Polyconvex neural-network surrogates with learned paired directional features for anisotropic hyperelastic homogenization
Abstract
Approximating the effective anisotropic hyperelastic response of a representative volume element (RVE) requires constitutive expressiveness, mechanical consistency, and computational efficiency. This work develops a polyconvex physics-augmented neural network (PANN) construction based on paired directional features whose orientations, stretch powers, and determinant exponents adapt to the RVE response. The features are composed with convex, componentwise nondecreasing cores, realized as an input-convex neural network (ICNN) or an input-convex Kolmogorov-Arnold network (ICKAN). The complete energies satisfy objectivity, an exactly stress-free reference, two-dimensional polyconvexity, and prescribed growth conditions by construction, and nonnegative energy is certified separately for the selected models. On a multicavity and a single-cavity periodic cell, these energies are compared with an Unconstrained energy, trained identically without the curvature and growth constraints, and with three hyperreduced projection-based reduced-order models (HPROMs), which serve as physics-based references. On the first cell, learning the features lowers the stress error sevenfold, and six features suffice on both. The constrained energies remain below one percent in stress error and in the displacement error of a tensile-coupon FE2 application with a full-order reference, whereas the Unconstrained energy and the HPROMs are more accurate inside the data. Beyond the data, however, the Unconstrained energy loses rank-one convexity, and under confined compression it stops below half of the 500 MPa that the constrained energies reach. These results quantify what the guarantees cost in accuracy and what they secure beyond the data.
Related
- Homogenization of anisotropic diffusion on pre-Sierpiński carpets
- A singular minimizer of a smooth strongly polyconvex functional in two dimensions
- Universal neural network approximation of symmetric polyconvex energies and envelopes
- Determination of hyperelastic parameters with non-trapping wavespeeds
- A globally defined compressible polyconvex isotropic energy satisfying the true-stress-true-strain monotonicity condition (TSTS-M++)
BibTeX
@misc{parga2026polyconvex,
title = {{Polyconvex neural-network surrogates with learned paired directional features for anisotropic hyperelastic homogenization}},
author = {S. Ares de Parga and A. Cornejo and J. A. Hernández and R. Rossi},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.12801},
eprint = {2610.12801},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.12801},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Polyconvex neural-network surrogates with learned paired directional features for anisotropic hyperelastic homogenization
AU - S. Ares de Parga
AU - A. Cornejo
AU - J. A. Hernández
AU - R. Rossi
PY - 2026
DA - 2026/10/08
PB - arXiv
DO - 10.48550/arxiv.2610.12801
UR - https://arxiv.org/abs/2610.12801
AB - Approximating the effective anisotropic hyperelastic response of a representative volume element (RVE) requires constitutive expressiveness, mechanical consistency, and computational efficiency. This work develops a polyconvex physics-augmented neural network (PANN) construction based on paired directional features whose orientations, stretch powers, and determinant exponents adapt to the RVE response. The features are composed with convex, componentwise nondecreasing cores, realized as an input-convex neural network (ICNN) or an input-convex Kolmogorov-Arnold network (ICKAN). The complete energies satisfy objectivity, an exactly stress-free reference, two-dimensional polyconvexity, and prescribed growth conditions by construction, and nonnegative energy is certified separately for the selected models. On a multicavity and a single-cavity periodic cell, these energies are compared with an Unconstrained energy, trained identically without the curvature and growth constraints, and with three hyperreduced projection-based reduced-order models (HPROMs), which serve as physics-based references. On the first cell, learning the features lowers the stress error sevenfold, and six features suffice on both. The constrained energies remain below one percent in stress error and in the displacement error of a tensile-coupon FE^2 application with a full-order reference, whereas the Unconstrained energy and the HPROMs are more accurate inside the data. Beyond the data, however, the Unconstrained energy loses rank-one convexity, and under confined compression it stops below half of the 500 MPa that the constrained energies reach. These results quantify what the guarantees cost in accuracy and what they secure beyond the data.
ER -