A novel LES-augmented machine learning algorithm for turbulent flow and bed morphodynamics prediction in large-scale environments

In this talk, I introduced a novel framework that integrates large-eddy simulation (LES) with machine learning to predict turbulent flow behavior and bed morphodynamics in large-scale natural environments. By leveraging high-fidelity simulation data to train a deep learning model, the approach enables accurate and efficient prediction of complex hydrodynamic and sediment transport processes. The presentation highlighted the model architecture, training strategy, and validation against LES results, demonstrating its potential for real-time forecasting and digital twin applications in environmental fluid mechanics.