Pith. sign in

REVIEW 2 major objections 4 minor 77 references

A Physics-Augmented Machine Learning Constitutive Model for Damage in Solids

T0 review · 2 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash

Pith's one-line read A physics-augmented neural network learns anisotropic damage from stress–strain data, reproducing direction-dependent softening in hydrogels.

desk verdict A genuinely new anisotropic damage architecture, undermined by a mismatch between the derived evolution laws and the final energy—worth review but not acceptance as is. read the letter →

arxiv 2508.05638 v1 pith:QNXAEYVH submitted 2025-07-23 physics.app-ph physics.data-an

classification physics.app-phphysics.data-an
keywords anisotropicdamagecontinuummechanicsdamage-inducedanisotropyinputconvexneuralnetworkspolyconvexitythermodynamicconsistencyMullinseffectdouble-networkhydrogels
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

Soft materials lose stiffness unevenly: cracks and broken polymer bonds leave a material softer in some directions than others, a family of behaviors known as damage-induced anisotropy. This paper proposes a machine-learning constitutive model that learns such direction-dependent degradation directly from stress–strain measurements. The central claim is that representing damage as a second-order tensor and writing the strain-energy density as a convex neural network of deformation and damage invariants yields a thermodynamically consistent, polyconvex model for compressible and incompressible orthotropic materials. On multiaxial experiments in double-network hydrogels, the anisotropic model reproduces equal-biaxial, unequal-biaxial, and planar responses where an isotropic damage model misses important stress components. If the claim holds, physics-safe learned constitutive equations of this type could replace hand-chosen degradation functions in simulations of soft-tissue and polymer failure.

What carries the argument

The load-bearing object is the free-energy representation $\bar\psi_e(C,L_i,\alpha_i,\alpha_0)$: an isotropic function of the right Cauchy–Green tensor $C$, structural tensors $L_i$ for material symmetry, and scalar amplitudes $\alpha_0,\alpha_i$ from the second-order damage tensor $D_{\rm aniso}=\sum_i \alpha_i L_i$. Polyconvex invariants $I_C=C:I$, $II_C=\mathrm{cof}(C):I$, $III_C=\det C$, $\tilde I^{(i)}=C:\tilde L^{(i)}$, $\tilde J^{(i)}=\mathrm{cof}(C):\tilde L^{(i)}$ feed non-decreasing input-convex neural networks, whose composition preserves convexity. The attenuation envelope $p(\alpha)$ is a trainable weighted sum of convex decreasing functions $(1-\alpha/a)^{q_j}$, and damage evolves through potentials $g_i=G_i(y_i)-G_i(r_i)$ with consistency and loading/unloading conditions. A two-stage training procedure first fits the unloading branches to learn the elastic energy, then fits damage evolution on the full cyclic data.

What would settle it

Train the model on a multiaxial protocol and then subject the material to non-proportional loading, such as uniaxial pre-stretch followed by shear at 45 degrees; if measurements of damage orientation (for example from X-ray scattering or birefringence) show the damage axes rotating away from the original symmetry axes, the co-axial assumption fails and the predicted stresses will diverge from the measured ones.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is a general data-driven anisotropic damage formulation. The Helmholtz free energy is written as an isotropic function of deformation invariants and structural tensors, with damage encoded by a second-order damage tensor whose principal directions are assumed to remain co-axial with the virgin material's symmetry axes. Non-decreasing input-convex neural networks parameterize the strain-energy potentials, so polyconvexity holds by construction; attenuation of energy under damage is a trainable weighted sum of convex decreasing functions $(1-\alpha/a)^{q_j}$, generalizing the classical $(1-d)$ degradation; and damage evolution is governed by conjugate thermodynamic forces through damage potentials with thresholds and KKT consistency conditions. Additional terms are included so that energy and stress vanish at the undeformed state, satisfying the normality condition. The authors demonstrate accurate recovery of synthetic isotropic, transversely isotropic, and compressible orthotropic responses, and show that on double-network hydrogel data the anisotropic model captures the dominant multiaxial behavior where an isotropic damage model falls short.

Load-bearing premise

The load-bearing premise, stated in Section 3.1, is that damage principal directions stay fixed and aligned with the material's original symmetry axes throughout loading; the paper also notes in Sections 2.6 and 2.7 that growth conditions are omitted and energy non-negativity is verified numerically rather than by construction.

Editorial extensions

If this is right

  • The same formulation covers incompressible and compressible orthotropic materials, so one architecture can model gels, elastomers, and soft tissue without switching constitutive forms.
  • Because polyconvexity and thermodynamic consistency are enforced by construction, the learned strain-energy function can be used directly in finite-element simulations of failure.
  • The nonlinear convex decreasing attenuation functions replace the classical $(1-d)$ degradation, so the model can represent richer damage evolution while keeping energy non-negative in the demonstrated range.
  • The decoupled training strategy learns elastic response from unloading branches and damage evolution from full cycles, reducing the cost of fitting path-dependent models and improving scalability.
  • For the double-network hydrogel validation, the anisotropic model reproduces the dominant multiaxial response in equal-biaxial, unequal-biaxial, and planar tests, whereas the isotropic model misses key stress components.

Reading between the lines

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

  • An implicit extension: adding independent structural tensors for damage axes would let the same architecture track rotating damage fabrics under shear or non-proportional loading, relaxing the co-axial assumption.
  • Because the authors note that multiple internal states can fit the same stress–strain data, pairing the model with damage-orientation measurements could disambiguate the hidden state and make predictions more physical.
  • The inclusion of $III_C$ in the potentials means volumetric damage is representable, pointing toward applications in rocks and concrete, though the reported examples are dominated by deviatoric loading.
  • A natural test of extrapolation is to train on one loading mode and predict another, such as training on uniaxial and planar data and testing on pure shear; the paper's experiments do not isolate this.
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

2 major / 4 minor

Summary. The manuscript proposes a data-driven constitutive framework for anisotropic damage in solids, built around a second-order damage tensor, an ICNN-parameterized strain energy with polyconvex invariant inputs, convex decreasing attenuation functions, a damage potential with KKT conditions, and additional normality-correction terms. Damage evolution is formulated through conjugate thermodynamic forces, and a decoupled training scheme is introduced to separate unloading-branch energy fitting from damage-evolution fitting. The framework is demonstrated on synthetic incompressible isotropic, transversely isotropic, and compressible orthotropic benchmarks, and on experimental double-network hydrogel data under multiaxial loading, where the anisotropic model is reported to outperform an isotropic damage model.

Significance. If the theoretical claims are fully established, this would be a useful contribution: it combines several nontrivial constraints (objectivity, polyconvexity, normality, and thermodynamic consistency) in a trainable constitutive model and demonstrates predictive capability on multiaxial damage data. The decoupled training scheme and the explicit treatment of damage-induced anisotropy are practically valuable. The paper also honestly acknowledges that non-negativity of the energy is verified numerically on the training range rather than proven globally. The main theoretical concern, detailed below, concerns the consistency between the damage evolution equations used in the implementation and the conjugate forces associated with the final energy, which is load-bearing for the central 'thermodynamically consistent as designed' claim.

major comments (2)
  1. [Sections 3.2, 4, 5.3, and 8, Eqs. (3.15), (4.4), (5.4), (5.12)-(5.13)] The damage evolution equations (3.15) and (4.4) are derived from the conjugate forces in Eq. (3.10), namely y_i = -p'_i(alpha_i) bar-psi_i and y_0 = -p'_0(alpha_0) bar-psi_0. However, the actual model defined in Section 5.3 includes the normality-correction term psi_stress, and the final conjugate forces in Eq. (5.4) are y_i = -p'_i(alpha_i) hat-psi_i^ICNN - d psi_stress/d alpha_i, with an analogous expression for y_0. Because psi_stress depends on alpha_i and alpha_0 through p_i and p_0, the extra derivatives d psi_stress/d alpha_i are generally nonzero. No corrected version of (3.15) or (4.4) is supplied, and the experimental validation in Section 8 explicitly states that the anisotropic model is trained using (3.15)_1 and (4.4). The alpha_i-update actually implemented is therefore not the derivative of the stated Helmholtz free energy. The dissipation inequality (2.6)_2 is established for the energy and evolution equations only when the conjugate forces are those of Eq. (3.10); it is not established for the model that includes psi_stress. The numerical dissipation plots in Sections 8 cannot close this gap unless the plotted dissipation is computed with the full conjugate forces and the corrected evolution equations. This is a load-bearing inconsistency for the central claim that the formulation is thermodynamically consistent by construction.
  2. [Section 3.1, Eq. (3.2), and Section 9] The damage tensor is assumed to remain coaxial with the virgin material symmetry axes and to have fixed principal directions throughout the damage process. The invariant basis in Eq. (3.7) is built from the fixed structural tensors L_i, so the model cannot represent a damage state whose principal directions rotate relative to the material axes, for example under non-proportional shear with evolving crack orientation. This is an explicit assumption, and Remark 2 sketches a generalization, but the abstract and concluding remarks present the model as a general anisotropic damage model for orthotropic materials without this qualification. Since the experimental validation is limited to biaxial and planar tests where the loading directions coincide with the symmetry axes, the scope of the model should be stated explicitly in the abstract and conclusions, or the generalized formulation of Remark 2 should be implemented and tested to support the unqualified claim.
minor comments (4)
  1. [Section 2.7 and Section 9] The paper correctly notes that non-negativity of the trained energy is verified only on the training range. The concluding sentence that 'numerical results across all examples demonstrate that the resulting strain energy density functions are non-negative' should be qualified as a data-range verification, not a global property, to avoid overstating the physics guarantee.
  2. [Section 5.3, Eq. (5.7)] The function 'softplus' is used without definition. Although it is standard in machine learning, a brief definition or reference would improve readability for the mechanics audience.
  3. [Section 8, experimental example] The text refers to 'Mullin's effect' in one place; this should be 'Mullins effect' for consistency with the literature and with the rest of the manuscript.
  4. [Section 8, experimental example] The comparison between the isotropic and anisotropic models would be clearer if a quantitative measure of fit error (for example, normalized RMS error per loading mode) were reported, since the qualitative improvement of the anisotropic model is otherwise assessed only visually from Figs. 5 and 6.

Circularity Check

1 steps flagged · score 5.0 of 10

Experimental 'prediction' is a training-data fit; the constitutive derivation itself is otherwise self-contained.

  1. fitted input called prediction [Sections 7-8 (Training scheme and hydrogel example) and Section 9 (Concluding remarks)]
    "The training data is only the stress-strain curves for all synthetic and the experimental cases. ... For the first model, we train an isotropic damage data-driven model, similar to (8.3), against all data. ... To obtain the results, we first applied the proposed decoupled training procedure, then performed full-parameter training for both the isotropic and anisotropic models. ... the anisotropic model successfully captures the dominant characteristics of the response."

    All parameters entering the reported stress predictions are optimized against the same experimental curves: Stage I fits the strain-energy potentials and per-cycle attenuation values to the unloading branches (Eqs. 7.3-7.5), Stage II fits the damage potentials, attenuation parameters, and energy scaling constants to the full loading-unloading cycles (Eqs. 7.6-7.7), and then a final full-parameter pass is performed. The 'predicted' hydrogel curves are therefore the fitted output of the loss function (7.2), not independent predictions. The concluding claim that the anisotropic model 'successfully captures' the response, and the comparison showing it outperforms the isotropic model, reduce by construction to the training objective on the same data.

full rationale

No load-bearing self-citation chain is present: the authors cite their own prior isotropic framework [1], but the anisotropic extension, the invariant architecture, the polyconvexity argument, and the normality-correction construction are derived in this paper from stated assumptions and standard results. The polyconvexity, objectivity, and normality steps are constructions rather than circular predictions. The one clear circular pattern is that the experimental 'validation' for the double-network hydrogel is performed on the same stress-strain curves used to train all model parameters, so the reported success is a training-fit comparison. A separate, non-circular correctness caveat must be weighed: the damage-evolution equations (3.15) and (4.4) are derived from conjugate forces (3.10) that omit the psi_stress terms introduced in Eq. (5.4); no corrected consistency equations are supplied for the final energy (5.12)-(5.13), so the implemented model is not literally the derivative of the stated free energy. This is an internal-consistency problem, not a circularity, but it undercuts the 'thermodynamically consistent, as designed' claim and is flagged here with location: Eqs. (3.15), (4.4) versus (5.4). The score of 5 reflects the partial reduction of the headline experimental claim to the training fit while the theoretical construction retains substantial independent content.

Assumptions & free parameters 7 free parameters · 7 assumptions · 0 invented entities

The framework introduces no new physical entities such as particles or forces. The damage variables alpha_0 and alpha_i are standard internal variables of continuum damage mechanics, and the generalized structural tensors L_tilde(i) in Eq. (3.5) are a representation device from [43]. The main burden of the ledger is therefore carried by the trainable parameters and the modeling assumptions listed above.

free parameters (7)
  • ICNN weights and biases for psi_0 and psi_i = not reported
    Trained in Stage I on unloading branches (Eq. 7.3) and used in the final energy expressions (5.12)-(5.13).
  • ICNN weights and biases for G'_0 and G'_i = not reported
    Trained in Stage II on full loading data (Eq. 7.6) and define damage evolution through (5.6)-(5.7).
  • Attenuation weights w_j and cutoffs a0, a_i in p(alpha) = not reported
    Trainable parameters in Eq. (5.5) that determine how much damage reduces the strain energy.
  • Per-cycle attenuation values P_i(alpha_i,cycle) = not reported
    Fitted constants in Stage I (Eq. 7.4) with the monotonicity constraint (7.5).
  • Energy scale constants C0 and C_i = not reported
    Introduced in Eq. (7.7) to resolve the scaling ambiguity between energy and attenuation identified in (7.8).
  • q_j exponents in Eq. (5.5) = fixed lists, e.g. [1, 1.5, 2, 3, ..., 200]
    Chosen by hand rather than trained; they control the expressiveness of the degradation functions.
  • beta in the loss function = not reported
    Hyperparameter weighting the normality residual penalty R in Eq. (7.2).
assumptions (7)
  • standard math Symmetric right Cauchy-Green tensor, cofactor, and determinant form admissible polyconvex arguments (Ball's theorem, Definition 4).
    Used in Section 2.4 to build polyconvex functionals from invariants.
  • standard math Any anisotropic scalar function can be represented as an isotropic function of C, structural tensors, and internal variables (Zhang and Rychlewski [40]).
    Invoked in Section 2.3, Eqs. (2.10)-(2.15), to include anisotropy via structural tensors.
  • domain assumption The virgin material is at most orthotropic, and the damage principal axes are co-axial with the material symmetry axes and fixed during evolution.
    Eq. (3.2) and surrounding text; restricts the range of anisotropic damage histories the model can represent.
  • domain assumption The Helmholtz free energy is convex in the damage variables, following generalized standard material theory.
    Invoked in Section 3.1 after Eq. (3.7) to guarantee well-posed damage evolution, following [73,74].
  • domain assumption Damage state remains frozen during unloading branches of cyclic tests.
    Justifies the decoupled Stage I training on unloading data in Section 7, before Eq. (7.3).
  • ad hoc to paper Non-negativity of the trained energy is checked numerically on the training range, not proven globally.
    Sections 2.7 and 3 acknowledge this limitation and do not provide a global proof.
  • domain assumption Growth condition at extreme deformations is not enforced.
    Section 2.6 states that such regimes are outside the numerical examples and are primarily of theoretical interest.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A Physics-Augmented Machine Learning Constitutive Model for Damage in Solids." pith.science (2026). https://pith.science/paper/QNXAEYVH

@misc{pith2026250805638,
  author       = {Pith},
  title        = {Pith review of: A Physics-Augmented Machine Learning Constitutive Model for Damage in Solids},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QNXAEYVH}},
  note         = {Machine review of arXiv:2508.05638}
}
read the original abstract

We propose a data-driven constitutive framework for anisotropic damage mechanics based on the second-order damage tensor approach for both compressible and incompressible materials. The formulation is thermodynamically consistent and satisfies the Clausius-Duhem inequality. The strain energy density potentials are expressed as isotropic functions of the right Cauchy-Green deformation tensor, along with structural tensors that encode anisotropy either present in the virgin material or resulting from damage. To guarantee the polyconvexity condition, non-decreasing convex neural networks with inputs that ensure polyconvexity are used to parameterize the strain energy density potentials. The model vanishes in the undeformed state, fulfilling the normality condition. In contrast to classical [1-d] damage models, the expressiveness of the new data-driven model is enhanced by employing a family of nonlinear, convex, decreasing functions to capture the effect of damage. Damage evolution is governed through a damage potential, where the corresponding threshold is defined in terms of the damage conjugate forces. As a special case of the general formulation, a new anisotropic generic format is introduced to predict constitutive responses under damage-induced anisotropy in initially isotropic materials. To reduce the computational burden during training, a decoupled training scheme is introduced, and its accuracy is demonstrated in all numerical examples. These include benchmarks for incompressible isotropic, transversely isotropic, and compressible orthotropic materials. The framework is also validated against experimental data capturing anisotropic Mullins-type damage.

Figures

Figures reproduced from arXiv: 2508.05638 by the authors.

Figure 1
Figure 1. (a) Generic architecture showing the strain-energy and attenuation functions [PITH_FULL_IMAGE:figures/full_fig_p015_1.png] view at source ↗
Figure 2
Figure 2. Validation of the model in recovering synthesized isotropic materials. The synthesized materials were [PITH_FULL_IMAGE:figures/full_fig_p016_2.png] view at source ↗
Figure 3
Figure 3. Validation of the model in recovering the damage behavior of synthesized transversely isotropic materials. [PITH_FULL_IMAGE:figures/full_fig_p017_3.png] view at source ↗
Figures from the paper (6 more)
Figure 4
Figure 4. Figure 4: Evaluation of the capability of the proposed data-driven model to predict the damage behavior of a synthesized [PITH_FULL_IMAGE:figures/full_fig_p019_4.png]
Figure 5
Figure 5. Figure 5: The isotropic data-driven damage model was used to predict the nominal stress for various experimental [PITH_FULL_IMAGE:figures/full_fig_p020_5.png]
Figure 6
Figure 6. Figure 6: The performance of the new anisotropic Data-driven damage model to predict the nominal stresses for various [PITH_FULL_IMAGE:figures/full_fig_p021_6.png]
Figure 7
Figure 7. Figure 7: The trained strain energy density functions corresponding to the anisotropic data-driven damage model [PITH_FULL_IMAGE:figures/full_fig_p022_7.png]
Figure 8
Figure 8. Figure 8: The dissipative energy predicted by the trained data-driven model [PITH_FULL_IMAGE:figures/full_fig_p022_8.png]
Figure 9
Figure 9. Figure 9: The components used to generate the synthetic data. From left to right: the first plot shows the load [PITH_FULL_IMAGE:figures/full_fig_p023_9.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

77 extracted references · 76 canonical work pages

  1. [1]

    Data-driven continuum damage mechanics with built-in physics,

    V . Tac, E. Kuhl, and A. B. Tepole, “Data-driven continuum damage mechanics with built-in physics,”Extreme Mechanics Letters, vol. 71, p. 102220, 2024

  2. [2]

    Comparison of discontinuous damage models of mullins-type,

    A. Ricker, N. H. Kröger, and P. Wriggers, “Comparison of discontinuous damage models of mullins-type,”Archive of Applied Mechanics, vol. 91, no. 10, pp. 4097–4119, 2021

  3. [3]

    A review on data-driven constitutive laws for solids,

    J. N. Fuhg, G. Anantha Padmanabha, N. Bouklas, B. Bahmani, W. Sun, N. N. Vlassis, M. Flaschel, P. Carrara, and L. De Lorenzis, “A review on data-driven constitutive laws for solids,”Archives of Computational Methods in Engineering, pp. 1–43, 2024

  4. [4]

    Rupture time under creep conditions,

    L. M. Kachanov, “Rupture time under creep conditions,”International journal of fracture, vol. 97, no. 1, pp. 11–18, 1999

  5. [5]

    Y . N. Rabotnov,Creep Problem in Structural Members. Amsterdam: North-Holland, 1969

  6. [6]

    The effect of creep constitutive and damage relationships upon the rupture time of a solid circular torsion bar,

    D. Hayhurst and F. Leckie, “The effect of creep constitutive and damage relationships upon the rupture time of a solid circular torsion bar,” Journal of the Mechanics and Physics of Solids, vol. 21, no. 6, pp. 431–432, 1973

  7. [7]

    Behavior of rubber under repeated stresses,

    W. Holt, “Behavior of rubber under repeated stresses,”Rubber Chemistry and Technology, vol. 5, no. 1, pp. 79–89, 1932

  8. [8]

    Effect of stretching on the properties of rubber,

    L. Mullins, “Effect of stretching on the properties of rubber,”Rubber chemistry and technology, vol. 21, no. 2, pp. 281–300, 1948

Show all 77 references
  1. [9]

    Theoretical model for the elastic behavior of filler-reinforced vulcanized rubbers,

    L. Mullins and N. Tobin, “Theoretical model for the elastic behavior of filler-reinforced vulcanized rubbers,” Rubber chemistry and technology, vol. 30, no. 2, pp. 555–571, 1957

  2. [10]

    Double-network hydrogels with extremely high mechanical strength,

    J. P. Gong, Y . Katsuyama, T. Kurokawa, and Y . Osada, “Double-network hydrogels with extremely high mechanical strength,” Advanced materials, vol. 15, no. 14, pp. 1155–1158, 2003

  3. [11]

    Induced anisotropy by mullins effect in filled elastomers subjected to stretching with various geometries,

    T.-T. Mai, Y . Morishita, and K. Urayama, “Induced anisotropy by mullins effect in filled elastomers subjected to stretching with various geometries,” Polymer, vol. 126, pp. 29–39, 2017

  4. [12]

    Distinctive characteristics of internal fracture in tough double network hydrogels revealed by various modes of stretching,

    T.-T. Mai, T. Matsuda, T. Nakajima, J. P. Gong, and K. Urayama, “Distinctive characteristics of internal fracture in tough double network hydrogels revealed by various modes of stretching,”Macromolecules, vol. 51, no. 14, pp. 5245–5257, 2018

  5. [13]

    Toughening elastomers with sacrificial bonds and watching them break,

    E. Ducrot, Y . Chen, M. Bulters, R. P. Sijbesma, and C. Creton, “Toughening elastomers with sacrificial bonds and watching them break,” Science, vol. 344, no. 6180, pp. 186–189, 2014

  6. [14]

    Mechanics of elastomeric molecular composites,

    P. Millereau, E. Ducrot, J. M. Clough, M. E. Wiseman, H. R. Brown, R. P. Sijbesma, and C. Creton, “Mechanics of elastomeric molecular composites,” Proceedings of the National Academy of Sciences , vol. 115, no. 37, pp. 9110–9115, 2018. 23 A PREPRINT

  7. [15]

    A micro-mechanically based continuum damage model for carbon black-filled rubbers incorporating mullins’ effect,

    S. Govindjee and J. Simo, “A micro-mechanically based continuum damage model for carbon black-filled rubbers incorporating mullins’ effect,” Journal of the Mechanics and Physics of Solids, vol. 39, no. 1, pp. 87–112, 1991

  8. [16]

    The constitutive equations of continuum creep damage mechanics,

    F. A. Leckie, “The constitutive equations of continuum creep damage mechanics,”Philosophical Transactions of the Royal Society of London. Series A, Mathematical and Physical Sciences, vol. 288, no. 1350, pp. 27–47, 1978

  9. [17]

    Anisotropic damage in elasticity and plasticity,

    J. Cordebois and F. Sidoroff, “Anisotropic damage in elasticity and plasticity,”Journal de mecanique theorique et appliquee, pp. 45–60, 1982

  10. [18]

    Damage induced elastic anisotropy,

    J. Cordebois and F. Sidoroff, “Damage induced elastic anisotropy,” in Mechanical Behavior of Anisotropic Solids/Comportment Méchanique des Solides Anisotropes: Proceedings of the Euromech Colloquium 115 Villard- de-Lans, June 19–22, 1979/Colloque Euromech 115 Villard-de-Lans, ...

  11. [19]

    How to use damage mechanics,

    J. Lemaitre, “How to use damage mechanics,” Nuclear engineering and design, vol. 80, no. 2, pp. 233–245, 1984

  12. [20]

    A continuous damage mechanics model for ductile fracture,

    J. Lemaitre, “A continuous damage mechanics model for ductile fracture,” Journal of Engineering Materials and Technology, vol. 107, no. 1, pp. 83–89, 1985

  13. [21]

    Application of continuous damage mechanics to strain and fracture behavior of concrete,

    J. Mazars and J. Lemaitre, “Application of continuous damage mechanics to strain and fracture behavior of concrete,” in Application of fracture mechanics to cementitious composites, pp. 507–520, Springer, 1985

  14. [22]

    Strain-and stress-based continuum damage models—i. formulation,

    J. C. Simo and J. Ju, “Strain-and stress-based continuum damage models—i. formulation,” International journal of solids and structures, vol. 23, no. 7, pp. 821–840, 1987

  15. [23]

    On energy-based coupled elastoplastic damage theories: constitutive modeling and computational aspects,

    J. Ju, “On energy-based coupled elastoplastic damage theories: constitutive modeling and computational aspects,” International Journal of Solids and structures, vol. 25, no. 7, pp. 803–833, 1989

  16. [24]

    A unified theory of elastic degradation and damage based on a loading surface,

    I. Carol, E. Rizzi, and K. Willam, “A unified theory of elastic degradation and damage based on a loading surface,” International journal of solids and structures, vol. 31, no. 20, pp. 2835–2865, 1994

  17. [25]

    Tensorial nature of damage measuring internal variables,

    F. Leckie and E. Onat, “Tensorial nature of damage measuring internal variables,” inPhysical Non-Linearities in Structural Analysis: Symposium Senlis, France May 27–30, 1980, pp. 140–155, Springer, 1981

  18. [26]

    Representation of mechanical behavior in the presence of changing internal structure,

    E. Onat and F. Leckie, “Representation of mechanical behavior in the presence of changing internal structure,” 1988

  19. [27]

    Description of anisotropic damage application to elasticity,

    F. Sidoroff, “Description of anisotropic damage application to elasticity,” inPhysical non-linearities in structural analysis: symposium senlis, France May 27–30, 1980, pp. 237–244, Springer, 1981

  20. [28]

    On an anisotropic damage theory,

    P. Ladevèze, “On an anisotropic damage theory,” in The CNRS International College on Failure Criteria of Structured Media, vol. 351, pp. 355–363, 1983

  21. [29]

    An anisotropic theory of elasticity for continuum damage mechanics,

    C. Chow and J. Wang, “An anisotropic theory of elasticity for continuum damage mechanics,” International Journal of fracture, vol. 33, pp. 3–16, 1987

  22. [30]

    Isotropic and anisotropic damage variables in continuum damage mechanics,

    J. Ju, “Isotropic and anisotropic damage variables in continuum damage mechanics,” Journal of Engineering Mechanics, vol. 116, no. 12, pp. 2764–2770, 1990

  23. [31]

    A new formulation of continuum damage mechanics (cdm) for composite materials,

    J. Maire and J. Chaboche, “A new formulation of continuum damage mechanics (cdm) for composite materials,” Aerospace Science and Technology, vol. 1, no. 4, pp. 247–257, 1997

  24. [32]

    Continuum damage modelling: approximation of crack induced anisotropy,

    S. Fichant, G. Pijaudier-Cabot, and C. La Borderie, “Continuum damage modelling: approximation of crack induced anisotropy,”Mechanics Research Communications, vol. 24, no. 2, pp. 109–114, 1997

  25. [33]

    On the formulation of anisotropic elastic degradation. i. theory based on a pseudo-logarithmic damage tensor rate,

    I. Carol, E. Rizzi, and K. Willam, “On the formulation of anisotropic elastic degradation. i. theory based on a pseudo-logarithmic damage tensor rate,”International Journal of Solids and Structures, vol. 38, no. 4, pp. 491–518, 2001

  26. [34]

    Murakami, Continuum damage mechanics: a continuum mechanics approach to the analysis of damage and fracture, vol

    S. Murakami, Continuum damage mechanics: a continuum mechanics approach to the analysis of damage and fracture, vol. 185. Springer Science & Business Media, 2012

  27. [35]

    A continuum theory of creep and creep damage,

    S. Murakami and N. Ohno, “A continuum theory of creep and creep damage,” in Creep in Structures: 3rd Symposium, Leicester, UK, September 8–12, 1980, pp. 422–444, Springer, 1981

  28. [36]

    Mechanical modeling of material damage,

    S. Murakami, “Mechanical modeling of material damage,” Journal of Applied Mechanics, vol. 55, pp. 280–286, 06 1988

  29. [37]

    A simple derivation of representations for non-polynomial constitutive equations in some cases of anisotropy,

    J.-P. Boehler, “A simple derivation of representations for non-polynomial constitutive equations in some cases of anisotropy,”ZAMM-Journal of Applied Mathematics and Mechanics/Zeitschrift für Angewandte Mathematik und Mechanik, vol. 59, no. 4, pp. 157–167, 1979

  30. [38]

    On representations of anisotropic invariants,

    I.-s. Liu et al., “On representations of anisotropic invariants,” International Journal of Engineering Science , vol. 20, no. 10, pp. 1099–1109, 1982. 24 A PREPRINT

  31. [39]

    Theory of representations for tensor functions—a unified invariant approach to constitutive equations,

    Q.-S. Zheng, “Theory of representations for tensor functions—a unified invariant approach to constitutive equations,” Applied Mechanics Reviews, vol. 47, no. 11, pp. 545–587, 1994

  32. [40]

    Structural tensors for anisotropic solids,

    J. M. Zhang and J. Rychlewski, “Structural tensors for anisotropic solids,” Archives of Mechanics, vol. 42, no. 3, pp. 267–277, 1990

  33. [41]

    Itskov, Tensor algebra and tensor analysis for engineers, vol

    M. Itskov, Tensor algebra and tensor analysis for engineers, vol. 2. Springer, 2009

  34. [42]

    Invariant formulation of hyperelastic transverse isotropy based on polyconvex free energy functions,

    J. Schröder and P. Neff, “Invariant formulation of hyperelastic transverse isotropy based on polyconvex free energy functions,” International journal of solids and structures, vol. 40, no. 2, pp. 401–445, 2003

  35. [43]

    A class of orthotropic and transversely isotropic hyperelastic constitutive models based on a polyconvex strain energy function,

    M. Itskov and N. Aksel, “A class of orthotropic and transversely isotropic hyperelastic constitutive models based on a polyconvex strain energy function,”International journal of solids and structures, vol. 41, no. 14, pp. 3833–3848, 2004

  36. [44]

    Anisotropic polyconvex energies on the basis of crystallographic motivated structural tensors,

    J. Schröder, P. Neff, and V . Ebbing, “Anisotropic polyconvex energies on the basis of crystallographic motivated structural tensors,” Journal of the Mechanics and Physics of Solids, vol. 56, no. 12, pp. 3486–3506, 2008

  37. [45]

    On damage induced anisotropy for fiber composites,

    A. Matzenmiller and J. L. Sackman, “On damage induced anisotropy for fiber composites,” International Journal of Damage Mechanics, vol. 3, no. 1, pp. 71–86, 1994

  38. [46]

    A framework for geometrically nonlinear continuum damage mechanics,

    P. Steinmann and I. Carol, “A framework for geometrically nonlinear continuum damage mechanics,”International Journal of Engineering Science, vol. 36, no. 15, pp. 1793–1814, 1998

  39. [47]

    A theoretical and computational framework for anisotropic continuum damage mechanics at large strains,

    A. Menzel and P. Steinmann, “A theoretical and computational framework for anisotropic continuum damage mechanics at large strains,” International Journal of Solids and Structures, vol. 38, no. 52, pp. 9505–9523, 2001

  40. [48]

    On the modelling of anisotropic elastic and inelastic material behaviour at large deformation,

    B. Svendsen, “On the modelling of anisotropic elastic and inelastic material behaviour at large deformation,” International Journal of Solids and Structures, vol. 38, no. 52, pp. 9579–9599, 2001

  41. [49]

    Using structural tensors for inelastic material modeling in the finite strain regime–a novel approach to anisotropic damage,

    S. Reese, T. Brepols, M. Fassin, L. Poggenpohl, and S. Wulfinghoff, “Using structural tensors for inelastic material modeling in the finite strain regime–a novel approach to anisotropic damage,” Journal of the Mechanics and Physics of Solids, vol. 146, p. 104174, 2021

  42. [50]

    Multilayer feedforward networks are universal approximators,

    K. Hornik, M. Stinchcombe, and H. White, “Multilayer feedforward networks are universal approximators,” Neural networks, vol. 2, no. 5, pp. 359–366, 1989

  43. [51]

    TensorFlow: Large-scale machine learning on heterogeneous systems,

    M. Abadi et al., “TensorFlow: Large-scale machine learning on heterogeneous systems,” 2015. Software available from tensorflow.org

  44. [52]

    JAX: composable transformations of Python+NumPy programs,

    J. Bradbury, R. Frostig, P. Hawkins, M. J. Johnson, C. Leary, D. Maclaurin, G. Necula, A. Paszke, J. VanderPlas, S. Wanderman-Milne, and Q. Zhang, “JAX: composable transformations of Python+NumPy programs,” 2018

  45. [53]

    Geometric deep learning for computational mechanics part i: Anisotropic hyperelasticity,

    N. N. Vlassis, R. Ma, and W. Sun, “Geometric deep learning for computational mechanics part i: Anisotropic hyperelasticity,”Computer Methods in Applied Mechanics and Engineering, vol. 371, p. 113299, 2020

  46. [54]

    Constitutive artificial neural networks: A fast and general approach to predictive data-driven constitutive modeling by deep learning,

    K. Linka, M. Hillgärtner, K. P. Abdolazizi, R. C. Aydin, M. Itskov, and C. J. Cyron, “Constitutive artificial neural networks: A fast and general approach to predictive data-driven constitutive modeling by deep learning,” Journal of Computational Physics, vol. 429, p. 110010, 2021

  47. [55]

    Polyconvex neural networks for hyperelastic constitutive models: A rectification approach,

    P. Chen and J. Guilleminot, “Polyconvex neural networks for hyperelastic constitutive models: A rectification approach,” Mechanics Research Communications, vol. 125, p. 103993, 2022

  48. [56]

    A mechanics-informed artificial neural network approach in data-driven constitutive modeling,

    F. As’ ad, P. Avery, and C. Farhat, “A mechanics-informed artificial neural network approach in data-driven constitutive modeling,” International Journal for Numerical Methods in Engineering, vol. 123, no. 12, pp. 2738– 2759, 2022

  49. [57]

    Learning hyperelastic anisotropy from data via a tensor basis neural network,

    J. N. Fuhg, N. Bouklas, and R. E. Jones, “Learning hyperelastic anisotropy from data via a tensor basis neural network,” Journal of the Mechanics and Physics of Solids, vol. 168, p. 105022, 2022

  50. [58]

    Fe ann: an efficient data-driven multiscale approach based on physics-constrained neural networks and automated data mining,

    K. A. Kalina, L. Linden, J. Brummund, and M. Kästner, “Fe ann: an efficient data-driven multiscale approach based on physics-constrained neural networks and automated data mining,” Computational Mechanics, vol. 71, no. 5, pp. 827–851, 2023

  51. [59]

    Data-driven tissue mechanics with polyconvex neural ordinary differential equations,

    V . Tac, F. S. Costabal, and A. B. Tepole, “Data-driven tissue mechanics with polyconvex neural ordinary differential equations,” Computer Methods in Applied Mechanics and Engineering, vol. 398, p. 115248, 2022

  52. [60]

    Neural networks meet hyperelasticity: A guide to enforcing physics,

    L. Linden, D. K. Klein, K. A. Kalina, J. Brummund, O. Weeger, and M. Kästner, “Neural networks meet hyperelasticity: A guide to enforcing physics,” Journal of the Mechanics and Physics of Solids , vol. 179, p. 105363, 2023

  53. [61]

    A new family of constitutive artificial neural networks towards automated model discovery,

    K. Linka and E. Kuhl, “A new family of constitutive artificial neural networks towards automated model discovery,” Computer Methods in Applied Mechanics and Engineering, vol. 403, p. 115731, 2023. 25 A PREPRINT

  54. [62]

    Benchmarking physics-informed frameworks for data-driven hyperelasticity,

    V . Taç, K. Linka, F. Sahli-Costabal, E. Kuhl, and A. B. Tepole, “Benchmarking physics-informed frameworks for data-driven hyperelasticity,”Computational mechanics, vol. 73, no. 1, pp. 49–65, 2024

  55. [63]

    Input convex neural networks,

    B. Amos, L. Xu, and J. Z. Kolter, “Input convex neural networks,” in International conference on machine learning, pp. 146–155, PMLR, 2017

  56. [64]

    Neural ordinary differential equations,

    R. T. Chen, Y . Rubanova, J. Bettencourt, and D. K. Duvenaud, “Neural ordinary differential equations,”Advances in neural information processing systems, vol. 31, 2018

  57. [65]

    Convexity conditions and existence theorems in nonlinear elasticity,

    J. M. Ball, “Convexity conditions and existence theorems in nonlinear elasticity,”Archive for rational mechanics and Analysis, vol. 63, pp. 337–403, 1976

  58. [66]

    Recovering mullins damage hyperelastic behaviour with physics augmented neural networks,

    M. Zlati´c and M. ˇCana ¯dija, “Recovering mullins damage hyperelastic behaviour with physics augmented neural networks,” Journal of the Mechanics and Physics of Solids, vol. 193, p. 105839, 2024

  59. [67]

    On a fully three-dimensional finite-strain viscoelastic damage model: formulation and computational aspects,

    J. C. Simo, “On a fully three-dimensional finite-strain viscoelastic damage model: formulation and computational aspects,” Computer methods in applied mechanics and engineering, vol. 60, no. 2, pp. 153–173, 1987

  60. [68]

    Thermodynamics with internal state variables,

    B. D. Coleman and M. E. Gurtin, “Thermodynamics with internal state variables,”The journal of chemical physics, vol. 47, no. 2, pp. 597–613, 1967

  61. [69]

    Quasi-convexity and the lower semicontinuity of multiple integrals,

    C. B. J. Morrey, “Quasi-convexity and the lower semicontinuity of multiple integrals,”Pacific Journal of Mathe- matics, vol. 2, no. 1, pp. 25–53, 1952

  62. [70]

    On a new class of elastic deformations not allowing for cavitation,

    S. Müller, T. Qi, and B. Yan, “On a new class of elastic deformations not allowing for cavitation,” inAnnales de l’Institut Henri Poincaré C, Analyse non linéaire, vol. 11, pp. 217–243, Elsevier, 1994

  63. [71]

    G. A. Holzapfel, Nonlinear Solid Mechanics: A Continuum Approach for Engineering. John Wiley & Sons, 2000

  64. [72]

    Polyconvex anisotropic hyperelasticity with neural networks,

    D. K. Klein, M. Fernández, R. J. Martin, P. Neff, and O. Weeger, “Polyconvex anisotropic hyperelasticity with neural networks,” Journal of the Mechanics and Physics of Solids, vol. 159, p. 104703, 2022

  65. [73]

    Lemaitre and J.-L

    J. Lemaitre and J.-L. Chaboche, Mechanics of solid materials. Cambridge university press, 1994

  66. [74]

    G. A. Maugin, The thermomechanics of nonlinear irreversible behaviours, vol. 27. World scientific, 1999

  67. [75]

    Modeling neurodegeneration in chronic traumatic encephalopathy using gradient damage models,

    L. Noël and E. Kuhl, “Modeling neurodegeneration in chronic traumatic encephalopathy using gradient damage models,” Computational Mechanics, vol. 64, no. 5, pp. 1375–1387, 2019

  68. [76]

    On the implementation of finite deformation gradient-enhanced damage models,

    R. Ostwald, E. Kuhl, and A. Menzel, “On the implementation of finite deformation gradient-enhanced damage models,” Computational Mechanics, vol. 64, pp. 847–877, 2019

  69. [77]

    Damage models for soft tissues: a survey,

    W. Li, “Damage models for soft tissues: a survey,”Journal of Medical and Biological Engineering, vol. 36, no. 3, pp. 285–307, 2016. 26

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.