Pith. sign in

REVIEW 4 major objections 4 minor 75 references

Physics-informed solution reconstruction in elasticity and heat transfer using the explicit constraint force method

T0 review · 4 major / 4 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read The paper claims that invisible 'constraint forces' introduced by data constraints—not the physics—can dominate PINN reconstructions, and that making those forces explicit and minimal yields interpretable, robust, data-consistent…

desk verdict Clear and useful diagnosis of why PINN solution reconstruction fails under model misspecification, plus a practical fix that deserves a serious referee — but the optimality principle is normative and one uniqueness claim is over-stated. read the letter →

arxiv 2505.04875 v1 pith:Q542443J submitted 2025-05-08 cs.CE cs.LG

Conor Rowan , Kurt Maute , Alireza Doostan This is my paper · ORCID
classification cs.CEcs.LG
keywords solutionreconstructionphysics-informedneuralnetworksconstraintforcesminimumforceprinciplemodelmisspecificationdataassimilationelasticityheattransfer
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 studies solution reconstruction—recovering a full field from sparse measurements—when the parameterized governing equation does not match the physics that generated the data. It argues that standard physics-informed neural network formulations fail three basic requirements: the reconstruction offers no physically interpretable quality measure, it depends on arbitrary choices of physics loss and constraint enforcement, and flexible discrepancy models can make the recovered parameters non-unique. The paper identifies 'constraint forces'—additional source terms that constraints inject into the system—as the hidden cause, and shows how their spatial form is dictated by the loss formulation. It proposes the explicit constraint force method (ECFM), which adds analyst-controlled source terms at measurement locations and selects physics parameters by minimizing the squared norm of the constraint forces. If correct, ECFM makes reconstructions predictable, customizable, and interpretable even when the model is misspecified, which matters for digital-twin and structural-health-monitoring settings.

What carries the argument

The central object is the 'constraint force': a source term that enters the governing equation solely to enforce measurement constraints, with magnitude carried by Lagrange multipliers or penalty terms in standard PINN formulations. ECFM replaces those implicit forces with explicit ones: chosen functions $\Gamma(x-x_i)$ centered at the measurements, scaled by coefficients $\lambda_i$, and an inner-loop system $R(\theta, \lambda | \epsilon)=0$ that determines solution parameters and force magnitudes for any physics parameters $\epsilon$. The outer-loop objective is the minimum constraint force principle, $\arg\min_\epsilon \tfrac{1}{2}\lambda:H\lambda$, where $H$ is the Gram matrix of the chosen force shapes; the value of this objective is the interpretable quality measure. The analysis shows that the strong-form loss implicitly minimizes constraint force, while the energy loss does not, which is why energy cannot be used directly as a reconstruction loss.

What would settle it

Construct a synthetic system with a known source term that is deliberately non-minimal in constraint-force norm, such as a large localized force away from the measurement locations, then run ECFM with smooth hat or radial-basis constraint forces. If the method reports a near-zero total constraint force and a confident reconstruction while the held-out field error is large, or if the recovered physics parameter is farther from the data-generating parameter than one obtained by directly minimizing data error, then the minimum constraint force principle is not selecting the objectively better reconstruction.

Watch

Extended reading notes

Core claim

The paper's central claim is that whenever a data constraint is enforced by adding a term to a physics-loss objective that cannot be driven to zero, the constraint injects a fictitious source term—a 'constraint force'—whose spatial distribution is determined by the choice of physics loss (strong form, weak form, or energy) and constraint method (penalty or Lagrange multiplier). That hidden term, not the physics, can dominate the reconstruction when the parameterized model is inconsistent with the true system. The proposed remedy is the explicit constraint force method (ECFM): introduce source terms $\Gamma(x-x_i)$ of analyst-chosen shape, centered at measurement points, with unknown magnitudes $\lambda_i$; for fixed physics parameters $\epsilon$, solve the constrained system $R(\theta, \lambda | \epsilon)=0$; then choose $\epsilon$ to minimize $\tfrac{1}{2}\lambda:H\lambda$, the integrated squared magnitude of the constraint forces. The paper argues that this minimum constraint force principle is what the strong-form loss enforces implicitly, and that ECFM makes reconstructions robust to the choice of loss and constraint method, gives a quality measure in physical units (force or heat flux), and matches model freedom to the number of measurements so parameters remain identifiable.

Load-bearing premise

That the 'right' physics parameters are the ones minimizing the squared magnitude of the fictitious source terms needed to satisfy the measurements; this principle is imposed as an optimality criterion rather than derived from data or from an independent definition of reconstruction error.

Editorial extensions

If this is right

  • Standard PINN reconstructions can be silently dominated by constraint forces rather than physics whenever the parameterized model cannot reproduce the data, so similar-looking formulations can yield different answers.
  • Reconstructions from ECFM agree across strong-form, weak-form, and energy inner-loop formulations, because the constraint force is fixed explicitly instead of being determined by the loss.
  • The minimized total constraint force gives an interpretable quality metric in source-term units, letting analysts tell whether the model is misspecified and by how much.
  • Matching constraint-force degrees of freedom to the number of measurements avoids the non-uniqueness that arises when discrepancy models have more parameters than constraints.
  • In smooth-misspecification settings ECFM and standard PINNs give similar reconstructions, so ECFM is most needed when missing physics is localized or when a quality measure is required.

Reading between the lines

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

  • The minimum constraint force principle can be read as a prior that favors explanations requiring the least fictitious forcing; if so, ECFM's recovered parameters inherit the biases of that prior, and the choice of $\Gamma(x-x_i)$ is a modeling decision that could be tuned.
  • The constraint-force distribution itself could serve as an estimator of the missing physics term; a testable extension is to feed the recovered force field back as an updated source-term parameterization and check whether the reconstruction improves.
  • For time-dependent problems the extension is not automatic: constraint-force magnitudes would become functions of time, and the minimum principle would need a temporal norm, which may change identifiability and the form of the inner-loop system.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 4 minor

Summary. The manuscript addresses solution reconstruction—estimating a full-field state from sparse measurements using parameterized governing equations—under model misspecification. It proposes three criteria for reconstruction methods (interpretability, robustness to numerical formulation, and data consistency), argues that standard PINN penalty and Lagrange-multiplier formulations violate them, and identifies induced 'constraint forces' as the mechanism. Section 5 derives constraint-force forms for strong-form and energy losses on a 1D model problem. Section 6 introduces ECFM, which appends analyst-chosen source terms Γ(x−x_i) to the governing equation and selects the physics parameters ε by minimizing (1/2)λ:Hλ subject to the inner-loop residual system R(θ,λ|ε)=0. Sections 7.1–7.3 present 1D elastic, 1D hyperelastic, and 2D heat-conduction examples, including comparisons with standard PINN formulations.

Significance. The Section 5 analysis is the paper's strongest contribution: it cleanly shows on a 1D model problem that Lagrange multipliers act as distributed hat forces under strong-form losses and point forces under energy losses, and that minimizing the strong-form loss is equivalent to minimizing the constraint force (Eq. (32)). This gives a concrete, mechanistic explanation of loss-dependence and motivates ECFM. The ECFM quality measure z(ε) is physically interpretable, and the numerical examples support the claims of predictability and customizability. The main caveat is that the minimum constraint force principle is a normative selection rule, not a derived optimality criterion; the paper should present it as such and test it against independent criteria. The 1D derivations are self-contained, and the reported examples are simple enough to be reproduced without external code.

major comments (4)
  1. [Section 6, Eq. (39)] The selection rule ε* = argmin_ε (1/2)λ:Hλ subject to R(θ,λ|ε)=0 is imposed as the 'minimum constraint force principle' rather than derived. Section 2 promises that different notions of optimality 'will be discussed below,' but the paper never returns to this point; consequently, the claims that ECFM reconstructions are 'optimal' and that z(ε) is an objective quality measure are self-referential with respect to this principle. Please add a benchmark comparison on the Section 7 examples in which ε_ECFM is compared with the ε minimizing an independent criterion (e.g., L2 error to a synthetic truth, cross-validated prediction error, or likelihood under a discrepancy model), and explicitly state conditions under which the minimum constraint force principle coincides with those criteria. If no such coincidence is claimed, the wording should be softened to describe the principle as a design choice.
  2. [Section 6, Eq. (38)] The text states that Eq. (38) 'always has a unique solution' and that the systems in Eqs. (41)–(43) are 'square by construction and therefore avoid issues with non-uniqueness.' Squareness alone does not imply invertibility: overlapping supports of Γ(x−x_i), collinear measurement rows, or symmetric measurement configurations can make the G block or the Schur complement rank-deficient. Section 4.3 also explicitly disclaims that respecting data consistency guarantees identifiability, which conflicts with the Section 6 claim. Please state sufficient conditions (e.g., distinct constraint locations, linearly independent Γ(·−x_i), invertible Schur complement) and verify them for the examples, or soften the uniqueness claim to 'the system is square and is nonsingular in the reported examples.'
  3. [Introduction and Section 4.2] Contribution 1 states that 'the variational form of a PDE cannot be used as a loss function,' and Section 4.2 repeats that the energy 'cannot be used as a loss function to reconstruct the solution in the same way that the strong form can.' This categorical claim is qualified by the paper's own later results: Section 5, Eq. (34) uses the energy with the minimum constraint force principle, and Section 7.1 notes that the weak-form and energy systems coincide. Please rephrase the claim as 'the energy cannot be used as a standalone joint objective over θ and ε without an additional selection criterion for ε' and adjust the contribution list accordingly.
  4. [Section 7.3, Eq. (53)] The text reports a maximum constraint violation of 0.002 'with no need to select a penalty hyperparameter,' but the inner-loop problem in Eq. (53) contains the penalty weight λ′_d, which is set to 1000 in the experiment. This weight is a free parameter and can affect the recovered λ(ε) and therefore the outer-loop objective. Please report the value of λ′_d in the experiment, study the sensitivity of the results to it, and correct the sentence; the claimed advantage over PINNs is not the absence of a penalty parameter but that the inner-loop objective can be driven to zero.
minor comments (4)
  1. [Section 5, Eqs. (30)–(32)] There is a sign/typographical inconsistency in the hat-function expression: Eq. (30) contains λ(I(x−x_c)−(1−x_c)x), while Eqs. (31) and (32) omit the factor x in the second term. The derivation through integration by parts supports the form in Eq. (30), so Eqs. (31) and (32) should be corrected.
  2. [Section 4.1] The text says 'The domain here is Ω ∈ [0, 1]'; this should read 'Ω = [0, 1].'
  3. [Appendix D, Eq. (Appendix D1)] The notation ℓ2(X; ǫ1) appears in the augmented residual system, but in Section 7.2 the Lamé parameter ℓ2 is not parameterized by ǫ; please harmonize the notation or clarify the dependence.
  4. [Section 7.3, Eq. (55)] The displayed equation for R_PINN has an unbalanced parenthesis: R_PINN(x) = −(∇²ŵ(x) + s(x), ...; please add the missing closing parenthesis.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: ECFM's objective is a disclosed normative principle, and the constraint-force equivalence is derived rather than assumed.

full rationale

The central derivation is in Section 5, where the strong-form loss is shown to be equivalent to minimizing the squared Lagrange-multiplier (constraint-force) magnitude: Eq. (32) follows from the stationarity conditions of Eq. (29), not from assuming the principle. ECFM (Eq. 39) adopts minimum constraint force as an explicitly proposed, postulational criterion (Contribution 4; Conclusion: 'we ... postulate the minimum constraint force as the notion of an optimal reconstruction from the measurement data'), so its optimality is conditional on that disclosure rather than a hidden input. The interpretability and data-consistency criteria are design desiderata; showing that ECFM satisfies them by construction is a property of the method, not an equation reduced to an input. The only self-citation, [46], is a standard reference for Lagrange-multiplier constraint enforcement and is not load-bearing. The assertion that squareness 'ensures Eq. (38) always has a unique solution' is not proven and is an identifiability/correctness concern, not a circularity. No fitted parameter is renamed as a prediction, and no derivation step equates its conclusion to its assumptions by definition.

Assumptions & free parameters 2 free parameters · 6 assumptions · 1 invented entities

The ledger shows that ECFM's central claims rest on several stated domain assumptions plus one normative principle (minimum constraint force). The main new freedom is the analyst-chosen constraint force basis, which is not learned from data. The physics parameters themselves are the intended outputs of the reconstruction, not additional free parameters introduced by the method.

free parameters (2)
  • Constraint force basis functions Gamma(x-xi) = analyst-chosen: piecewise linear hats (1D), radial basis with p=25 (2D), wide hats with p=20 (hyperelastic)
    The shape of the source terms used to enforce constraints is not learned or derived. It controls the smoothness of the reconstruction, and Section 8 admits little guidance exists for choosing it.
  • Inner-loop penalty weight lambda'_d = 1000
    Used in the 2D neural network example (Eq. 53) to balance residual and data mismatch. The paper says the value is not critical, but it is still a user-chosen hyperparameter.
assumptions (6)
  • domain assumption True system obeys a time-independent BVP with known Dirichlet boundaries and known domain geometry.
    Section 2, Eq. (1); the entire reconstruction setup assumes this.
  • domain assumption Parameterized operator G and source b are known up to constant parameters epsilon, and may be misspecified.
    Section 2, Eq. (3); this is the model-misspecification setting studied.
  • domain assumption Measurement noise is symmetric with finite variance, and thresholds are set via alpha.
    Section 2, Eq. (2); used to convert stochastic constraints into deterministic inequalities.
  • ad hoc to paper Minimum constraint force principle: optimal epsilon minimizes (1/2) lambda:H lambda.
    Sections 5 and 6, Eq. (37). The optimal physics parameters are defined as minimizers of this norm; it is a normative criterion, not derived from data or an independent error metric.
  • ad hoc to paper Constraint force basis Gamma is selected by the analyst.
    Section 6, Eq. (35). The forms (hats, RBFs) and widths are not implied by the measured system.
  • domain assumption Square inner-loop system Eq. (38) is nonsingular for the chosen discretizations.
    Section 6 after Eq. (39); the authors assert uniqueness but provide no rank proof for general Gamma and measurement layouts.
invented entities (1)
  • Explicit constraint force source terms Gamma(x-xi)
    purpose: Enforce data constraints and absorb model misspecification with controllable, analyst-chosen spatial support.
    No independent measurement identifies the spatial form or magnitude of these forces; they are constructs of the method, though they carry physical units. The paper acknowledges their choice is arbitrary in Section 8.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Physics-informed solution reconstruction in elasticity and heat transfer using the explicit constraint force method." pith.science (2026). https://pith.science/paper/Q542443J

@misc{pith2026250504875,
  author       = {Pith},
  title        = {Pith review of: Physics-informed solution reconstruction in elasticity and heat transfer using the explicit constraint force method},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/Q542443J}},
  note         = {Machine review of arXiv:2505.04875}
}
read the original abstract

One use case of ``physics-informed neural networks'' (PINNs) is solution reconstruction, which aims to estimate the full-field state of a physical system from sparse measurements. Parameterized governing equations of the system are used in tandem with the measurements to regularize the regression problem. However, in real-world solution reconstruction problems, the parameterized governing equation may be inconsistent with the physical phenomena that give rise to the measurement data. We show that due to assuming consistency between the true and parameterized physics, PINNs-based approaches may fail to satisfy three basic criteria of interpretability, robustness, and data consistency. As we argue, these criteria ensure that (i) the quality of the reconstruction can be assessed, (ii) the reconstruction does not depend strongly on the choice of physics loss, and (iii) that in certain situations, the physics parameters can be uniquely recovered. In the context of elasticity and heat transfer, we demonstrate how standard formulations of the physics loss and techniques for constraining the solution to respect the measurement data lead to different ``constraint forces" -- which we define as additional source terms arising from the constraints -- and that these constraint forces can significantly influence the reconstructed solution. To avoid the potentially substantial influence of the choice of physics loss and method of constraint enforcement on the reconstructed solution, we propose the ``explicit constraint force method'' (ECFM) to gain control of the source term introduced by the constraint. We then show that by satisfying the criteria of interpretability, robustness, and data consistency, this approach leads to more predictable and customizable reconstructions from noisy measurement data, even when the parameterization of the missing physics is inconsistent with the measured system.

Figures

Figures reproduced from arXiv: 2505.04875 by the authors.

Figure 1
Figure 1. Different techniques rely on different amounts of da [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Measurement data for the 1D linearly elastic BVP (l [PITH_FULL_IMAGE:figures/full_fig_p010_2.png] view at source ↗
Figure 3
Figure 3. In this example, minimizing the energy (Π) ˆ over ǫ will lead to poor interpolations (left), whereas the minimum strong form loss (which we call Z, to distinguish it from the loss involving the strong form and penalty in Eq. (8)), is obtained when the physics parameter is correct, leading to good interpolations (right). It is interesting to note that energy and strong form loss produce very different interpolations … view at source ↗
Figures from the paper (12 more)
Figure 4
Figure 4. Figure 4: For the 1D second-order linear model problem with a single unknown physics parameter and a single constraint, we have shown that Lagrange multipliers behave like constraint forces with a spatial form dictated by the choice of physics loss. It is also possible to back o…
Figure 4
Figure 4. Figure 4: Remedying the issues with the energy objective by u [PITH_FULL_IMAGE:figures/full_fig_p016_4.png]
Figure 5
Figure 5. Figure 5: Comparing the performance of the strong and weak fo [PITH_FULL_IMAGE:figures/full_fig_p020_5.png]
Figure 6
Figure 6. Figure 6: Comparing the performance of the strong and weak fo [PITH_FULL_IMAGE:figures/full_fig_p020_6.png]
Figure 7
Figure 7. Figure 7: Constraint forces using the energy loss are point fo [PITH_FULL_IMAGE:figures/full_fig_p021_7.png]
Figure 8
Figure 8. Figure 8: Comparing the performance of the strong and weak fo [PITH_FULL_IMAGE:figures/full_fig_p021_8.png]
Figure 9
Figure 9. Figure 9: A sample of measurement data to use in the solution r [PITH_FULL_IMAGE:figures/full_fig_p023_9.png]
Figure 10
Figure 10. Figure 10: By controlling the constraint force, the ECFM lea [PITH_FULL_IMAGE:figures/full_fig_p024_10.png]
Figure 11
Figure 11. Figure 11: Temperature field computed using a neural network [PITH_FULL_IMAGE:figures/full_fig_p025_11.png]
Figure 12
Figure 12. Figure 12: Radial basis constraint force functions Γ(x − xi) for C = 9 constraints. Here, the width is given by p = 25. argmin θ,ǫ 1 2 Z  ∇ · (A˜ (x; ǫ)∇wˆ) + s(x) 2 dΩ + λd 2 X C i=1  wˆ(xi ; θ) − vi 2 . (54) 26 [PITH_FULL_IMAGE:figures/full_fig_p026_12.png]
Figure 13
Figure 13. Figure 13: Reconstructing the solution with the strong form [PITH_FULL_IMAGE:figures/full_fig_p027_13.png]
Figure 14
Figure 14. Figure 14: When using ECFM, the source terms introduced to en [PITH_FULL_IMAGE:figures/full_fig_p028_14.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

75 extracted references · 47 canonical work pages

  1. [1]

    Abueidda, Seid Koric, Rashid Abu Al-Rub, Corey M

    Diab W. Abueidda, Seid Koric, Rashid Abu Al-Rub, Corey M. Parrott, Kai A. James, and Nahil A. Sobh. A deep learning energy method for hyperelasticity and viscoe lasticity. European Journal of Mechanics - A/Solids , 95:104639, September 2022. arXiv:2201.08690 [cs]

    arXiv 2022

  2. [2]

    Jagtap Ameya D

    Ameya D. Jagtap Ameya D. Jagtap and George Em Karniadakis George Em Karniadakis. Extended Physics- Informed Neural Networks (XPINNs): A Generalized Space-Ti me Domain Decomposition Based Deep Learning Framework for Nonlinear Partial Differential Equations. Communications in Computational Physics , 28(5):2002– 2041, January 2020

    work page 2002

  3. [3]

    Anagnostopoulos, Juan Diego Toscano, Nikol aos Stergiopulos, and George Em Karniadakis

    Sokratis J. Anagnostopoulos, Juan Diego Toscano, Nikol aos Stergiopulos, and George Em Karniadakis. Residual- based attention in physics-informed neural networks. Computer Methods in Applied Mechanics and Engineering , 421:116805, March 2024. 33

    work page 2024

  4. [4]

    Mininni, Martín Obligado, and Patricio Clark Di Leoni

    Sofía Angriman, Pablo Cobelli, Pablo D. Mininni, Martín Obligado, and Patricio Clark Di Leoni. Assimilation of statistical data into turbulent flows using physics-info rmed neural networks. The European Physical Journal E , 46(3):13, March 2023

    work page 2023

  5. [5]

    Neural Operators for Accelerating Scientific Simula tions and Design, January 2024

    Kamyar Azizzadenesheli, Nikola Kovachki, Zongyi Li, Mi guel Liu-Schiaffini, Jean Kossaifi, and Anima Anand- kumar. Neural Operators for Accelerating Scientific Simula tions and Design, January 2024. arXiv:2309.15325 [cs]

    arXiv 2024

  6. [6]

    Physics and equalit y constrained artificial neural networks: Application to forward and inverse problems with multi-fidelity data fus ion

    Shamsulhaq Basir and Inanc Senocak. Physics and equalit y constrained artificial neural networks: Application to forward and inverse problems with multi-fidelity data fus ion. Journal of Computational Physics , 463:111301, August 2022

    work page 2022

  7. [7]

    Berrone, C

    S. Berrone, C. Canuto, M. Pintore, and N. Sukumar. Enforc ing Dirichlet boundary conditions in physics-informed neural networks and variational physics-informed neural n etworks. Heliyon, 9(8):e18820, August 2023

    work page 2023

  8. [8]

    Stress field prediction in fiber-reinforced composite materials using a deep learning approach

    Anindya Bhaduri, Ashwini Gupta, and Lori Graham-Brady. Stress field prediction in fiber-reinforced composite materials using a deep learning approach. Composites Part B: Engineering , 238:109879, June 2022

    2022

Show all 75 references
  1. [9]

    Javier Bonet and Richard D. Wood. Nonlinear Continuum Mechanics for Finite Element Analysis . Cambridge University Press, Cambridge, 2 edition, 2008

    2008

  2. [10]

    Physics-Informed Neural Networks for Heat Transfer Problems

    Shengze Cai, Zhicheng Wang, Sifan Wang, Paris Perdikar is, and George Em Karniadakis. Physics-Informed Neural Networks for Heat Transfer Problems. Journal of Heat Transfer , 143(060801), April 2021

    2021

  3. [11]

    D eep least-squares methods: an unsupervised learning- based numerical method for solving elliptic PDEs

    Zhiqiang Cai, Jingshuang Chen, Min Liu, and Xinyu Liu. D eep least-squares methods: an unsupervised learning- based numerical method for solving elliptic PDEs. Journal of Computational Physics , 420:109707, November 2020

    2020

  4. [12]

    Yifan Du and Tamer A. Zaki. Evolutional Deep Neural Netw ork. Physical Review E , 104(4):045303, October 2021. arXiv:2103.09959 [physics]

    2021 arXiv

  5. [13]

    The Deep Ritz method: A deep learnin g-based numerical algorithm for solving variational problems, September 2017

    Weinan E and Bing Yu. The Deep Ritz method: A deep learnin g-based numerical algorithm for solving variational problems, September 2017. arXiv:1710.00211 [cs]

    2017 arXiv

  6. [14]

    Wagner, Annamaria Scherzl, Ma rco Klein, Norbert Hoffmann, and Merten Stender

    Svenja Ehlers, Niklas A. Wagner, Annamaria Scherzl, Ma rco Klein, Norbert Hoffmann, and Merten Stender. Data Assimilation and Parameter Identification for Water Wa ves Using the Nonlinear Schrödinger Equation and Physics-Informed Neural Networks. Fluids, 9(10):231, October 2024

    2024

  7. [15]

    Digital twins in mec hanical and aerospace engineering

    Alberto Ferrari and Karen Willcox. Digital twins in mec hanical and aerospace engineering. Nature Computational Science, 4(3):178–183, March 2024. Publisher: Nature Publishing G roup

    2024

  8. [16]

    Zahr, and Jian-Xun Wang

    Han Gao, Matthew J. Zahr, and Jian-Xun Wang. Physics-in formed graph neural Galerkin networks: A unified framework for solving PDE-governed forward and inverse pro blems. Computer Methods in Applied Mechanics and Engineering, 390:114502, February 2022

    2022

  9. [17]

    Jimack, and René de Bor st

    Yousef Ghaffari Motlagh, Peter K. Jimack, and René de Bor st. Deep learning phase-field model for brittle fractures. International Journal for Numerical Methods in Engineering , 124(3):620–638, 2023

    2023

  10. [18]

    Tartakovsky

    QiZhi He, David Barajas-Solano, Guzel Tartakovsky, an d Alexandre M. Tartakovsky. Physics-informed neural networks for multiphysics data assimilation with applicat ion to subsurface transport. Advances in Water Resources, 141:103610, July 2020

    2020

  11. [19]

    Flow field rec onstruction from sparse sensor measurements with physics-informed neural networks

    Mohammad Yasin Hosseini and Yousef Shiri. Flow field rec onstruction from sparse sensor measurements with physics-informed neural networks. Physics of Fluids , 36(7):073606, July 2024

    2024

  12. [20]

    Jagtap, George Em Karniadakis, and Kenji Kawaguchi

    Zheyuan Hu, Ameya D. Jagtap, George Em Karniadakis, and Kenji Kawaguchi. When Do Extended Physics- Informed Neural Networks (XPINNs) Improve Generalization ? SIAM Journal on Scientific Computing , 44(5):A3158–A3182, October 2022. arXiv:2109.09444 [cs]

    2022 arXiv

  13. [21]

    Jagtap, Ehsan Kharazmi, and George Em Karniada kis

    Ameya D. Jagtap, Ehsan Kharazmi, and George Em Karniada kis. Conservative physics-informed neural networks on discrete domains for conservation laws: Applications to forward and inverse problems. Computer Methods in Applied Mechanics and Engineering , 365:113028, June 2020. 34

    2020

  14. [22]

    NSFnets (Navier-Stokes Flow nets): Physics- informed neural networks for the incompressible Navier-St okes equations

    Xiaowei Jin, Shengze Cai, Hui Li, and George Em Karniada kis. NSFnets (Navier-Stokes Flow nets): Physics- informed neural networks for the incompressible Navier-St okes equations. Journal of Computational Physics , 426:109951, February 2021

    2021

  15. [23]

    Physics-informed ne ural network for modeling dynamic linear elasticity, January 2024

    Vijay Kag and Venkatesh Gopinath. Physics-informed ne ural network for modeling dynamic linear elasticity, January 2024. arXiv:2312.15175 [cs]

    2024

  16. [24]

    Kharazmi, Z

    E. Kharazmi, Z. Zhang, and G. E. Karniadakis. Variation al Physics-Informed Neural Networks For Solving Partial Differential Equations, November 2019

    2019

  17. [25]

    hp-VPINNs: Variational Physics-Informed Neural Networks With Domain Decomposition

    Ehsan Kharazmi, Zhongqiang Zhang, and George Em Karnia dakis. hp-VPINNs: Variational Physics-Informed Neural Networks With Domain Decomposition. Computer Methods in Applied Mechanics and Engineering , 374:113547, February 2021

    2021

  18. [26]

    Zavlanos

    Reza Khodayi-Mehr and Michael M. Zavlanos. VarNet: Var iational Neural Networks for the Solution of Partial Differential Equations, December 2019. arXiv:1912.07443 [ cs]

    2019 arXiv

  19. [27]

    Review of Stru ctural Health Monitoring Methods Regarding a Multi- Sensor Approach for Damage Assessment of Metal and Composit e Structures

    Christoph Kralovec and Martin Schagerl. Review of Stru ctural Health Monitoring Methods Regarding a Multi- Sensor Approach for Damage Assessment of Metal and Composit e Structures. Sensors (Basel, Switzerland) , 20(3):826, February 2020

    2020

  20. [28]

    Krishnapriyan, Amir Gholami, Shandian Zhe, Ro bert M

    Aditi S. Krishnapriyan, Amir Gholami, Shandian Zhe, Ro bert M. Kirby, and Michael W. Mahoney. Characterizing possible failure modes in physics-informed neural network s, November 2021. arXiv:2109.01050 [cs]

    2021 arXiv

  21. [29]

    I. E. Lagaris, A. C. Likas, and D. G. Papageorgiou. Neura l-network methods for boundary value problems with irregular boundaries. IEEE transactions on neural networks , 11(5):1041–1049, 2000

    2000

  22. [30]

    Lagaris, A

    I.E. Lagaris, A. Likas, and D.I. Fotiadis. Artificial ne ural networks for solving ordinary and partial differential equations. IEEE Transactions on Neural Networks , 9(5):987–1000, September 1998. Conference Name: IEEE Transactions on Neural Networks

    1998

  23. [31]

    NH-PINN: N eural homogenization-based physics-informed neural network for multiscale problems

    Wing Tat Leung, Guang Lin, and Zecheng Zhang. NH-PINN: N eural homogenization-based physics-informed neural network for multiscale problems. Journal of Computational Physics , 470:111539, December 2022

    2022

  24. [32]

    Physical informed neural net- works with soft and hard boundary constraints for solving ad vection-diffusion equations using Fourier expansions, December 2023

    Xi’an Li, Jiaxin Deng, Jinran Wu, Shaotong Zhang, Weide Li, and You-Gan Wang. Physical informed neural net- works with soft and hard boundary constraints for solving ad vection-diffusion equations using Fourier expansions, December 2023

    2023

  25. [33]

    Fourier Neural Operator for Parametr ic Partial Differential Equations, May 2021

    Zongyi Li, Nikola Kovachki, Kamyar Azizzadenesheli, B urigede Liu, Kaushik Bhattacharya, Andrew Stuart, and Anima Anandkumar. Fourier Neural Operator for Parametr ic Partial Differential Equations, May 2021. arXiv:2010.08895 [cs]

    2021 arXiv

  26. [34]

    Deep Nitsche Method: Deep Ritz Method with Essential Boundary Conditions

    Yulei Liao and Pingbing Ming. Deep Nitsche Method: Deep Ritz Method with Essential Boundary Conditions. Communications in Computational Physics , 29(5):1365–1384, January 2021

    2021

  27. [35]

    Inve stigating deep energy method applications in thermoelasticity

    Kuan-Chung Lin, Kuo-Chou Wang, and Cheng-Hung Hu. Inve stigating deep energy method applications in thermoelasticity. Engineering Analysis with Boundary Elements , 159:302–314, February 2024

    2024

  28. [36]

    Deep Ritz met hod with adaptive quadrature for linear elasticity

    Min Liu, Zhiqiang Cai, and Karthik Ramani. Deep Ritz met hod with adaptive quadrature for linear elasticity. Computer Methods in Applied Mechanics and Engineering , 415:116229, October 2023

    2023

  29. [37]

    DeepONe t: Learning nonlinear operators for identifying differential equations based on the universal approximatio n theorem of operators

    Lu Lu, Pengzhan Jin, and George Em Karniadakis. DeepONe t: Learning nonlinear operators for identifying differential equations based on the universal approximatio n theorem of operators. Nature Machine Intelligence , 3(3):218–229, March 2021. arXiv:1910.03193 [cs]

    2021 arXiv

  30. [38]

    Maclaren and Ruanui Nicholson

    Oliver J. Maclaren and Ruanui Nicholson. What can be est imated? Identifiability, estimability, causal inference and ill-posed inverse problems, July 2020. arXiv:1904.028 26 [math]

    2020

  31. [39]

    Physics Informed Neural Networks for heat conduction with phase change, October 2024

    Bahae-Eddine Madir, Francky Luddens, Corentin Lothod é, and Ionut Danaila. Physics Informed Neural Networks for heat conduction with phase change, October 2024. arXiv: 2410.14216 [math]

    2024 arXiv

  32. [40]

    Makridakis, Aaron Pim, and Tristan Prye r

    Charalambos G. Makridakis, Aaron Pim, and Tristan Prye r. A Deep Uzawa-Lagrange Multiplier Approach for Boundary Conditions in PINNs and Deep Ritz Methods, Novembe r 2024. arXiv:2411.08702 [math]. 35

    2024 arXiv

  33. [41]

    Manav, R

    M. Manav, R. Molinaro, S. Mishra, and L. De Lorenzis. Pha se-field modeling of fracture with physics-informed deep learning. Computer Methods in Applied Mechanics and Engineering , 429:117104, September 2024

    2024

  34. [42]

    Arif Masud, Sharbel Nashar, and Shoaib A. Goraya. Physi cs-constrained Data-Driven Variational method for discrepancy modeling. Computer Methods in Applied Mechanics and Engineering , 417:116295, December 2023

    2023

  35. [43]

    Self-Adaptive P hysics-Informed Neural Networks using a Soft Attention Mechanism

    Levi McClenny and Ulisses Braga-Neto. Self-Adaptive P hysics-Informed Neural Networks using a Soft Attention Mechanism. Journal of Computational Physics , 474:111722, February 2023. arXiv:2009.04544 [cs]

    2023 arXiv

  36. [44]

    Messenger and David M

    Daniel A. Messenger and David M. Bortz. Weak SINDy For Pa rtial Differential Equations. Journal of Computa- tional Physics , 443:110525, October 2021. arXiv:2007.02848 [math]

    2021 arXiv

  37. [45]

    A deep energy method for finite deformation hyperelasticity

    Vien Minh Nguyen-Thanh, Xiaoying Zhuang, and Timon Rab czuk. A deep energy method for finite deformation hyperelasticity. European Journal of Mechanics - A/Solids , 80:103874, March 2020

    2020

  38. [46]

    Constrained or Uncon- strained? Neural-Network-Based Equation Discovery from D ata

    Grant Norman, Jacqueline Wentz, Hemanth Kolla, Kurt Ma ute, and Alireza Doostan. Constrained or Uncon- strained? Neural-Network-Based Equation Discovery from D ata. Computer Methods in Applied Mechanics and Engineering, 436:117684, March 2025. arXiv:2406.02581 [cs]

    2025 arXiv

  39. [47]

    Hwang, an d Beomjin Kwon

    Nagahiro Ohashi, Nam Phuong Nguyen, Leslie K. Hwang, an d Beomjin Kwon. Multiple scale method integrated physics-informed neural networks for reconstructing tran sient natural convection. Physics of Fluids , 36(12):123625, December 2024

    2024

  40. [48]

    Hamprecht, Yoshua Bengio, and Aaron Courville

    Nasim Rahaman, Aristide Baratin, Devansh Arpit, Felix Draxler, Min Lin, Fred A. Hamprecht, Yoshua Bengio, and Aaron Courville. On the Spectral Bias of Neural Networks , May 2019. arXiv:1806.08734 [stat]

    2019 arXiv

  41. [49]

    Raissi, P

    M. Raissi, P. Perdikaris, and G.E. Karniadakis. Physic s-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear p artial differential equations. Journal of Computational Physics, 378:686–707, February 2019

    2019

  42. [50]

    Hidden fluid mechanics: Learning velocity and pressure fields from flow visualizations

    Maziar Raissi, Alireza Yazdani, and George Em Karniada kis. Hidden fluid mechanics: Learning velocity and pressure fields from flow visualizations. Science (New York, N.Y.) , 367(6481):1026–1030, February 2020

    2020

  43. [51]

    En-Ze Rui, Zheng-Wei Chen, Yi-Qing Ni, Lei Yuan, and Gua ng-Zhi Zeng. Reconstruction of 3D flow field around a building model in wind tunnel: a novel physics-informed neu ral network framework adopting dynamic prioritization self-adaptive loss balance strategy. Engineering Appli...

    2023

  44. [52]

    Nonlinear Optimization

    Andrzej Ruszczynski. Nonlinear Optimization. Princeton University Press, 2006. ISBN: 9780691119151

    2006

  45. [53]

    Sahin, M

    T. Sahin, M. von Danwitz, and A. Popp. Solving Forward an d Inverse Problems of Contact Mechanics using Physics-Informed Neural Networks. Advanced Modeling and Simulation in Engineering Sciences , 11(1):11, May

  46. [54]

    Jakob G. R. von Saldern, Johann Moritz Reumschüssel, Th omas L. Kaiser, Moritz Sieber, and Kilian Oberleith- ner. Mean flow data assimilation based on physics-informed n eural networks. Physics of Fluids , 34(11):115129, November 2022. arXiv:2208.03109 [physics]

    2022 arXiv

  47. [55]

    Randomized Neural Networks wit h Petrov-Galerkin Methods for Solving Linear Elasticity Problems, August 2023

    Yong Shang and Fei Wang. Randomized Neural Networks wit h Petrov-Galerkin Methods for Solving Linear Elasticity Problems, August 2023. arXiv:2308.03088 [math ]

    2023 arXiv

  48. [56]

    Deep Petrov-Galer kin Method for Solving Partial Differential Equations, January 2022

    Yong Shang, Fei Wang, and Jingbo Sun. Deep Petrov-Galer kin Method for Solving Partial Differential Equations, January 2022. arXiv:2201.12995 [math]

    2022 arXiv

  49. [57]

    PFNN: A penalty-free neura l network method for solving a class of second-order boundary-value problems on complex geometries

    Hailong Sheng and Chao Yang. PFNN: A penalty-free neura l network method for solving a class of second-order boundary-value problems on complex geometries. Journal of Computational Physics , 428:110085, March 2021

    2021

  50. [58]

    DGM: A deep learning algorithm for solving partial differential equations

    Justin Sirignano and Konstantinos Spiliopoulos. DGM: A deep learning algorithm for solving partial differential equations. Journal of Computational Physics , 375:1339–1364, December 2018. arXiv:1708.07469 [q-fin]

    2018 arXiv

  51. [59]

    Mean flow reconstr uction of unsteady flows using physics-informed neural networks

    Lukasz Sliwinski and Georgios Rigas. Mean flow reconstr uction of unsteady flows using physics-informed neural networks. Data-Centric Engineering, 4:e4, January 2023

    2023

  52. [60]

    Enhanced physics-informed neural networks with Augmented Lagrangian relaxation method (AL-PINNs)

    Hwijae Son, Sung Woong Cho, and Hyung Ju Hwang. Enhanced physics-informed neural networks with Augmented Lagrangian relaxation method (AL-PINNs). Neurocomputing, 548:126424, September 2023. 36

    2023

  53. [61]

    Sukumar and Ankit Srivastava

    N. Sukumar and Ankit Srivastava. Exact imposition of bo undary conditions with distance functions in physics- informed deep neural networks. Computer Methods in Applied Mechanics and Engineering , 389:114333, February 2022

    2022

  54. [62]

    Fei Tao, He Zhang, Ang Liu, and A. Y. C. Nee. Digital Twin i n Industry: State-of-the-Art. IEEE Transactions on Industrial Informatics , 15(4):2405–2415, April 2019. Conference Name: IEEE Trans actions on Industrial Informatics

    2019

  55. [63]

    Digital twin modeling for structural strength moni- toring via transfer learning-based multi-source data fusi on

    Bo Wang, Zengcong Li, Ziyu Xu, Zhiyong Sun, and Kuo Tian. Digital twin modeling for structural strength moni- toring via transfer learning-based multi-source data fusi on. Mechanical Systems and Signal Processing , 200:110625, October 2023

    2023

  56. [64]

    Mo, Bassam Izzuddin, and Chul-Woo Kim

    Jiaji Wang, Y.L. Mo, Bassam Izzuddin, and Chul-Woo Kim. Exact Dirichlet boundary Physics-informed Neural Network EPINN for solid mechanics. Computer Methods in Applied Mechanics and Engineering , 414:116184, September 2023

    2023

  57. [65]

    Resp ecting causality for training physics-informed neural networks

    Sifan Wang, Shyam Sankaran, and Paris Perdikaris. Resp ecting causality for training physics-informed neural networks. Computer Methods in Applied Mechanics and Engineering , 421:116813, March 2024

    2024

  58. [66]

    Learnin g the solution operator of parametric partial differential equations with physics-informed DeepOnets, March 2021

    Sifan Wang, Hanwen Wang, and Paris Perdikaris. Learnin g the solution operator of parametric partial differential equations with physics-informed DeepOnets, March 2021. ar Xiv:2103.10974 [cs]

    2021 arXiv

  59. [67]

    Data fusion a pproaches for structural health monitoring and sys- tem identification: Past, present, and future

    Rih-Teng Wu and Mohammad Reza Jahanshahi. Data fusion a pproaches for structural health monitoring and sys- tem identification: Past, present, and future. Structural Health Monitoring, 19(2):552–586, March 2020. Publisher: SAGE Publications

    2020

  60. [68]

    Structu ral health monitoring for mechanical structures using multi-sensor data

    Jia-Wei Xiang, Zhi-Bo Yang, and Jose L Aguilar. Structu ral health monitoring for mechanical structures using multi-sensor data. International Journal of Distributed Sensor Networks , 14(9):1550147718802019, September

  61. [69]

    Barajas-Solano, and Alexandre M

    Yu-Hong Yeung, David A. Barajas-Solano, and Alexandre M. Tartakovsky. Physics-Informed Machine Learning Method for Large-Scale Data Assimilation Problems. Water Resources Research , 58(5):e2021WR031023, 2022. _eprint: https://onlinelibrary.wiley.com/doi/pdf/10. 1029/2021WR031023

    2022

  62. [70]

    Wea k Adversarial Networks for High-dimensional Partial Differential Equations

    Yaohua Zang, Gang Bao, Xiaojing Ye, and Haomin Zhou. Wea k Adversarial Networks for High-dimensional Partial Differential Equations. Journal of Computational Physics , 411:109409, June 2020. arXiv:1907.08272 [math]

    2020 arXiv

  63. [71]

    O perator Learning for Reconstructing Flow Fields from Sparse Measurements: an Energy Transformer Approach, January 2025

    Qian Zhang, Dmitry Krotov, and George Em Karniadakis. O perator Learning for Reconstructing Flow Fields from Sparse Measurements: an Energy Transformer Approach, January 2025. arXiv:2501.08339 [physics]

    2025 arXiv

  64. [72]

    Wei Zhou and Y. F. Xu. Data-Guided Physics-Informed Neu ral Networks for Solving Inverse Problems in Partial Differential Equations, July 2024. arXiv:2407.10836 [cs]

    2024 arXiv

  65. [73]

    Cor recting model misspecification in physics-informed neural networks (PINNs)

    Zongren Zou, Xuhui Meng, and George Em Karniadakis. Cor recting model misspecification in physics-informed neural networks (PINNs). Journal of Computational Physics , 505:112918, May 2024. 37

    2024

  66. [2018]

    Publisher: SAGE Publications

  67. [2024]

    arXiv:2308.12716 [math]

    arXiv

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.