The motion of mono-dispersed spherical bubbles rising along an inclined flat plate is investigated experimentally. An inclination angle of the flat plate is changed to control the bubble Reynolds number set to be 100. As experimental conditions, the number density of bubbles and the boundary condition on the bubble surface (free-slip and no-slip conditions) are varied, and their effects on the motion of the bubbles are analyzed. MgSO4 solution and Triton X-100 solution are used to achieve free-slip and no-slip bubble surfaces, respectively. In both cases, bubble coalescences are almost inhibited. When the number density of bubbles becomes higher and the rising velocity along an inclined flat becomes slower, the bubbles tend to be horizontally arranged in lines and such arrangements pile up to the bubble cluster. The radial pair distribution function and the conditional average of the relative velocity of two bubbles are calculated to evaluate the spatial distribution and bubble-bubble interaction quantitatively. The results shows that the bubble pair configuration of next to each other frequently appears; a horizontally-oriented (side-by-side) configuration is stable compared to a vertically-oriented (tandem) one. A qualitative tendency of the clustering motion is similarly confirmed when the no-slip condition is imposed on the bubble surface, however, quantitative differences occur due to the difference in the wake structure between these two conditions.

We experimentally examine the dynamic motion of single bubbles rising in hydrophobically modified alkali-soluble emulsion polymer (HASE) solutions with pH = 7.0. In this study, 1.80wt%TT935, 0.551wt% TT615 and 0.550wt%TT615 HASE solutions are used and we focus on the velocity jump discontinuity of the rise velocity. From experimental observations, we confirm that the velocity jump occurs for all the HASE solutions as a critical bubble volume is exceeded. We present that bubbles rising in TT615 HASE solutions may exhibit the velocity jump twice. It is found that the large-scale velocity jump for all the HASE solutions is closely related to the bubble shape: the velocity jump occurs when a threadlike long tail is formed at the rear of the bubble.

In a fundamental research of bubbly flows, it is required a technique for the bubble generation control in various liquid such as silicone oil, chemicals and so on. We have developed a bubble generation method that uses a slit elastic tube and an acoustic pressure wave in gas phase. Elastic silicone tubes were used for the bubble generation section in the early study. However, the silicone tube was deformed and expanded in a silicone oil, and bubble generation control was failed. In this study, four types of elastic tubes were tested as the bubble generation section to achieve the bubble generation control in a silicone oil. As a result, the fluorine-contained rubber tube that contains none of the plasticizer held original shape in the silicone oil. And its elastic response to the acoustic pressure wave was enough to control bubbles. On the contrary, large acoustic pressure was required to generate bubble using this tube compared to the silicone tube.

Air bubble slowly released from an underwater nozzle was recorded with high-speed camera and measured by hydrophone in order to observe sound radiation. From experimental data the acoustic frequency which was called 'Breathing mode' was close to one calculated from Minnaert's equation and the lower frequency called 'Deformation mode' was 10~55% of the above basic frequency. CFD analysis was done with two models. One was the approximating bubble shape just after pinched off and another model was to be released from nozzle as same as experiment. There were some differences between two models such as pressure and mode shapes.

Effects of the translational motion of a vapor bubble on its cavitation inception is numerically investigated on the basis of bubble dynamics in which nonequilibrium evaporation or condensation at the bubble wall is accurately considered on a molecular level. The bubble dynamics is applied to the problem that a vapor bubble nucleus formed at a separation point on a circular cylinder surface expands there, then detaches, and moves downstream in the high-speed water flow. The locus of the moving vapor bubble after the detachment is traced, and infinite growth condition of the bubble, i.e., inception condition, investigated in detail.

We experimentally and numerically study a shock-bubble interaction problem in gelatin gels. The nucleation of spherical gas bubbles is triggered by focusing an infrared laser pulse into gassupersaturated gels and the radius of the nucleated bubble is controlled by mass transfer of the dissolved gases. A spherical shock is then generated, through rapid expansion of plasma induced by the laser focusing, in the vicinity of the gas bubble. The shock-bubble interaction is recorded by a CCD camera with flash illumination of a nanosecond green laser pulse. The observation captures cavitation inception in the gelatin under tension that results from acoustic impedance mismatching at the bubble wall. We measure the probability of cavitation with varying the bubble radius and the standoff distance. The threshold pressure for the cavitation inception is defined by the probability equal to one half and is predicted, through comparisons to Euler flow simulations, at −25 MPa.

We study, with experiments and theory, effects of elasticity on quasi-static growth of spherical bubbles under gas supersaturation in gelatin gels. An optical system is developed to trigger gas bubble nucleation by focusing a laser pulse into gas-supersaturated gelatin gels and visualize the subsequent bubble growth driven by the mass influx of the dissolved gases; the growth rate is found to be hindered by the gel elasticity. To quantify effects of the gel elasticity on the bubble growth, the Epstein–Plesset theory is extended to the case of bubbles in linear elastic materials of Voigt type. It is found from comparisons between the experiments and the proposed theory that the modeling with linear elasticity is applicable to the case with gels of low gelatin concentrations.

Microbubble aeration is used to dissolve gases into water and is an important technique in agriculture and industry. We can measure concentration of dissolved oxygen (DO) in aerated water by commercial DO meters. However, there do not exist commercially available techniques to measure concentration of dissolved nitrogen (DN). In the present study, we propose a method to measure DN in aerated water with the aid of EpsteinPlesset-type analysis ; particularly, we are interested in how the fractions of DN and DO are altered before and after aeration. Gas-supersturated water is produced by applying aeration with micron-sized air bubbles and is then stored in a glass container. The diffusion-driven growth of a bubble nucleated at the glass surface is recorded with a video camera. The bubble growth rate is compared to the extended Epstein-Plesset theory that models mass transfer of both DO and DN into the surface-attached bubble. Given the DO measurements, we infer the DN level from fitting between the experiment and the theory. The fitted DN level is found to exhibit a deviation from Henry's prediction.

The aim of this study is to simulate the dynamics (collapse and rebound) of a spherical bubble filled with condensable and non-condensable gas. For bubble motion, we utilize the Fujikawa and Akamatsu equation; for the gas-mixture flow in the bubble, we utilize the Boltzmann equation for gas mixture. In this paper, we show that the numerical method for the equations and numerical results for the bubble collapse problems.

This study investigates general criterion of cavitation occurrence at a sudden acceleration of a liquid. We conduct an experiment as follows: A test tube filled with a liquid falls and hits a floor. The liquid is accelerated quickly and negative pressure is generated inside the liquid. When the negative pressure is large enough, cavitation can happen. In previous study, Ca number is proposed as a cavitation criterion for explaining experimental results installing a SS400 floor. In our preliminary experiments installing an ABS floor, however, results disagree with Ca number. It suggests unsuitable expression of pressure inside the liquid in Ca number. Therefore, this study aims to reformulate Ca number applicable for various floors. We measure acceleration of the test tube time by an acceleration sensor and estimate pressure fluctuation in the liquid. The reformulated number Ca∗ agree well with cavitation criteria for both floors.

To clarify a relationship between non-condensable gases and generation of residual microbubble after the collapse of a single bubble generated by laser or spark, the single bubble dynamics and concentration of the dissolved hydrogen gas were analyzed in ultrapure water. The projected cross-sectional area of the residual microbubbles per potential energy of the bubble was 0.86 mm² J^-1 in the laser case and 3.5 mm² J^-1 in the spark case. The dissolved H₂ concentrations per pulse were 0.23 µgL^-1J^-1 in case of the laser and 0.69 µgL^-1J^-1 in case of the spark. The amount of the generated residual microbubbles showed a strong relationship with the dissolved H₂ concentration. It is possible to conclude that part of the non-condensable gases such as hydrogen generated by the laser and spark form residual microbubbles.

Ozone has a strong oxidation capability. It is often used for water treatment system. However, ozone has the demerit that it is difficult to be dissolved in water. Therefore, we used micro- bubble in order to dissolve ozone in water. The diameters of the micro- bubbles are less than 50μm. The micro- bubble has features that are slow ascent rate and high inner pressure. In this study, we conducted a water treatment for cutting oil using the micro- bubble ozone in order to improve the efficiency of the ozone treatment. The experiment shows BOD increased by the micro- bubble ozone.

Micro-bubbles are used in the fields of aqueous environment, food, health care and marine areas. These conventional methods do not control the position of the micro-bubbles.This paper aims to actively control the behavior of micro-bubbles. The method of controlling the behavior of micro-bubbles is to give electrical field.The surfactants were added to the micro-bubbles. The electrical charge was given on surface of the micro-bubbles. Three kinds of surfactants were used. These are anionic, cationic and non-ionic surfactant. Result showed the velocity of micro-bubbles depended on the surfactants in the electric field condition.The anionic moved micro-bubbles in the positive electrode direction. The cationic moved micro-bubbles in the negative electrode direction.The non-ionic did not have an effect on the micro-bubbles.

Laser sustained plasma (LSP) wind tunnels have been attracting attention as a new high-enthalpy wind tunnel. But operating limits for a stable plasma generation are not clear. This study numerically investigates the dependence of the plasma generation on the working gas flow rate. As a result, the LSP was found to be maintained at a mass flow rate between 0.6-6.0 g/s, when the laser power is 700 W and the working gas is argon. Moreover, it was found that the electron diffusion is a dominant mechanism for maintaining the LSP in the case of the flow rate under 3.0 g/s. However, in the case of the flow rate over 3.0 g/s, an another but unknown mechanism might maintain the LSP. The value of the laser absorption coefficient strongly depends on the mass flow rate, and can be categorized into the three region according to the mass flow rate.

Strong acoustic waves are generated from the rocket plumes in the supersonic jet condition. The acoustic waves are significantly affected by temperature of jets. In this study, we perform large eddy simulations of the hot and cold jets that generate the acoustic waves. The temperature ratio of chamber to ambient air is set to be 4.0 and 1.0 for the hot and cold jets, respectively. High frequency Mach waves are radiated from the region close to the nozzle exit. As closer to the potential core, the frequency of radiated Mach waves continuously becomes low. In the hot jet case, we confirmed that the shorter potential core length, the larger angle of Mach waves, and the higher sound pressure level, as compared with those in the cold jet case.

Underwater micro shock wave generated from metal surface by irradiation of single pulse laser has been investigated and observed with shadowgraph method. In order to produce underwater micro shock wave, flat and concave glasses coated with 100 nm thick titanium have been used. In this experiment, we used relatively low laser intensity, ranged from 0 to 200 GW/m² and shock wave was generated due to photo-acoustic effect in metal induced by laser irradiation. Shock wave was focused immediately using concave surface and we confirmed the focusing effect by optical observation and pressure measurement. The strength of shock wave from concave surface has attained approximately twice of flat one.

Vortex tube can be used for local cooling and heating device. A compressed gas supplied through tangential nozzles makes high swirl velocity field in vortex tube. A compressible and turbulent vortex flow makes temperature separation. Energy transfer process that vortex flow experiences is related to temperature separation. In order to improve temperature separation efficiency, it's important to clarify the principle of energy transfer in vortex tube. In the present study, we focus on heat transfer in vortex flow, then, CFD analysis with large eddy simulation (LES) technique was employed to investigate temperature field and influence of turbulent thermal diffusion in vortex tube.

Temperature measurement of the DC discharged field is crucial to understand the fluid dynamic effect such as shock wave modulation due to the interaction with the discharged field. For the temperature determination, spectrum matching method based on the emission spectrum is promising as one of the non-intrusive methods. For using the spectrum matching method, database of the theoretical spectrum of chemical species is required. For the method, our research group has developed a theoretical spectrum calculation code for molecular nitrogen and ion of the molecular nitrogen. Then, based on the code, we have constructed the database. Especially for the molecular nitrogen spectrum, the presently developed code was verified by the comparison with the published results. And we obtained the fairly good agreement between them. Hereafter, for the application of the spectrum matching method to the temperature determination from the experimental spectroscopic data, the fitting code should be developed.

The Mars landing system has been developed with a disk-gap-band (DGB) parachute in supersonic flow regime. The Mars' thin density creates a scaling challenge that is hard to overcome in windtunnel tests. Therefore, a computational analysis will require for parachute design. We use NURBS basis functions for spatial discretization in both structural and fluid mechanics computations. The parachute shape becomes smoother than what we get from a typical finite element discretization.

Shock and detonation waves due to the explosion is caused by the rapid change in pressure. When these waves propagate across a layer of liquid, a portion of the liquid turned into fog that is called as aerosol. Aerosol is very light and readily diffuse into the atmosphere. If the aerosol contains hazardous materials, such as the radioactive materials and toxic substances, it may cause serious environmental pollution especially to the soil and health risk to the human. Thus, this is an important issue to be solved in the safety analysis. However, it may possible to alleviate this risk if we have a detailed insight about the shock-induced aerosol formation. In this study, the interaction between shockwave and water surface is simulated, and CFD analyses are performed to investigate the flow features and aerosol formation. The compressibleInterFoam solver of OpenFOAM is used to simulate this problem.

Brayton cycle of CO₂ has been recognized as an attractive new method for power generations. Supercritical CO₂ is partially condensed in the compressor due to the near-critical condition. The nonequilibrium condensation model for the high pressure condition is applied to the condensation of H₂O in a nozzle and the condensation of CO₂ in a compressor cascade. The numerical result in a nozzle is in a good agreement with the experimental results. The flow of CO₂ in a cascade is condensed between the blades because of the temperature decrease.

When gas-turbine engines are exposed in sandy environment, sand particles are ingested from the engine inlet and they would impinge and erode the wall surfaces, which results in the engine failure. Therefore, sand erosion phenomenon is one of serious problems in gas-turbine engines and anti-erosion coating techniques have been developed recently. In this paper, numerical simulations of the jet flows including sand particles impinging on a plate are performed. The jet flow impinges on the coated ceramic material composite substrate and the particles erode its surface. We simulated two types of coated CMC, the Coating A and the Coating B. It is found that due to larger Young's modulus, the erosion rate of the Coating A substrate is smaller than that of the Coating B substrate.

We previously constructed a new cavitation model called “Multi-process Cavitation Model”, which takes the various elementary processes in cavitation into account, and farily good results were obtained for cold water cavitation while a poor result was done for a cryogenic cavitation with thermodynamic effect. In this paper, we revisited the application of this model thoroughout investigation of the model parameters. We conducted a cavitation simulation on single hydrofoil, and found that the appropriate modeling of expasion/shrinkage is rather more important than the evaporation/condensation rate that causes thermodynamic effect.

Cavitation experiments was conducted with high temperature and high pressure cavitation tunnel to investigate thermodynamic effect on tip leakage cavitation. NACA0009 hydrofoil with tip clearance is chosen for a test body. In the experiments results, at T_∞ = 90 ^oC, the large amplitudes are observed at the higher frequencies than 50 Hz. These can be considered because collapse time of cavity was shortened by thermodynamic effect. Moreover, the temperature depression inside tip leakage cavity increases with the increase of the thermodynamic parameter.

The aim of this study is to clarify effects of dissolved air on cavitating flow around an isolated Clark Y hydrofoil. We conducted observations of instantaneous cavity shapes by high-speed camera, measurements of lift and drag forces and number density distribution of bubble nuclei in different conditions of dissolved air at two angles of attack. As a result, the number density of bubble nuclei increased with growth of cavity in high dissolved air condition. On the other hand, such increase is not observed regardless of cavity development in low dissolved air condition. Inception cavitation number and characteristics of lift and drag forces are different between the conditions of dissolved air particularly in the case of small angle of attack.

A pump intake model study is required in case of the intake geometry deviating from the ideal ones. Criteria of submerged vortices in model test based on TSJ Standard are to check the existence of air core or bubbles entrained into suction bell inlet. The visible air core or bubbles are considered as incipient cavitation mixing water vapor and air bubble dissolved in water. The purpose of this paper is to verify the aggregation phenomenon of air bubbles to submerged vortices. Steady and transient numerical analysis was done using SST and LES model. The submerged vortex visualized by air bubble void fraction is located similar to measured result.

In high speed turbopump inducers for rocket engines, cavitation occurs and often causes flow instabilities. In the present study, we tried to suppress all cavitation instabilities by adding swirl breakers to an original circumferential groove of a casing in wide ranges of flow rates in consideration of coming reusable rockets. The casing with swirl breakers succeeded in the suppression of almost all cavitation instabilities without degradations of pump and suction performances by suppressing backflow vortex cavitation. The swirl breaker drastically decreased circumferential velocities near the tip of the leading edge of the inducer and also created a circumferential strong vortex which suppressed a backflow. This is a mechanism of the suppression of backflow vortex cavitation.

This paper presents study for high performance of very low specific-speed centrifugal pump. The disc friction loss is one of the main causes of the performance degradation of the very low specific speed centrifugal pump. In order to examine the influence of the number of the balance hall giving a leakage flow to back shroud clearance, its number was changed with 4,2,1 and numerical flow analysis was performed. With a decrease in number of the balance hall, the flow rate in the impeller decreased, and the head increased. On the other hand, the shaft power did not change much, and therefore the efficiency was increased. Consequently, the highest efficiency was obtained in one balance hole, it was improved up to 2.7% efficiency.

We investigated complex behavior of rotating flow induced by a pair of corotating disks mounted in a non-axisymmetric enclosure. A flow model used in the experiment was equipped with a shroud opening and an arm-like obstacle, which was inserted into the area of rotating flow. Refractive index matching was applied to the working fluid. We measured two-dimensional velocity fields in planes parallel to the disks using particle image velocimetry. The measurement was performed at four different angles of the arm insertion at two different image planes. The circumferential component of the flow velocity exhibits local accelerations and decelerations caused by the shroud opening and arm insertion during the revolution of the disks. Both of the shroud opening and the arm insertion were found to induce the flow behavior, but in different mechanisms and regions of the flow.

Heat transfer rates from a heater to operation gas rise, and the output of the stirling engine is known to improve when I make movement of the operation gas turbulence. There, I prepared for three kinds of piston shapes of cylindrical and no cylinder and chevron shape, and possessed a stirling engine and performed an experiment to measure the output. As a result, no cylinder came to have lowest output as expected, but cylindrical turned out high in the output unlike expectation in cylindrical and chevron shape. Furthermore, I understood that differences between cylindrical and output of chevron shape shrank as number of revolutions rose.

The twisted Savonius turbine is a kind of vertical axis wind turbine (VAWT) which have twisted shape, and the output mechanism is still unknown clearly. Many researches show the flow field around the turbine has vertical components along with the rotational axis, and we think this is a key to understand the output mechanism. In order to obtain the three dimensional flow at the wake of the turbine, we applied three dimensional color-PTV and spatio-temporal LER for interpolation of vectors. As a result of these analysis, we found the three dimensional vorticity distribution.

A propeller fan in the outdoor unit of air conditioner is a key component in order to ventilate the heat exchanger of the heat pump. From a viewpoint of energy conservation, it is important to increase the blowing efficiency of the propeller fan. Moreover, it is necessary to reduce its weight from a viewpoint of resource saving. Therefore, turbo-machinery ribs are put along its continuous surface of enlarged blades by removing the center boss. As a results, it makes possible to reduce the weight while avoiding the stress on the leading edge of the wing, and develop the highly efficient propeller fan by using suction effect to inner side.

In this study, a multi-stage transonic axial compressor has been investigated using large-scale detached eddy simulation (DES). The complicated flow field of stator hub in the compressor was analyzed using data mining techniques, extracting important flow phenomena. The data mining techniques applied in the present study include vortex identification based on the critical point theory and topology data analysis of the limiting streamline pattern visualized by the line integral convolution (LIC) method. In the stator cascade, the incoming flow in the vicinity of the hub is subject to influences of the leakage flow and the secondary flow on the hub, and eventually accumulates near the filled clearance section on the blade suction side. As a result, a focal-type boundary layer separation is caused there, and separation vortex generates huge loss. In the aft-part of the passage, the separation vortex interacts with the leakage flow, generating further complicated flow field.

Interaction between surge and rotating stall in an axial flow compressor was investigated from the viewpoint of an unsteady inner flow structure. The aim of this study was to unravel the mechanism of the interaction between the surge and rotating stall, and unsteady behavior under the coexisting phenomena. The main feature of the tested compressor is a shock tube connected in series to the compressor outlet through a diaphragm, slits, and a concentric duplex pipe. The unsteady characteristics and the internal flow velocity fluctuations were measured in detail. The result suggests that the key factor that determines the unsteady behavior of surge is the transformation of the internal flow structure, particularly between the stall cell and the entire circumferential stall.

A centrifugal fan which will be equipped for HVAC on automotive was developed to aim high-efficient and environmental friendly device. In order to achieve the application for EV, especially, we have a fundamental demand of improvement of efficiency, performance, and noise reduction. In the present research for this sake, PIV measurement and CFD analysis are conducted for a fan of new concept in order to improve efficiency of HVAC system from the point of view of fluid dynamics. Then, verification of CFD analysis is reported by compering velocity fields visualized in terms of both methods.

The characteristics and the transition process of rotating stall behavior, which were followed by the decline of flow rate, in centrifugal compressor with vaned diffuser were investigated through experimental analysis. Measurements showed that impeller and diffuser rotating stall occurred at 55 Hz and 25 Hz, during off-design point at ϕ=0.14. Furthermore, stage stall, which rotated within impeller and diffuser passages, occurred at 22 Hz and ϕ=0.10. According to the measurement of the pressure and velocity at the diffuser inlet, the flow field was divided into three behaviors at ϕ=0.10. At the “pre-stage stall” condition, the diffuser stall was enlarged and hub-side local blockage emerged behind the scroll tongue in vaneless space. This blockage induced a static pressure rise on the shroud wall and a deceleration of radial velocity on the hub side. The diffuser stall moved into the impeller passage because of the blockage and caused the stage stall.

To improve the efficiency and to reduce the noise of centrifugal blowers, the authors investigated the flow in a vaned diffuser using compressible flow simulation. In this study, the effect of the gap between the impeller and the casing was focused on. From the results of flow simulation, it was confirmed that the reverse flow generated near the leading edge of the diffuser vanes streamed into the gap of the impeller and the casing. Additionally, the secondary flow at the semi-vaneless diffuser region which was observed in a CFD model with the gap was stronger compared to a CFD model without the gap.

This paper deals with the aerodynamic shape optimization of the return channel for a multi-stage centrifugal compressor. Utilizing the 3 dimensional Computational Fluid Dynamics, CFD, and automated optimizer, optimal geometry of return channel is investigated. Through this optimization process, the aerodynamic loss and de-swirl performances are much improved. The method how to design the return channel with high performance is also discussed in this paper.

In this study, we developed the new centrifugal compressor test facility which could make the wet gas conditions (LMF is up to 15%) and analyzed the influence of the mist flow on the fluid performance of the centrifugal compressor. As a result, we confirmed the adiabatic efficiency decreased with increasing in LMF. The main cause was increase in the shaft power. Moreover, we assumed that the liquid film on the inner wall of the impeller was discharged from the trailing edge at a circumferential speed by the centrifugal force. Then we confirmed that an accuracy of the predicted shaft power increase at the maximum LMF of 15% based on the assumption was 3.7%

Air conditioners designed for an energy saving are required in a market. Improvement of a fan efficiency and uniformization of a velocity distribution in heat exchanger are necessary for improvement of a performance for energy saving. The velocity distribution and pressure-loss of heat exchanger and shaft power of the centrifugal fan were evaluated with Computational Fluid Dynamics (CFD). An appropriate diameter of the fan was determined by the result of the evaluation. Reducing the diameter to a certain value reduced the shaft power by approximately 4 % compared with a basic diameter. However, further reducing the diameter increase the shaft power in approximately 6 % compared with a basic diameter. These were caused by the change of the fan efficiency and the velocity distribution. In particular, we found the velocity distribution was related with the outlet height of the fan and a distance between the fan and the heat exchanger.

The laminar-turbulent transition of a boundary layer induced by a jet injection in the inlet region of a circular pipe was experimentally investigated. The jet was periodically injected radially from a small hole in the inlet region into the pipe flow. The turbulence induced by the jet within the boundary layer developed into turbulent patches which then grew in the axial, circumferential and radial directions downstream. A single hot-wire probe was used in the measurements. The axial velocity component was measured at four downstream stations, one within the inlet region and the other within the development region. Output voltage of the hot-wire was digitized and data was processed by a PC. An ensemble-averaging based on the signal from the jet trigger was performed. Mean and fluctuating velocity in the axial direction and intermittency factor which is the fraction of turbulence were obtained.

An isolated turbulent oblique band is investigated at a low bulk Reynolds number Re_m = 440 in plane channel flow with a very large numerical domain. The downstream edge of the band propagates in the streamwise and spanwise directions to quasi-periodically generates disturbances which will be left behind to decay in time. Long time integration shows that the band keeps roughly constant length in time without continuously shrinking, extending or splitting, suggesting the presence of an isolated equilibrium turbulent band.

In the final stage of laminar to turbulent transition, it was known that the regions called ‘turbulent spot' appear. When these spots develop downstream and the boundary layer fill with them, the transition process is finished. Previous studies have shown that development of spot was arranged by the energy of jet unrelated these shapes. The purpose of this study was to clarify growth and development proses of the spot using some shapes jet with same energy. From results obtained, the spot created by sharp jet grew faster than that created by flat-shape jet. In the spot created by flat jet, hair-pin type vortex induced secondarily was weak and wider than first one. The wide hair-pin type vortex could not create a strong low-speed shear layer between its legs. Consequently, growth of the spot created by flat-shape jet became slow.

In a dissipative system, there exists the (global) attractor which has finite fractal dimensions. The flow on the attractor can be parametrized by a finite number of parameters (Temmam 1987). For low-Reynolds-number turbulence in plane Couette flow, we construct precise low-dimensional governing equations on the attractors using machine learning.

Direct numerical simulation of a spectral method was performed to study the relaminarization process of the plane Couette flows under spanwise system rotation. As increasing the rotational effect, large-scale intermittency of the turbulence appear in the near wall region. Then the turbulent stripe consisting of the turbulent and quasi-laminar regions near the wall, while homogeneous turbulence was observed in the channel center. The small scale structure is almost same in the entire channel except wall vicinity consisting of viscous sub-layer and buffer layer. As increasing the rotational effect, turbulent spot was observed, though the homogeneous turbulence was observed in channel center. In the intermediate state, friction coefficient largely oscillates with time.

The consistency between the mean velocity profile and the friction factor measured at the Hi-Reff at AIST, NMIJ was investigated. The velocity profile data was fitted to a velocity profile form based on the log law, and an equation for the friction factor was derived by integration. The derived equation for the friction factor accurately represents the friction factor data. The deviation from the friction factor data is less than 0.6%. This small deviation indicates the high reliability of the measurement data at the Hi-Reff. Based on the equations for the friction factor derived using the mean velocity profile, the best-fitting constants for the friction factor data are also proposed. For the Reynolds number region above 10⁶, the Kármán constant is estimated to be 0.383.

High Reynolds number turbulent boundary layer measurement is performed in the RTRI's Large-scale low-noise wind tunnel in Maibara Japan. The momentum thickness Reynolds number is up to approximately 7×10⁴ . The mean velocity profile shows a clear log-law region with Kármán constant 0.38. The turbulent intensity profile also shows the log-law scaling predicted by attached eddy model. These data will be used as the fundamental data base made in Japan which is also applicable to industrial application including railway technology.

Using an efficient spectral method, we perform direct numerical simulation of turbulent square-duct flow at high Reynolds numbers to investigate the dependence of turbulent secondary flow on the Reynolds number. Similar to previous studies, our simulation also shows that the intensity of the secondary flow can be most likely scaled by the bulk mean velocity. We further investigate the distribution of Reynolds stress terms in the averaged equations of the cross-streamwise velocity, which is the cross-streamwise driving force of the secondary flow to find that the driving force is oriented to the corner and its curl (driving torque) is localized in the corner region as the Reynolds number increases. The distribution of the driving force is observed to scale with the kinematic viscosity and the friction velocity. These results suggest that the secondary flow of the order of the friction velocity is induced near the corners. We also discuss a possibility that the intensity of the whole secondary flow may be scaled by the friction velocity.

Direct numerical simulations (DNSs) are used to analyze a flat-plate turbulent boundary layer with separation and reattachment. In the present study, particular attention is given to unsteady low-frequency motion observed in the pressured induced separation bubble for Reθ=900 where Reθ is the Reynolds number based on the inlet momentum thickness and freestream velocity.

Spatial growth of transient turbulence in rectangular-duct flow for the cross-sectional aspect ratio As = 3, 5, 7 is measured experimentally by using a multi-PIV (Particle Image Velocimetry) system. As an indicator of the spatial growth, we estimate the area and its growth rate along the duct axis for transient turbulence on the duct midplane. The comparison of the growth rates for As = 3, 5, 7 shows that the growth for As = 3 exhibits a different feature from the others. For As = 5, 7 an equilibrium state of turbulence is observed only at a specific value of the Reynolds number Re, while for As = 3 an equilibrium state is observed in a finite range of Re.