New algorithms of 3D particle tracking velocimetry (3D PTV) based on a tomographic reconstruction approach have been developed and tested by using synthetic images of unsteady 3D flows. The new algorithms are considered not only in the tomographic reconstruction process of the fluid volume with particles but also in the subsequent process of individual particle detection and validation. In particular, the tomographic reconstruction accuracy is boosted up by using a new recursive validation scheme through which many of ghost particles can be removed effectively. The particle detection process includes the particle mask correlation operator and the dynamic threshold scheme to extract individual particle centroids from the reconstructed intensity clusters of the fluid volume. The overall reconstruction accuracy is checked by the synthetic image data sets with different particle density and different volume thickness.
We performed direct numerical simulations (DNS) of turbulent channel flow of surfactant solution to investigate characteristics of modulated wall turbulence and turbulent drag reduction. In the simulations, the effect of surfactant solution was represented by the modified Bird-Carreau model for shear viscosity. Our simulations reproduced experimental features of turbulent flow of surfactant solution. One result was a drag reduction rate of 37%, which is in the range of low drag reduction. Considering instantaneous structures and turbulence statistics, our scale analysis found that modulation of turbulent flow of surfactant solution for the Newtonian fluid could be generally normalized with the local variable viscosity. Some additional corrections are, however, needed to apply this to predicting the dissipation rate, which the local viscosity affects directly in turbulent flow.
In this study, we used a 1/5-scale model tunnel to investigate the temperature characteristics of backlayering thermal fumes in addition to the heat release rate and the longitudinal ventilation velocity in a tunnel fire. The ventilation flow through the model tunnel provided a sufficient turbulent flow as the Reynolds number based on ventilation velocity and tunnel height was 40,000 or more. The governing scaling parameter in this experiment was the dimensionless parameter Q*/Fr. In addition to Froude's and Reynolds' similarity laws, the similarity of the thermal characteristics in the tunnel wall was considered in making the model tunnel. We clarify the conditions under which the backlayering thermal fumes maintain a stratified layer. Empirical equations of the thermal fume height in the longitudinal ventilation are derived from the experimental data. The evacuation safety based on the height of thermal fumes is examined in detail.
We numerically simulated the aerodynamic noise generated from the unsteady flow around a generic model of automobile rear-view mirrors in this study. This generic model is made by cutting a hemispherical cylinder into half lengthwise, and placed on a flat plate. The governing equations employed in the present numerical simulation are the compressible Navier-Stokes equations, where no additional models of sound generation and propagation are used. Results show that acoustic pressure waves are generated from the disturbed wake of the mirror model and propagate into the far field. In addition, the spatial distribution of the sound pressure level (SPL) in the far field has a similar tendency to that of the experimental data. These results show that useful data on aerodynamic noise can be obtained by the simple computational method presented in this study, and utilized to make a preliminary design to reduce the aerodynamic noise.
The lattice Boltzmann method is used to examine the effect of tripping rods on the lift force acting on a circular cylinder subjected to flows with a Reynolds number ranging from 100 to 500. The ratios of the diameter of the tripping rod to that of the cylinder are 0.08, 0.10, and 0.11. The angular position of the tripping rod is varied from 20° to 120°. Numerical results show that the lift force is smaller than that without the tripping rod if the angular position ranges from 20° to 40°. Further, the numerical simulation shows that, if the angular position is greater than 50°, the lift force surpasses that without the tripping rod and is maximized if the angular position is around 80° for a case with a Reynolds number of 100 and 60° for a case with a Reynolds number of 500.
Laser Doppler Velocimetry (LDV) measurements in a 90 degree elbow of which the curvature radius coincides with its inner diameter were examined for the cases of inflow from a long pipe, short pipe and swirl generator. Ensemble averaged flow distribution at the Reynolds number of 320000 based on the inner pipe diameter and bulk velocity shows that shortening the upstream pipe length to 4.9D from 10D induces the flow separation downstream of the elbow. Detailed observation suggests that shortening upstream pipe weakens the Prandtl's secondary flow of the first kind. Our swirl generator induced a swirling inflow with the non-dimensional angular momentum of 0.12 based on the inner pipe diameter and bulk velocity. The circumferential velocity distribution formed a shape like a Rankine combined vortex at the elbow inlet, and the accelerated axial velocity was observed at the vortex center. The axial velocity distribution however was found to be almost the same as that of the non-swirl inflow case in the latter half of the elbow. Frequency analyses showed that the Strouhal number by vortex shedding from the boundary layer occurring at the inner side of the elbow become 0.5, except for 0.6 in the case of the long pipe. The change of the Strouhal number is probably related with the boundary layer width and the local flow velocity.
As part of the development of a flow-induced vibration evaluation methodology for the primary cooling piping in Japan sodium-cooled fast reactor, important factors were discussed in evaluating the flow-induced vibration for the hot-leg piping. To investigate a complex flow near the inlet of the hot-leg piping, a steady-state numerical analysis was carried out for the reactor upper plenum flow, which was simulated in a 1/10-scale reactor upper plenum experiment. Based on this analysis, experimental conditions on swirl inflow and deflected inflow that were identified as important factors were determined for flow-induced vibration experiments simulating only the hot-leg piping. In this study, the effect of the swirl inflow on flow pattern and pressure fluctuation onto the pipe wall was investigated in a 1/3-scale hot-leg pipe experiment. The experiment has indicated less significant for the pressure fluctuations, while the flow separation region was slightly influenced by the swirl inflow. Computational fluid dynamics simulation with a U-RANS approach results also appear in this paper. Through the simulations under the swirl inflow conditions of 0% and 5%, the validity of the U-RANS simulation was confirmed by comparison to the 1/3-scale hot-leg piping experiments.
In numerical fluid dynamic simulations, adaptive mesh refinement (AMR) approaches have recently been growing in popularity, particularly when the target flow field includes complex turbulent shear flows such as jets, mixing layers, and shear layers in separated regions. In the AMR approach presented in this paper, for numerical simplicity and practicality we adopt a block-based method that uses a structured mesh in each block, a body-fitted coordinate system and a self-similar tree-based hierarchical data structure. We also implement measures that address memory/communication reduction and load balancing. As application examples we solve a separated flow around an airfoil, a transonic flow around a reentry capsule, and a coaxial jet flow. The examples demonstrate that the AMR approach is effective for capturing complex turbulent shear flows, although numerical issues such as scalability remain to be addressed for larger simulations.
It is presented through numerical analysis that accelerated motions of a two-dimensional fish-like foil by fluid forces can be modeled from impulsive starts to terminal states. Time histories of drag coefficient and foil speed depend on a frequency and motion of a foil and amplitude of its caudal fin. Three models that take thrust, inertial drag and viscous drag into account for the time histories of foil speed and drag coefficient are suggested. It is indicated that the model using viscous drag is in good agreement with numerical solutions in various conditions for small Reynolds numbers. Thrust, drag and terminal velocity can be summarized by functions that are almost power laws of the foil frequency even for various wavy lateral oscillations of the midline of a foil.
This paper discusses comparisons of first and second-order slip boundary conditions on the range of Knudsen numbers of some typical MEMS as O(10-3)<Kn<O(10-1) by numerical simulations using the continuum model based on the Navier-Stokes (N-S) equations. Slip flows are solved by the Constrained Interpolation Profile (CIP) method. A numerical method for a simple but general form of a slip boundary condition on a simple planar wall is also introduced. Numerical solutions are in very good agreement with exact solutions of the N-S equations with slip boundaries in two-dimensional Poiseuille flows in the hard sphere (HS) model. The present numerical method is next applied to two and three-dimensional rid-driven cavity flows, employing slip coefficients for the Bhatnagar-Gross-Krook (BGK) model. The results indicate differences of slip effects near walls between the first and second-order slip boundary can also affect whole streamlines and velocity distributions in a closed region with increasing the Knudsen number.
In this study, the backlayering distance of thermal fumes in a tunnel fire was examined by using a large-scale model tunnel. A 1/5-scale model was constructed taking into consideration both the similarity of thermal characteristics in the wall and Froude similarity. Experimental parameters were the heat release rate of the fire source and the longitudinal ventilation velocity. The following conclusions were obtained. A new expression was proposed to calculate the dimensionless backlayering distance based on the dimensionless heat release rate and Froude number. Constants included in this expression were obtained from experimental data by using the least-squares method. The backlayering distance calculated by the new expression was compared with that calculated by previous expressions developed by other researchers. The new expression has the ability to calculate not only the backlayering distance but also the critical velocity.
In order to clarify the mechanism of cavitation cloud shedding, the shedding frequencies of cavitation clouds were investigated by high-speed observations and image analysis. The results indicate that the shedding frequency follows the similarity law for unsteady vortex flow. The ranges of parameters considered were from 0.4 to 2.0 mm for the nozzle throat diameter, from 10 to 30MPa for the injection pressure, and from 0.01 to 0.05 for the cavitation number. The similarity law is discussed with reference to the Strouhal number, which depends on the shedding frequency of the cavitation cloud, the width of the cavitating jet and the jet velocity at the nozzle exit. Formulas describing the relationships between the shedding frequency of the cavitation cloud and the investigated parameters were developed. In addition, the Strouhal number for cavitation cloud shedding was shown to be independent of the investigated parameters and constant, with a value of 0.18. In fact, the cavitation cloud shedding was governed by the constant Strouhal number.
A numerical method for liquid-vapor two-phase flows accompanied by condensation/evaporation is proposed. The phase change model at the vapor-liquid interface, which had been theoretically derived by Sone and Onishi, is coupled with an interface capturing scheme. The model gives the condensation/evaporation rate as boundary conditions. The interface is captured by the volume-of-fluid method associated by the reconstruction scheme, and the scheme to obtain the pressure with consideration for thermodynamics in the vicinity of the interface is employed. These methods are coupled and arranged on the cylindrical coordinate system to appropriately deal with the liquid film flow around a cool cylinder in a natural convection of vapor. The development of a liquid film due to the condensation, the flow in a thin layer and a drop of the distilled liquid are successfully reproduced in our result. The influence of the volumetric conservation is confirmed by comparing the results of condensation case and a non-condensation case. Besides the Nusselt number is in good agreement with a theoretically estimated value.