We have developed hybrid gas channels for polymer electrolyte fuel cells (PEFCs) that can arbitrarily change the ratio of the flow rates between interdigitated and parallel flows. We study gas channel structure and the above ratio by analyzing overpotentials to improve the cell performance. The flow rates of the two flows are controlled by valves of the inlets and outlets of high and low pressure gas channels in the cathode. The flow coefficients of the valves (CV values) are measured beforehand for calibration. Differences between total flow rate and each the flow rate give the rate of in-plane flow passing through the GDL under the rib of the separator. Increasing the flow rate through the GDL under the rib of the cathode separator improves the cell performance although the effect of the flow rates in the high and low pressure gas channels of the cathode on the cell performance is small. This can be explained by that large amount of oxygen is consumed at the electrode (catalyst layer) under the rib where the electrode reaction takes place predominantly.

抄録全体を表示

We have developed hybrid gas channels for polymer electrolyte fuel cells (PEFCs) that can arbitrarily change the ratio of the flow rates between the interdigitated and parallel flows. We analyze flooding under the rib of the cathode separator using the interdigitated flow to find methods to improve the cell performance. The ratio of the rate of in-plane flow passing through the gas diffusion layer (GDL) under the rib of an operating cell to the rate of the in-plane flow of dry air without operation gives the cross-sectional area ratio of through-pores (A/A₀) and flooding ratio (1-A/A₀) of the GDL. Quantitative analyses of these ratios clarify that the concentration over-potential increases with increase in the flooding ratio. Increasing pressure difference between high and low pressure channels for the interdigitated flow beyond the bubble point of the GDL pores enables an increase in current density up to 2 A/cm² when the utilization of the air is 40%. Such current density has hardly been accomplished by the conventional PEFCs.

抄録全体を表示

The effects of the areal weight and thickness of the GDL of a polymer electrolyte fuel cell (PEFC) have been analyzed using a cell with hybrid gas channels that can arbitrarily change the ratio of the flow rates between the interdigitated and parallel flows. We clarify that a decrease in the areal weight of the GDL reduces the pressure loss in in-plane direction of the GDL owing to the enhanced exhaust of accumulatd product water in liquid phase. Increased pore diameter also contributes to reduce the flow resistivity. On the other hand, 30% increase in the thickness of the GDL decreases the cross-sectional area ratio of through-pores (A/A₀) due to further flooding in the GDL, leading to the decrease in the pressure loss of about 70% of that expected. In both the cases, cell voltage fluctuation from the instability of the exhaust of the liquid water under the operation at high air utilization remains to be solved.

抄録全体を表示

The Hamiltonian particle method for incompressible fluid flows has been developed by authors and proved to have excellent properties in conservation of mechanical energy as well as linear and angular momenta. In this study, standing waves in a rectangular tank are simulated using this method. The result is compared to analytical solutions which are obtained from the Stokes perturbation expansion. Agreement between the numerical and the analytical solutions is fairly good at least in the initial two or three cycles. The numerical result is also compared to the one obtained using the moving-particle semi-implicit (MPS) method. The supremacy of the Hamiltonian particle method compared to the MPS method is confirmed.

抄録全体を表示

Laminar flows between the bottom surface of a moving rectangular body and the plane wall are investigated experimentally using a towing wind tunnel facility where the model is carried in still air by a linear-motor actuator. The Reynolds number based on the moving speed and the distance between the bottom of the moving body and the fixed plane wall is about 170. In the transient stage of the flow development including the acceleration stage immediately after the body starts to move, the flow is driven not only by viscous shearing stress but also pressure gradient so that the maximum flow velocity exceeds the speed of the moving body. After the flow becomes steady, the normal-to-wall velocity distribution tends to the linear profile of the Couette flow except near the leading edge of the moving body where the flow is still affected by pressure gradient. In addition, the same rectangular-body is carried between two side-walls set normal to the plane wall. For such a nearly two-dimensional configuration, the flow develops to a Couette-Poiseuille flow.

抄録全体を表示

The unused biomass such as residues left on forest floor and unused portions of farm crops is considered as an important domestic and renewable energy resource. The present study addresses the forestry biomass derived from logging and thinning operations in forests. Since most of the forests in Japan locate in the mountain region, the forestry biomass has not been used continually due to the difficulty in carrying residues out of the forest. In order to promote the utilization of forestry biomass, the efficiencies for the transportation across steep slopes have to be improved. In the present study, the new method to transport wood chips by means of liquid film flow in sloped pipeline is proposed. The transportation limit of wood chips is investigated by using the test section imitating the sloped rough terrain pipeline. The wood chips are stagnated in the downstream region of hydraulic jump, which causes blockage of wood chips in the tube. To predict the flow rate of transportation limit of wood chips at a given liquid flow rate, analytical model is presented. From the comparison with the experimental results, it is found that the transportation limit in the sloped pipeline with horizontal tube can be predicted by the present analytical model.

抄録全体を表示

This paper details experimental studies on the flow field around a low-pressure linear turbine cascade whose solidity is changeable. Highly loaded low-pressure turbine (LPT) blades or ultrahigh lift airfoils are one of the key paths to successful future aero-engines, however those blades are usually accompanied with separation bubble, eventually leading to the increase in aero dynamic loss. The purpose of this study is to clarify the effect of incoming wakes upon the aerodynamic loss of the rotor blade cascade, through the measurements of wake-affected boundary layers upon including separation bubble for low Reynolds number conditions and/or low solidity conditions. Cylindrical bars on the timing belts work as wake generator to emulate upstream stator wakes that impact the rotor blade. Hot-wire probe measurement is conducted over the blade suction surface to understand to what extent and how the incoming wakes affect the boundary layers containing separation bubble. In particular, from the viewpoint of providing database for blade designing, this paper focuses on time-averaged characteristics of the separated boundary layer influenced by the wake passing.

抄録全体を表示

In this study, the interaction vortex flow features around a pair of parallel arranged bluff cylinders were observed by visualizing water flow experiment at the range of the gap ratio G/d=0-3. It was obtained that the result of established wind tunnel test and the result of this water tank test agreed about the characteristics of vortex shedding when varying the distance of circular cylinder gap. The flow pattern and vortex shedding frequency of another type bluff cylinder (triangular and square cylinder) were also investigated. As a result of the experiment, it was shown that the flow pattern of wake flow was divided into three kinds (coupled vortex streets, biased gap flow and single vortex street) regardless of the cylinder section shape and the cylinder size. Then, the region of the appearance of flow pattern was shown about each case. In the case where two each other independent vortex streets were formed, three typical flow patterns of vortex formation (in-phase coupled vortex streets, out-of-phase coupled vortex streets and complication coupled vortex streets) were observed. It was known that three configuration of vortex formation appear intermittently and alternatively.

抄録全体を表示

In this study, numerical analysis of cavitating three-blade flat-plate cascade is performed with paying attention to cavitation surge which is a type of cavitation instabilities. A numerical method employing “locally homogeneous gas-liquid two-phase compressible medium model” is applied to solve the unsteady cavitating flowfield. From results of numerical analysis, few typical tendency of cavitation surge which is well known in experiments was reproduced numericaly, that frequency of cavitation surge decreases with lowering σ, and that cavitation surge is easy to occur at low flow rate condition. Additionaly, mechanism of pulsation phenomenon of cavitation surge in flat-plate cascade was clarified, that collapse pressure of cloud cavity in downstream of cascade propagates to upstream on contact with re-entrant jet which is flowing backward inside a sheet cavity and causes oscillation in upstream pressure, also that negative angle of attack and choking of cascade throat are considered to cause of oscillation in flow rate. About dynamic transfer function of cavitation, certain degree of possibility was confirmed in quasi-steady mass flow gain factor M_quasi and cavitation compliance K_quasi for prediction of occurrence limit of cavitation instabilities. Additionally, for prediction of frequency of cavitation surge, it was confirmed that local cavitation compliance K_local is more available than quasi-steady K_quasi.

抄録全体を表示

In previous investigations, a series of wind tunnel tests was conducted on the vibration and scattering behavior of full-sized model of roof tiles, which were used widely for roofings of Japanese wooden dwellings. As a result, the mechanism of vibration and scattering of roof tiles could be understood. This paper describes a water leak test of roof tiles. The 100 roof tiles were set up on 10 lines and 10 rows on the pitched roof in the downstream of the flow from the wind tunnel. The flow velocity was increased gradually from the low velocity to the high velocity, and the effects of the wind on the roof tiles were investigated by the water leak test. The experimental apparatus for the purpose of reproducing the wind and the rain was used. The situation of the water leak test was filmed by video camera. Two accelerometers were simultaneously used. The roof tiles which showed significant vibration at any velocity were found in a series of experiments and the accelerometers were attached to the two neighboring roof tiles on the model roof. The water leak phenomena in the experiment by using the artificial rainfall apparatus are discussed in this paper. Some mechanisms with relatively large amplitudes at low-frequency vibrations, which caused water leak phenomena were made clear.

抄録全体を表示

In this study, the influence of a drag-reducing surfactant on the turbulent boundary layer was extensively investigated under different solution concentrations using a two-component laser-Doppler velocimetry system. The surfactant solution used here was a mixture of cetyltrimethyl ammonium chloride (CTAC) with sodium salicylate as counterion, which was dissolved in tap water. The concentrations tested were 100 and 150 ppm, and the previous data on surfactant solution of 65 ppm were also used.The temperature of surfactant solution was T=20°C.It is found that the drag reduction ratio DR becomes larger downstream, and decreases with the increase of concentration from C=65 to 150 ppm. Regardless of solution concentrations, the value of the mean velocity scaled by the friction velocity increases with increasing the amount of drag reduction, while the mean velocity scaled by the free-stream velocity, which are collapsed for the different Reynolds number and drag reduction ratio, are about in the middle between the mean velocity profile of water and the Blasius laminar profile. For the present experimental conditions tested, the maximum of streamwise turbulence intensity near the wall is not related to the amount of DR directly and seems to be affected by the low Reynolds number effect strongly, while the location of the maximum becomes more distant from the wall with the increase of DR. The additional maximum of streamwise turbulence intensity near the center of the boundary layer, which is observed for all the concentration tested, and the wall-normal location of additional maximum are almost constant and are independent of the amount of drag reduction.

抄録全体を表示

In this study, the influence of a drag-reducing surfactant on the turbulent boundary layer was extensively investigated under different solution temperatures using a two-component laser-Doppler velocimetry system. The surfactant solution used here was a mixture of cetyltrimethyl ammonium chloride (CTAC) with sodium salicylate as counterion, which was dissolved in tap water. The temperature of surfactant solution was T=20, 25, 30, and 35°C. It is found that the drag reduction ratio DR for C=65 and 100 ppm becomes smaller with the increase of the temperature from T=25 to 35°C, and the DR at T=20°C is smaller than DR at T=25°C. We introduced three distinct regimes of drag reduction, which were referred to as low drag reduction (LDR), high drag reduction (HDR), and maximum drag reduction (MDR) regimes. The distributions of the mean velocity, the wall-normal turbulence intensity, and the Reynolds shear stress could be clarified by these three regimes, while the streamwise turbulence intensity profile could not be explained.

抄録全体を表示

The adoption of the contra-rotating rotor has been proposed in response to a demand for the axial flow pump with higher specific speed. The internal flow field and blade rows interactions between the front and rear rotors should be considered in the design for higher performance and more stable operation, but has not been clarified yet. The flow field in contra-rotating axial flow pump was measured at various flow rates with the LDV and wall pressure measurements. In the present paper, the experimental and numerical results in the flow field between front and rear rotors are shown at the design flow rate and the flow behaviors are discussed, related on blade loading, blade tip leakage flow, wake and blade rows interaction.

抄録全体を表示

Cooled wakes of mercury and water are numerically simulated, and computed results are shown and discussed. As a result, effects of fluid kinds on wake patterns, i.e. the streaklines, streamlines, equi-vorticity lines, and isothermal lines, are elucidated as follows : (1) Effects of the fluid kinds on the cooled wake and the cooled vortex street are extremely strong, because the dominant action is the temperature diffusion in mercury, and the temperature advection in water. (2) The temperature distribution is extremely different between mercury and water. In mercury, the temperature varies in the whole wake, and the isotheral lines draw concentric circles. In water, a number of the temperature islands are generated, and advect heat to the downstream. (3) The cooled vortex street occurs in mercury and water. The shape of the vortex spiral of water is vague and is not circular. The vortex spiral of mercury rotates as a vortex with smooth and clear streaklines. (4) The motions of two separated shear layers of vorticity are extremely different between mercury and water. (5) The Richardson numbers of onsets of the cooled vortex street and symmetric wake in water are extremely larger than those in mercury.

抄録全体を表示

Void fraction of nitrogen gas-water two-phase flow in a circular tube from 1 to 10 mm in diameter was measured by neutron radiography. The measured void fraction results were compared with Ishii's drift flux model based on Mishima-Ishii's flow regime criteria. The void fraction was predicted well in annular flow region except for α<0.8. Mishima-lshii flow regime map was modified by a churn to annular flow transition criterion of α=0.8. The void fraction was predicted well by Ishii's drift flux model in annular flow region in every diameter tubes.

抄録全体を表示

A correlation of pool boiling heat transfer coefficient in Ammonia/Water mixtures was proposed on the basis of heat transfer data measured by the present experiment. The correlation was derived using both the fraction difference between vapor and liquid, y₁-x₁, and the temperature difference between dew and bubble points, ΔT_E It can predict the heat transfer coefficients of the mixtures during pool boiling in the entire fraction range within an accuracy of±20 percent. It was clarified that the coefficient more strongly depends on y₁-x₁ than on ΔT_E for Ammonia/Water mixtures.

抄録全体を表示

As heat generation in satellites increases, securing sufficient radiator panel area is an important problem. Deployable radiator, whose radiator panels are deployed post-launch in space to increase the effective radiator panel area of the satellite, is becoming an important thermal control technology. Reservoir Embedded Loop Heat Pipe (RELHP) is applied to the deployable radiator for thermal transport device. This paper presents the heat transport dynamic characteristics of RELHP using radiant cooling condenser and liquid forced convection cooling condenser by experimental study. It was found that heat leak into liquid line, flexible line and reservoir increases length of subcooling region in the condenser. In case of radiant cooling condenser, sub-cooling region length is shorter than liquid forced convection cooling condenser. Furthermore, vapor temperature is mainly decided by radiation capacity of radiator panel, because liquid temperature return into evaporator is rises with increasing radiator panel temperature. In addition to that, time length from start-up until steady state is longer than in case of using liquid forced convection cooling condenser, because radiator panel has large heat capacity.

抄録全体を表示

A new type of thermosyphon was experimentally examined mainly for its thermal conductance. It consisted of parallel installed two conventional thermosyphons connected by two U tubes. A conventional single thermosyphon was also tested to compare heat transfer characteristics with the new type. Water and ethanol were used as the working fluid. Experiments were carried out with three different cooling temperatures, 25, 37.5, and 50°C and two charge ratios, 0.5 and 0.9. The following findings are obtained. The relationship between thermal conductance and heat transfer rate is similar between the new type and the conventional thermosyphons, but the magnitude of the conductance in the former one is much larger than that in the latter due to larger maximum heat transfer rate. The higher coolant temperature gives the higher conductance. The Imura's boiling correlation and Nusselt equation for condensation heat transfer can predict the conductance within -20 to +60%, and then the major reason of the prediction error is in the condensation heat transfer.

抄録全体を表示

Conventional geothermal development has been targeted by either the geothermal fluid reservoir in which steam and hot water are dominant or a deep-seated hot dry rock. However, these geothermal resources are regionally restricted therefore, long-term research development has been necessary to ascertain and establish the achievement of a power generation system by geothermal energy. From the view point of this research background, the development of shallow geothermal resources has been extensively carried out recently because shallow geothermal resources are not only exploited comparatively with ease but also can be found anywhere. The present study aims to investigate the heat extraction characteristics of a coaxial heat exchanger, which is the downhole heat exchangers extracting geothermal energy. From the computed results it was found that the heat extraction performance using a coaxial heat exchanger under the forward circulation mode depended on the inner tube material of the coaxial heat exchanger in the case where the length of the coaxial heat exchanger was comparatively long. Further, it was found that there was not much difference in the heat extraction rate among the metal tubes as long as they were used as the inner tube material.

抄録全体を表示

In order to utilize shallow geothermal energy resources which can be found anywhere, a vertical U-tube downhole heat exchanger has come to be widely used around the world. This is because this resin U-tube is moderate in price, and it is comparatively easy to insert and to set up in the downhole after drilling a well. In utilizing the U-tube downhole heat exchanger, multiple U-tubes are practically inserted in the downhole for the purpose of extracting more geothermal energy from a stratum. In the present study, a numerical simulation has been performed to clarify the heat extraction characteristics of both vertical double and triple U-tube downhole heat exchangers. The numerical results showed that the heat extraction rates from a stratum for both vertical double and triple U-tube downhole heat exchangers increase with increasing the flow rate of the heat extraction medium of water. Furthermore it was shown that the heat extraction rate for a vertical double U- tube downhole heat exchanger depends on the well depth, in other words, the total length of the U-tube ; and the heat extraction rate increases with increasing well depth.

抄録全体を表示

The purpose of this research is to verify the validity of the rotation mechanism theory of the SMA engine by experiment. The main results obtained by this research are as follows. We manufactured the equipment for the experiment to measure simultaneously rotation speed and torque, and obtained a maximum output power 1.2W. We compared the experiment values of the rotation speed and the torque with the theoretical values, and partially verified the validity of the theory. We drew up the empirical formula which predicts the maximum output power of the SMA engine.

抄録全体を表示

The ignition of fuel spray is affected by various factors, i.e., temperature, pressure, equivalence ratio, droplet diameter, number density of droplets, air velocity, ignition energy, and so on. In order to clarify the effects of spray characteristics on spray ignition phenomenon, there have been many researches. However, the effect of drop size distribution on spray ignition has not been studied yet because it is difficult to control drop size distribution independently with other spray characteristics. In this study, it is aimed to clarify the effect of drop size distribution of fuel spray on ignition phenomenon, using drop size distribution control technique developed by the authors The results of this study clearly show that the ignition probability decrease with the increase of standard deviation of drop size distribution under the constant mean drop size and concentration condition. In addition, the decrease ratio of ignition probability decreases by the decrease of average fuel spray concentration. The results also show that flame area decrease with the increase of the standard deviation of drop size distribution.

抄録全体を表示

Nano meter size particulate matter (PM) and polycyclic aromatic hydrocarbons (PAHs) have harmful effects for human health. Investigations of fundamental characteristic of formation behavior of PAHs in combustion process were required for the reduction of nano-PM emission. In this paper, PAHs characteristics in diffusion flames of paraffin fuels were investigated using laser diagnostic and gas sampling methods. Methane (CH₄), propane (C₃H₈) and hexane (C₆H₁₄) were used as the test fuels. A laser induced fluorescence (LIF) method was used to obtain the spatial distributions of PAHs which were precursor molecules of soot. Spatial distributions of soot were also measured by a laser induced incandescence (LII) method. A transition region from PAHs to soot was defined from the results of LIF and LII. Further, PAH concentrations in the test flames were measured using a gas chromatography (GC).Profiles of PAHs concentrations of test flames were compared. As the results, it was found that the low carbon number fuels such as propane and methane are remained to some fuel at downstream part. And, these some fuels were changed to large PAHs at downstream part.

抄録全体を表示

We have analysed the diffusion impedance of a polymer electrolyte fuel cell (PEFC) by electrochemical impedance spectroscopy (EIS). As a result, we derive oxygen diffusion distance δ adjacent to the cathode electrode (catalyst layer) assuming planar oxygen diffusion in a thin water layer unlike the conventional flooded-agglomerate model and gas phase diffusion model in a gas diffusion layer. The experimental results and their analyses show that δ agrees with a value reported from our previous ex-situ measurement. Moreover, δ gives reasonable concentration overpotentials. Increasing current density and gas humidification brings about the growth of δ. It supports that δ corresponds to the thickness of the accumulated product water in liquid state at the cathode electrode. Thus EIS analysis of the diffusion impedance can be used to diagnose the water accumulation at the cathode electrode. This result enables the prediction of the flooding and drying-up of PEFCs and assures their stable operation.

抄録全体を表示

In the present work performance of a novel ignition system based on imploding detonation was studied in order to improve ignition ability and combustion enhancement of gas engines using a natural gas for power generation. Deflagration generated in the pre-chamber was introduced into the imploding chamber whose cross-sectional area decreases in the propagation direction and then transition to detonation occurs. Flame jet emanating from a nozzle fixed at the implosion center is expected to have superior ignition ability due to high temperature and pressure inherent to imploding detonations. The experimental results show that maximum pressure at the implosion center was 300 to 350 times higher than the initial pressure. The burning time in the main combustion chamber of 350 mm in diameter ranged from 2.5 ms to 4 ms and decreased with increase in the nozzle diameter under the condition that the nozzle orifice was choked. From the estimation of burning velocity and cell size of a stoichiometric methane-air mixture under the gas engine condition, it was requested that the nozzle diameter must be larger than 4.0 mm and that the volumetric ratio of the pre-combustion chamber to the main combustion chamber is less than 0.01, in order to detonate the mixture in the pre-combustion chamber more easily with comparison to the present test.

抄録全体を表示

In order to clarify the mechanism of radiant heat transfer in the combustion chamber of DI diesel engines, radiant heat fluxes at various points on the piston head of a rapid compression and expansion machine were measured by using thin-film thermocouples. The ambient oxygen volume fraction, the swirl ratio, and the diameter and the number of the fuel injection nozzle hole were varied as experimental parameters. Measured radiant heat fluxes were compared with the total heat fluxes measured at the similar measurement locations to the radiant heat fluxes, the flame temperature and KL-factor obtained by the two-color method. In addition, in order to investigate the effects of the flame distribution and the reflectivity of the wall on the spatial distribution of the radiant heat loss, numerical simulations of the radiant heat transfer between the luminous flame and the chamber wall were also carried out via Monte-Carlo method. The results showed that the magnitude of the radiant heat flux is smaller than that of the total heat flux by one order, regardless of the changes in the oxygen volume fraction, swirl ratio and fuel injection nozzles. It is also shown that the spatial distribution of the radiant heat loss inside the piston cavity is uniform due to high reflectivity of the chamber wall made of aluminum alloy.

抄録全体を表示

Particle size distribution, mean diameter, and number density of nano-PM (Particulate Matter) in the exhaust gas from a single cylinder DI diesel engine with high rates of cooled EGR (Exhaust Gas Recirculation) were studied using SMPS (Scanning Mobility Particle Sizer), and the effects of the exhaust catalyst and fuel injection quantity (engine load) on these characteristics were investigated. With increasing EGR, the peak of the particle size distribution shifts to larger diameters, and the mean diameter and number density increase till the smoke concentration reaches a peak value, and then these values decrease as smoke decreases with further increases in EGR. In the ultra-high EGR smokeless low temperature combustion regime, the nuclei mode particles are dominant and show a peak around 35 nm, the mean diameter and the number density of the particles reach levels lower than those without EGR. Exhaust oxidation catalysts are effective to reduce to some degree the particle number density in a wide EGR range but the effect on the mean diameter is small. More fuel injected into the cylinder causes a larger mean diameter when smoke concentration is high due to EGR, but in the smokeless combustion regime the mean diameter does not change significantly with fuel injection quantity. The reduction in the number density by the catalyst is less significant with a greater fuel injection quantity.

抄録全体を表示

We propose a new formulation for the attractive term of van der Waals type EOS written in the compression factor form. The formulation has three model constants at the critical temperature. One is determined from experimental data, and the other two are given by satisfying constraints on the critical compression factor and its first derivative at the critical point. An original temperature dependence function is also developed. The EOS along with the new formulation has the ability to reproduce the gas phase fluid property within 0.5% of the relative error as compared with experimental data.

抄録全体を表示

According to the Huygens' principle for a wave-front propagation, any point on a wave front may be regarded as the source of secondary waves and the surface that is tangent to the secondary waves can be used to determine the future position of the wave front. Therefore, the flamelet velocity-vector is defined as a vector perpendicular to it.However, it seems that this definition is not well understood. In the present paper, the definition of velocity-vector of the wave front is discussed. There is a difference between the particle velocity and the front velocity. The particle velocity, in general, has a component parallel to the front. The front velocity, by definition, is normal to the front. For fronts that propagate from particle to particle, it is impossible to introduce a unique definition of front velocity that has a component parallel to the front. It has been shown that the flamelet velocity-vector is the sum of the burning velocity-vector of the flamelet and the component of the gas velocity-vector parallel to the flamelet velocity-vector.

抄録全体を表示