A numerical method is developed to simulate a bubbly flow around a circular cylinder using a boundary-fitted coordinate system. In the present study, the number density model of a bubbly flow is employed, where the equations of conservation and the group motion of bubbles are solved directly. The three-dimensional simulations have been conducted under laminar flow conditions for the liquid Reynolds numbers from 100 to 2000. The void fraction profile behind the cylinder shows good agreement with that of the experiment and another simulation carried out at the liquid Reynolds number of 430. The Strouhal numbers at Reynolds numbers of 200 and 500 are presented. In the lower Reynolds number case, the Strouhal number does not depend on the variation of the void fraction. On the other hand, the local maximum value of the Strouhal number is observed in the higher Reynolds number case, similar to that observed in the experiment. The bubble accumulation in the Kármán vortices is observed in the case of the greater relaxation time of the bubble motion. Such structures have also been observed in the experiment conducted behind a bluff body.
This study deals with the heat storage characteristics of latent-heat microcapsule slurry consisting of a mixture of fine microcapsules packed with latent-heat storage material and water. The heat storage operation for the latent-heat microcapsules was carried out by the direct-contact heat exchange method using hot air bubbles. The latent-heat microcapsule consisted of n-paraffin as a core latent-heat storage material and melamine resin as a coating substance. The relationship between the completion time of latent-heat storage and some parameters was examined experimentally. The nondimensional correlation equations for temperature efficiency, the completion time period of the latent-heat storage process and variation in the enthalpy of air through the microcapsule slurry layer were derived in terms of the ratio of microcapsule slurry layer height to microcapsule diameter, Reynolds number for airflow, Stefan number and modified Stefan number for absolute humidity of flowing air.
The program for the 200T/D coal gasification pilot plant was initiated in 1986 and successfully completed in 1996. In this program, a two-stage pressurized air-blown entrained flow coal gasifier was adopted, jointly developed by Central Research Institute of Electric Power Industry (CRIEPI) and Mitsubishi Heavy Industry, Ltd. using a 2T/D bench scale gasifier. In the 200T/D pilot plant, domestic Taiheiyo coal and Australian Moura, Warkworth coal were used for test operations. The commissioned tests of these three types of coals have been carried out in a 2T/D gasifier, prior to the pilot plant operation. The gasification characteristics of 2T/D and 200T/D gasifiers were evaluated. At the same gasifier air ratio, per pass fixed carbon conversion efficiency of the 2T/D gasifier is about 10% lower than that of the 200T/D gasifier and the heat absorption rate to the gasifier wall of the 2T/D gasifier is about 2% higher than that of the 200T/D gasifier.
The critical heat flux (CHF) is studied in a confined space both experimentally and analytically. The confined space is consisted of two horizontal surfaces with the lower surface heated and the upper mesh screen. The observation shows that the behaviors of the vapor mushroom over the mesh screen have not significant differences with that in the conventional pool boiling. However, the CHF reduces with the decrease of the space gap when it is narrower than someone critical value. The theoretical analyses show that the reduced CHF is related to the earlier dryout of macrolayer during the period of vapor mushroom because the initial thickness of the macrolayer reduces with the decrease of the space gap. However, for larger enough space, the CHF is reached due to the crisis of the local heat transfer by the microlayer model but not the macrolayer dryout. Analyses show that the heat transfer crisis (CHF) caused by the macrolayer dryout occurs only in advance of that by the microlayer model and the conventional macrolayer dryout model is a special case of the microlayer model. Theoretic prediction of CHF agrees reasonable well with experimental data.
In this paper, the authors present liquid film characteristics during transition from countercurrent annular two-phase flow to flooding in a vertical tube. Time variations of liquid film thickness at four different heights in the tube are measured. Their power spectra are used to examine interfacial wave characteristics. The instability of a liquid-gas interface in countercurrent annular two-phase flow is studied using linear and nonlinear methods. In the linear analysis, the Rayleigh equation is applied. In the nonlinear analysis, the long-wave analysis considering the base harmonic and harmonics is performed. The critical wave number obtained from experiments is compared with the stability maps. In the inviscid analysis, most of the critical wave numbers for each flow regime are located in the unstable region obtained from the theoretical instability map; therefore, this analysis cannot explain the transition to flooding. The wave number obtained from the present experiment decreases with increasing gas flow rate. In the nonlinear analysis, the critical wave number increases, and ultimately, the critical wave number from the experiment is found in the unstable region immediately before flooding. This analysis is found to be consistent with the experimental results.
Jets issuing through small holes in a wall into a freestream have proven effective in the control of boundary layer separation. Longitudinal (streamwise) vortices are produced by the interaction between the jets and the freestream. This technique is known as the vortex generator jet method of separation or stall control because it controls separation in the same general way as the well-known method using solid vortex generators. For the vortex generator jets, the shape and the downstream development of longitudinal vortices are varied by issuing jets discretely (pulsed jets). In order to understand the reasons why the vortices for the pulsed jet case behave in a manner different from those for the steady jet case, velocity measurements were carried out in various phases of pulsed jets. The longitudinal vortices produced by pulsed jets have a stronger influence of jet pitch angle in comparison with the steady jet case. Pulsed jets promote the interaction between adjacent vortices and therefore the shape and the downstream development of longitudinal vortices are different from those of the steady jet case.
The interaction of a straight diffusing vortex tube with a background uniform shear flow is investigated analytically. The vortex tube is aligned with the axial uniform shear flow βyez and simultaneously the tube undergoes the cross-axial weak shear αyex. In the limit of |α|/β<<1, the asymptotic solution to the Navier-Stokes equation is obtained to examine energy dissipation as well as axial and cross-axial vorticity components around the vortex tube. Vortex lines of the axial shear are wrapped and stretched by the tube to intensify the azimuthal vorticity. This vorticity stretching and intensification are found to be enhanced (or reduced) when the vorticity of the cross-axial shear is anti-parallel (or parallel) to that of the tube, which leads to the enhancement (or reduction) of the total energy dissipation around the vortex tube. It is shown that at the initial time of evolution βt<0.623(Γ/2πν)1/3 the energy dissipation associated with the tube itself is dominant, while at a later time βt>0.623(Γ/2πν)1/3 the dissipation of the wrapped vortex layers dominates that of the tube, where Γ is the circulation of the tube and ν is the kinematic viscosity.
There exists the serious problem of flue gas imbalance at the entry of horizontal gas-pass in many large tangentially fired utility boilers used presently in China. In this paper, the experimental and computational investigations on residual swirl and flue gas imbalance are carried out based on an isothermal model of 600 MW tangentially fired boiler. The Porosity method and the improved non-uniform QUICK scheme are presented in 3-D numerical simulation in order to simulate the platen superheater correctly and reduce the pseudo-diffusion effectively. The numerical results agree well with the experiment. The calculated results of flow field in furnace, diameter of actual tangential circle, the distribution of swirling intensity along height of furnace and the effects of the structures of nose and the platen superheater on residual swirl and velocity deviation are presented. The forming cause and influencing factors of flue gas imbalance are discussed. Meanwhile, the means to diminish the deviation are introduced.
A technique for measurement of thermodynamic variables with high sensitivity is necessary for analyses of highly rarefied gas flows, gas-surface interaction, and so on. REMPI (Resonantly Enhanced Multiphoton Ionization) is a powerful optical tool because of its high sensitivity even in highly rarefied gas flows and its ability to measure nonequilibrium amongst internal (translational, vibrational, and rotational) energy modes. In this paper, a REMPI system is developed and REMPI spectra are measured on the center line of a free-molecular flow. The fundamental properties of the REMPI signal are also described. A method using a Boltzmann plot with the spectral lines of multiple branches is proposed, and the ratio of the electronic transition dipole moments in the two-photon Hönl-London factors for the O and P branches is determined.
This paper describes quantitative three-dimensional measurement method for flow field of a rotating Rayleigh-Benard convection in a cylindrical cell heated below and cooled above. A correlation method for two-dimensional measurement was well advanced to a spatio-temporal correlation method. Erroneous vectors, often appeared in the correlation method, was successfully removed using Hopfield neural network. As a result, calculated 3-D velocity vector distribution well corresponded to the observed temperature distribution. Consequently, the simultaneous three-dimensional measurement system for temperature and flow field was developed.
An unsteady heat flux method was used to determine the thermophysical properties of Polymethyl methacrylate (PMMA) and cellulose, both of which are commonly used as standard sample materials in flame spread tests. The analysis is based on an analytical solution to a heat conduction equation with periodic heat flux. The effective thermal conductivity and diffusivity for a packed bed of each sample material were simultaneously determined from the amplitude and phase delay of the temperature response for periodic heat flux. Experiments were performed for a range of heating rates, particle sizes and apparent densities. The variation in thermophysical properties with temperature was obtained, and the effects of pyrolysis for cellulose, and phase transition for PMMA, on the thermophysical properties of these materials were investigated.
To consider the nonreflection property of the turbulence statistics in a rotating system, a new method has been proposed to introduce turbulence helicity into an isotropic turbulent field for direct numerical simulation (DNS). Effects of the helicity and the rotation on turbulence statistics and vortex structure are investigated by theoretical analysis and DNS of the homogeneous decaying turbulence. Although DNS results show that both rotation and the helicity inhibit energy decay, theoretical analysis reveals the occurrence of very different mechanisms for the rotation and the helicity for producing the inhibition effect. The helicity directly decreases the energy transfer while the rotation suppresses the energy transfer due to the so-called scrambling effect. The other effect of system rotation is that rotation elongates the vortex structure along the rotation axis, however, the presence of the helicity appears to weaken this tendency.
The equilibrium composition and spectral absorption coefficient of SiC ablation layer plasmas have been calculated for temperature of 5000 to 7000 K, layer thickness of 0 to 7.5 mm, and pressure of 0.1 to 1.0 MPa. The radiations included were molecular bands, atomic lines, and continuum processes. The absorption coefficient thus calculated was applied to a simplified shock layer model for the Jupiter entry probe to investigate the effectiveness of the ablation layer in reducing the radiative heating from a shock layer to a body surface. It was found that the SiC ablation layer is very effective to protect the body from radiative heating and that the photoionization processes of atomic carbon and silicon were mainly responsible for radiative absorption at high photon energy range. Furthermore the molecular carbon bands were effectively absorptive in relatively low photon energy range, Particularly at low temperatures.
The flow around a cylindrical body standing on the sand is computed numerically and the movement of the sand is investigated. The numerical method employed in this study can be divided into three parts: (1) calculation of the flow around the pole using the marker and cell (MAC) method with a generalized coordinate system; (2) estimation of the sand transfer caused by the flow through friction; (3) determination of the shape of the ground. Since the computational area changes at step (3), this procedure has to be repeated at each time step. Results show that the horseshoe vortex scoops out the ground in front of the pole. When the pole stands vertically on a flow, it generates a big horseshoe vortex to make the scouring force strong. On the other hand, when the pole has a cross section that is flatter along the flow, such as a lens or ellipse, the scouring force becomes weaker. Furthermore, it is found that the horseshoe vortex becomes smaller when the pole has a conical base.
A space-time finite element method was presented for solving time-dependent and two-dimensional heat conduction problems. Linear hexadedron elements in a space-time domain were used as the finite elements. In this paper we present a discretization technique in which finite element approximations are used in time and space simultaneously for a relatively large time period called a time slab. The weighted residual process is used to formulate a finite element method for a space-time domain based upon the continuous Galerkin method. Numerical illustrations have indicated that the unstructured space-time algorithm provides more rapid convergence to the analytical solutions than the usual finite element analysis with discretization in space only.
Two numerical procedures in the Direct Simulation Monte Carlo(DSMC) method, applying particle flux conservation at inflow/outflow pressure boundaries, have been developed to treat the two most important boundary conditions encountered in micromechanical devices involving gaseous flows. The first one is for both specified pressures at inlet and exit; while the second one is for specified mass flow rate and exit pressure. Both numerical procedures have been tested on short and long microchannels in the slip and transitional regimes. Excellent agreement has been found between the current results and the previous reported numerical results as well as the experimental data for the first type of boundary conditions. Finally, the developed numerical procedures have been applied to backward-facing micro-step gaseous flows to demonstrate its general applicability in more complicated flows.
In order to study unsteady flowfield around high speed trains passing by each other, a three-dimensional inviscid numerical method based on three types of domain decomposition techniques is developed. Roe’s FDS scheme is used for the space discretization, and LU-SGS method is adopted for the time integration. After validation of the code to a single track train/tunnel interaction problems with three dimensional tunnel configuration, the numerical simulations of the trains passing by on the double-track are performed for the 5 different cases using 3 basic parameters; e.g. nose shape, existence of tunnel, and train length. After the parametric study, variational parametric studies are carried out to understand the effects of the velocity of the train, the gap between the train and the blockage ratio. Firstly, train/tunnel interaction problems for double track railway system are investigated and aerodynamics loads histories during the crossing events—train/train interaction problem—are presented and discussed.
This paper deals with viscoelastic flows due to a rotating disc enclosed in a cylindrical casing with variable axial clearances. The behavior of viscoelastic flows due to a rotating disc has been characterized based on an experimental study using a flow visualization technique and numerical simulation using several models for constitutive equations. The pattern of secondary flow for the region of competition between the elastic force and the inertial force is a radially arranged double-cell structure (Type DC 1') which differs from Type DC 1. Type DC 2 was reproduced qualitatively by numerical simulation with Giesekus and Phan Thien-Tanner models, but not by numerical simulation with the upper convected Maxwell model or power-law model. Furthermore, the profiles of the tangential velocity component Vθ and the radial velocity component Vr obtained by numerical simulation with the Giesekus and Phan Thien-Tanner models for small H/R agreed with those measured using particle tracking velocimetry (PTV).