A physics-data combined neural network-based finite volume parametric reduced order model

View Full Paper

Ask AI

Bookmark

Authors

Discussion

Analysis shows improved stability and accuracy for incompressible flows using new parametric reduced order model.

Proposed model achieves high accuracy in predicting steady-state for incompressible flows, ensuring stability across various scenarios.
Key evidence indicates that using proper orthogonal decomposition yields effective predictions in complex flow configurations.
Assessment using a finite volume parametric reduced order model introduces mass conservation through a specialized loss mechanism.
Results demonstrate strong generalization ability in different test cases, highlighting the model's robustness in fluid dynamics applications.

Pan et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark