Dynamic Mode Decomposition of a Fluid Flow

Abstract:

Dynamic Mode Decomposition (DMD) was first introduced in the fluid mechanics community for analyzing behavior of nonlinear systems. DMD algorithm processes empirical data generated by nonlinear dynamics and calculates eigenvalues and eigenmodes (“DMD modes”) of an approximate linear model. In fluid dynamics, the DMD modes can be used for many purposes, such as predicting future behavior, identifying periodic behavior of fluid flows, and isolating special patterns in the flow. Computationally, DMD can be interpreted as an Arnoldi-like method based on the Koopman operator.