Gas-liquid two-phase slug in a concentric annulus with a rotating inner cylinder has been experimentally investigated. The flow patterns identified on the basis of visual observations are classified into four categories, i.e., slug flow, spiral-slug flow, peripherally elongated slug flow and ring-form slug flow. Such a flow regime map is expressed as a function of the total volumetric flux and the rotational speed. The rising velocity of a large bubble approximately coincides with Nicklin et al.'s correlation at relatively low rotational speed, while the rising velocity is lower than that of Nicklin et al.'s correlation at high rotational speed. This is closely related to the void fraction distribution along the radial direction in the annulus. The ratio of the kinetic energy loss to the potential one in the slug flow regime is well correlated by a linear function of the volumetric flux ratio of two phases as in the case of a bubbly flow regime. This ratio of energy losses is well expressed using the drift-flux model modified by taking into account the large bubble size in the annulus.

抄録全体を表示

In the present study, we report experimental results obtained from air-water two-phase flows in the downstream region of a rotating pipe which consists of three independent rotating parts with a total length of 3m. The pressure drop characteristics associated with the change in flow behavior due to effect of the pipe rotation are examined. Particular attention is paid to the length of the rotating pipe on the pressure drop characteristics. From the results of experiment, it is revealed that the effect of the rotating pipe becomes significant in the downstream of the entrance. A correlation based on the consideration of bubble fusion number applied to the bubble fusion model (proposed in the previous report) can be used to estimate the friction multiplication factor φ_L² satisfactory in bubble flow and plug flow regimes for the developed rotating two-phase flows.

抄録全体を表示

An experimental study was conducted to investigate the structures of a pure wake (no injection), a weak wake (some injection), a momentumless wake (injection adjusted to provide a thrust which exactly cancels the model's drag) and a weak jet (more injection than necessary to cancel the drag) by means of flow visualizations/hot-wire measurements in a wind tunnel. The weak wake/jet and momentumless wake were generated behind a two-dimensional airfoil model with mass injection through a slit along its rear end. The structures for the momentumless wake are neither wakelike nor jetlike, but instead burst out into the free stream at random intervals. Similarity of the turbulent intensities was observed. Turbulent quantities show that the momentumless wake decays downstream of the model very quickly and is analogous to isotropic turbulence.

抄録全体を表示

A numerical study has been performed on turbulent flow developing in a 90 deg curved duct of retangular cross section and an aspect ratio of 6. The ratio of bend mean radius of curvature to hydraulic diameter is 2.04 and straight ducts of 7.5 and 25.5 width are attached to the inlet and outlet of the bend, respectively. In addition to this straight duct. two-dimensional lateral contraction of the wind tunnel with an area ratio of 6 is considered as the numerical domain as well as the experimental apparatus. Calculated results were compared with the experimental data measured for Reynolds stress components in detail. In this calculation, an algebraic Reynolds stress model was adopted in order to predict anisotropic turbulence precisely, and a boundary-fitted coordinate system was introduced as the method of coordinate transformation. As a result of this examination, the present method can predict normal and shear stresses distributions of the inlet straight duct but cannot reproduce the characteristic destabilizing effect and shows an increase in the turbulence intensity close to the outer wall in the curved duct. The discrepancy may be caused by use of the wall function as the boundary condition. In the outlet straight duct. the calculated value of normal and shear stresses are smaller than those of the experiments.

抄録全体を表示

In this paper we describe numerical analysis to investigate the three-dimensional turbulent structure and fluid flow behavior in a square-to-circular transition duct. The hydraulic diameter, 40mm, was the same in the square inlet and round exit sections, resulting in a cross-sectional area reduction of 21.5 percent. The transition takes place over two hydraulic diameters. In the calculation, an algcbraic Reynolds stress model combined with a boundary-fitted coordinate system is applied to a square-to-circular transition duct in order to solve anisotropic turbulent flow precisely. The calculated results are compared with the experimental results available to examine the validity of the present method. As a result of calculation, it is found that the present method predicts streamwise and horizontal mean velocities well but the calculated distributions of the normal stresses show a discrepancy compared with the experiment. Although the flow was fully developed at the inlet to the transition duct, as stated in the experimental report, the experimental results of normal stresses differ greatly from those of flow, while calculated results are similar to those for fully developed flow. In this transition duct, two kinds of the secondary flow exist, i.e., one is the secondary flow of the second kind in the square duct and the other is that of the first kind in the transition duct induced by the pressure gradient. Moreover, the secondary flow of the transition duct disappears when the flow enters the duct with a circular cross section. The present method can predict these features quantitatively.

抄録全体を表示

Turbulence structure near a shear-free. non wavy gas-liquid interface (free surface) with stable stratification is investigated by means of direct numerical simulation. The influence of stable stratification on the turbulence structure and scalar transfer near the free surface is discussed. Profiles of the turbulence intensities near the free surface indicate that redistribution of the turbulence kinetic energy from the verical to the horizontal direction (internal energy exchange) occurs in unstratified flow. However, this redistribution is restricted by the stable stratification. These results strongly suggest that vertical motions near the free surface are no longer dominant in stratified flow. As a consequence, scalar transfer across the free surface is deactivated due to the restriction of the vertical motions near the free surface.

抄録全体を表示

The laminar boundary layer in axial flow along a long thin cylinder with constant temperature is investigated by expanding in powers of a variable ξ that was used by Seban & Bond [J. Aero. Sci., 18 (1951), 671.]. The series solutions for the skin friction, the local Nusselt number and the other boundary-layer parameters are obtained up to about 20 terms, and by utilizing Pade approximants, their singular points on a complex plane are evaluated. Euler transformations then yield more convergent expansions, and by reflecting the known asymptotic behavior of the solutions on the extended series solutions, reasonable solutions are obtained for sufficiently large ξ. It is clarified that the solutions by the method of analytic continuation using Pade approximants are also reasonable. Pade approximants are applied further to velocity and temperature distributions.

抄録全体を表示

In a previous paper, we proposed a simple, low-storage finite element scheme for three-dimensional flow analysis of non-Newtonian fluids. The scheme is based on the GSMAC (generalized simplified marker and cell) finite element method, and employs rational Runge-Kutta time integration in order to maintain stability for high viscosity flow. In the present work, the flow of Carreau fluids through an axisymmetric abrupt contraction is calculated to confirm the applicability of the present scheme at low shear rates. The numerical results agree with numerical solutions calculated by Kim et al., and by means of comparing the results for two contraction ratios (1 : 4, 1 : 2), we investigate how the contraction ratio affects the velocity profiles at the plane of contraction and the pressure drop distance.

抄録全体を表示

The oscillation of a flexible tube in the vicinity of a copula with a rigid tube was studied experimentally in order to understand the basic behavior of the artery in the vicinity of a copula with a vascular graft. In this study, the emulsion was used as a substitute for blood because of its properties similar to blood. The radial displacement of the flexible tube and the flow rate of the emulsion were measured using a high sensitive gap sensor and an electromagnetic flowmeter respectively. The experimental results are summarized as follows : (1) The non-Newtonian property of fluid has the effect of increasing the amplitude of the flexible tube. (2) If a rigid tube is installed in the upstream side of a flexible tube, the amplitude of the flexible tube is increased markably. This effect becomes clear with the decrease in the stiffness of the flexible tube.

抄録全体を表示

In the present study, the wall shear stress and the flow structure in the right angle branch have been studied experimentally in laminar steady flow using a laser Doppler velocimeter. The side branch, 14mm in diameter, bifurcates at a right angle from the trunk, 24mm in diameter, and both upstream and downstream corners at the entrance of the side branch are square edged. In the upstream trunk, the wall shear stress at the upstream corner acquires a very large magnitude owing to the deflection of the core flow. Consequently, the change of wall shear stress around the upstream corner is comparable to that around the flow divider. There is a periodical change of velocity along the separation surface in the side branch. There is a high probability that this phenomenon is similar to that of the Karman vortex street.

抄録全体を表示

In this paper we describe the acoustic analysis of fluid flow in a duct with four cylinders. The sound field is considered to be related to the oscillating lift forces which act on the cylinders, which are experimentally measured. The Helmholtz acoustic equation associated with oscillating fluid force is applied to the flow past four circular cylinders. The boundary element method for the numerical solution of the sound field is applied. The sound pressure inside a cavity with a one-way movable wall is calculated in order to test the method. The sound pressures on the four cylinder surfaces and the wall surface are evaluated.

抄録全体を表示

Three-dimensional steady and unsteady flows through the bifurcation of the porcine renal artery were analyzed numerically using the finite element method. In the present calculations, the geometry of bifurcation model was determined in accordance with the measured data. The arterial wall was assumed to be rigid and the non-Newtonian effect of blood was not taken into consideration. The calculated results showed that the present geometrical model had an effect to decrease the length of separated flow region. In addition, it was shown that the present geometrical model flattened the distribution of wall shear stress, which has important implications relative to the concept of WSSG.

抄録全体を表示

The frequency response of pressure drop of ER fluids between two parallel-plate electrodes to sine wave changes in the applied electric field is investigated at constant flow rates. The ER fluids used in the present study consist of cellulose suspended in silicone oil. Unipolar and bipolar wave electric fields, and constant ones are used. Amplitudes of the electric fields are 0.4 and 0.7kV/mm and DC component is 1.0kV/mm for unipolar electric fields. The frequency of the electric fields is varied from 0.1 to 1000Hz. The central values, the amplitudes and the phase angles of the pressure drop are measured. Dependence of the pressure drop on electrical excitation frequency is investigated. Furthermore, the microscopic behavior of ER suspension structure in two fixed electrodes is visualized for stationary ER fluids with a CCD camera under application of electric fields. The effect of electrical change on the pressure drop is explained from the results of flow visualization.

抄録全体を表示

A cavity arranged by means of varying its length and depth was placed either with flush against the lip line of a rectangular nozzle allowing zero distance between cavity and nozzle or with its opening displaced at some distance from the nozzle lip line. The jet was operated in the range of Mach number from 0.05 to 0.15. Significant jet mixing enhancement can be achieved through deep cavity forcing. We present an improved expression for predicting the frequency of deep cavity resonance. The effects of cavity forcing on mixing enhancement are assessed by examining flow visualization and velocity profiles of the excited jet.

抄録全体を表示

Experimental and numerical investigations are carried out to study the striation and roughness formation mechanism on the cutting surface in the machining process using an abrasive water jet. The results of two-dimensional numerical simulations which are based on a simple model are compared with experimental results. The agreement between the two is satisfactory. It is also shown that although the surface roughness increases with the distance from the nozzle, the mechanism by which the material is cut and the striations formed is fundamentally the same throughout the kerf.

抄録全体を表示

In this paper, an explicit time-dependent full Navier-Stokes solution is used to analyze flow field of a two-dimensional ejector-nozzle system. In order to investigate its performance, a parametric study was performed using three controlling parameters duct to nozzle area ratio, nozzle pressure ratio and Reynolds number. Results show that there is an optimum area ratio for efficient pumping of secondary flow. At high area ratios, a free stream flows directly through the mix duct without giving adequate pumping so that secondary flow in the ejector-nozzle system is reduced. On the other hand, at low area ratios, secondary flow is blocked by the jet boundary. A larger nozzle pressure ratio will result in higher efficiency in pumping. However, due to interaction between the duct wall and shock cell structure, a decrease in efficiency of the ejector-nozzle system occurs. The pumping rate increases with Reynolds number while the loss of the system decreases with it.

抄録全体を表示

In the previous papers, we showed that a plane jet discharging into a channel exhibits selfsustained flip-flop motion which changes from periodic to chaotic oscillation at a certain Reynolds number. The present report deals with feedback control of the flip-flop jet based on the OGY method which is known to be useful in changing an unstable orbit into a stable orbit in the phase space of a nonlinear system. The method is first tested using model equations of both one-and two-dimensional systems, and is confirmed to work fairly well. The OGY method is applied to the numerical database of the flip-flop jet. It is found that for a certain range of control parameters, the unstable chaotic state can be reduced to a stable periodic state by the feedback input of short-interval and small-amplitude velocity perturbation.

抄録全体を表示

Direct numerical simulations of a spatially developing plane jet have been conducted to clarify the mechanism of interaction between the two shear layers in a plane jet. Following conclusions were obtained. (1) When the jet is width half of the vortex spacing in the flow direction, interaction between two shear layers occurs. (2) The phase angle of the perturbations in the two shear layers shifts to π, and the arrangement of vortex structure changes to that of Karman's vortex except in the case of an initial phase angle of 0. (3) Due to the effect of the interaction during roll-up, the fluctuation energy shows maximum value when the initial phase angle is π. (4) The jet shows asymmetric behavior during roll-up when the initial phase angle is π/2. (5) When the jet width is one-quarter of the vortex spacing in the flow direction, tearing of the vortices occurs in the pairing process.

抄録全体を表示

Numerical simulations of viscoelastic flows through an eccentric four-to-one abrupt contraction are carried out using the Giesekus model. The SMAC (Simplified-Marker-and-Cell) method is used to analyze the three-dimensional flows. The velocity profiles along the path line passing through the center of the exit exhibit an overshoot near the entry section, and at high Weissenberg numbers an undershoot follows the overshoot. The magnitude of the stress along the same path line has a peak near the entry section, and its slow relaxation process indicates that a large downstream length is necessary for fully developed stress conditions to exist. The peak is lower than that for the flow through the concentric four-to-one abrupt contraction ; the decrease in the peak amplitude is understood to be due to the distortion of the path line in the eccentric geometry. A corner vortex, the height of which is a maximum at the widest corner, grows as the Weissenberg number increases. Furthermore, the tangential flow toward the widest section inside the vortex is determined.

抄録全体を表示

A new pressure-correction method is developed in order to calculate the flow field involving both subsonic and supersonic flows. The method is applied to the turbulent flow computation code, in which governing equations including the standard k-εturbulence model are solved in the physical component tensor form with a boundary-fitted coordinate system. The MUSCL-type TVD scheme is also used not only for the estimation of the convection term of the transport equations but also for the estimation of the mass residual of the pressure correction equation. In order to examine the accuracy of the present pressure-correction method, two-dimensional supersonic flow through convergent-divergent nozzles and two-dimensional transonic flow through a channel with the shock/turbulent boundary-layer interaction are computed. The computed results are compared with the experimental data and good agreement between them is obtained.

抄録全体を表示

A high-order space-accurate scheme of hyperbolic conservation laws usually requires data obtained at five grid points. An implicit version of the scheme constitutes an expensive pentadiagonal system on the left-hand side of the scheme when the scheme is written in ⊿-form of the algorithm. In order to make the left-hand side of the scheme an efficient tridiagonal system, various types of approximation are often applied to flux-Jacobian matrices on the left-hand side of the scheme. In this paper, these approximations are examined and assessed with respect to actually permissible Courant number, shock and contact-discontinuity profiles, and numerical accuracy of the solution using a one-dimensional shock tube problem. The effect of subiteration procedure at each time step, which is introduced in the developed scheme, is also assessed. The results show that implicit schemes with subiteration are favorable in terms of both stability and accuracy of the solution.

抄録全体を表示

A newly developed, inclined double-acting wave energy pump which can be designed effectively to inclined beach bottoms and the variation of wave periods has been experimentally investigated. In order to improve the pump performance the reciprocating moving parts, a pair of float-pistons have to be tuned with wave periods and take advantage of horizontal forces of the waves, which provides an idea of variable geometry. The pump, with variable axis angle, is operated in double-action mode in which water is pressurized by both up and down motions of the float-pistons. Model test pumps with rectangular and cylindrical floats were tested to determine the effect of the pump axis angle (θ), wave period (T_w) and some kinds of horizontal wave energy converters with an inclined plate. It is shown in this paper that an optimum pump axis angle exists in terms of the pump performance, and the combination of the cylindrical type float with the inclined wave energy converter results in increased pump efficiency on a large scale.

抄録全体を表示

A spiral grooved vacuum pump and a compound molecular pump (the combination of a spiral grooved pump and a turbomolecular pump) are widely used in the thin-film industry for processes such as semiconductor production. Pumping performance is high at pressures below 1000 Pa and low at pressures above 1000 Pa when the clearance between rotor and stator is on the order of 0.1mm, which is the practical value for industrial use. The double spiral grooved vacuum pump is thought to have better pumping performance at such high pressures than the conventional spiral grooved vacuum pump. The aim of this study is to investigate the feasibility of use of the double spiral grooved vacuum pump at pressures above 1000Pa. A double spiral grooved vacuum pump with a rotor of 150mm diameter and 190mm length has been designed and manufactured. Its pumping performance has been tested by experiments. The test results show the improvement in the performance at pressures above 1000Pa compared to the conventional spiral grooved vacuum pump.

抄録全体を表示

We describe a method of calculating yawed inflow effects on horizontal axis wind turbine rotors. Yawed inflow brings about asymmetric induced velocity distributions on the rotor plane due to an inclined vortex wake construction. For the calculation of yawed inflow phenomena, an inviscid aerodynamic model based on pressure distributions is adopted. Under the assumptions of an incompressible inviscid irrotational flow, the pressure perturbation satisfies the Laplace equation and acts as an acceleration potential function. The rotor blades are represented by spanwise and chordwise pressure distributions composed of analytical first-order asymptotic solutions to the Laplace equation. Integration of the accelerations determines the velocities in the rotor plane. With these velocities the aerodynamic loads can be calculated. An advantage of this model is that evaluation of the relatively complicated unstationary effects from the vorticity in the wake is not necessary in the calculation procedure.

抄録全体を表示

The critical heat flux (CHF) under conditions of high subcooling and high velocity in small diameter channels was systematically evaluated to obtain a reliable CHF correlation which can be used to predict CHFs higher than 50 MW/m² in the design of high performance nuclear research reactors, the accelerator target and the first wall and diverter of the fusion reactor etc.. The major test conditions of the existing CHF experiments investigated in this study were a channel diameter or gap of 0.3 to 4mm, a channel length-to-diameter ratio of 15 to 97, a pressure of 0.1 to 7.1 MPa, an inlet water subcooling of 13 to 210°C and a water mass flux of 124 to 90000kg/(m²·s), resulting in a CHF of 4.2 to 224MW/m². The existing experimental data were compared with the analytical CHF solution which was derived by the author for the channel flow under high subcooling and high velocity of water at atmospheric pressure. As a result, the effect of the channel diameter on CHF was clarified quantitatively and a correlation of CHF was newly proposed, which can be used to predict the existing CHF data within an error of about ±35 percent, reflecting the effect of the pressure properly.

抄録全体を表示

Experimental studies have been made on heat transfer on a horizontal platinum wire during nucleate pool boiling in nonazeotropic refrigerant binary mixtures at pressures of 0.25 to 0.7 MPa and at heat fluxes up to CHF. The boiling feature of the mixtures and the single-component substances was observed by photography. The relationship between the boiling behavior and the reduction of heat transfer coefficient in binary mixtures is discussed in order to propose a correlation useful for predicting the present experimental data over a wide range of low to high heat fluxes. It is shown that the correlation is applicable to alcoholic mixtures. The physical meaning of k, which was introduced to evaluate the effect of heat flux on the reduction of heat transfer coefficient, is clarified based on the measured nucleate pool boiling heat transfer data and the visual observation of the boiling features.

抄録全体を表示

We present an efficient inverse analysis method based on a sensitivity coefficient algorithm to estimate unknown boundary conditions of multidimensional heat conduction problems. Sensitivity coefficients were used to represent the temperature response of a system under unit loading conditions. The proposed method, coupled with sensitivity analysis in the finite element formulation, is capable of estimating unknown both temperature and heat flux on the surface provided that temperature data are given at discrete points in the interior of a solid body. Inverse heat conduction problems are referred to as ill-posed ones because minor inaccuracy or error in temperature measurement causes a drastic effect on the predicted surface temperature and heat flux. To verify the accuracy and validity of the new method, two-dimensional steady-state and transient problems are considered and their surface temperature and heat flux are evaluated. From a comparison with exact solutions, the effects of measurement accuracy, number and location of measuring points, a time step, and regularization terms are discussed.

抄録全体を表示

Experimental study has been conducted on local heat (mass) transfer characteristics in the entrance region of rectangular ducts with/without rib-roughened long side walls using the naphthalene sublimation technique. Three rib heights and four rib pitches have been tested. In the smooth duct, the laminar boundary layers are first formed near the duct entrance and then begin to transit to turbulent ones from the duct corners. The turbulent region develops from the corner toward the duct center as the flow proceeds downstream ; therefore, the local mass transfer rates change in both the streamwise and spanwise directions. In the rib-roughened duct as well, the local mass transfer rates vary in not only the streamwise but also the spanwise directions very near the duct entrance ; the location of the maximum local mass transfer rate caused by the flow reattachment moves upstream as the adjacent wall is approached. It has been found that the influence of rib height on the local mass transfer rate is more pronounced near the corner than in the duct center region.

抄録全体を表示

Investigation of characteristics of cold water pipe flow with surfactant is important to develop a cold heat energy transport system. Both flow drag and heat transfer reductions by the Toms effect occur due to a rod-like micell structure of surfactant. When a counter-ion (sodium salicylate) was added to water solution containing of dodecyltrimethylammonium chloride (DATC) surfactant in the present study, the rod-like micell structure of DTAC was formed by connection of spherical micells of DTAC. Moreover, the reduction characteristics of flow resistance and heat transfer were influenced by the amount of the counter-ion. Useful nondimensional correlative equations for flow resistance and heat transfer are derived in terms of various nondimensional parameters.

抄録全体を表示

Numerical simulation of turbulent heat transfer over a blunt flat plate accompanying the separation and reattachment has been carried out by using the large eddy simulation (LES) technique. The dynamic Smagorinsky model (DSM) is used for modeling the subgrid-sacle (SGS) heat flux, and the SGS turbulent Prandtl number is calculated as a function of space and time. Numerical calculations were made for three Reynolds numbers at Prandtl number 0.7 and are compared with previous experimental data and the direct numerical simulation (DNS) data. The present results using DSM show better agreement with experimental data than using Smagorinsky model (SM) and agree well with DNS data. It is shown that LES using DSM is applicable to the engineering prediction of the temperature field with the separated and reattached flow regions over a wide range of Reynolds number.

抄録全体を表示

An experimental study has been performed to determine the melting heat transfer characteristics of slush ice in a horizontal cylindrical capsule. Slush ice, which is a mixture of fine ice particles and ethylene-glycol aqueous solution adopted as the testing material, was heated by the cylindrical capsule wall. The shape of the mush-ice interface, melting rate, temperature distribution, and local/average heat transfer coefficient at the heated wall were extensively observed and determined under a variety of conditions of heat flux and initial concentration of the aqueous binary solution. The results revealed that the melting heat transfer of slush ice is markedly effected by free convection in the double-diffusive layers owing to the thermal and solutal buoyancy forces.

抄録全体を表示

The heat transfer and its enhancement of an air-water dispersed flow in a heated vertical pipe have been studied. The axial and circumferential wall temperature distributions were measured for various mist ratios and wall heat fluxes. The measured wall temperature increased sharply at a particular point in the direction of the stream with a notable variation in the circumferential profile. This sharp rise was thought to be caused by the breakdown of the water film rather than by its dryout. A separate experiment in which no heating was used was carried on to estimate the droplet deposition velocity and the water-film flow rate. Numerical analysis taking into account the heat and mass transfer from the water film was performed to estimate the mean wall temperature, and good agreement was obtained with the experimental results in the area where the entire inner surface of the pipe was covered with the water film. In this area, the rate of heat transfer was approximately seven times higher than that for a single phase air flow. This enhancement was shown to be mainly due to the evaporation of the water film. The mechanism of the heat transfer enhancement is discussed in detail using the numerical analysis results.

抄録全体を表示

In the present report, first, heat transfer coefficients of oscillatory flow at inner surfaces of the heating and cooling regions of oscillation controlled heat transport tubes (OCHTs) are investigated numerically. The numerical simulation is conducted under the following three conditions for the tube walls at the heating and cooling regions : isothermal, extremely thin and actual wall systems. Based on the numerical results and Hausen's correlating equation for laminar flow heat transfer in tubes, a correlating equation of the heat transfer coefficient is developed which can be generally applied to these three conditions. Next, using this correlating equation and the author's simplified model of overall thermal resistance in OCHTs, heat transport rates are predicted, and it is found that the predicted results are in good agreement with the numerical results. Finally, numerical simulation is conducted also to compare the heat transport rates of OCHTs with those of the conventional forced circulation type under the same pumping power. The results indicate that there exist oscillatory flow regions in which the heat transport rate of the phase shifted OCHTs is larger than that of the conventional circulation type.

抄録全体を表示

A numerical analysis of the ice melting heat transfer phenomenon in a horizontal cylindrical capsule immersed in water is presented. This phenomenon is caused by a combination of the natural convection heat transfer of water around the capsule and the ice melting heat transfer inside the capsule. When capsule material has low heat conductivity, heat conduction in the capsule itself must be considered. This combined heat transfer problem has been studied numerically. Simulation is carried out by solving the Navier-Stokes equation, adhering to the Boussinesq approximation coupled to the energy equation for copper capsule and acrylic capsule. As a result, the ice melting characteristics, temperature profiles, flow patterns, local Nusselt number profiles, average Nusselt number profiles, and ice melting rate were shown.

抄録全体を表示

Experimental and numerical studies of the effects of side-wall temperature on the flow pattern of Rayleigh Benard cells in a rectangular enclosure with an aspect ratio of 6 : 2 : 1 (length : width : height) are presented. Experiments were carried out using air at Rayleigh number≤16000. We have demonstrated that the flow pattern always consists of a steady even or odd number of roll-cells with axes almost parallel to the short or long side of the enclosure for all side-wall temperatures. The results of numerical calculations based on the steady three dimensional Boussinesq equations are in quantitative agreement with the observed flow patterns.

抄録全体を表示

In this paper we present a model of the dynamics of a two-heat-exchanger network connected in parallel. The model is represented using the inner admittances, which represent the dynamics of each heat exchanger, in series. The series connection of the inner admittances indicates the complexity of the two-heat-exchanger network. The complexity mainly depends on the characteristic of the confluent process. The fluid temperature behavior in the pipe between the heat exchangers and the confluent junction markedly influences the characteristic of the confluent process. We investigate the dynamics of the two-heat-exchanger network with respect to the effect of the piping and the confluence.

抄録全体を表示

In this work, we consider the absorption refrigerator driven by waste steam from fuel cells. The absorption refrigerator we consider does not use steam as a heat source from a boiler which can supply constant heat, but uses waste steam from fuel cells. Therefore the steam flow rate or pressure changes greatly as disturbance. In this work we aim to clarify the effect of the change in steam flow rate or pressure on the performance of the absorption refrigerator driven by waste steam. In this research, first a high precision dynamic simulation model is constructed to enable examination of the dynamic characteristics. In order to simplify the model, the delay in factors which can be ignored is considered as the lumped parameter system model. It was verified that the simulation model matched the results of experimental data to a high degree. With this simulation model, a detailed dynamic characteristics were clarified.

抄録全体を表示

Experimental study has been carried out to clarify the characteristics of liquid film thickness of an air water annular two-phase flow in a double tube bend (inner dia. of the outer tube is 25mm, outer dia. of the inner tube is 16mm and bend radius is 250mm). The distribution of liquid film thickness on the outer tube varies with both the air and liquid flow rate, whereas on the inner tube, it varies only with the air flow rate. The average liquid film thickness of the inner tube approaches an asymptotic value, which is determined by the air flow rate, with the increase of liquid flow rate. The average liquid film thickness of a cross section in the double tube bend coincides with that of a coiled single tube of the same hydraulic diameter. It was confirmed that the film inversion phenomenon on the outer tube occurs when the centrifugal force acting on the gas phase exceeds that of the liquid film.

抄録全体を表示

Solidification of an aqueous solution is studied using a laser interferometry technique combined with an optical microscope. In order to measure the concentration distribution in an aqueous NaCl solution near an ice crystal and three-dimensional shape of the ice crystal under a known temperature field, a directional solidification stage is used, which is composed of low-and high-temperature blocks and a moving bed. In this experiment ice crystals showed four shapes : flat, treelike, swordlike and needlelike. As a result of the interferometry experiment, it was observed that concentration increases in bulk solution for thick samples, swordlike ice crystal has an asymmetric triangular cross section, and that treelike crystal has a flat and low top. Also, ice formation response to injection of a high concenttration solution was studied.

抄録全体を表示

We present a new methodology for inverse analysis of time-dependent two-phase Stefan problems. The problem considered is that of determining the time change of the phase-change interface from several observed temperatures. In the present method, imaginary heat sources arranged in an imaginary domain are introduced, and then the phase change interface is identified as the isothermal surface at the melting temperature by controlling the imaginary heat source intensities. Using the delta-function type imaginary heat sources and their corresponding Green's functions, which are precalculated numerically, it is shown that the phase-change interface is determined noniteratively at each time step. Some numerical examples demonstrate the capability of the proposed method.

抄録全体を表示

We study the transition mechanism governing the three-dimensional evolution of a coflowing fuel jet by means of HSMAC simulations which can handle the density change. In the purely axisymmetric case without any perturbations, calculations provide only Kelvin-Helmholz-type vortices. The spatial and temporal calculations triggered by an azimuthal perturbation, however, provide a detailed picture of orderly three-dimensional transition processes leading to the concentration. reorientation, and stretching of the vorticity. Furthermore, the azimuthal vorticity shows the helical structure in the braid region of vortex rings. The streamwise vorticity, which indicates azimuthally several pairs of vortices with opposite rotation, is produced in the braid region immediately upstream of the core of vortex ring. The streamwise vorticity is stretched in the braid region under the tension of vortex ring, penetrates inside the downstream core, wraps round the upstream core and finally connects each core. The streamwise vortices with opposite rotation induce the radial ejection flow.

抄録全体を表示

The soot formation process and the structure of a fuel-rich premixed flame using porous medium have been investigated by one-dimensional numerical analysis. The combustion reaction is represented by the Arrhenius law, while the reaction rate of soot formation is based on Tesner's model which has been modified to include the oxidation of soot. A novel model for soot formation is developed by taking coalescence and surface growth into account ; the coalescence probability is statistically evaluated for polydispersed soot particles. On the basis of this model, the following features of the soot formation process, especially those in the distribution of the soot particle size, are clarified. Immediately after the beginning of soot formation, surface growth is dominant. At the downstream, however, soot particles grow mainly by coalescence, but terminate if they collide with radicals. The flame structure is also discussed in terms of energy recirculation between the upstream porous medium and polydispersed soot particles.

抄録全体を表示

The hydrogen concentration of a hydrogen/nitrogen mixture at a stability limit of a diffusion flame formed in a downstream region of a backward-facing step is experimentally measured as functions of the step height, the distance from the step to the fuel injection point, the freestream velocity and the fuel injection velocity. The stabilization conditions can be divided into three types depending on the fuel injection point. When the fuel injection point is located near the reattachment point, the hydrogen concentration at the stability limit increases more significantly than that in other regions due to shear turbulence. Furthermore, in this region, the stabilization conditions can be divided into two types as a function of l^* and l_r. where l^* is the distance from the step to the center of the reattachment region and l_r is the distance from the step to the reattachment point : the flame as l^*/l_r>1 is more stable than that as l^*/l_r<1.

抄録全体を表示

A multi impingement-wall haed attached to the cylinder head near the center of the combustion chamber of a DI diesel engine has been proposed by authors. The relative uniform sprays have been realized due to the effect of impingement of sprays by means of this head. Photographs of these impinging sprays were taken in the atmosphere at the 1 MPa with room temperature and the local Sauter mean diameter of these sprays was detected in the ambient atmosphere. It is clear from these experiments that the effects of head is very extinguished and the distribution of droplets is very uniform.

抄録全体を表示

In order to develop a natural gas fueled spark ignition engine with high thermal efficiency and clean exhaust gases, combustion characteristics of natural gas and air mixtures were examined using a rapid compression combustor. We concentrated on lean mixtures because of their potential for high efficiency and low pollutant emissions. To elucidate the effect of compression-induced physical aspects on the combustion process, the charge was kept quiescent before the start of the compression process. The results showed that an increased charge pressure increased the time required for combustion. A high compression ratio or piston speed tended to shorten the combustion time, but their effects were rather weak. An increased humidity in mixtures measurably increased the combustion time. The stratified charge, which was examined using the soap bubble method, markedly improved the combustion process of lean mixtures.

抄録全体を表示

Fractal burning model was applied to the simulation of combustion in a spark-ignition engine. This model was examined under various conditions of swirl intensity in a disk-type combustion chamber. The computed results of mass fraction burned were in good agreement with experimental ones. In this model, the following relations were assumed. (1) The fractal dimension D₃ is expressed as a function of (ρ/ρ_r)(u'/S_L), where ρ/ρ_r, u', and S_L are density ratio, turbulence intensity, and laminar burning velocity, respectively. (2) The outer cutoff scale is equal to the flame radius, and the inner cutoff scale is proportional to Kolmogoroff scale.

抄録全体を表示

Cycle-to-cycle fluctuation in combustion of a lean mixture and a nitrogen-diluted in a sparkignition engine was investigated. From the results given in the first report, it was clarified through experimentation that the duration of 1% combustion was a dominant factor for determining the fluctuation of mean effective pressure. In this study, the fluctuations of 1% combustion duration and indicated mean effective pressure were computed as the function of fluctuations both in the equivalence ratio of the mixture and in the turbulence characteristics of the cylinder charge. The computed results were in good agreement with the experimental results obtained under well-controlled conditions of mixture composition in the cylinder.

抄録全体を表示

In order to contribute to international scientific and technical fields toward the twenty first century, it is necessary for us to achieve original developments in science and technology. Creative challenges in education, therefore, should be introduced into colleges and universities, particularly on the grounds that individual talents can be developed not only by passive and uniform-oriented education but also by means of creative challenges. For this purpose, creative activities aimed at developing creative talents of students, which allowed them to choose topics freely were introduced. The findings obtained by investigating and creating about their topics were presented and discussed by the students. In this report, the trial results of creative activities are discussed.

抄録全体を表示

Experimental investigations were conducted of heat transfer from a single row of circular cylinders, placed in the transverse direction of flows of air and water. Local and average Nusselt numbers for a cylinder at cylinder spacings L/d from 1.3 to 3.3 were measured precisely in the Reynolds number range from 10² to 5.0×10⁴. Average Nusselt numbers are well correlated with the Reynolds numbers based on mean velocity in the flow channel between cylinders. The effect of the interaction between wake flows behind cylinders on average Nusselt number is relatively small.

抄録全体を表示