On unsteady cavitation flow of a high-speed submerged water jet based on data-driven modal decomposition

Abstract

Harnessing the destructive effect of cavitation, high-speed submerged cavitating water jets have a wide range of engineering applications. However, the characteristics of complex unsteady flow patterns due to turbulence and super-cavitation are still unclear. This paper employs the data-driven modal decomposition to study the mechanism of the unsteady turbulent cavitation flow. The simulation is based on the stress-blended eddy simulation (SBES) turbulence model coupled with the Schnerr-Sauer cavitation model. The internal evolution and frequency characteristics of cavitation flow are analyzed. Proper orthogonal decomposition (POD) and dynamic mode decomposition (DMD) are utilized to identify the flow pattern of turbulent cavitation flow, including the vapor and the velocity fields. In results, the upstream part of the cavitating jet is characterized by growth and shedding behaviors, while the downstream part by decay and collapse behaviors. POD modes extract the most energetic structures within the flow field, with low-order modes capturing the majority of the energy. The vapor-based POD Mode 2 and Mode 3 reveal the behavior pattern of cavitation cloud shedding. While DMD identifies the shedding behavior and collapse behavior of cavitation through the decoupling of frequency domain.