Numerical Simulation of Cavitating Flow in a Francis Turbine

The 3-D unsteady Reynolds averaged Navier-tokes equations based on the pseudo-homogeneous flow theory and a vapor fraction transport-equation that accounts for non-condensable gas are solved to simulate cavitating flow in a Francis turbine. The calculation results agreed with experiment data reasonably. With the decrease of the Thoma number, the cavity first appears near the centre of the hub. At this stage the flow rate and the efficiency change little. Then the cavity near the centre of the hub grows thick and the cavities also appear on the blade suction side near outlet. With further reduce of the Thoma number the cavitation extends to the whole flow path, which causes flow rate and efficiency decrease rapidly.

Electrohydrodynamic Gas Pump in a Vertical Tube

The characteristics of an electrohydrodynamic (EHD) gas pump were experimentally evaluated in this study. Experiments were conducted using positive DC voltage (13.0 kV – 30.0 kV), applied to a thin corona wire of diameter 0.51 mm. The ground plate used in the study was mounted on the inner surface of a cylinder and had a width of 12.7 mm and a total exposed surface area of 1.52 × 10−3 m2. The pumps tested were identical except for their spacing between the emitting electrode and the ground plate, which varied from 0.66D to 1.33D. The results show that the current on the ground plate increases as the applied voltage increases. It is also observed that the applied voltage at which flow was first detected in the cylinder correlates well with the electrode spacing. It appears that the four electrodes placed along the surface of the pipe were able to disturb the boundary layer enough to create a uniform flow profile within the pipe. As a result, flow as high as 2.5 m/s was observed in the cylinder with an electrode spacing of L/D = 1.33. The results also show that the ionic wind velocity increases with an increasing EHD Reynolds number for L/D = 0.66 and 1.0, but decreases with the EHD Reynolds number for L/D > 1.

Effect of Blade Angle of Attack and Hub to Tip Ratio on Mass Flow Rate in an Axial Fan at a Fixed Rotational Speed

In this investigation an attempt is made to find the best hub to tip ratio, the maximum number of blades, and the best angle of attack of an axial fan with flat blades at a fixed rotational speed for a maximum mass flow rate in a steady and turbulent conditions. In this study the blade angles are varied from 30 to 70 degrees, the hub to tip ratio is varied from 0.2 to 0.4 and the number of blades are varied from 2 to 6 at a fixed hub rotational speed. The results show that, the maximum flow rate is achieved at a blade angle of attack of about 45 degrees for when the number of blades is set equal to 4 at most rotational velocities. The numerical results show that as the hub to tip ratio is decreased, the mass flow rate is increased. For a hub to tip ratio of 0.2, and an angle of attack around 45 degrees with 4 blades, a maximum mass flow rate is achieved.

Numerical Study of Dead-End Micro Polymer Electrolyte Membrane Fuel Cell

In this paper, 3D full size simulation on single cell dead-end micro PEMFC is carried out using Computational Fuel Cell Dynamics (CFDC) analysis. The active area in this Micro PEMFC is about 10 cm2, producing 1A of current under standard condition (25 °C and 1 atm). The dead end anode configuration is achieved by increasing the air flow rate well above stoichometric at the cathode to obtain the complete depletion of hydrogen at anode exit. It is also assumed that the presence of water is only in the vapor phase. Different types (single serpentine and triple serpentine) of gas channel design in dead-end anode are studied and results are compared. The polarization curves for both designs as well as contour plots for the different cell region are presented.

Simulation of Flow Field on a Large Scale Wind Turbine

Large scale wind turbines and wind farms continue to evolve mounting 94.1GW of the electrical grid capacity in 2007 and expected to reach 160.0GW in 2010 according to World Wind Energy Association. They commence to play a vital role in the quest for renewable and sustainable energy. They are impressive structures of human responsiveness to, and awareness of, the depleting fossil fuel resources. Early generation wind turbines (windmills) were used as kinetic energy transformers and today generate 1/5 of the Denmark’s electricity and planned to double the current German grid capacity by reaching 12.5% by year 2010. Wind energy is plentiful (72 TW is estimated to be commercially viable) and clean while their intensive capital costs and maintenance fees still bar their widespread deployment in the developing world. Additionally, there are technological challenges in the rotor operating characteristics, fatigue load, and noise in meeting reliability and safety standards. Newer inventions, e.g., downstream wind turbines and flapping rotor blades, are sought to absorb a larger portion of the cost attributable to unrestrained lower cost yaw mechanisms, reduction in the moving parts, and noise reduction thereby reducing maintenance. In this work, numerical analysis of the downstream wind turbine blade is conducted. In particular, the interaction between the tower and the rotor passage is investigated. Circular cross sectional tower and aerofoil shapes are considered in a staggered configuration and under cross-stream motion. The resulting blade static pressure and aerodynamic forces are investigated at different incident wind angles and wind speeds. Comparison of the flow field results against the conventional upstream wind turbine is also conducted. The wind flow is considered to be transient, incompressible, viscous Navier-Stokes and turbulent. The k-ε model is utilized as the turbulence closure. The passage of the rotor blade is governed by ALE and is represented numerically as a sliding mesh against the upstream fixed tower domain. Both the blade and tower cross sections are padded with a boundary layer mesh to accurately capture the viscous forces while several levels of refinement were implemented throughout the domain to assess and avoid the mesh dependence.

Design of Small Wind Turbine

The project has been completed, and all of the aforementioned objectives have been achieved. An anemometer has been constructed to measure wind speed, and a wind vane has been built to sense wind direction. An LCD module has been acquired and has been programmed to display the wind speed and its direction. An H-Bridge circuit was used to drive a gear motor that rotated the nacelle toward the windward direction. Finally, the blade pitch angle was controlled by a swash plate mechanism and servo motors installed on the generator itself. A microcontroller has been programmed to optimally control the servo motors and gear motor based on input from the wind vane and anemometer sensors.

Wall Normal Jet Produced by DBD Plasma Actuator With Doughnut-Shaped Electrode

Properties of coaxial annular jets produced by a dielectric barrier discharge (DBD) plasma actuator with a doughnut shaped electrodes were investigated under atmospheric pressure and room temperature. The actuator consists of two circular electrodes sandwiching a thin dielectric layer. By applying 0 – ±3.3 kV between the electrodes at radio frequencies, the plasma jet is formed near the inner edge of the top electrode. The radial jet runs toward the center of the electrode and then impinges at the center to generate a wall normal annular jet. The evolution of the wall normal jet was observed precisely using particle image velocimetry (PIV) system. It was found that characteristic velocities increase in proportion to the bursting frequency and inversely proportional to the inner diameter of the electrode at the surging time of the voltage at 5.0 × 10−6sec.

A Condensed Phase Computational Fluid Dynamics Model for Polymer Melt Flow

This paper presents a condensed phase computational fluid dynamics (CFD) based tool for modeling the processes of melting, flow and gasification of thermoplastic materials exposed to a high heat flux. Potential applications of the tool include investigating the behavior of polymer materials commonly used in personal computers and computer monitors if exposed to an intense heat flux, such as occurs during a fire. The finite-volume based model uses a three-dimensional body-fitted time dependent grid formulation to solve the unsteady Navier Stokes equations. A multi-grid method is used to accelerate convergence at each time step. Sub-models are included to describe the temperature dependent viscosity relationship and in-depth gasification and absorption of thermoplastic materials, free surface flows and surface tension. A series of test cases have been performed and the model results are compared to experimental data to investigate the impacts of different sub-models, boundary conditions, material properties and problem configurations on the accuracy, efficiency and applicability of the modeling tool.

Drag Reduction for Blunt Body With Cross Flow by Extremum-Seeking Control

The active flow control, which can adapt to variation of flow velocity and/or direction, is an effective technique to achieve drag reduction. The present study has investigated a separated shear layer and established two control systems; the system reduces drag force and lift force by controlling the separated shear layer to reattachment for variation of flow velocity and /or direction. The adaptive control system to the variation of flow velocity was constructed by using a hot wire anemometer as a sensor to detect flow separation. The system to flow direction was constructed by using pressure transducers as a sensor to estimate drag force and lift force. The extremum-seeking control was introduced as a controller of the both systems. It is indicated from the experimental results that adaptive drag/lift control system to various flow velocity ranging from 3 to 7 m/s and various flow direction ranging from 0 to 30 deg. was established.

Simulations to Determine Laminar Loss Coefficients for Flow in Circular Ducts With Arbitrary Planar Bifurcation Geometries

The goal of this study was to determine laminar stagnation pressure loss coefficients for circular ducts in which flow encounters a planar bifurcation. Flow conditions and pressure losses in these laminar bifurcations are of interest in microfluidic devices, in porous media, and in other networks of small ducts or pores. Until recently, bifurcation geometries had been studied almost exclusively for turbulent flow, which is often found in fluid supply and drain systems. Recently, pressure loss coefficients from simulations of a few arbitrary bifurcation geometries in two-dimensions have been published — the present study describes the extension of these two-dimensional simulations to three-dimensional circular ducts. The pressure loss coefficients determined in this study are intended to allow realistic simulation of existing laminar flow networks or the design of these networks. This study focused on a single inlet duct with two outlet ducts, which were allowed to vary in diameter, flow fraction, and angle — all relative to the inlet duct. All ducts considered in this study were circular with their axes in a common plane. Laminar stagnation pressure loss coefficients were determined by simulating incompressible flow through 475 different geometries and flow condition combinations. In all cases, the flow was laminar in the inlet and outlet ducts with a Reynolds number of 15 in the inlet duct. Simulations of the dividing flow geometries were done using FLUENT and a custom written computer code, which automated the process of creating the three-dimensional flow geometries. The outputs, pressure and velocity distributions at the inlet and outlets, were averaged over the circular ducts and then used to calculate pressure loss coefficients for each of the geometries and flow fraction scenarios simulated. The results for loss coefficient for the geometries considered ranged from 2.0 to 70. The loss coefficient for any geometry increased significantly as the outlet flow fraction increased. A consistent increase in loss coefficient was also observed as a function of decreasing outlet duct diameter. Less significant variation of the loss coefficient was observed as a function of the angles of the outlet ducts.