3D numerical modeling of non-isotropic turbulent buoyant helicoidal flow and heat transfer in a curved open channel

A 3D non-isotropic algebraic stress/flux turbulence model is employed to simulate turbulent buoyant helicoidal flow and heat transfer in a rectangular curved open channel. The prediction shows that, unlike the isothermal flow, there are two major and one minor secondary flow eddies in a cross section of thermally stratified turbulent buoyant helicoidal flow in a curved open channel. The results compare favorably with available experimental data. The thermocline in a curved channel is thicker than that in a straight channel. All of these is the result of complex interaction between the buoyant force, the centrifugal force and the Reynolds stresses. The turbulent flow in a curved channel is obviously non-isotropic: the turbulence fluctuations in vertical and radial directions are lower in magnitude than that in the axial direction, which illustrates the suppression of turbulence due to buoyant and centrifugal forces. The results are of significant practical value to engineering works such as the choice of sites for intake and pollutant-discharge structures in a curved river.     

Heat transfer in a turbulent particle-laden channel flow

The objective of this paper is twofold: (i) to present and analyze particle temperature statistics in turbulent non-isothermal fully-developed turbulent gas–solid channel flow for a large range of particle inertia in order to better understand particle heat transfer mechanisms; (ii) to examine the performance of a recent Probability Density Function (PDF) model provided by Zaichik et al. (2011) [1]. In order to achieve such objectives, a Direct Numerical Simulation (DNS) coupled with a Lagrangian Particle Tracking (LPT) was used to collect fluid and particle temperature statistics after particles reach a statistically stationary regime. A non-monotonic behavior of particle temperature statistics is observed as inertia increases. The competition between different mechanisms (filtering inertia effect, preferential concentration, production of fluctuating quantities induced by the presence of the mean velocity and/or mean temperature gradients) are responsible for such a behavior. This competition is investigated from the exact transport equations of particle temperature statistical moments, fluid statistics conditionally-averaged at particle location, and instantaneous particle distribution in the flow field. Using these data, the accuracy of a PDF model is also assessed in the second part. From this assessment, it is seen that, despite the assumptions made, the model leads to a satisfactory prediction of most of the particle temperature statistics for not too high particle inertia.     

Experimental study on buoyant flow stratification induced by a fire in a horizontal channel

Experiments were carried out in a reduced-scale horizontal channel to investigate the fire-induced buoyant flow stratification behavior, with the effect of the velocity shear between the hot buoyant flow and the cool air flow considered. This shear intensity was controlled and varied by changing the exhaust rate at the ceiling with one of the end of the channel opened. The flow pattern was visualized by the aid of a laser sheet. The horizontal traveling velocity, vertical temperature profile and stratification interface height of the buoyant flow were measured. The stratification pattern was found to fall into three regimes. Buoyancy force and inertia force, as the two factors that dominate the buoyant flow stratification, were correlated through the Froude number and the Richardson number. At Region I (Ri > 0.9 or Fr < 1.2), the buoyant flow stratification was stable, where a distinct interface existed between the upper smoke layer and the lower air layer. At Region II (0.3 < Ri < 0.9 or 1.2 < Fr < 2.4), the buoyant flow stratification was stable but with interfacial instability. At Range III (Ri < 0.3 or Fr > 2.4), the buoyant flow stratification becomes unstable, with a strong mixing between the buoyant flow and the air flow and then a thickened smoke layer.     

Calculation of turbulent flow and heat transfer in a porous-baffled channel

This study presents the numerical predictions on the turbulent fluid flow and heat transfer characteristics for rectangular channel with porous baffles which are arranged on the bottom and top channel walls in a periodically staggered way. The turbulent governing equations are solved by a control volume-based finite difference method with power-law scheme and the k-ε turbulence model associated with wall function to describe the turbulent structure. The velocity and pressure terms of momentum equations are solved by SIMPLE (semi-implicit method for pressure-linked equation) method.The parameters studied include the entrance Reynolds number Re (1×10⁴-5×10⁴), the baffle height (h=10, 20 and 30 mm) and kind of baffles (solid and porous); whereas the baffle spacing S/H are fixed at 1.0 and the working medium is air. The numerical calculations of the flow field indicate that the flow patterns around the porous- and solid-type baffles are entirely different due to different transport phenomena and it significantly influences the local heat transfer coefficient distributions. Relative to the solid-type baffle channel, the porous-type baffle channel has a lower friction factor due to less channel blockage.Concerning the heat transfer effect, both the solid-type and porous-type baffles walls enhanced the heat transfer relative to the smooth channel. It is further found that at the higher baffle height, the level of heat transfer augmentation is nearly the same for the porous-type baffle, the only difference being the Reynolds number dependence. As expected, the centerline-averaged Nusselt number ratio increases with increasing the baffle height because of the flow acceleration.     

Numerical simulations of a second-order chemical reaction in a plane turbulent channel flow

We analyze numerical simulations of a second-order chemical reaction (Da = 1) in a fully developed turbulent plane channel flow at a low Reynolds number (Re_τ = 180). The reactive plume is formed when a reactant A is released through a line source into the channel flow doped with reactant B. Two different inlet pre-mixing conditions and line source heights are considered. Direct Numerical Simulations (DNS) and Stochastic Fields (SF) methods have been used and compared for these different conditions. The results obtained using SF are sensitive to the particular value of the turbulent Schmidt number (Sc_T) selected to model the turbulent dispersion. It has been found that if a representative value of Sc_T extracted from DNS is used in the SF method, both DNS and SF, give similar results.     

Thermal characterization of turbulent flow in a channel with isosceles triangular ribs

The article presents an experimental investigation on turbulent heat transfer and friction loss behaviors of airflow through a constant heat-fluxed channel fitted with different heights of triangular ribs. The rib cross-section geometry used in the present study was isosceles triangle. Two rib arrangements, namely, in-line and staggered arrays, were introduced. Measurements were carried out for a rectangular channel of aspect ratio, AR = 10 and height, H = 30 mm with three uniform rib heights, e = 4, 6 and 8 mm (e/H = 0.13, 0.2 and 0.26) and one non-uniform rib height, e = 4,6 mm (e/H = 0.13,0.2) alternately for a single rib pitch, P = 40 mm. The flow rate in terms of Reynolds numbers based on the inlet hydraulic diameter of the channel was in a range of 5000 to 22,000. The experimental results show a significant effect of the presence of the ribs on the heat transfer rate and friction loss over the smooth wall channel. The uniform rib height performs better than the corresponding non-uniform one. The in-line rib arrangement provides higher heat transfer and friction loss than the staggered one for a similar mass flow rate. In comparison, the largest e/H rib with inline array yields the highest increase in both the Nusselt number and the friction factor values but the lowest e/H rib with staggered array provides the best thermal performance.     

Bubble dynamics of a pressure-driven cavitating flow in a micro-scale channel using a high density pseudo-potential Lattice Boltzmann method

Abstract

A single-component multiphase solver based on Lattice Boltzmann method has been developed and was used to study dynamics of a single cavitating bubble subjected to pressure-driven flow in a two-dimensional channel. Simulations were performed with and without contact to the wall. A pseudopotential model coupled with Peng-Robinson equation of state was implemented to incorporate inter-particle force interaction. The solver was validated by comparing the simulated densities with the theoretical co-existence curves at different temperatures for water. Additionally, the contact angle obtained at various adhesive parameters is also validated at 583?K for water. The dynamics of a single cavitating bubble in a two-dimensional channel subjected to a pressure gradient is studied. Displacement of this bubble at different aspect ratios (5,10) and Reynolds numbers (1–30) when placed along the channel centerline and at off-center positions were studied. Moreover, bubble growth is computed at various contact angles for different aspect ratios and Reynolds number. Dynamic contact angles and contact lengths during the flow are estimated. As the aspect ratio increases, the bubble appears to be more elongated with lower contact angles.     

Large-eddy simulation of turbulent heat convection in a spanwise rotating channel flow

In this paper, we investigate the effects of the Coriolis force in a heated plane channel flow subjected to spanwise rotation using the method of large-eddy simulation. We present both the general and simplified transport equations for the resolved turbulent stresses, which are essential for understanding the unique pattern of turbulent kinetic energy production in a rotating system. Numerical simulations are performed using primarily two dynamic subgrid-scale stress models and one dynamic subgrid-scale heat flux model; namely, the conventional dynamic model (DM) and a novel dynamic nonlinear model (DNM) for closure of the filtered momentum equation, and an advanced dynamic full linear tensor thermal diffusivity model (DFLTDM) for closure of the filtered thermal energy equation. The turbulent flow field studied herein is characterized by a Reynolds number Re_τ = 150 and various rotation numbers Ro_τ ranging from 0 to 7.5. In order to validate the LES approach, turbulent statistics obtained from the simulations are thoroughly compared with the available experimental results and direct numerical simulation (DNS) data. A detailed comparative study has been conducted in order to evaluate the performance of the DM and DNM in terms of their prediction of characteristic features of the velocity and temperature fields and their capability of reflecting both forward and backward scatter of kinetic energy between the filtered and subgrid scales.     

A spatially advancing turbulent flow and heat transfer in a curved channel

Direct numerical simulation was performed for a spatially advancing turbulent flow and heat transfer in a two‐dimensional curved channel, where one wall was heated to a constant temperature and the other wall was cooled to a different constant temperature. In the simulation, fully developed flow and temperature from the straight‐channel driver was passed through the inlet of the curved‐channel domain. The frictional Reynolds number was assigned 150, and the Prandtl number was given 0.71. Since the flow field was examined in the previous paper, the thermal features are mainly targeted in this paper. The turbulent heat flux showed trends consistent with a growing process of large‐scale vortices. In the curved part, the wall‐normal component of the turbulent heat flux was twice as large as the counterpart in the straight part, suggesting active heat transport of large‐scale vortices. In the inner side of the same section, temperature fluctuation was abnormally large compared with the modest fluctuation of the wall‐normal velocity. This was caused by the combined effect of the large‐scale motion of the vortices and the wide variation of the mean temperature; in such a temperature distribution, large‐scale ejection of the hot fluid near the outer wall, which is transported into the near inner‐wall region, should have a large impact on the thermal boundary layer near the inner wall. Wave number decomposition was conducted for various statistics, which showed that the contribution of the large‐scale vortex to the total turbulent heat flux normal to the wall reached roughly 80% inside the channel 135^° downstream from the curved‐channel inlet. © 2009 Wiley Periodicals, Inc. Heat Trans Asian Res; Published online in Wiley InterScience ( www.interscience.wiley.com ). DOI 10.1002/htj.20275     

Scaling heat transfer in fully developed turbulent channel flow

An analysis is given for fully developed thermal transport through a wall-bounded turbulent fluid flow with constant heat flux supplied at the boundary. The analysis proceeds from the averaged heat equation and utilizes, as principal tools, various scaling considerations. The paper first provides an accounting of the relative dominance of the three terms in that averaged equation, based on existing DNS data. The results show a clear decomposition of the turbulent layer into zones, each with its characteristic transport mechanisms. There follows a theoretical treatment based on the concept of a scaling patch that justifies and greatly extends these empirical results. The primary hypothesis in this development is the monotone and limiting Peclet number dependence (at fixed Reynolds number) of the difference between the specially scaled centerline and wall temperatures. This fact is well corroborated by DNS data. A fairly complete qualitative and order-of-magnitude quantitative picture emerges for a complete range in Peclet numbers. It agrees with known empirical information. In a manner similar to previous analyses of turbulent fluid flow in a channel, conditions for the existence or nonexistence of logarithmic-like mean temperature profiles are established. Throughout the paper, the classical arguments based on an assumed overlapping of regions where the inner and outer scalings are valid are avoided.     

Prediction of turbulent flow structure in a fully developed rib-roughened narrow rectangular channel

Fully developed turbulent water-flow structure over one-side repeated-ribs in narrow two-dimensional rectangular channels was investigated experimentally by Particle Image Velocimetry (PIV) and analytically by the standard κ-ɛ and nonlinear κ-ɛ turbulent models. Two rib-pitch to height ratios (p/k) of 10 and 20 were investigated while the rib height was held constant at 4 mm. The rib height-to-channel equivalent diameter ratio (k/De) was 0.1. The streamwise mean velocity and turbulent kinetic energy distributions at six selected axial stations from the center rib for the two Reynolds number (Re) of 7,000 and 20,000 were obtained and compared with the predicted one. The performance ability in predicting separating and reattaching turbulent water-flow between the standard κ-ɛ and nonlinear κ-ɛ models had yielded no clear conclusion. A large-scale turbulent eddy was generated by the rib promoter and then propagated into the mainstream flow, which led to the deformation of the velocity profile. The turbulent kinetic energy was increased about two times higher at p/k = 20 than that at p/k =10 under the two Reynolds numbers. The effect of the p/k value and the Reynolds number (Re) on reattachment length (XR) was investigated and showed that the p/k and Re had no significant effect on the reattachment length beyond a critical value of Re = 15,000 where XR was found to be approximately 4 times of the rib height under water-flow condition.     

Direct numerical simulation of turbulent flow in a vertical channel with buoyancy orthogonal to mean flow

The effect of buoyancy on turbulent air flowing horizontally between two differentially heated vertical plates has been investigated using direct numerical simulation. Grashof number ranges between zero and 4.0 × 10⁶ and the Boussinesq approximation is used in the buoyant term. In this particular configuration, the buoyancy forces result in skewed mean velocity profiles with non-zero anti-symmetric spanwise component W. The resulting flow has the features of three-dimensional turbulent boundary layer flows. In particular, suppression of the primary Reynolds stress in near-wall region is observed. Titled streaks with significant destruction of the associated vortical structures are highlighted. The induced mean spanwise strain enhances the turbulence intensities in the channel core. The effect of buoyancy on mean quantities and second-order statistics including Reynolds stress components and turbulent heat fluxes are presented and analyzed with detailed budgets of their transport equations which are believed to be helpful to testify and improve turbulence models incorporating buoyancy effect.     

Direct simulation of turbulent heat transfer in swept flow over a wire in a channel

We investigate heat transfer characteristics of a turbulent swept flow in a channel with a wire placed over one of its walls using direct numerical simulation. This geometry is a model of the flow through the wire-wrapped fuel pins, the heat exchanger, typical of many civil nuclear reactor designs. The swept flow configuration generates a recirculation bubble with net mean axial flow. A constant inward heat flux from the walls of the channel is applied. A key aspect of this flow is the presence of a high temperature region at the contact line between the wire and the channel wall, due to thermal confinement (stagnation). We analyze the variation of the temperature in the recirculation bubble at Reynolds number based on the bulk velocity along the wire-axis direction and the channel half height of 5400. Four cases are simulated with different flowrates transverse to the wire-axis direction. This configuration is topologically similar to backward-facing steps or slots with swept flow, except that the dominant flow is along the obstacle axis in the present study and the crossflow is smaller than the axial flow, i.e., the sweep angle is large. The temperature field is simulated at three different Prandtl numbers: 10^?2, 10^?1 and 1. The lower value of Prandtl number is characteristic of experimental high-temperature reactors that use a molten salt as coolant while the high value is typical of gas (or water vapor) heat exchangers. In addition, mean temperature, turbulence statistics, instantaneous wall temperature distribution and Nusselt number variation are investigated. The peak Nusselt number occurs close to the reattachment location, on the lee side of the wire, and is about 50–60% higher compared to the case without crossflow. The high temperature region follows the growth of the recirculation bubble which increases by about 65% from the lowest to highest amount of crossflow. Particular attention is devoted to the temperature distribution on the walls of the channel and the surface of the wire. The behavior of the heat-flux across the mean dividing streamline of the recirculation bubble is investigated to quantify the local heat transfer rates occurring in this region.     

Enhancement of heat transfer in a turbulent channel flow with insertion of a flat body

Experiments have been performed for turbulent channel flow obstructed with a flat body. The local heat transfer coefficient and the wall static pressure were measured on two kinds of flat bodies for which the trailing edge shape differed. The length of the body, the thickness of the body, and the distance between the wall and the body were changed in several steps. The total performance between heat transfer and pressure drop was estimated under conditions of equal pumping power. The total performance hardly changed, even if the trailing edge shape and length of the bodies were different. The averaged heat transfer coefficient increased with increasing thickness of the bodies. However, as the friction factor increased, the performance became worse. When a comparatively thin body was installed near the heating surface, good performance was obtained. © 2003 Wiley Periodicals, Inc. Heat Trans Asian Res, 32(4): 354–366, 2003; Published online in Wiley InterScience ( www.interscience.wiley.com ). DOI 10.1002/htj.10100     

Temperature wall modelling for large-eddy simulation in a heated turbulent plane channel flow

Industrial flows are often wall-bounded, characterized by a high Reynolds number turbulence and a strong unsteadiness. Large-eddy Simulations applied to this kind of flows require a heavy mesh resolution which is out of reach of the nowadays computational power. Wall models are usually used to alleviate this constraint. However, very few of them are dedicated to the temperature field. Besides, most of these wall models are based on equilibrium laws which are not able to take into account the unsteadiness of the flow in the near-wall region. In this study, an original thermal wall model, inspired from the TBLE wall model of Balaras et al. [E. Balaras, C. Benocci, U. Piomelli, Two-layer approximate boundary conditions for large-eddy simulations, AIAA J. 34 (6) (1996) 1111–1119], is developed and implemented in the CEA (French Atomic Center) Trio_U code and assessed on a heated and turbulent plane channel flow configuration. The investigated friction Reynolds numbers are 395, 1020 and 4800, and the Prandtl number is taken equal to 0.71.     

Simultaneous PIV/LIF measurements of a transitional buoyant plume above a horizontal cylinder

Results from an experimental investigation of the simultaneous temperature and velocity fields above an evenly heated horizontal cylinder with a Rayleigh number of 9.4E7 is presented. Ensemble averaged two-dimensional velocity and temperature fields, velocity fluctuations, temperature variance, and velocity–temperature correlations are computed from 2700 instantaneous data sets from simultaneous laser-induced fluorescence (LIF) and particle image velocimetry (PIV). The vertical velocity and temperature field in the plume are compared with similarity solutions from turbulent planar plumes. The production of turbulent kinetic energy (TKE) due to mean shear and buoyancy is evaluated and shows that the production of TKE is dominated by velocity shear.     

Thermophoretic deposition of small particles in a direct numerical simulation of turbulent channel flow

This paper presents results for the deposition rate of small particles on the walls of a turbulent channel flow. The results were obtained by direct numerical simulation of a horizontal turbulent channel flow. A temperature profile typical of ceramic oxide aerosol reactors was imposed across the channel. Thermophoresis played an important role in the deposition of particles for the range of conditions that were studied. An interaction between turbophoresis and thermophoresis was found to play an important role in the deposition process.