In hydraulic systems, transient flow often occurs and may results in cavitation in pipelines. In this paper, the Computational Fluid Dynamics (CFD) method based on the Fluent software was used to investigate the cavitation flow in pipeline; the density-pressure model was incorporated into the continuity equation by using further development of UDF (user defined function), which reflects the variable wave speed of the transient cavitation flow, and the related algorithms were established based on weakly compressible fluid Reynolds Average Navier-Stokes (RANS) techniques. Firstly, the numerical simulations of the transient non-cavitation and cavitation flows caused by the fast closing valve in the reservoir-pipe-valve system were carried out by using the grid slip technique. The simulation results can enrich the flow field information such as velocity, pressure and vapor volume fraction. Through the evolution process of the pressure field, the propagation characteristics of pressure waves can be analyzed qualitatively and quantitatively. Through the evolution process of the velocity field, it can be seen that the velocity distribution in the wall area changes rapidly and has a high gradient, which mainly depends on the viscosity. However, the change of the velocity distribution in the core region is related to the velocity distribution of the history of the past time, which mainly depends on the diffusion. The formation, development and collapse of the cavity can be successfully captured, and it can be clearly and visually observed that the uneven distribution of vapor cavity in the direction of pipe length and pipe diameter, and the vapor cavity move slowly along the top of the pipe wall. Rarefaction wave’s propagation into pressure decreasing region and pressure increasing region can lead to different results of cavitation flow. The accuracy and reliability of the weakly compressible fluid RANS method were verified by comparing the calculated results with the experimental data.

Hydraulic transients caused by a fast-closing needle valve and accompanying water hammer-induced cavitation-vortex flow are numerically investigated using the shear stress transport (SST) k–ω model with a pressure–density equation considering the weak compressibility of water and the new cavitation models based on the Rayleigh–Plesset equation that takes into account the surface tension and kinematic viscosity effect on the bubble radius change rate, respectively. The predicted pressure evolutions exhibit reasonable agreement with the experimental data, and the interfacial surface tension-based model is the most accurate in predicting transient pressure evolutions, followed by the kinematic viscosity-based model and the Singhal cavitation model, based on three proposed error indices with minimum of 0.0515, 0.1337, and 0.0035. The wave speed in the upper part of the cavitation trailing edge is higher than that in the lower part, resulting in the evolution of inclination angle of the trailing edge. According to the convective transport mechanism of vorticity, the change rate of vorticity in non-cavitation regions is mainly dominated by the stretching–twisting term and the turbulent fluctuation–diffusion term. In cavitation regions, the dilatation–shrink term makes significant contributions to the change rate of vorticity, and the baroclinic-torque term mainly acts near the edge of the cavitation region.

The RNG k‐ε turbulence model combined with cavitation model was used to simulate unsteady cavitating flows inside a double‐suction centrifugal pump under different flow rate conditions based on hexahedral structured grid. The numerical external characteristic performances agree well with the experimental performances. The predicted results show that the turbulence kinetic energy and the turbulence dissipation rate inside the impeller at design flow rate are lower than those at other off‐design flow rates, which are caused by various vortexes. Based on frequency‐domain analyses in the volute casing, the blade passing frequency is the dominant one of the pressure fluctuations except the vicinity of volute tongue for all operating cases, and the dominant frequency near the volute tongue ranges from 0 to 0.5 times the blade passing frequency for other off‐design points, while the blade passing one near the volute tongue is the dominant one of the pressure fluctuations at design point. The increase of flow rate reduces the pressure fluctuations amplitude. For cavitation cases, the blade loading of the middle streamline increases a bit during the initial stage, but, for serious cavitation, the blade loading near the blade inlet reduces to 0 and even negative values, and the serious cavitation bubbles block the blade channels, which results in a sharp drop in pump head. Under noncavitation condition, the predicted power related to the pressure in the impeller channels increases from the inlet to the exit, while, under different cavitation conditions at the design flow rate, these power‐transformation distributions in the impeller channels show that these power conversions are affected by the available NPSH_a and the corresponding work in leading regions of the blades increases increases gradually a bit, and then it increases sharply in the middle regions, but it decreases in the blade trailing regions and is greatly influenced by secondary flows.

This work uses a commercial computational fluid dynamics code to predict three-dimensional (3D) vortex flows in a large centrifugal-pump station under construction in China and proposes relevant vortex-eliminating schemes. Because of the complex nature of the vortex flows in sumps, different turbulence models, namely, standard k–ε, re-normalization group k–ε and realizable k–ε models, are first used to investigate their feasibility in predicting flows in a small physical model of an open pump sump, and various vortex streamlines and strength in the sump are predicted, analysed, and compared with the experimental data. The comparisons show that the realizable k–ε model predicts the position and strength of free-surface, sidewall-attached, and floor-attached vortices more accurately than the other two models. Then, the realizable k–ε model is used here to investigate 3D vortex flows in a large pumping-station sump. All the various vortices, such as free-surface, wall-attached vortices, are successfully predicted. Thus, based on the information of location, shape, size, and strength of the calculated vortices, three types of vortex-eliminating devices are proposed and their corresponding vortex suppression effects are analysed. These results will be used as reference for the safe and stable operation of the Hui–Nan–Zhuang pumping station in the future.

Transient flow characteristics and dissipation mechanism in pressurized pipeline were investigated based on 1D friction models and 3D turbulence models, where the pressure–density model was combined into the 3D continuity equation allowing for the elasticity of the fluid and the pipes. The applicability of 3D realizable k–ε and 3D SST (shear stress transport) k–ω turbulence models was verified with comparison to 1D traditional water hammer models and the experimental data for fast closing of the valve in the reservoir–pipe–valve system. The valve closure rule was instantaneously carried out using the grid slip CFD (computational fluid dynamics) technique. The SST k–ω turbulence model has the highest accuracy in predicting the pressure attenuation of transient flows. The 3D detailed flow field confirms that the asymmetric flows induced by the change of valve opening within approximately three-fourths of the pipe inner diameter before the valve are captured. In the pressure wave cycles, the unsteady inertia, axial pressure gradient, viscous shear stress and turbulent shear stress mainly influence the velocity variations. During the pressure wave propagation, the viscous and turbulent dissipation are critical in the pressure attenuation in the wall region; the viscous dissipation is mainly concentrated in the viscous sublayer, while the turbulent dissipation increases to the maximum values at y+ = 13–23.