Using Proper Orthogonal Decomposition and Dynamic Mode Decomposition Methods for Comparing CFD Results Experimental Measurements

A method for the quantitative comparison of numerical and/or experimental
data of unsteady aerodynamics around static and oscillating bodies is introduced. It is
based on Proper Orthogonal Decomposition (POD) and Dynamic Mode Decomposition
(DMD) to extract the dominant structures of the unsteady flow. The proposed method
is applied to spatio-temporal data for the flow around a 4:1 rectangular cylinder. Exper-
imental data are obtained from wind tunnel testing and two dimensional Time-resolved
Particle Image Velocimetry (Tr-PIV) measurements, while unsteady Reynolds Averaged
Navier-Stokes (uRANS) are used to compute numerical results. It is demonstrated that
the two approaches are complementary and represent a powerful tool that enables the
analysis and the quantitative comparison of the main spatial (POD) and temporal (DMD)
characteristics of unsteady aerodynamic data.