September 10, 2025APL Machine LearningOpen Access

Physics-informed multi-fidelity surrogate modeling of fluid flow in porous media

View Full Paper

Ask AI

Bookmark

The multi-fidelity surrogate model predicts fluid flow behavior in porous media efficiently and accurately.
Integration of physics-informed neural networks and the Navier–Stokes equations improves model fidelity and accuracy.
Low-fidelity and high-fidelity data contribute to model training, achieving a computational speed increase of approximately 103×.
Model performance maintains reasonable accuracy, with velocity deviations within 5%–20% and pressure deviations ≤10%.

Discussion

No takes yet. Share an insight, caveat, or question.

Research reveals a multi-fidelity surrogate model enhances fluid flow predictions in porous media, indicating significant computational efficiency.

Behrou et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark

Research reveals a multi-fidelity surrogate model enhances fluid flow predictions in porous media, indicating significant computational efficiency.