computational Mechanics, 66(6):1377–1398, September 2020

    2020

  6. [14]

    A. Sky, J. Hale, A. Zilian, and S. Bordas. Intrinsic mixed-dimensional beam-shell-solid couplings in linear Cosserat continua via tangential differ- ential calculus.Computer Methods in Applied Mechanics and Engineering, 432:117384, December 2024

    2024

  7. [15]

    Hansbo and M

    P. Hansbo and M. Larson. Nitsche’s finite element method for model cou- pling in elasticity.Computer Methods in Applied Mechanics and Engineer- ing, 392:114707, March 2022

    2022

  8. [16]

    H. B. Dhia and G. Rateau. Analyse math´ ematique de la m´ ethode Arlequin mixte.Comptes Rendus de l’Acad´ emie des Sciences - Series I - Mathemat- ics, 332(7):649–654, Apr 2001

    2001

  9. [17]

    H. B. Dhia and G. Rateau. The Arlequin method as a flexible engineering design tool.International Journal for Numerical Methods in Engineering, 62(11):1442–1462, 2005. 25

    2005

  10. [18]

    Qiao, Q.D

    H. Qiao, Q.D. Yang, W.Q. Chen, and C.Z. Zhang. Implementation of the arlequin method into abaqus: Basic formulations and applications.Ad- vances in Engineering Software, 42(4):197–207, Apr 2011

    2011

  11. [19]

    Hackbusch.Elliptic differential equations, volume 18 ofSpringer series in computational mathematics

    W. Hackbusch.Elliptic differential equations, volume 18 ofSpringer series in computational mathematics. Springer, Berlin, 1992

    1992

  12. [20]

    L. C. Evans.Partial differential equations. AMS Press, 1999

    1999

  13. [21]

    Brezzi and M

    F. Brezzi and M. Fortin.Mixed and hybrid finite element methods. Springer, Berlin, 1991

    1991

  14. [22]

    Boffi, F

    D. Boffi, F. Brezzi, and M. Fortin.Mixed Finite Element Methods and Applications, volume 44 ofSpringer Series in Computational Mathematics. Springer Science & Business Media, Berlin, Heidelberg, July 2013

    2013

  15. [23]

    Chapelle and K

    D. Chapelle and K. J. Bathe. The inf-sup test.Computers & Structures, 1993

    1993

  16. [24]

    K J. Bathe. The inf-sup condition and its evaluation for mixed finite ele- ment methods.Computers & Structures, 79:243–252, May 2013

    2013

  17. [25]

    Runions, M

    A. Runions, M. Fuhrer, B. Lane, P. Federl, A. Rolland-Lagan, and P. Prusinkiewicz. Modeling and visualization of leaf venation patterns. InACM SIGGRAPH 2005 Papers, pages 702–711. ACM, 2005

    2005

  18. [26]

    embedded

    A. Runions, B. Lane, and P. Prusinkiewicz. Modeling trees with a space colonization algorithm. In David Ebert and Stephane Merillou, editors, Eurographics Workshop on Natural Phenomena, pages 63–70. The Euro- graphics Association, 2007. A Implementation details We provide in t...

    2007

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.