Data-driven and physical property-based hydro-acoustic mode decomposition

Abstract

A data-driven and physical property-based hydrodynamic and acoustic mode decomposition method combining dynamic mode decomposition and Helmholtz decomposition is proposed. It allows decomposition and fast prediction of hydrodynamic and acoustic components of the flow field. The method is tested by a two-dimensional subsonic open cavity flow and a supersonic cold jet, and the hydrodynamic and acoustic features are revealed. For the cavity flow, it is found that the acoustic velocity inside the cavity is composed of several pairs of standing waves. The propagating trajectory of the acoustic waves in the cavity is well captured. The dynamic relation between the hydrodynamic and acoustic motion is investigated. For the supersonic jet, the method successfully identifies the screech in the far field and the “trapped wave” within the potential core.