Inverse Calculation of Flame Impingement Heat Transfer

Inverse calculations are presented here for the estimation of heat transfer from an impinging flame on a flat surface. This work is a preliminary exercise for estimating heat transfer from an impinging plasma jet, where direct measurements can be very difficult and costly, and the correlations based on air or water jet impingement measurements may not be applicable because of the very high temperature (and property) gradients. As the gas flame impinges on an initially cold flat plate, the temperature evolution on the backside is recorded using an infrared camera. The time–temperature data thus obtained are then compared with those predicted by a finite volume method based code. The code uses a polynomial series for estimating the convection coefficient, which varies with radial distance. The coefficients of this polynomial are treated as a set of parameters to be estimated through the Levenberg-Marquardt approach. The results obtained so far indicate that it may be possible to use such an approach for estimating heat transfer from a plasma jet.

Numerical Modeling of Flow Over a Dam Spillway

Free surface flows are frequently encountered in hydraulic engineering problems including water jets, weirs and around gates. An iterative solution to the incompressible two-dimensional vertical steady Navier-Stokes equations, comprising momentum and continuity equations, is used to solve for the priori unknown free surface, the velocity and the pressure fields. The entire water body is covered by a unstructured finite element grid which is locally refined. The dynamic boundary condition is imposed for the free surface where the pressure vanishes. This procedure is done continuously until the normal velocities components vanish. To overcome numerical errors and oscillations encountering in convection terms, the SUPG (streamline upwinding Petrov-Galerkin) method is applied. The solution method is tested for different discharges onto a standard spillway geometries. The results shows good agreement with available experimental data.

Two-Fluid Large-Eddy Simulation Approach for Gas-Particle Turbulent Flows Using Equilibrium Assumption

This article illustrates a two-fluid large-eddy simulation (LES) approach for gas-particle turbulent flows. The equilibrium assumption in which the velocity of particles is approximated in terms of the velocity and acceleration of the gas phase, is made for the development of gas-particle LES formulation in this study. A filtered Eulerian velocity field is defined for particles and expressed in terms of the temporal and spatial derivatives of the gas-phase filtered velocity field. Also, filtered particle concentration defined in the Eulerian framework is governed by a transport equation with a closure problem resulted from filtering the particle concentration nonlinear convection term and in the form of subgrid-scale particle flux. A Smagorinsky kind of formulation is used to model the subgrid-scale particle flux and close the transport equation of the filtered particle concentration. The developed gas-particle LES formulation is implemented in a homogeneous shear turbulence configuration and results are discussed. It is shown that the equilibrium assumption is valid for sufficiently small particle time constants through conducting the direct numerical simulation of the same configuration.

Velocity and Pressure Measurements Along a Row of Confined Cylinders

The results of flow experiments performed in a cylinder array designed to mimic a VHTR Nuclear Plant lower plenum design are presented. Pressure drop and velocity field measurements were made. Based on these measurements, five regimes of behavior are identified that are found to depend on Reynolds number. It is found that the recirculation region behind the cylinders is shorter than that of half cylinders placed on the wall representing the symmetry plane. Unlike a single cylinder, the separation point is found to always be on the rear of the cylinders, even at very low Reynolds number. Boundary layer transition is found to occur at much lower Reynolds numbers than previously reported.

Effect of Divergence Angle on Flow Through Inlet Section of Diffuser of a Gas Turbine Combustion Chamber: The CFD Approach

In the combustor inlet diffuser section of gas turbine engine, high-velocity air from compressor flows into the diffuser, where a considerable portion of the inlet velocity head PT3 − PS3 is converted to static pressure (PS) before the airflow enters the combustor. Modern high through-flow turbine engine compressors are highly loaded and usually have high inlet Mach numbers. With high compressor exit Mach numbers, the velocity head at the compressor exit station may be as high as 10% of the total pressure. The function of the diffuser is to recover a large proportion of this energy. Otherwise, the resulting higher total pressure loss would result in a significantly higher level of engine specific fuel consumption. The diffuser performance must also be sensitive to inlet velocity profiles and geometrical variations of the combustor relative to the location of the pre-diffuser exit flow path. Low diffuser pressure losses with high Mach numbers are more rapidly achieved with increasing length. However, diffuser length must be short to minimize engine length and weight. A good diffuser design should have a well considered balance between the confliction requirements for low pressure losses and short engine lengths. The present paper describes the effect of divergence angle on diffuser performance for gas turbine combustion chamber using Computational Fluid Dynamic Approach. The flow through the diffuser is numerically solved for divergence angles ranging from 5 to 25°. The flow separation and formation of wake regions are studied.

Simulation of the Flow Condition Near the Intake of a Pump Sump

It is known that the flow condition in the pump sump is very complicated, which usually performs as several types of vortex, water wave of free surface, vibration, noise, and etc. To make clear the flow condition experimentally and numerically is very important to develop the performance and operating stability of the pump station. As one of the projects with Hitachi Industries ltd. Co., Japan, the investigation experimentally on the internal flow condition of the pump sump has been carried out in Tsinghua University. In this paper, we introduce the simulation results, which can show more detail information near the intake of the pump sump. The simulation is focused on the area near the intake, together with the extension of up-stream and down-stream. The calculation research includes two parts: steady simulation and unsteady simulation by VOF model, which is provided by the commercial software of Fluent. Through the steady simulation, the distribution of free water surface (water height) near the intake, as well as the flow condition inside the intake, were obtained. Comparing with the experimental data, a good agreement was observed. After analyzing the unsteady calculation results by VOF, four stages of the development of air-entraining vortex, and the wave characteristic of free water surface, were obtained, which were in accord with the experiment visualizing.

Numerical Jet Atomization: Part II — Modeling Information and Comparison With DNS Results

The main objective of our work is to develop direct numerical simulation tools for the primary break up of a jet. Results can help to determine closure relation in the ELSA model [1] which is based on a single-phase Eulerian model and on the transport equation for the mean liquid/gas interface density in turbulent flows. DNS simulations are carried out to obtain statistical information in the dense zone of the spray where nearly no experimental data are available. The numerical method should describe the interface motion precisely, handle jump conditions at the interface without artificial smoothing, and respect mass conservation. We develop a 3D code [2], where interface tracking is ensured by Level Set method, Ghost Fluid Method [3] is used to capture accurately sharp discontinuities, and coupling between Level Set and VOF methods is used for mass conservation [4]. Turbulent inflow boundary conditions are generated through correlated random velocities with a prescribed length scale. Specific care has been devoted to improve computing time with MPI parallelization. The numerical methods have been applied to investigate physical processes that are involved in the primary break up of an atomizing jet. The chosen configuration is close as possible of Diesel injection (Diameter D = 0.1 mm, Velocity = 100m/s, Liquid density = 696kg/m3, Gas density = 25kg/m3). Typical results will be presented. From the injector nozzle, the turbulence initiates some perturbations on the liquid surface, that are enhanced by the mean shear between the liquid jet and the surrounding air. The interface becomes very wrinkled and some break-up is initiated. The induced liquid parcels show a wide range of shapes. Statistics are carried out and results will be provided for liquid volume fraction, liquid/gas interface density, and turbulent correlations.

Natural Convection in a Horizontal Cavity Partially Occupied by a Porous Media Saturated by a Coolant Liquid

The natural convection is studied in a cavity witch the lower half is filled with a porous media that is saturated with a first fluid (liquid), and the upper is filled with a second fluid (gas). The horizontal borders are heated and cooled by uniform heat fluxes and vertical ones are adiabatic. The formulation of the problem is based on the Darcy-Brinkman model. The density variation is taken into account by the Boussinesq approximation. The system of the coupled equations is resolved by the classic finite volume method. The numerical results show that the variation of the conductivity of the porous media influences strongly the flow structure and the heat transfer as well as in upper that in the lower zones. The effect of conductivity is conditioned by the porosity which plays a very significant roll on the heat transfer. The structures of this flow show that this kind of problem with specific boundary conditions generates a complex flow structure of several contra-rotating two to two cells, in the upper half of the cavity.

Nanoparticle Precipitation in a T-Mixer: Coupling Experimental and Numerical Investigations of the Fluid Dynamics With the Solid Formation Processes

Mixing and consequently fluid dynamic is a key parameter to tailor the particle size distribution (PSD) in nanoparticle precipitation. Due to fast and intensive mixing a static T-mixer configuration is capable for synthesizing continuously nanoparticles. The flow and concentration field of the applied mixer is investigated experimentally at different flow rates by Particle Image Velocimetry (PIV) and Laser Induced Fluorescence (LIF). Due to the PIV measurements the flow field in the mixer was characterized qualitatively and the mixing process itself is quantified by the subsequent LIF-measurements. A special feature of the LIF set up is to detect structures in the flow field, which are smaller than the Batchelor length. Thereby a detailed insight into the mixing process in a static T-Mixer is given. In this study a CFD-based approach using Direct Numerical Simulation (DNS) in combination with the solid formation kinetics solving population balance equations (PBE) is applied, using barium sulfate as modeling material. A Lagrangian Particle Tracking strategy is used to couple the flow field information with a micro mixing model and with the classical theory of nucleation. We found that the DNS-PBE approach including macro and micro mixing, combined with the population balance is capable of predicting the full PSD in nanoparticle precipitation for different operating parameters. Additionally to the resulting PSD, this approach delivers a 3D-information about all running subprocesses in the mixer, i.e. supersaturation built-up or nucleation, which is visualized for different process variables.

Effects of Particle-Particle Collisions on Particle Concentration in a Turbulent Channel Flow

This study is concerned with the effect of inter particle collisions on the particle concentration in turbulent duct flows. The time history of the instantaneous turbulent velocity vector was generated by the two-way coupled direct numerical simulation (DNS) of the Navier-Stokes equation via a pseudospectral method. The particle equation of motion included the Stokes drag, the Saffman lift, and the gravitational forces. The effect of particles on the flow is included in the analysis via a feedback force on the grid points. Several simulations for three classes of particles (28 μm Lycopodia, 50μm glass and 70μm copper) and different mass loadings were performed, and the effect of inter particle collisions on the particle concentration was evaluated and discussed. It was found that the particle-particle collisions reduce the tendency of particles to accumulate near the wall. This might be because collisions decorrelate particles with coherent eddies which are responsible for accumulation of particles near the wall. The spatial distribution of particles at the channel centerplane was compared with the experimental results of Fessler et al. (1994). The simulation results showed that the copper and glass particles had a random distribution while Lycopodium particles showed a non-random distribution with bands of particles that were preferentially concentrated.