Unsteady Flow

Unsteady flow research focuses on accurately modeling and predicting fluid motion that changes over time, a crucial aspect of many natural and engineering systems. Current efforts leverage machine learning, particularly neural networks like physics-informed neural networks (PINNs), convolutional autoencoders (CAEs), and transformers, often combined with reduced-order modeling techniques to improve computational efficiency and prediction accuracy, even with limited data. These advancements are improving the ability to simulate complex flows, such as those around moving bodies or in turbulent regimes, with applications ranging from aerodynamic design to environmental modeling. The development of more efficient and accurate models is driving progress in understanding and predicting unsteady flow phenomena.