Metraipolis › archive › dc_f053df75cc54

Only Linear Constraints Survive Coarse-Graining: Evaluating Physics-Constrained Neural Operators on Stochastically Forced Turbulence — arXiv:2610.07811

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.07811
License (as archived)
CC0-1.0
Archived
2026-10-07T16:59:03.908Z
SHA-256 of the archived text
ccdb4e2e2f6883c37f7ee0f5ac45f9194e74bfb47cdff1d1ac358170b609af11
Stable URL of this page
https://metraipolis.com/documento/dc_f053df75cc54

How to cite

Only Linear Constraints Survive Coarse-Graining: Evaluating Physics-Constrained Neural Operators on Stochastically Forced Turbulence — arXiv:2610.07811 — arXiv (en), https://arxiv.org/abs/2610.07811 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f053df75cc54 (Metraipolis archive dc_f053df75cc54, sha256 ccdb4e2e2f6883c37f7ee0f5ac45f9194e74bfb47cdff1d1ac358170b609af11)
BibTeX
@misc{dcf053df75cc54,
  title = {Only Linear Constraints Survive Coarse-Graining: Evaluating Physics-Constrained Neural Operators on Stochastically Forced Turbulence — arXiv:2610.07811},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.07811},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f053df75cc54; sha256 ccdb4e2e2f6883c37f7ee0f5ac45f9194e74bfb47cdff1d1ac358170b609af11},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.07811. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

A physics-informed loss adds a discretised PDE residual to the data term, but is only strictly valid if the equations being enforced are closed on the fields being fitted. This fails on coarse-grained, stochastically forced data: nonlinear terms in the PDE generally do not commute with the filter, so the coarse-grained fields do not satisfy the original equations, and driving their residual to zero encodes an implicit closure into the learned operator. Under normalised, translation-invariant filtering on a periodic domain, constant-coefficient linear constraints remain valid on the filtered grid. For the 2D incompressible Navier-Stokes equations, two such constraints are continuity and global momentum balance. We enforce these exactly by applying a closed-form projection on the output of a Fourier neural operator (FNO) at $O(N \log N)$ cost. On two-dimensional isotropic turbulence, truncated so that its forcing lies entirely beyond the cutoff wavenumber, the projection achieves a continuity error of $5.0\times10^{-7}$ against 0.51 for a plain FNO and 0.17 for a physics-informed FNO, and a global momentum balance error of $8.6\times10^{-11}$ against $7.5\times10^{-4}$ and $8.3\times10^{-4}$. The projection only costs an extra 2% of training time, and takes the fraction of rollouts whose energy remains bounded at 448 steps from 0.18 to 0.87. On resolved Kolmogorov flow, test-time optimisation costs approximately 3,300$\times$ as much per trajectory as the projection, while leaving maximum divergence three orders of magnitude higher. Nonlinear constraints can also be enforced once their unclosed terms are modelled: closing the global energy balance with a constant subgrid flux, fitted from the coarse training data alone, removes the 20.9% energy deficit that enforcing the unclosed balance produces and keeps every free-running rollout bounded over 2,000 steps.

Tema degli esploratori: fisica e astronomia (elenco arXiv physics.comp-ph, nuovi lavori del giorno).
Categorie arXiv: Fluid Dynamics (physics.flu-dyn) ; Computational Physics (physics.comp-ph).
Pagina del lavoro: https://arxiv.org/abs/2610.07811 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-07T16:31:14.787Z dalla pagina https://arxiv.org/list/physics.comp-ph/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).