CFD Simulation Of Flow Research Articles

Modeling Heat Transfer and Skin Friction Frequency Responses of a Cylinder in Cross-Flow—a Unifying Perspective

ABSTRACTThe dynamic behavior of skin friction and heat transfer of a cylinder in pulsating cross-flow is investigated. Existing analytical solutions are presented as transfer functions versus frequency, known from control theory. New results for Reynolds number ranges where no appropriate model exists until now are derived. These results are obtained by the combination of computational fluid dynamics and system identification (CFD/SI). In the CFD/SI approach, time series for fluctuations of skin friction and the rate of heat transfer are generated by imposing broad-band inlet velocity fluctuations in a CFD simulation of laminar flow across a cylinder. Direct numerical simulations are conducted for mean flow Reynolds numbers between 0.1 and 40, solving the incompressible Navier–Stokes equations in a two-dimensional domain using a finite volume approach. The SI framework allows to estimate transfer functions for the response of skin friction and heat transfer to velocity fluctuations from time series data. Available analytical models for the dynamic behavior of skin friction and heat transfer usually match the simulated data up to a point, but do not give any dependence on Reynolds number. This shortcoming is addressed in this work. The identified models especially excel at Reynolds numbers of order 10.

A mechanistic model for expansion loss coefficient in upward vertical annular flow

• Differential effect of flow deceleration on gas, droplets and liquid film postulated. • Developed semi-analytical expressions for velocity profile correction factor in pipe flow. • Carried out CFD simulations of gas-droplet flow in a rough-walled pipe expansion. • Effect of droplet deceleration on pressure variation elucidated. • Mechanistic correlation developed for expansion loss coefficient for annular flow. Gas–liquid annular flow draws different responses from its three constituents, namely, the liquid film, the entrained liquid droplets and the gas core, as they flow through a diverging section in a pipe. The resulting change in the pressure profile is a combination of several effects associated with the dynamic interactions among these three fields. Accurate simulation of the response using Eulerian–Eulerian computational fluid dynamics is not feasible because the processes are inherently complex and a framework of relevant and validated constitutive relations describing the physical processes is not yet available. In the present work, a simpler approach is adopted by studying the interactions individually in idealized settings, and bringing together the separate effects into a phenomenological model for pressure loss in upward vertical annular flow. The overall pressure change is expressed as a sum of three contributors: change in area of cross-section available for gas flow, change in the effective interfacial roughness leading to peaking of the velocity profile, and droplet-gas momentum exchange in the immediate downstream of the expansion. Using air–water experimental data from two expansion ratios and three half-angles of the diverging sections, a mechanistic correlation is proposed to evaluate the overall pressure loss coefficient.

Explosion Characteristics of Hydrogen for CFD Modelling and Simulation of Turbulent Gas Flow

Renewable energies became more and more important in the last years. Hydrogen as a promising energy carrier is a perfect candidate to supply the energy demand of the world. The state of the hydrogen gas (turbulences and point concentrations) has a significant impact on the gas explosion indices. A gas cloud is formed by a partial-pressure method in gas explosion experiments in the spherical 20.0∙10-3 m3 chamber. Gas in the chamber reaches an uniform state beyond in hundreds of ms. The absolute pressure for gas dispersion should be higher than 0.01 MPa for the H2 of concentration larger than 30 vol. % of fuel. The initial temperature also influences turbulent gas flow before ignition, especially in the case of the gases lighter-than-air.

Unification of particle velocity distribution functions in gas-solid flow

Continuum models have been extensively used in CFD simulation of gas-solid flow, which is usually closed with a kinetic theory of granular flow for particulate phase stresses and a correlation for interphase drag force. Particle velocity distribution function (PVDF) is the cornerstone of kinetic theory of granular flow and the PVDFs in gas-solid flow can be Maxwellian, bimodal and/or non-Maxwellian with an overpopulated high-energy tail, depending on the physical nature of gas-solid flow. Here we theoretically show that the different PVDFs can be unified under the umbrella of a recently proposed kinetic theory framework (Wang et al., 2016) and the root of different PVDFs lies in the different characteristics of mesoscale structures. Validation of theoretical analysis is then carried out by comparing the theoretical prediction with experimental and DNS data available in literature. The findings of present study offer a possibility of developing a unified kinetic theory for gas-solid flows.

A consistent steady state CFD simulation method for stratified atmospheric boundary layer flows

An effective methodology for CFD simulations of stratified atmospheric boundary layer flows is presented, based on the Poiseuille-type zero pressure gradient boundary layer. The concept of precursor inflow generation is applied in the scope of Reynolds Averaging Navier-Stokes (RANS) simulations, and combined with mean temperature control via explicit global heat correction in the energy equation. We thus obtain solutions that are fully consistent with the equation system in a one-dimensional computational domain, which then serve as inlet profiles for the three-dimensional calculation.The resulting inflow profiles are furthermore utilized for turbulence model validation and the calibration of the modeling parameters. This yields an extension of the standard k-ε turbulence model, incorporating buoyancy induced turbulence production and dissipation terms. Additionally, modified turbulent wall functions are applied, which consistently model wall roughness and heat transfer, in agreement with the similarity law. We demonstrate that the obtained turbulence model reproduces the expected analytical profiles from the Monin-Obukhov similarity theory, and also that the representation of turbulent kinetic energy shows a realistic evolution over a wide range of stratification parameters.The presented method is then applied to study flow over inhomogeneous surfaces, among others a cosine shaped hill and ridge, where lee wave patterns are observed in the expected range of Obukhov lengths. The final validation demonstrates quantitatively good agreement of the simulation results at a complex on-shore site in Germany with data from a 200 m met mast.

CFD simulations of single-phase flow in pulsed disc and doughnut columns: Axial dispersion and pressure drop

ABSTRACTCFD simulations of pulsed disc and doughnut columns are performed to understand the effects of operating and geometric parameters on axial dispersion and pressure drop in single-phase flow. CFD simulations have been carried out using a two-step approach. In the first step, the flow field is obtained by solving the continuity and the momentum equations along with the equations of the standard k–ε model of turbulence. In the second step, the species transport equation is additionally solved to obtain the residence time distribution and hence the Peclet number and axial dispersion coefficient. The computational approach is validated by comparing its predictions with the experimental data reported in the literature and then used for detailed parametric analysis.

Numerical study of Tallinn storm-water system flooding conditions using CFD simulations of multi-phase flow in a large-scale inverted siphon

The numerical experiments are carried out for qualitative and quantitative interpretation of a multi-phase flow processes associated with malfunctioning of the Tallinn storm-water system during rain storms. The investigations are focused on the single-line inverted siphon, which is used as under-road connection of pipes of the storm-water system under interest. A multi-phase flow solver of Computational Fluid Dynamics software OpenFOAM is used for simulating the three-phase flow dynamics in the hydraulic system. The CFD simulations are performed with different inflow rates under same initial conditions. The computational results are compared essentially in two cases 1) design flow rate and 2) larger flow rate, for emptying the initially filled inverted siphon from a slurry-fluid. The larger flow-rate situations are under particular interest to detected possible flooding. In this regard, it is anticipated that the CFD solutions provide an important insight to functioning of inverted siphon under a restricted water-flow conditions at simultaneous presence of air and slurry-fluid.

CFD simulation and experimental study of water-oil displacement flow in an inclined pipe

CFD simulation and experimental study of water-oil displacement flow in an inclined pipe

VirtuaSchlieren: A hybrid GPU/CPU-based schlieren simulator for ideal and non-ideal compressible-fluid flows

A schlieren post-processing tool for CFD simulations of ideal and non-ideal compressible-fluid flows is presented. The software VirtuaSchlieren simulates light propagation across a non-homogeneous medium to predict the schlieren images from an actual measurement apparatus. Trajectories of a large number of light rays—in the order of tens of millions—are reconstructed by numerical integration, from the light source to the screen plane. To this purpose, kd-tree search algorithms are implemented to retrieve the position of each ray within the CFD computational grid at each time step. The local value of the refraction index was retrieved from the interpolated value of the density at the center of each cell. The simple Gladstone–Dale and the Lorentz–Lorenz models are implemented to compute the value of the refractive index. Two search algorithms are evaluated, namely, the approximate nearest neighbor (ANN) and a simplified kd-tree search technique. The latter is implemented on both CPU and GPU architectures. The hybrid GPU/CPU implementation was successfully tested against a reference experimental schlieren image of the supersonic flow around a conical body. Numerical simulations of the supersonic expansion of non-ideal flows are also presented.

Evaluating Sampling Strategies for Visualizing Uncertain Multi-Phase Fluid Simulation Data

With Eulerian Method of Moment (MoM) solvers for the CFD simulation of multi-phase fluid flow, the positions of bubbles or droplets are not modeled explicitly, but through scalar fields of moments. These fields can be interpreted as probability density functions describing the distribution of bubble locations. To enable intuitive visualization that allows direct visual comparisons between simulation and physical experiment, explicit instances of the bubble distribution are required. In this work, we examine one sampling-based method for obtaining sets of bubble positions from density fields. Based on an example dataset, we study the influence of the main parameter, the kernel size, on the resulting bubble set. We identify a tradeoff between numerical accuracy and temporal continuity for the visualization.

Calibration of drag models for mesoscale simulation of gas–liquid flow through packed beds

We propose a systematic approach for calibrating the interphase drag models used in mesoscale CFD simulations of gas-liquid flow through packed beds. The method uses packing specific liquid holdup and pressure drop data to find model coefficients that both minimize relative error in the two-phase momentum equations and ensure correct limiting single phase behavior of the model. The approach is demonstrated for three phenomenological drag models from the literature using data from a random Bialecki ring packing under counterflow conditions. Calibration results in a 2–3-fold reduction in the mean relative error for predicted pressure gradient/liquid holdup, relative to predictions with “standard” Ergun coefficients, when an idealized set of momentum equations are solved. The calibrated models are then implemented into a two-fluid CFD solver and used to perform a two-dimensional time accurate simulation of the same packed bed at high gas and liquid flowrates. The predicted mean flow properties are in very good agreement with macroscale experimental data, while the instantaneous liquid distributions show significant transient, axial and lateral nonuniformities.

Application of a porous media model for the acoustic damping of perforated plate absorbers

Perforated panel, or plate, absorbers are commonly employed to reduce sound pressure levels across a broad range of applications including the built environment, industrial installations and propulsion devices. The acoustic performance of a perforated plate absorber depends upon a number of parameters such as physical geometry of the absorber, acoustic spectrum and sound pressure level of the acoustic source. As a consequence, experimental determination of acoustic properties is often required on an individual basis in order to optimise performance.Computational simulation of a perforated plate absorber would alleviate the necessity for experimental characterisation. Fundamentally this can be achieved by the direct numerical solution of the underlying governing equations, the compressible form of the Navier-Stokes equations. The numerical methodology is available and routinely implemented as a Computational Fluid Dynamics solver. However, the numerical simulation of flow through a perforated plate with a large number of very small orifices would require significant computational resource, not routinely available for engineering design simulations.In this paper, a porous media model, implemented as a sub-model within a CFD solver, is investigated and validated against a number of well-acknowledged acoustic experiments undertaken in an impedance tube, for a sound pressure wave incident normal to a perforated plate. The model expresses the underlying governing equations within the perforated plates in terms of a pseudo-physical velocity representation. Comparison between three dimensional, compressible, laminar flow CFD simulations and experimental data, demonstrate that the porous model is able to represent acoustic properties of perforated plate absorbers in linear and non-linear absorption regimes and also the inertial effect in the presence of a mean bias flow.The model significantly reduces the computational resource required in comparison to full geometric resolution and is thus a promising tool for the engineering design of perforated plate absorbers.

Sensitivity analysis on chaotic dynamical systems by Non-Intrusive Least Squares Shadowing (NILSS)

This paper develops the Non-Intrusive Least Squares Shadowing (NILSS) method, which computes the sensitivity for long-time averaged objectives in chaotic dynamical systems. In NILSS, we represent a tangent solution by a linear combination of one inhomogeneous tangent solution and several homogeneous tangent solutions. Next, we solve a least squares problem using this representation; thus, the resulting solution can be used for computing sensitivities. NILSS is easy to implement with existing solvers. In addition, for chaotic systems with many degrees of freedom but few unstable modes, NILSS has a low computational cost. NILSS is applied to two chaotic PDE systems: the Lorenz 63 system and a CFD simulation of flow over a backward-facing step. In both cases, the sensitivities computed by NILSS reflect the trends in the long-time averaged objectives of dynamical systems.

Unsteady simulation of hypersonic flow around a heat flux probe in ground testing conditions

In order to investigate the accuracy of the rebuilding code for the free stream conditions and the total enthalpy in the Longshot Hypersonic facility at the von Karman Institute (VKI), a series of unsteady CFD simulations of axisymmetric hypersonic flow over a heat flux probe have been performed. The free stream and the total conditions provided by experiments as a function of time were used as unsteady inlet boundary conditions for those simulations. The full duration (15ms) of the test window is simulated by means of an implicit time accurate finite volume solver. The numerical method and the time integration procedure used during this investigation are described. The numerical results are within 6% of the experimental data for both the stagnation pressure and the heat flux at the reference test time conditions.

CFD simulation of the magnetohydrodynamic flow inside the WCLL breeding blanket module

The interaction between the molten metal and the plasma-containing magnetic field in the breeding blanket causes the onset of a magnetohydrodynamic (MHD) flow. To properly design the blanket, it is important to quantify how and how much the flow features are modified compared with an ordinary hydrodynamic flow. This paper aims to characterize the evolution of the fluid inside one of the proposed concepts for DEMO, the Water-Cooled Lithium Lead (WCLL), focusing on the central cell of the equatorial outboard module. A preliminary validation was required to gauge the capability of ANSYS CFX to deal with MHD problems. The buoyant and pressure-driven fully developed laminar flows in a square duct were selected as benchmarks. Numerical results were compared with theoretical solutions and an excellent agreement was found. The channel analysis was realized on a simplified version of the latest available design geometry, developed by ENEA, for M≤1000. The simulation highlighted the formation of high velocity jets close to the baffle and the onset of an asymmetrical potential distribution.

New generalized expressions for forced convective heat transfer coefficients at building facades and roofs

Previous research indicated that the surface-averaged forced convective heat transfer coefficient (CHTC) at a windward building facade can vary substantially as a function of building width and height. However, existing CHTC expressions generally do not consider the building dimensions as parameters and are therefore strictly only applicable for the building geometry for which they were derived. Most CHTC expressions also categorize facades only as either windward or leeward. This indicates the need for new and more generally applicable CHTC expressions. This paper presents new generalized expressions for surface-averaged forced CHTC at building facades and roofs that contain the reference wind speed, the width and the height of the windward building facade as parameters. These expressions are derived from CFD simulations of wind flow and forced convective heat transfer for 81 different isolated buildings. The 3D Reynolds-averaged Navier-Stokes equations are solved with a combination of the high-Re number realizable k-ε model and the low-Re number Wolfshtein model. First, a validation study is performed with wind-tunnel measurements of surface temperature for a reduced-scale cubic model. Next, the actual simulations are performed on a high-resolution grid with a minimum near-wall cell size of 400 μm to resolve the entire boundary layer, including the viscous sublayer and the buffer layer, which dominate the convective surface resistance. The new CHTC expressions are analytical formulae (trivariate polynomials) that can easily be implemented in Building Energy Simulation (BES) and Building Envelope Heat-Air-Moisture (BE-HAM) transfer programs. The accuracy of the expressions is confirmed by in-sample and out-of-sample evaluations.

Influence of Bypass on Thermo-Hydraulics of VVER 440 Fuel Assembly

Abstract The paper deals with CFD modelling and simulation of coolant flow within the nuclear reactor VVER 440 fuel assembly. The influence of coolant flow in bypass on the temperature distribution at the outlet of the fuel assembly and pressure drop was investigated. Only steady-state analyses were performed. Boundary conditions are based on operating conditions. ANSYS CFX is chosen as the main CFD software tool, where all analyses are performed.

CFD Simulation of Natural Convection Flow in Pressurized Tanks Exposed to Fire

CFD Simulation of Natural Convection Flow in Pressurized Tanks Exposed to Fire

CFD simulation of flow around angle of attack and sideslip angle vanes on a BAe Jetstream 3102 – Part 1

CFD modelling techniques are exploited to investigate the local velocity field around angle of attack and sideslip angle sensors fitted to the nose of a modified BAe Jetstream 3102 small airliner. Analysis of the flow angularity at the vane locations has allowed the vanes response to varying flight conditions to be predicted and errors in the readings to be quantified. Subsequently, a more accurate calibration of the system is applied to the current configuration on the Jetstream, and a better understanding of the position error with respect to the vane locations is obtained.The above aircraft was acquired by Cranfield University in 2003 with subsequent flow angle vane modifications taking place in 2005. The aircraft is currently in operation with the National Flying Laboratory Centre (NFLC) for research and demonstration purposes.

CFD simulation of flow around angle of attack and sideslip angle vanes on a BAe Jetstream 3102 – Part 2

A previous study analysing the local flow around angle of attack and sideslip angle vanes on a BAe Jetstream 3102 turboprop is extended to study the additional effects of bank angle. A full matrix of CFD simulations is carried out to investigate how the introduction of a bank angle affects vane performance for a range of flight conditions. An updated calibration method to convert the raw vane readings into true values of angle of attack and sideslip, incorporating a correction factor as a function of the bank angle, is presented. The results are shown to be accurate for a wide range of flight configurations. Uncertainty analysis indicates that raw vanes reading errors should be below ±0.1° to ensure that total calibration errors are restricted to less than 2% for angle of attack and 5% for sideslip angle.