Large eddy simulation of turbulent flow over a backward-facing step with a neural-network-based subgrid-scale model

View Full Paper

Ask AI

Bookmark

Authors

Discussion

Analysis demonstrates improved predictions for turbulent flow in various grid resolutions, suggesting advances in neural network modeling.

Improved predictions of SGS stress tensor using neural network models enhance turbulent flow simulations.
Results show large eddy simulation using neural networks outperforms the dynamic Smagorinsky model.
Method involved training a neural network with filtered direct numerical simulation data at Reynolds number 5100.
Enhanced grid resolution applicability indicates robustness in modeling turbulent flow predictions.

Park et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark

Analysis demonstrates improved predictions for turbulent flow in various grid resolutions, suggesting advances in neural network modeling.