Journal of Hydrodynamics, Ser. B

A general framework (methodology and procedures) for verification and validation (V&V) of large eddy simulations in computational fluid dynamics (CFD) is derived based on two hypotheses. The framework allows for quantitative estimations of numerical error, modeling error, their coupling, and the associated uncertainties. To meet different needs of users based on their affordable computational cost, various large eddy simulation (LES) V&V methods are proposed. These methods range from the most sophisticated seven equation estimator to the simplest one-grid estimator, which will be calibrated using factors of safety to achieve the objective reliability and confidence level. Evaluation, calibration and validation of various LES V&V methods in this study will be performed using rigorous statistical analysis based on an extensive database. Identification of the error sources and magnitudes has the potential to improve existing or derive new LES models. Based on extensive parametric studies in the database, it is expected that guidelines for performing large eddy simulations that meet pre-specified quality and credibility criteria can be obtained. Extension of this framework to bubbly flow is also discussed.

The unsteady behaviors, such as surging, heaving and pitching motion, which often occur during the advancing of supercavitating vehicle, has significant effect on the stability of supercavitaty and the trajectory of the vehicle. This paper presents a 3-dimentional numerical simulation of periodically forced-pitching of supercavitating vehicle. Based on the finite volume method and the pressure-based segregate algorithm, in the framework of Mixture multiphase model, associated with dynamic mesh method, the Reynolds-Averaged Navier-Stokes equations are solved for the ventilated cavitating flow field in a cavitation tunnel. For both steady-state and dynamic cases, the numerical results agree with the experimental results very well. When the vehicle is pitching periodically, the profile of the supercavity doesn’t vary significantly. The pressure inside the cavity fluctuates slightly during the pitching motion, while the pressure fluctuates significantly at the rear of the down-line of the vehicle. The hydrodynamic forces of the vehicle oscillated periodically, but not linear related to the pitching motion.

The interactions between the bubbles and the particles near structures are important issues for the applications of the cavitation in the fluid machinery. To study the hidden microscopic mechanisms, a numerical method for simulating the laser-generated bubble between the solid wall and a particle is developed in this paper with considerations of the viscosities and the compressibility of the gas and the liquid phases, as well as the surface tension between them. The gas-liquid interface is tracked by the coupling level set and the volume of fluid (CLSVOF) method. The numerical results clearly reveal that the particle can influence the cavitation bubble behaviors. The potential damage of the nearby structures is numerically quantified in terms of the wall pressure, which helps better understand the synergetic effects of the particle on the cavitation. The effects of three dimensionless parameters on the wall pressure are also investigated, especially, on the peak pressure, namely, γ1 (defined as L1 / Rmax, where L1 is the distance from the center of the initial bubble to the solid wall and Rmax is the maximum bubble radius), γ2 (defined as L2 / Rmax, where L2 is the distance from the lower surface of the spherical particle to the initial bubble center) and θ (defined as Rp / Rmax, where Rp is the spherical particle radius). Further numerical results show that these parameters play a dominant role in determining the peak pressure. When γ1 < 1.00, the peak pressure on the solid wall during the bubble collapse is mainly resulted from the liquid jet. When γ1 > 1.00, the peak pressure is caused by the shock wave. With the increase of θ or decrease of γ2, the peak pressure increases. When γ2 > 2.00, the effect of the particle on the bubble behavior can be neglected.

This paper investigates the unsteady structures and the hydrodynamics of cavitating flows. Experimental results are presented for a Clark-Y hydrofoil fixed at α = 0°, 5° and 8°, for various cavitation numbers, from subcavitating flow to supercavitation. The high-speed video camera and the particle image velocimetry (PIV) are applied to observe the transient flow structures. Statistics of the cavity lengths, the velocity and vorticity distributions, as well as the turbulent intensities are presented to quantify the unsteady process. Meanwhile, the dynamic measurement system is used to record the dynamic characteristics. The experimental results show that the flow structures and the hydrodynamics of the cavitation vary considerably with various combinations of angles of attack and cavitation numbers. Under various conditions, the cavitation can be generally grouped as the inception cavitation, the sheet cavitation, the cloud cavitation and the supercavitation. The cloud cavitation exhibits noticeable unsteady characteristics. Experimental evidence indicates that the hydrodynamics are clearly affected by the cavitating flow structures, the amplitude of the load fluctuation is much higher in the cloud cavitating cases.