Towards Long-Term predictions of Turbulence using Neural Operators
It addresses the problem of developing efficient surrogate models for turbulent flow simulations in fluid dynamics, though it is incremental with hybrid methods.
This paper tackled predicting turbulent flows using Neural Operators, finding that U-NET structures like U-FNET outperformed standard FNO in accuracy and stability, especially at higher Reynolds numbers, with regularization terms being crucial for stable predictions.
This paper explores Neural Operators to predict turbulent flows, focusing on the Fourier Neural Operator (FNO) model. It aims to develop reduced-order/surrogate models for turbulent flow simulations using Machine Learning. Different model configurations are analyzed, with U-NET structures (UNO and U-FNET) performing better than the standard FNO in accuracy and stability. U-FNET excels in predicting turbulence at higher Reynolds numbers. Regularization terms, like gradient and stability losses, are essential for stable and accurate predictions. The study emphasizes the need for improved metrics for deep learning models in fluid flow prediction. Further research should focus on models handling complex flows and practical benchmarking metrics.