Unsteady Flow Structure on a Centrifugal Pump: Experimental and Numerical Approaches

Both experimental and numerical studies of the unsteady pressure field inside a centrifugal pump have been carried out. The unsteady patterns found for the pressure fluctuations are compared and a further and more detailed flow study from the numerical model developed will be presented in this paper. Measurements were carried out with pressure transducers installed on the volute shroud. At the same time, the unsteady pressure field inside the volute of a centrifugal pump has been numerically modelled using a finite volume commercial code and the dynamic variables obtained have been compared with the experimental data available. In particular, the amplitude of the fluctuating pressure field in the shroud side wall of the volute at the blade passing frequency is successfully captured by the model for a wide range of operating flow rates. Once the developed numerical model has shown its capability in describing the unsteady patterns experimentally measured, an explanation for such patterns is searched. Moreover, the possibilities of the numerical model can be extended to other sections (besides the shroud wall of the volute), which can provide plausible explanations for the dynamic interaction effects between the flow at the impeller exit and the volute tongue at different axial positions. The results of the numerical simulation are focused in the blade passing frequency in order to study the relative effect of the two main phenomena occurring at that frequency for a given position: the blade passing in front of the tongue and the wakes of the blades.

Computational Aerodynamics of a Winston-Cup-Series Race Car

Since the goal of racing is to win and since drag is a force that the vehicle must overcome, a thorough understanding of the drag generating airflow around and through a race car is greatly desired. The external airflow contributes to most of the drag that a car experiences and most of the downforce the vehicle produces. Therefore, an estimate of the vehicle’s performance may be evaluated using a computational fluid dynamics model. First, a computer model of the race car was created from the measurements of the car obtained by using a laser triangulation system. After a computer-aided drafting model of the actual car was developed, the model was simplified by removing the tires, roof strakes, and modification of the spoiler. A mesh refinement study was performed by exploring five cases with different mesh densities. By monitoring the convergence of the computed drag coefficient, the case with 2 million elements was selected as being the baseline case. Results included flow visualization of the pressure and velocity fields and the wake in the form of streamlines and vector plots. Quantitative results included lift and drag, and the body surface pressure distribution to determine the centerline pressure coefficient. When compared with the experimental results, the computed drag forces were comparable but expectedly lower than the experimental data mainly attributable to the differences between the present model and the actual car.

A Remote Area Power Supply Using Wind Power and Cold Thermal Storage

A remote area power supply using cold thermal storage and wind as the energy source is proposed. The primary objective is to provide a renewable energy remote area power supply with cheaper and more robust storage than lead-acid batteries. The proposal amalgamates a vapour compression refrigeration system with a Rankine cycle engine, both using the same working fluid. A tank of freezing brine acts as the condenser in the Rankine cycle and as the evaporator in the refrigeration cycle but also provides the “energy storage”. Analysis of the system indicates that it is practical and that its performance is comparable with existing battery based systems.

Stochastic Estimation and Non-Linear Wall-Pressure Sources in a Separating/Reattaching Flow

Simultaneous wall-pressure and PIV measurements are used to study the conditional flow field associated with surface-pressure generation in a separating/reattaching flow established over a fence-with-splitter-plate geometry. The conditional flow field is captured using linear and quadratic stochastic estimation based on the occurrence of positive and negative pressure events in the vicinity of the mean reattachment location. The results shed light on the dominant flow structures associated with significant wall-pressure generation. Furthermore, analysis based on the individual terms in the stochastic estimation expansion shows that both the linear and non-linear flow sources of the coherent (conditional) velocity field are equally important contributors to the generation of the conditional surface pressure.

An Approach to Parallel Computing in an Eulerian-Lagrangian Two-Phase Flow Model

An approach to parallel solution of an Eulerian-Lagrangian model of dilute gas-solid flows is presented. Using Lagrangian treatments for the dispersed phase, one of the principal computational challenges arises in models in which inter-particle interactions are taken into account. Deterministic treatment of particle-particle collisions in the present work pose the most computationally intensive aspect of the simulation. Simple searches lead to algorithms whose cost is O(N2p) where Np is the particle population. The approach developed in the current effort is based on localizing collision detection neighborhoods using a cell-index method and spatially distributing those neighborhoods for parallel solution. The method is evaluated using simulations of the gas-solid turbulent flow in a vertical channel. The instantaneous position and the velocity of any particle is obtained by solving the equation of motion for a small rigid sphere assuming that the resulting force induced by the fluid reduces to the drag contribution. Binary particle collisions without energy dissipation or inter-particle friction are considered. The carrier flow is computed using Large Eddy Simulation of the incompressible Navier-Stokes equations. The entire dispersed-phase population is partitioned via static spatial decomposition of the domain to maximize parallel efficiency. Simulations on small numbers of distributed memory processors show linear speedup in processing of the collision detection step and nearly optimal reductions in simulation time for the entire solution.

Physics of Tip Clearance Flow in Turbomachinery (Keynote Paper)

In a recent advanced aerodynamic design of turbomachinery, the physical interpretation of three-dimensional flow field obtained by a numerical simulation is important for iterative modifications of the blade or impeller geometry. This paper describes an approach to the physical interpretation of the tip clearance flow in turbomachinery. First, typical flow phenomena of the tip clearance flow are outlined for axial and radial compressors, pumps and turbines to help comprehensive understanding of the tip clearance flow. Then, a vortex-core identification method which enables to extract the vortical structure from the complicated flow field is introduced, since elucidation of the vortical structure is essential to the physical interpretation of the tip clearance flow. By use of the vortex-core identification, some interesting phenomena of the tip clearance flows are interpreted, especially focussing on axial flow compressors.

PIV Measurements in a Confined Jet

This paper reports PIV measurements made at three locations in an axisymmetric, confined jet that is approaching a free surface from below. The apparatus consists of a tank 40.5 cm × 40.5 cm at its base and 61 cm high. A 9 mm diameter nozzle is centered in the base of the tank and directs a jet of water upwards. The jet produced has a top-hat velocity profile with a maximum deviation of 0.32% of the mean and an axial relative turbulence intensity of 0.60%. The water is removed from the tank by an overflow around the perimeter of the tank. The PIV measurements achieved a spatial resolution of between 0.425–1.08 mm. The measurements show details of the velocity field in three regions of the flow; at the jet exit, near the surface on the centerline of the jet, and near the top corner of the tank. The centerline velocity remains at the exit velocity until ≈5D from the exit. The axial confinement of the jet begins to significantly influence the centerline velocity at ≈13D from the free surface. All entrained fluid is deflected downward from the horizontal surface flow as it approaches the overflow around the perimeter of the tank. This creates a large recirculation region in the upper region of the tank driven by the downward flow along the wall and the upward flow of the jet itself at the center of the tank.

Towards Practical Flow Sensing and Control via POD and LSE Based Low-Dimensional Tools (Keynote Paper)

We present the application of Proper Orthogonal Decomposition (POD) and Linear Stochastic Estimation (LSE) based low-dimensional methods to the flow over a backward facing ramp with an adjustable flap above the ramp which allows for dynamic variation of the adverse pressure gradient. There is a range of flap angles where the flow is incipiently separated so that this relatively simple experiment can be used to flush out ideas for active feedback separation control strategies. The study utilized a combination of PIV and multi-point wall pressure measurements to estimate the full velocity field (mean plus fluctuating) from a modified complementary technique. Specifically we want to identify a low-dimensional mean flow to observe when the profiles are inflectionary, i.e., the incipient condition, just from wall pressure. We demonstrate via this method, that a reasonable estimate of the low dimensional full velocity field can be obtained. This is important for practical active feedback flow control strategies since from wall pressure we can estimate the state of the flow without resorting to probes in the flow.

Numerical and Experimental Evaluation of Lift Forces in Flows Around Surface-Mounted Bluff Bodies

The flow around a surface-mounted bluff body with cuboid shape is investigated. Therefore, the velocity field including the distribution of the turbulent kinetic energy is computed and compared with experimental Laser Doppler Anemometry data. Several different turbulence models, namely the standard k-ε model, the Wolfshtein two-layer k-ε model and a Large-Eddy approach are validated. Since the Large-Eddy model remains the only model representing the flow accurate, it is chosen for further investigations. The pressure distribution on the body and on the carrying surface around the body is analysed. The lift coefficients are computed for Reynolds numbers, ranging from 1.1 × 104 up to 4.4 × 104. The lengths of the separation zone above and the recirculation zone downstream the body are evaluated.

Numerical Simulation of the Two-Phase Flow in Centrifugal Pump Impellers

The numerical simulations of three types of two-phase flow in centrifugal pump impellers are described. First, the liquid-solid particle flow is modeled by an Euler-Lagrangeian approach assuming a mass concentration less than 5% and particle diameters being less than 1000 microns. The empirical erosion model to predict the local and total wear is calibrated by measurements. Second, the influence of the relative air contents on the head-drop is simulated assuming a relatively small volume fraction and applying a simple one-fluid model. The mixture is characterized by a common density depending on the flow field. Finally, the cavitating flow is studied by implementing the Rayleigh equation into the numerical procedure describing the transient process of bubble growth and collapse. The developed simulation tools are applied to predict the three types of two-phase flows in impellers. Within the defined ranges of application the simulation results agree fairly well with the experimental data.