Investigation of Dynamic Mode Decomposition for ۲D Flow Field Estimation

Reduced order models (ROMs) are recognized for enhancing the computational efficiency of numerical and experimental approaches in studying complex phenomena such as nonlinear dynamics. Among the recently developed ROMs, Dynamic Mode Decomposition (DMD), one of the stated methods among the newly developed ROMs, demonstrates a high range of adaptability and applicability in fluid dynamics and other data-centric applications. This study employs a fast and accurate technique to capture a proper approximation between the DMD snapshots. As a case study, this innovative algorithm is applied to the lid-driven cavity fluid flow.

کلیدواژه ها:

Dynamic Mode Decomposition ، Computational Fluid Dynamics ، Supervised Learning