On the flow structure of cloud cavitating flow around an axisymmetric body near the free surface

Yiwei Wang,Yiwei Wang,Xianxian Yu,Chenguang Huang,Chenguang Huang,Xiaocui Wu,Xiaocui Wu

doi:10.1088/1742-6596/656/1/012174

Abstract

The influence of the free surface on the cavitating flow is an important issue involved in the design of high speed surface vehicles. In the present paper, unsteady cavitating turbulent flow around an axisymmetric body near the free surface was investigated by both launching experiment and LES simulation. The vortex motion induced by cavity shedding under the effect of the free surface is emphatically analyzed by comparing with the submerged condition. The vortex shedding process around the projectile is not synchronized, while the asymmetric characteristic in collapse process is more remarkable, with the generation of multiple vortex ring structures.

Highlights

Characteristics of cavitating flow, especially unstable structures of cloud cavitating flow, always have close relationship with the vortex motion [1]
The present paper focuses mainly on the influence of the free surface on the flow structure of the cloud cavitating flow
The volume of fluid (VOF) method and the modified high resolution interface capturing (HRIC) scheme are used to capture the interfaces of the free surface and cavity profile

Summary

Introduction

Characteristics of cavitating flow, especially unstable structures of cloud cavitating flow, always have close relationship with the vortex motion [1]. On the simulation for engineering applications, solving Reynolds-averaged Navier–Stokes (RANS) equations and turbulence models with physical modifications is the main way. Approaches such as modified RNG k-ε turbulence model [2,3,4,5,6] are widely used. Large eddy simulation (LES) method is adopted for the research on cavitating flows, which can capture considerable details of large-scale eddies in the flow field with high accuracy [7,8,9,10,11,12,13,14]. The cavitation shedding horse-shoe structures were captured numerically by Ji et al [8, 14]

Methods

Results

Conclusion