Seckin Gokaltun

Lattice Boltzmann computations of incompressible laminar flow and heat transfer in a constricted channel

A multi-population thermal lattice Boltzmann method (TLBM) is applied to simulate incompressible steady flow and heat transfer in a two-dimensional constricted channel. The method is validated for velocity and temperature profiles by comparing with a finite element method based commercial solver. The results indicate that, at various Reynolds numbers, the average flow resistance increases and the heat transfer rate decreases in a constricted channel in comparison to a straight channel. The effect of the constriction ratio is also investigated. The results show that the presented numerical model is a promising tool in analyzing simultaneous solution of fluid flow and heat transfer phenomena in complex geometries.

Numerical Study of Axisymmetric Gas Flow in Conical Micronozzles by DSMC and Continuum Methods

Gas flow in conical micronozzles was numerically studied by DSMC and continuum methods. Experimental data of helium flow in a converging-diverging nozzle with throat diameter of 1 mm were used as the reference problem. In the experiment, the throat Reynolds number ranged from 0.16 to 22.69 and the flow experienced the transition from slip to transitional regimes at the nozzle entrance. 2D axisymmetric models were used for both DSMC and continuum methods with slip wall boundary condition applied in the latter one. The two different approaches were validated with the experimental data and compared with each other for performance evaluations. For throat Reynolds number larger than 10, the two numerical methods predicted nozzle exit thrust values in agreement with the experimental data. Pressure and Mach number contours were similar in pattern, but different in magnitude and smoothness of distributions. DSMC method predicts larger thrust value compared to continuum solution for Re < 10 where the flow is in transition regime.

Lattice Boltzmann Method for Steady Gas Flows in Microchannels With Imposed Slip Wall Boundary Condition

In this paper, we use a lattice Boltzmann method (LBM) for simulation of rarefied gas flows in microchannels at the slip flow regime. LBM uses D2Q9 lattice structure and BGK collision operator with single relaxation time. The solid wall boundary conditions used in this paper are based on the idea of bounceback of the non-equilibrium part of particle distribution in the normal direction to the boundary. The same idea is implemented at inlet and exit boundaries as well as at the wall surfaces. The distribution functions at the solid nodes are modified according to imposed density and slip velocity values at the wall boundaries. Simulation results are presented for microscale Couette and Poiseuille flows. The results are validated against analytical and/or experimental data for the slip velocity, nonlinear pressure drop and mass flow rate at various flow conditions. It was observed that the current application of LBM can accurately recover the physics of microscale flow phenomena in microchannels. The type of boundary treatment used in this study enables the implementation of coupled simulations where the flow properties at the regions near the wall can be obtained by other numerical methods such as the Direct Simulation Monte Carlo method (DSMC).© 2007 ASME

Verification and Validation of CFD Simulation of Pulsating Laminar Flow in a Straight Pipe

In this paper, verification and validation analysis for pulsatile flow in a straight pipe is presented. Numerical results were obtained using the finite volume method, and optimum grid density was determined using the Grid Convergence Index (GCI) calculation for the steady flow case. Based on the GCI calculation procedure we have developed a similar calculation for Time Convergence Index for the unsteady simulations. Transient computations were performed to obtain time dependent solutions of pulsating flow using three dierent time-steps and TCI was calculated to check time-step convergence. The numerical solution for the unsteady case was verified by comparing with accurate analytical solutions of Navier-Stokes equations for the pulsating flow in a pipe. Finally the numerical results were validated with experimental data. The results indicate that there is good agreement between the present results and analytical and experimental solutions. This paper stands as a good example for verification studies of unsteady CFD simulations where we have shown that TCI calculation can be applied in a similar procedure to grid convergence index calculation.

A Numerical Approach for the Simulation of Internal Nozzle Flow in a Pressure Swirl Atomizer Using Different Turbulent Models and Towards an Effective Inlet Weber Number

VOF Multiphase model is used to simulate the flow inside a pressure-swirl-atomizer. The capability of the Reynolds Stress Model and variants of the K-e and K-ω models in modeling of turbulence has been investigated in the commercial computational fluid dynamics (CFD) software FLUENT 6.3. The Implicit scheme available in the volume-of-fluid (VOF) model is used to calculate the interface representation between phases. The atomization characteristics have been investigated as well as the influence of the inlet swirl strength of the internal flow. The numerical results have been successfully validated against experimental data available for the computed parameters. The performance of the RNG K-e model was found to be satisfactory in reducing the computational cost and introducing an effective Weber number for the flow simulated in this study.Copyright

Statistical Modeling of Rarefied Gas Channel Flows

Lattice Boltzmann method (LBM) and direct simulation Monte Carlo (DSMC) method are used for analysis of moderate Knudsen number phenomena. Simulation results are presented for pressure-driven isothermal rarefied channel flow at various pressure ratios. Analytical equations for non-linear pressure distribution and velocity profiles along the channel axis are used to verify the present LBM and DSMC results. We conclude that the LBM method can be used as an alternative model to DSMC simulations.Copyright

Verification and Validation Studies for a Piloted Methane-Air Jet Flame

In this paper, verification and validation analysis for a turbulent jet flame is presented. The numerical simulations were performed for a piloted methane-air jet flame in an axisymmetric burner (jet Reynolds number of 22,400). Numerical results were obtained using the finite volume method on structured grids. The verification of the numerical solutions was performed by calculating the Grid Convergence Index (GCI). A set of three different grids is used to calculate the discretization uncertainty where each grid was generated by doubling the number of cells in each direction of the coarser grid. The value of GCI was used to calculate the observed order of convergence of the numerical method for local values of temperature, mass fraction of species and local velocity at various points along the centerline of the flow domain. A detailed chemical mechanism was used with 16 species and 41 reactions. Numerical simulations showed that the fine grid result was in the asymptotic range of convergence. The observed order of convergence was found to be between 1.8 and 4.5. The error band for the Richardson extrapolated value was found to be below 1.3%. The effect of k-e turbulence parameter C2e was investigated and a value of 1.8 was found to be appropriate to match the fine grid results with the experimental data. Finally the numerical results were validated with experimental data using the local measurements of temperature and species mass fractions. Comparison of the computational results with experimental data showed that simulations correctly predict the trends observed in the measurements. Good agreement between experiments and simulations was obtained with the finest grid, which predicted peak temperature within 3.0% of experiment. For the main reaction products of CO2 and H2 O, the peak values are captured within 0.33% and 3% of experiment, respectively.Copyright

Verification and Validation Studies for a Laminar Non-Premixed Methane/Air Flame

In this paper, verification and validation analysis for a nonpremixed methane/air laminar flame is presented. Numerical results were obtained using the finite volume method on structured grids. The verification of the numerical solutions was performed by using the Grid Convergence Index (GCI) and Richardson extrapolation techniques. A set of three different grids is used to calculate the error due to discretization where each grid was generated by doubling the number of cells in each direction of the coarser grid. The local value of GCI was used to calculate the observed order of convergence of the numerical method for local values of temperature and mass fractions of reaction products at various points along the flow domain. The largest error band at the finest grid solution was observed to be 4.6% for the static temperature, 0.5% for the mass fraction of methane and 2.9% for the mass fraction of water vapor. Finally the numerical results were validated with experimental data using the local measurements of temperature and species mass fractions. The results indicate that there is relatively good agreement between the present results and experimental data although a simple one-step reaction model was used for the methane/air combustion. The average deviation was found to be around 25%, 21% and 10% for temperature, methane mass fraction and water vapor mass fraction respectively.© 2006 ASME