Content area
Full Text
1. Introduction
Cavitation is a dynamic phase-change phenomenon that occurs in liquids when the static pressure drops below the vapor pressure of liquid (Brennen, 1995; Knapp et al., 1970). It is well known that the unsteady cavitation in turbo-machinery and marine control surfaces leads to problems such as material damage, vibration, noise and reduced efficiency (Arndt, 1981; Joseph, 1995). Actually, cavitation physical mechanisms have not been well understood due to the complex, unsteady flow structures associated with turbulence and phase change. There are significant computational issues in regard to stability, efficiency, and robustness of the numerical algorithm for turbulent unsteady cavitating flows (Chen and Heister, 1996; Delannoy and Kueny, 1990; Kubota et al., 1992; Kunz et al., 2000; Merkle et al., 1998; Senocak and Shyy, 2004a,b; Singhal et al., 2002).
Cavitating flows are generally relatively high Reynolds number flows and hence the turbulence modeling plays an important role in the capture of unsteady behavior. Large-eddy simulations (LES) are capable to capture the complex turbulent small-scale features in cavitating flows (Dittakavi et al., 2010; Ji et al., 2013; Qin, 2004; Wang and Ostoja-Starzewski, 2007). However, it is fundamentally difficult to find grid independent LES solutions unless one explicitly assigns a filter scale. Moreover, the computational cost of LES is very high. Conventional eddy-viscosity Reynolds-averaged Navier-Stokes (RANS) turbulence models, such as the k-ε model and the k-ω model, have been widely used in many turbulent flows of practical interest. However, many researchers (Coutier-Delgosha et al., 2003; Wu et al., 2005) have indicated that high eddy viscosity of the original Launder-Spalding version of the k-ε model may dampen the vortex shedding behavior and excessively restrain the cavitation instabilities.
To improve the eddy viscosity models’ capability of predicting unsteady cavitating flows, many approaches have been developed. One category of approaches, called hybrid approaches, combines the advantages of RANS and LES. Many versions of RANS/LES hybrids, such as detached-eddy simulations (DES, Huang et al., 2012; Kinzel et al., 2007), filter-based turbulence model (FBM, Johansen et al., 2004; Wu et al., 2005), partially averaged Navier-Stokes (Girimaji, 2006; Huang and Wang, 2011; Ji et al., 2012), etc., have been used to simulate cavitating flows and obtain expected results....