The cooling holes on gas turbine blades act as stress concentration regions, which become the starting points for cracks in thermal barrier coatings (TBCs) during applying the cyclic mechanical and thermal stresses. To evaluate the damage behavior of the cooling holes in TBC system, a three-dimensional finite element model possessing a cooling hole was established. The inelastic constitutive equation developed by our research group for the TBC’s top layer (TC) was employed to calculate the damage in TC layer. The effects of displacement load, thermal load, and their phase differences were investigated. The simulation results indicated that damage was concentrated on the TC layer around the cooling hole due to stress concentration. Furthermore, out-of-phase loading resulted in significantly higher damage to the TC layer compared to in-phase loading. Additionally, as the displacement load increased, the location of maximum damage shifted from near the TC/substrate interface to the surface of the TC layer. Moreover, the region of damage concentration around the cooling hole varied as the displacement load increased. This study provides new insights into the damage behavior of cooling holes under the interaction of mechanical and thermal stresses, facilitating the service life evaluation and working condition optimization of the gas turbine.

Resin materials are widely used in industrial fields as adhesives, coatings, etc. When used as adhesives or coatings, there is an adhesive interface between different materials, it is necessary to understand the adhesion mechanism to ensure their reliability. Because there are still many unclear points in the adhesion mechanism., it is necessary to evaluate the bonding strength by quantitative test methods to understand the adhesion mechanism. However, it isn’t easy to evaluate the bonding strength using conventional methods due to the influence of the specimen size and shape. In this study, the test method to evaluate the bonding strength using a small column resin formed on the surface of the adherend was proposed, and the effect of column diameter and shape on bonding strength obtained in this method was discussed. The main conclusions are summarized as follows. (1) The bonding strength decreased as the column diameter decreased by increasing the deformation of the resin column at the contact portion to the shear tool. (2) The bonding strength of the cut-out columnar specimen increased due to reduced deformation of the resin column. (3) The bonding strength of the quadrangular prism specimen is independent of the bonding area.

Resins and light metals are being used to reduce the weight of transportation equipment such as automobiles and airplanes. In the case of light metals, magnesium alloys are attracting attention because they have the lowest density among practical metals. It is about one-third lighter than aluminum. Further, the specific strength and specific rigidity of magnesium are higher than those of aluminum and resin. However, compared to other practical metals, the wear resistance and corrosion resistance of magnesium alloys are quite low. In the present study, surface modification of magnesium alloy was performed by shot lining. Shot lining is a method of joining dissimilar materials using shot peening. Dissimilar laminate materials with excellent wear resistance and corrosion resistance were joined together, and the improvement effect was investigated. Test materials were corrosion-resistant metal foils such as pure titanium and pure nickel, hard alumina powder, and thermoplastic resin sheet. Shot peening was performed using a centrifugal machine using cast steel blasting materials. The bondability of dissimilar laminated sheet was evaluated by bending tests. Corrosion resistance test was evaluated by immersion in supersaturated aqueous sodium chloride solution. In addition, the wear resistance was evaluated by the amount of wear using an abrasion tester. It was found that the surface properties of magnesium alloys were improved by joining dissimilar laminated sheet.

In this study, fatigue properties were investigated using joints made by friction stir welding to two types of rolled aluminum alloys, AA6111-T4 and AA5052-H34. Microstructural observation showed that the crystal grains in the joint were finer than those in the base metal. The results of Vickers hardness tests showed a decrease in hardness around the joint, which was attributed to annealing caused by frictional heat during joining. Static test results showed that the tensile strength of each joint was much lower than the tensile strength of the base metal, confirming that failure occurred from the upper plate joint. In fatigue tests, a difference in fatigue life of each joint was observed below a certain test force amplitude, and a higher fatigue life was observed in the joint with AA6111 as the upper plate. The fatigue crack propagated from the tip of the slit toward the upper plate, reached the top surface of the joint, and ruptured the joint.

In a previous study, we have clarified that addition of aluminum to gray cast iron and heat treatment to from an oxide layer on the surface improved the erosion resistance of the iron to aluminum alloy melt. The purpose of this study is to clarify the mechanism of the improvement of the erosion resistance. Erosion resistance were evaluated by immersing the test piece in AC4A aluminum alloy melt and stirred. The erosion time of ductile cast iron was shorter than that of the gray cast iron. The shear strength of the oxides layer on the cast iron specimens was measured using a tensile tester. The shear strength of oxide layer of ductile cast iron with 3mass% Al was smaller than that of the gray cast iron with 3mass% Al, suggesting that the shear strength of the oxide layer has an influence on the soaked time.

Multi-materialization is a process of combining structural materials with different properties to create components and products with superior overall properties, and is attracting particular attention in the lightweighting of automobiles and aircraft. The joining of dissimilar metals is an example of multi-materialization. However, joined dissimilar metals form oxide films and intermetallic compounds in the boundary layer, which often reduce the strength of the joint. In this study, a combination of aluminum-plated steel and aluminum alloy were chosen to contact the similar metals with each other and they were joined by forging. The interfacial joint strength was measured by tensile shear tests, and the results were discussed based on the morphology and elemental analysis on the fracture surface.

Processing-induced defects affect the performance and reliability of multilayered ceramics. They are introduced from the raw powder, or during tape casting, drying, thermo-compression, binder burnout, and sintering. Using synchrotron X-ray multiscale tomography, we investigated how the microstructural heterogeneities and various defects evolve in the sintering of tape-cast alumina laminate. The strength-limiting defects were identified as large defects at the layer interface and peculiar crack-like defects formed around large inclusions contained in the raw powder. Although a slurry might be uniform in colloidal processing, the particle packing in the tape was heterogeneous. This heterogeneity determined the spatial distribution of pores and formed complex pores in later stages. Moreover, we found the formation of flake-like sheets in slurry due to the self-assembly of polyhedral alumina crystals. This phenomenon has the potential to be applied in the manufacture of 2D micro-sheets controlled to the thickness of a single particle.

This study focuses on the freeze-casting method, which introduces anisotropy into ceramic porous microstructures, and evaluates the impact of different cooling temperatures on the microstructural and mechanical properties. The microstructural observations indicate that the degree of microstructural anisotropy varies with different cooling temperatures. The degree of anisotropy was quantified using the tortuosity factor, which was derived from the three-dimensional microstructural images reconstructed using deep learning technology. Moreover, the yield stress was measured through compression testing. The results reveal that the introduction of anisotropy significantly enhances the compressive strength of porous materials. The findings of this study indicate that there are optimal freezing temperatures that balance ion conductivity and compressive strength.

In predicting the performance of solid oxide fuel cells, the technology to generate the three-dimensional microstructure of porous electrodes of importance. However, obtaining this microstructure poses a challenge due to the considerable amount of time required. In this study, we constructed a conditional generative adversarial network (CGAN) that generates two-dimensional porous images with different porosities using two-dimensional real-microstructure data as training data. Furthermore, our findings indicate that by combining this methodology with existing deep learning technologies, it is possible to generate three-dimensional microstructures from two-dimensional cross-sectional images with minimal computational cost. Additionally, the validity of the constructed network was evaluated through principal component analysis on the generated images.

The fabrication of the electrode in solid oxide fuel cells (SOFCs) typically involves a high-temperature sintering process. It is well-known that the microstructures resulting from the sintering process are correlated with the power generation performance and durability of SOFCs. Moreover, the microstructures prior to sintering may also have a considerable influence on the sintered microstructures and their related properties. Nevertheless, the precise impact of the heterogeneity of the initial ceramic powder on the fundamental properties of the sintered materials is not currently understood. In this study, initial LSC (lanthanum strontium cobaltite) powder, which is a promising cathode material, was prepared with and without initial grinding by a planetary ball mill. Subsequently, the initial microstructures were reconstructed using the focused ion beam-scanning electron microscope (FIB-SEM) technique. To investigate the impact of initial powder heterogeneity on the microstructural and mechanical properties, the mesoscale simulation techniques, including Kinetic Monte Carlo (KMC) simulations, voxel-based finite element analysis, and peridynamic approach, were performed on the obtained microstructural images. The results of our numerical analysis demonstrate that the variation in microcracking stress is significantly affected by the heterogeneity of the initial powder. Consequently, the use of a heterogeneous initial powder results in a higher probability of microcracking.

Generally, thin aluminum sheet fabrication involves multiple processes and is expensive. In this paper, we investigate the conditions under which thin aluminum alloy A6061 sheets can be fabricated by the twin roll casting method, which can directly fabricate thin sheets from molten metal. The effects of roll speed on sheet continuity, surface properties, and thickness were investigated, and the characteristics of twin-roll casting were evaluated by observing the microstructure of the sheets and conducting Vickers hardness tests. The experimental apparatus was newly designed and fabricated to study continuous casting of thin sheets under low-speed and high-pressure conditions. The roll material is CCM-A, a copper alloy. Experiments were conducted under three conditions of roll peripheral speeds of 3, 5, and 7 m/min, and thin sheets could be produced under all conditions. The surface of the fabricated sheets consisted of metallic luster and cloudy areas, and fine solidification cracks were observed in some places.

This paper reports on the vertical twin roll casting of aluminum alloy A7075. Aluminum alloy A7075 belongs to the highest strength alloys and its sheets are produced by rolling and extrusion. Although A7075 is used in the aerospace industry and for special applications, it has a high susceptibility to solidification cracking and is difficult to obtain sheets easily. Twin-roll casting can produce thin sheets directly from molten metal, making it possible to produce sheets other than conventional sheet alloys. In this study, experiments were conducted using a miniaturized apparatus to study the continuous casting of A7075 thin plates under low-speed and high-pressure conditions suitable for continuous casting of A7075. Because of the smaller size and narrower roll width of this equipment, an increase in load per unit width is expected. The effects of roll speed on sheet continuity, surface properties, thickness, microstructure, and Vickers hardness were investigated. Experiments were conducted under five conditions of roll peripheral speeds of 1, 3, 5, 7, and 10 m/min. Thin sheet fabrication was possible under all conditions except 1 m/min in this experiment, and thin sheet fabrication of aluminum alloy A7075 by vertical twin roll casting was feasible.

Additive manufacturing (AM) technology is a forming method that does not require molds and can create complex shapes as a single unit. Fused deposition modeling (FDM) is a most popular AM technology because the molding equipment is inexpensive. It has been reported that the mechanical properties of objects formed by FDM depend on the layering configuration. In order to estimate the mechanical properties of specimens formed by FDM using lamination theory, the effect of lamination configuration on the mechanical properties was investigated in this study. As a result, the lamination configurations in which the printing direction was at an angle to the loading direction, such as [90]₂₂, [±45]₁₁, and [0/60/120]₇, were significantly deformed due to the rotation of the filaments within the specimen. The tensile strength of [0/90]₁₁ and [±45]₁₁ was higher than that of [0]₂₂. Therefore, the mechanical properties may be increased by aligning the printing direction with the loading direction and by having layers printed in different directions.

Susceptibility of stress corrosion cracking (SCC) on Al-Zn-Mg alloy using friction stir method was evaluated. Alternating salt-water immersion test was conducted using specimen with base metal and stirred material. As a result of immersion test, cracks occurred on specimen with base metal after 5.5 days immersion. On the other hand, no cracks occurred on specimen with stirred material. It was concluded stirred region of Al-Zn-Mg alloy shows low susceptibility against SCC.

The behavior of microbubbles is controlled using an electric field. First, an aqueous solution containing ionic surfactants (cationic and anionic) was prepared. Using an SPG membrane (pore size 2.0 μm), microbubbles of uniform size were generated in this aqueous solution. This resulted in the generation of microbubbles with a charge on the surface. After that, an attempt was made to move the microbubbles using an electric field in a small water tank. The small tank is equipped with electrodes. Negatively charged microbubbles moved to the anode side, and positively charged microbubbles moved to the cathode side. As the concentration of the anionic surfactant increased, the movement speed toward the anode increased. When the critical micelle concentration was reached, the movement speed became constant. The speed also increased with an increase in the concentration of the cationic surfactant, and when the concentration exceeded the critical micelle concentration, the speed became constant.

Numerical flow analysis was conducted using fluid dynamics software to investigate the discharge of water from a vertical hole. The analysis captured the discharge of water in the vertical hole as the air jet from the drill tip pushed up the water, and the discharge of water mixed with the air by vacuum suction. It was also confirmed that the amount of residual water over time varied depending on the amount of air jet and vacuum suction. Numerical calculations were performed by changing the clearance of the drill shaft and comparing the results, and it was found that the effect of the clearance on the amount of residual water was small.

The purpose of this study is to control the jet diffusion by using the deforming nozzle. The shape of polypropylene nozzle can change from square shape into the cruciform shape or the octagonal one smoothly. The velocity measurement of the jet flow was conducted using X-type hot wire sensors and the constant temperature anemometer. The basic characteristics of the jet flow, such as the velocity distribution, the half value width and the turbulent intensities were obtained. Moreover, to catch the flow feature of the jet near the nozzle the flow visualization was carried out by using the Laser Light Sheet and the high speed video camera. As a result, it has been found that the diffusion of the jet issuing from the deforming nozzle is promoted, comparing with the case of the square nozzle jet.

We have studied an impinging jet interfering with a Couette flow by using the numerical simulation, then found that the jet angle toward the Couette main flow affects the heat transport mechanism and the periodical vortex structure. However, this periodicity has been captured by just visual check or basic frequency analysis, thus the qualitative and quantitative discussion is still not enough. In this research, we try applying the phase analysis on the time series of flow velocity with the aim of a more essential understanding of this flow field dynamics. To be specific, we apply Takens’s theorem and the recurrence plot to analyze the attractor or periodic pattern of the main flow direction velocity fluctuation. As a result, we found that there is no attractor when the jet flows opposite to Couette main flow, while there is some limit-cycle or strange attractor when the jet flows same direction as the main flow. Especially, the unique and clear strange attractor is reconstructed only when the jet flows orthogonally to the main flow.

A flow field clustering technique was developed under the assumption that all the obtained flow fields were generated by generators expressed by Gaussian mixture distributions. Each generator was considered to have informative features of the flow fields. This technique was applied to a flip-flop flow around diamond-shaped cylinders arranged in a staggered formation, obtained experimentally by two-dimensional PIV (particle image velocimetry) at the Reynolds number of 2000. A temporal ordering of the generators indicated that the flow in this study was periodic. The flow fields of the generators captured fluctuations around wakes from the diamond-shaped cylinders. It was found that the flip-flop flow was maintained both by the mixing between the main flow and the wake on one side of the wake, and a locally enlarged main flow on the other side of the wake.

Flow visualization observation was carried out for the aspect of the vortex flow when oscillating a cylinder with large amplitude low oscillation frequency comparatively supposing a pulsating flow using the closed circuit water channel by the tracer pouring streak method. The pulsating flow was produced by oscillating the cylinder in the direction of the flow by making main flow velocity into 0.0654 m/s. The used cylinders are two kinds of circular cylinders, a square cylinder and a bluff body, and those projection-width are 10mm and 16mm, 16.97mm and 15mm, respectively. Oscillating half amplitude is 40mm and oscillation frequency was gradually varied in 0 to 0.16Hz. As a result, it was shown by the existence of movement of the separating point that the aspects of the vortex shedding in one cycle of oscillation differ. The cylinder position when the vortex shedding is observed was investigated. When the velocity ratio became large, it became clear that the numbers of vortices at the time of acceleration and a slowdown differed.

Turbulence phenomena in two-dimensional pipes were investigated using the Navier-Stokes equations of motion, to which a term was added that quickly converged divergence of flow field to zero. To prevent naturally occurring turbulence from diverging the solution, the viscosity of the cell was increased where the values of the product of the Reynolds stress term included in the equations of motion and the square of the vorticity of turbulent velocity was large. This enabled the calculation of the pipe friction coefficient and the mean velocity distribution that matched the experimental values.

A large eddy simulation (LES) for fully-developed turbulent flows through several isosceles-trapezoid ducts using the immersed boundary (IB) method is carried out with three Reynolds numbers defined by the bulk streamwise-velocity and duct height in this study. The present numerical results show that there are differences between low and high Reynolds number cases. In the case of the large difference between top and bottom wall-length at high Reynolds number, the flow near the top wall is stabilized while that near the bottom one is unstable due to the weak secondary flow near the wall in comparison to the low-Reynolds-number case. Moreover, in the case of the short top wall at low Reynolds number, the number of the circulation flows decreases because of the disappearance of the vortices near the top wall.

This paper presents a reduced-order model (ROM) based on proper orthogonal decomposition (POD) for a dataset containing flow around a cylinder at various Reynolds numbers to represent the Reynolds number dependence. Even when the dataset contains the numerous flow fields with different Reynolds numbers, the ROM using the POD modes cannot accurately predict the flow field. To address this problem, a novel ROM framework is proposed that does not lack accuracy when using datasets containing enough flow fields. Our ROM obtains a set of POD modes from a single Reynolds number flow field in the dataset. To predict the flow field by the ROM, we select only two of these sets of POD modes with Reynolds numbers close to the predicted conditions and perform POD for those modes in the second stage. Our ROM accurately describes the Reynolds number dependence and is significantly faster than the conventional methods, even accounting for the computational time to perform a second stage POD.

To prevent efficiency loss due to collision of water flow with the main shaft, atmospheric air is naturally sucked in through a hole in the casing of a cross-flow turbine using the negative pressure in the runner chamber. In this study, performance changes due to the flow-rate of intake air were investigated under two different effective head conditions. The values of the effective suction head NPSHa were adjusted so that the volume change of air in the runner would be the same for both head. The following results were made clear. Even if without suction air, a small gas phase area is formed in the runner due to air released from the water. If Q_air/Qw is greater than 0.04, the performance such as efficiency based on runner height and runner chamber pressure were almost same even if the air volume increases further.

This study investigates the effect of floater pitch angle on the wake characteristics of upwind and downwind wind turbines. To that end, wind tunnel experiments are performed with model wind turbine fitted on bases which are used to simulate floater pitch angles. Results are compared for static pitch angles of 0° and 10°. The mean wake profiles shift upward for upwind turbine and downward for downwind turbine, when the pitch angle is 10°. Since the rotor projected area-with respect to the incoming wind- decreases with pitch angle, wind speed in the wake is also higher when the floater pitch is larger. However, for the pitch angle considered in this study, the floater pitch has no significant effect on the turbulence profiles of the wake.

A regenerative cooling is one of the key technologies to realize a reusable hypersonic airbreathing engine. In the case of a liquid hydrocarbon fuel, its cooling capability can be improved by thermal decomposition due to endothermic reaction. Also, the coolant (fuel) should keep supercritical state to avoid phase change, because the gaseous coolant covers the inner wall surface of cooling channel, which leads reduction in heat transfer efficiency. For effective application of the endothermic thermal decomposition to the regenerative cooler, well-validated simulation model is required, which can estimate the heat absorption performance with sufficient accuracy. In the present study, the thermal cracking test was carried out, and after the numerical simulation was performed regarding the experimental results as the validation data. The 100 % n-dodecane was adopted as the surrogate for the aviation fuel of Jet A-1. The experiment employed the Fuel-Heating Test Apparatus (FHTA) which made it possible to thermally decompose the n-dodecane under supercritical state. The reacting CFD simulation was implemented, which consisted of Reynolds-Averaged Navier-Stokes (RANS) equations with previously developed chemical reaction mechanism called JetSurf_comp model. This simulation adopted the supercritical value of physical property for n-dodecane referred from the REFPROP database. The comparison between experimental and simulation results showed that the simulation overestimated the mole fractions of 1-butene and 1-pentene products and underestimated the conversion rate. In addition, the cross-sectional distribution of fluid temperature and residual time indicated that the thermal decomposition almost occurred at vicinity of the inner wall so that the temperature nearby inner wall mainly determined the cracking composition ratio.

Due to growing concerns about environmental issues such as global warming, liquefied natural gas and liquefied hydrogen are being recognized as promising clean energy sources for efficient transportation and storage. While the demand for liquefied gases is expected to increase in the future, facilities for cryogenic fluid flow testing are scarce internationally. This study aims to develop a cryogenic fluid flow test facility using liquid nitrogen, based on a weighing system and differential pressure control. As a result, at a mass flow rate of 14.2 kg/min, the difference in mass flow rate calculated upstream and downstream was 0.08 %.

Mixing and diffusion phenomena between the jet and the surrounding fluid are very important in connection with improving the combustion efficiency of liquid fuels and promoting various chemical reactions. In this study, we focus on the coaxial circular jet, which is the basic flow of the air-fuel mixture system in a combustor. The coaxial circular jet consists of an inner circular jet and an outer annular jet, and it is believed that the diffusion and mixing of the jets can be actively promoted, controlled, and optimized by changing their velocity ratios. In this study, we measured the concentration field near the exit of a coaxial double-circular nozzle in liquid using an absorbance-based concentration measurement system in order to optimize the mixing and diffusion of the coaxial circular jet. Correlations between the circular and annular jets and the ambient fluid were also calculated.

The complexity or randomness of the relaminarization process of an accelerated turbulent boundary layer was analyzed by use of Kolmogorov complexity. Streamwise fluctuating velocity was measured at two points separated in the spanwise direction. The velocity was measured by two single hot-wire probes simultaneously. Velocity data was compressed using a compression program running on a Windows PC. From the compressed data, the approximated Kolmogorov complexity, named AK, and normalized compression distance, NCD was obtained. AK of two-point velocity difference increased as the spatial separation increased. The similarity of velocity between the two points affected this result. AK of two-point velocity difference decreased downstream due to the relaminarization. NCD between the two-points velocity was large when both velocity data are within turbulent flow because of the dissimilarity between both turbulent data. The behavior of NCD between two-point velocity difference corresponded to the velocity difference itself. From the NCD also, the relaminarization was confirmed.

The laminar-turbulent transition of a mixing layer induced by oscillating flat plates at an exit of a two-dimensional nozzle was experimentally investigated at Re = 3000. The mixing layer was formed between the jet which issued from the nozzle and the surrounding quiescent fluid. The plates oscillated vertically in relation to the mean flow. The oscillation frequency, 5 Hz, was two orders of magnitude smaller than the fundamental frequency of the velocity fluctuation in the natural transition process. Three types of plate oscillation were tested: stationary, symmetrical and anti-symmetrical oscillation states. The amplitude of the plates was 1.5 mm and 3.0 mm. Various statistics, such as fluctuating velocity and that with plate oscillating frequency in the streamwise component were measured by hot-wire anemometers. The role of the oscillation modes and amplitudes on the turbulent transition was also examined by smoke visualization. The turbulent transition in the mixing layer was promoted in the order of anti-symmetrical oscillating, symmetrical oscillating, and stationary states. The contribution of fluctuating velocity promoted extensively the anti-symmetrical oscillation but locally the symmetrical oscillation. The periodical oscillation in the mixing layer prevailed more extensively in the anti-symmetrical than in the symmetrical oscillation modes.

The operation range of centrifugal pumps is severely limited by unsteady phenomena such as diffuser rotating stall (DRS) at non-design points, which has become a problem in the industrial world. In a previous study, an attempt was made to suppress DRS by slitting the diffuser vanes to expand the operating range of the pump, but it was found that there is a trade-off between the effect of DRS suppression and pump efficiency. In this study, holes were drilled near the front edge of the diffuser vane to onset the jet flow to develop a new method to suppress DRS with minimal loss of efficiency. The effect of increasing the number of holes drilled in a stepwise manner on DRS suppression was experimentally investigated. It was found that the DRS suppression effectiveness increased as the number of holes was increased.

In this study, the airflow direction control performance of three piezoelectric fans in a staggered and parallel arrangement was experimentally investigated by airflow smoke visualization and cross-sectional air velocity distribution measurement experiments with a hot-wire anemometer. As a result of comparing the airflow direction control capability of fan units with the central fan-driven in opposite phase and those driven synchronously, it was found that the phase difference between the fans did not contribute to airflow direction control and airflow increase. On the other hand, even with the synchronously driven piezoelectric fan unit, its performance as a blower is greatly decreased in a narrow housing. Still, the airflow direction control provided by the staggered fan could be regained by installing an intake port in the top panel. However, the flow characteristics also show that the air mass blown by the fan travels along the ceiling of the narrow channel.

The pectoral fin plays the role of turning, surfacing, and descending in a fish-type underwater research vehicle. This study considered the possibility of applying a wing with LEP that imitates the pectoral fin of a humpback whale, which has good maneuverability, to its pectoral fin. We clarified the effect of wing with LEP on wing performance during flapping motion through lift measurements. In the case of flapping motion, time variation of lift force was found to be greatly affected by the lift caused by the pitching motion regardless of leading edge shape. It was also found that the time-averaged lift force is larger than that in the case of heaving or pitching motion. In the case of baseline with flapping motion, the amplitude of the lift fluctuation was found to be the largest when the phase difference PD was PD=0° and the smallest when PD=180°. In the case of flapping wing with LEP, the lift fluctuation was found to be smaller compared with the baseline case. These results indicate that the wing with LEP is likely to be suitable for a fish-type underwater survey machine.

Flow conditions near the casing tongue have a significant impact on the performance and stable operation of the centrifugal pump, so the internal flow conditions have been measured by the experiment. In the present paper, the internal flow near the casing tongue is clarified by PIV measurement results.

Diffuser Rotating Stall (DRS), a fluid instability in diffuser-type centrifugal pumps, often limits the operating range, prompting the industry to seek solutions. However, a DRS suppression method that does not reduce efficiency has not been implemented in practice. The authors have successfully suppressed DRS while maintaining efficiency by applying a non-axisymmetric diffuser that combines long and short vanes. However, an issue arose: unstable head flow characteristics at high flow rates. In this study, two new types of diffusers were introduced: an axisymmetric diffuser with equal numbers of short and long vanes, and an axisymmetric diffuser consisting solely of short vanes. Performance tests were conducted, and DRS occurrence was confirmed by measuring static pressure fluctuations at the inlet of the vaned diffuser. Performance tests showed that increasing the number of short vanes in the non-axisymmetric diffuser shifts the generated flow rate of the unstable head characteristic toward the high flow rate region. The original efficiency was maintained in all short vane configuration models, and the DRS was suppressed in some models. An analysis based on the stagger angle, focusing on vane shape, revealed a different trend from previous reports, indicating that DRS cannot be predicted based on the stagger angle.

Industrial turbopumps for pumping liquefied natural gas must ensure stable performance over a wide flow range from the standpoint of operating conditions and maintenance. When a low-solidity fan-type inducer (auxiliary impeller) is attached to the pump to improve its suction performance，instability phenomena caused by cavitation occur at lower flow rates than the design point. A typical example is cavitation surge，a type of self-excited oscillation caused by intermittent volume changes of cavitation. In this study，blade end plates were attached to the inducer for suppressing this instability phenomenon to investigate the suppression effect of pressure oscillation and to visualize the cavitation behavior. Numerical analysis was also conducted to analyze the internal flow of the pump. The results showed with the addition of blade end plates to the fan-type inducer，the pulsation phenomena caused by cavitation could be controlled up to lower flow rates compared to the basic design.

The purpose of this study is to compare predictions based on conservation law models with experimental results and to control the boundary layer transition by changing the bluntness of the leading edge of the flat plate. In the experiment, the transition was measured from thermocouple temperature measurements in a high-speed wind tunnel at Mach 2. The bluntness promoted a delay in the transition. As a result, the transition positions were backward to a smaller range than predicted by Newtonian flow theory. In addition, the larger bluntness, the larger transition Reynolds number.

The flow in the bronchial tube was experimentally investigated using a human bronchial tube model. Resin bronchus model that is made by 3D printer is provided for the experiment. Microbubbles were used as tracers to visualize oscillatory flow. Measurement areas were around the first branch of the bronchial model. Velocity distribution is measured by PIV. The measurement results were phase averaged in synchronization with the oscillatory flow. Above the first branch in the inspiratory phase, the peak of velocity was observed near the center of the channel. The area of slow speed stream was observed on the left side of the channel. In the right bronchus underside of the first branch, the peak of velocity was observed on the left side of the channel. This is because the high speed stream observed above the first branch flows into the left side of channel. In the left bronchus underside of the first branch, the peak of velocity was not observed as pronounced as velocity distribution in the right bronchus. This is because the slow speed stream observed above the first branch flows into the channel. Above the first branch in the expiratory phase, the peak of velocity was not observed as pronounced as the velocity distribution in the inspiratory phase. This is presumably because the flow from the right bronchus and the flow from the left bronchus, which have the almost the same velocity, flow into the channel separately. In the right bronchus underside of the first branch, the peak of velocity was observed on the left side of the channel. In the left bronchus underside of the first branch, the peak of velocity was not observed as pronounced as the velocity distribution in the right bronchus.

Turbulence subjected to system rotation provides significant insights into real-world flow phenomena. Therefore, it is indispensable to better understand the effects of rotation on turbulence. In order to investigate the effect of rotation on turbulence diffusion under the other effects being eliminated, our previous study performed a set of DNS of oscillating grid turbulence in systems at rest and subjected to rotation. Also, two mode analysis methods, Proper Orthogonal Decomposition (POD) and Dynamic Mode Decomposition (DMD) methods, were applied to the velocity data to extract dominant flow structures. However, these mode analyses provided the insight into flow structures in a qualitative manner. Therefore, for the sake of the quantitative evaluations, this study applies the idea of the statistical analysis based on turbulence kinetic energy to the POD and DMD modes. The results show that the statistics well express the characteristics of the flow structure, which had previously been evaluated qualitatively, in terms of turbulence production, diffusion, and dissipation.

Cold spray (CS) is one of the thermal spraying methods which makes coatings by spraying solid particles on substrate surfaces. Solid particles are sprayed onto a substrate at high speed to form a coating accelerated by a supersonic gas flow through a convergent-divergent nozzle. Therefore, it is important to understand the gas flow conditions inside the nozzle in which the gas flow is accelerated. However, commercial CS nozzles are fabricated from cemented carbide to prevent abrasions of the nozzle, making it impractical to drill pressure taps to diagnose the flow. Therefore, the authors have proposed a simple method for estimate the flow conditions using the outer surface temperature of the CS nozzle and the temperature recovery factor of the internal flow. To verify the validity of the method, the Mach number was compared with that calculated from the wall static pressure. As a result, the Mach number shows reasonable agreement, however, they differ out of the allowable limit in the downstream region of the nozzle. In the previous study, this was attributed to an increase in wall surface temperature due to heat transfer from atmosphere. Numerical simulation implies another possible cause of the disagreement by Ansys Fluent. Using the SST k-ω model, Mach number decreased during the process and then became almost constant.

Small thrusters installed on fishing vessels are often operated at high rotational speed near the draft surface, and are therefore prone to cavitation. Therefore, it is important to clarify the occurrence of cavitation and its effect on thrust performance at each-rotational speed, but there have been few studies focusing on these issues. In this study, we conducted experiments using a high-speed camera to visualize cavitation and measure thrust force, and cavitation analysis on a small thruster with a propeller diameter of 190[mm]. This paper reports on the thrust performance and cavitation flow in the high cavitation number region (σ_n=6.3 to 14.2).

In recent years, environmental issues have led to the need to improve the aerodynamic performance of vehicles. The objective of this study is to analyze the effects of changes in vehicle configuration and engine cooling airflow on aerodynamic characteristics. The vehicle shape was changed from box-shaped to fastback and notchback, with the rear bottom flip-up, and the intake height for the engine cooling air was also changed. Totally 270 configurations were investigated through wind tunnel experiments. As a result, the known results were obtained: drag rapidly increases and lift also increases when the rear window angle is around 20° to 30° for side window angles of 0° to 15°. However, for side window angle of 30°, the tendency for drag to rapidly increase is no longer observed. When the air intake was enlarged, drag and lift increased while maintaining the above-mentioned tendency. When the diffuser angle was changed, drag and lift were minimized at 15°. However, when the engine room was not included, the known result was obtained: drag is minimized at a diffuser angle of less than 5°. It was shown by CFD that changes in aerodynamic characteristics are linked to changes in the flow structure around the vehicle body.

The effects of various velocity fluctuations of upper flow on turbulent structure of atmospheric boundary layer were experimentally investigated. The atmospheric boundary layer was reproduced in a wind tunnel. In order to produce various velocity fluctuations in the upper flow, three kinds of turbulence grids were set to the entrance of the test section. The streamwise and spanwise velocities were simultaneously measured by using hot-wires anemometer. Turbulence statistics were estimated. The vertical profiles of Reynolds shear stress, Ruw and Ruv for M25 and M50 decrease as z/L increase, on the other hands, those for M75 increase. The ratio of the integral length scale, _x,g/_x,a, for M75 were higher than those for M25, M50 at z/L=0.67, on the other hands, _y,g/_y,a, _z,g/_z,a, for M75 were lower than those for M25, M50 at 0.50≦z/L≦0.67. The joint probability density functions of u and w for M75 show the strong negative correlation in the upper region of the atmospheric boundary layer. It is suggested that the effects on the turbulent structure of the atmospheric boundary layer depends on the scale of velocity fluctuations of the upper flow.

Gas-liquid agitation by turbine type impellers having shrouds structurally (closed impellers) was studied by visual observation of the flow behavior of gas-liquid mixture in the impeller region. Impellers were used in designs of the 6 and 12 flat blades of standard width (6F and 12F) and the 6 flat blades of tapered width (6T). The way in formation of the gas cavities and dispersion as the gas bubbles through the closed impellers differed from that through the open impellers without the shrouds. In comparison of the flow behavior among the impeller designs, the 6T closed impeller was suggested to be operable for more effective gas-liquid dispersion.

It was known that in the process of breakup of a vortex structure, the transition to turbulent spots rapidly progresses triggered by the generation of vortices in the width direction. However, since multiple hairpin vortices exist even just before the breakup, it is not possible to confirm whether the vortices generated in the width direction are the direct cause of the breakup. In addition, it is interesting to observe whether reconnection occurs of vortices when multiple hairpin vortices are in the vicinity. Therefore, to simplify this phenomenon, we artificially generated two hairpin vortices close to each other in the laminar boundary layer and investigated behaviors that occur between these vortices pair. From the results obtained, it was found that even a hairpin vortex that should disappear on its own develops into a turbulent spot when two hairpin vortices are in close proximity and proceed downstream. Furthermore, it was confirmed that the heads of the two hairpin vortices merge when they were close together within twice the width of the hairpin vortex.

It has known that hairpin vortices have suitable spacing in the flow direction. Therefore, it may be possible to artificially manipulate the hairpin vortex to narrow the spacing in the flow direction, to delay the transition to turbulence and to reduce turbulent frictional drag. The purpose of this study was to investigate the differences in the downstream development of the hairpin vortex by deforming it uniformly in the width direction by an obstruction placed on the wall surface. The results obtained show that the boundary layer is also unstable for obstacles with a height same to the displacement thickness of the boundary layer and that the development is slightly weakened for obstacles with a height 10 times the viscous length. It was found that in order to deform only the hairpin vortex, the height must be limited to less than the exclusion thickness and the aspect ratio of the obstacle must be considered.

This study investigates the oscillation phenomenon in a locally heated cylindrical flow channel simulating low-temperature ozone CVD with an impinging jet stream. The results show that when the heated substrate is placed in the downward direction (+g configuration), a periodic oscillating flow with a period of about 2.7 seconds is observed between the gas inlet and the heated substrate under specific conditions (d_H = 50 mm, Q = 6 SLM, and Q = 9 SLM). On the contrary, when the heated substrate was placed face up (-g configuration), no oscillations were observed, resulting in a stable flow. We used the experimental setup to visualize the flow pattern and analyze the periodic nature of the flow in different configurations. The results highlight the importance of substrate orientation in determining flow stability in CVD processes, with upward facing configurations promoting more stable flow conditions. This study demonstrates how the interaction of natural and forced convection can lead to complex flow dynamics depending on the orientation of the heated substrate.

The driving force of the cylinder blade wind turbine is the steady lift force generated by the longitudinal vortex flowing out of the gap between the cylinder blade and the downstream ring-plate. This study aims to clarify the effect of shape factors on the steady lift force, focusing on the ring-plate width W and the center diameter of the ring-plate D. The effect of D on the steady lift force is small when the cylinder blade diameter d and W are sufficiently small. However, as W increases, the steady lift force decreases when D is large and the ring-plate close to the cylinder blade tip, and when D is small and the ring-plate close to the axis of rotation. By comparing the steady lift force in the range where they are approximately constant, the lift force increases in proportion to the increase in W, the rate of increase being greater for larger d.

Centrifugal pumps designed for small-flow operation, used in various applications such as medical, food processing, chemical and manufacturing, are known to have lower efficiency than general industrial pumps. To overcome this problem, a compact double-rotating single-axle screw pump has been developed based on the principle of the single-axle screw pump. In this pump, the stator rotates passively as the rotor rotates. Therefore, two structural gaps exist between the casing and the stator and between the stator and the rotor. Previous studies have shown that pump performance can be improved by sealing grease between the casing and stator. In this study, with the aim of investigating the seal persistence of grease sealed in the cylindrical gap between the casing and the stator, the seal persistence of grease was investigated when the pump was operated for a long time with the pump pressure fixed at an arbitrary pressure as the initial condition and when the pump pressure was continuously changed from low to high pressure and repeatedly loaded.

We numerically investigated the changes in the collection efficiency of a cyclone separator depending on the hopper shape, that is, the particle collection section. We performed LES (large-eddy simulation), using a two-way model that considers the interaction between solid particles and gas. We tested various hopper shapes, including cylindrical and conical shapes. While there were almost no changes in the flow and particle behavior inside the cyclone separator, the flow within the hopper differed, leading to a bias in the particle collection locations. In terms of the collection efficiency, the conical hopper was slightly inferior compared to the others. We will discuss the behavior and causes of particles that were not collected in the hopper and flowed out from the cyclone separator.