Modal Decomposition of Turbulent Supersonic Cavity

TL;DR

This paper investigates the flow dynamics of a Mach 3 supersonic cavity using LES, POD, and DMD techniques to understand self-sustained oscillations and flow structure coherence.

Contribution

It introduces a combined use of LES, POD, and DMD to analyze the flow mechanisms in supersonic cavity oscillations, providing new insights into vortex dynamics and flow coherence.

Findings

Identification of discrete vortices along the shear layer.

Revealed flow structure coherence in spanwise and streamwise directions.

Dynamic Mode Decomposition uncovers the self-sustained oscillation mechanism.

Abstract

Self-sustained oscillation in Mach 3 supersonic cavity with a length-to-depth ratio of 3 is investigated using wall modeled Large Eddy Simulation (LES) methodology for ReD = 3.39*10^5. The unsteady data obtained through computation is utilized to investigate the spatial and temporal evolution of the flow field, especially second invariant of velocity tensor; while the phase averaged data is analyzed over a feedback cycle to study the spatial structures. This analysis is accompanied by the Proper Orthogonal Decomposition (POD) data, which reveals the presence of discrete vortices along the shear layer. The POD analysis is performed in both the spanwise and streamwise planes to extract the coherence in flow structures. Finally, Dynamic Mode Decomposition (DMD) is performed on the data sequence to get the dynamic information and deeper insight into the self-sustained mechanism.