The nozzle guide vane, which is a stationary part of a gas turbine, is a critical component of gas turbine engines because it must operate under harsh conditions with high pressure and temperature. Unfortunately, when a gas turbine runs for a long time, the turbine vane is subjected to repeated thermal load. This increases the possibility of fatigue damage and crack failure, thereby reducing the vane material's lifespan. In practice, the risk of failure at the trailing edge (TE) of a turbine vane is very high due to the reasons of shape configuration and cooling performance. The TE damage disturbs the flow physics of compressible air passing the vane TE, resulting in flow phenomena and heat convection. The study aims to numerically investigate the effects of damaged surfaces at the TE of a turbine vane on its flow behavior using computational fluid dynamics (CFD) with the SST k-w turbulence model. To simplify the simulation, the effects of the TE failure are presented by using two basic patterns, i.e., long (continuous) cutback damage, and two-short (discrete) cutback damage. To complete the investigation, a further study on the effects of repaired surfaces is included as well. The numerical results show the effects of damaged and repaired surfaces on flow behavior, particularly the vortex formation and the level of turbulent kinetic energy (TKE) in the TE region. Specifically, the damaged vane surface significantly increases the TKE level in the TE region, particularly the two-short damaged surface, which TKE shoots up to 7000-8000 m²/s². Meanwhile, TKE in the normal and long damaged case is around 1500 and 4000 m²/s². With the restoration of the vane surfaces, it can reduce the TKE level in the TE region. For instance, TKE is uniformly around 1750 m²/s² for the long repaired surface.

Optical fiber liquid film sensors utilizing the reflection intensity at the gas-liquid interface exhibit excellent environmental robustness. However, this technique needs to improve its sensitivity when measuring liquid film thicknesses approaching the fiber core diameter. In this study, we propose a method for accurately measuring liquid film thickness around the fiber core diameter by changing the spatial intensity distribution of laser light emitted from the fiber. We employ stepped-index-type optical fibers (SI-type) instead of the commonly used graded-index-type fibers (GI-type) to alter the spatial distribution and investigate the relationship between reflection intensity and liquid film thickness. The reflection intensity of SI-type optical fibers exhibits a linear response over a range approximately eight times the core radius, significantly enhancing sensitivity for thin films. In addition, we use a theoretical model for the spatial intensity distribution of light emitted from SI-type optical fibers, demonstrating a close match between the predictions and experimental results, and discussing the linear response. Finally, we demonstrate the liquid film thickness measurement of the liquid jet impinging surface.

In this study, the dispersion of droplets and the associated risk of airborne disease transmission are investigated under various seating arrangements and airflow conditions, encompassing both flat and inclined settings. The numerical framework known as “CUBE” integrates a fully compressible Navier–Stokes solver with a Lagrangian droplet dynamics model, thus facilitating large-scale parallel simulations. This framework employs a building-cube method-based meshing to model the Eulerian mesh, with the same number of cells decomposed in each cube unit. It also disaggregates the Lagrangian marker particle data within each cube unit, thus increasing the parallel computing efficiency. The simulation results reveal that both droplet dispersion patterns and infection risk are substantially influenced by the seating layout and direction of the wind. Notably, wind coming from an angle of 90° significantly reduces both localized droplet dispersion and overall infection risk compared with that from an angle of 0°. Therefore, strategic seating configurations effectively mitigate infection risk under various environmental and airflow conditions. This knowledge is vital for designing public domains to curb the transmission of airborne illnesses such as COVID-19.

The present paper focuses on the synchronization behavior of vortex shedding in in-line tube banks with and without acoustic resonance. We have studied the characteristics of acoustic resonance generated by in-line tube banks with a tube pitch ratio of 2.8 in the flow direction. The single high peak in the sound pressure level spectrum was formed at the gap velocity from 5.0 m/s to 48.0 m/s. As the gap velocity increased, the peak frequency increased. The Strouhal number became approximately 0.145. At the gap velocity of 23.0 m/s, the acoustic resonance of the first mode occurred in the transverse direction. The flow patterns and vortex shedding of in-line tube banks at acoustic resonance have been experimentally investigated. The vortex flow field in the tube banks was measured by PIV with and without the acoustic resonance phenomenon. The alternative vortices were shed from each tube in tube banks with and without the acoustic resonance. A pair of vortices has been created at the top and bottom of the jet stream flowing between the gaps of two tubes. On the other hand, the phase-aligned alternative vortices were occasionally shed from each tube in tube banks. No acoustic resonance occurred, the phase-aligned alternative vortices were shed for 11.7 % of the 0.4 s period. When acoustic resonance occurred, the proportion of phase-aligned alternate vortices increased from 11.7 % to 31.0 %. The rate of increase in wall sound pressure is approximately equal to the rate of increase in vortex synchronization. The increase in wall sound pressure due to acoustic resonance is primarily due to the synchronization of vortices in the tube banks during the occurrence of weak resonance phenomena.

Biomechanics is a field for mechanical study of the structure and function of living organisms, which finds many applications in various fields such as engineering, medicine and biology. Research in blood, systems showed that the blood volume is effectively changed by the strength and direction of shear stress on the vessel wall, also inducing changes in the vessel diameter. Most studies about wall shear stress in bends or branches are based on macroscopic single-phase flow, and more studies should be necessary to investigate wall shear stress distribution by multi-phase flow considering blood as a suspension. In this study, we focused on the flow characteristics of particle suspension, and aimed to analyze a flow field with asymmetrical differences in the shear stress on each wall of a two-dimensional (2D) parallel plates flow, i.e., a flow field with an asymmetric velocity profile. A 2D curvilinear channel was used to reproduce this flow field, and the regularized lattice Boltzmann method and the virtual flux method were used as the computational methods. Our results confirmed that the equilibrium positions were either two asymmetric points or a single point, depending on the radius of curvature of the channel and Reynolds number. The radial pressure gradient was found to be a major factor for the presence of a single equilibrium position. Furthermore, when the radius of curvature was decreased, the equilibrium position shifted closer to the inner wall side, and the relative viscosity of the suspension became higher because of the decreasing the distance between the particle and the inner wall.

In the present study, numerical simulation of cavitating flow using Multi-Process model, which is one of the homogeneous cavitation models and consists of multiple transport equations of bubbly flow characteristics defined on the basis of the moment method, is conducted for a centrifugal pump toward more accurate prediction. The experiment is also conducted for the same pump to obtain the hydraulic performance including the axial thrust force characteristics as well as the visual aspects of cavitation for the validation of numerical simulation. It is found that the present simulation can qualitatively reproduce the degradation of pump suction performance as well as the change in axial thrust force under cavitation at the design flow rate; thrust forces acting on the impeller and the balance drum decreases with the head drop. The cavity patterns leading to the head drop are also well reproduced; the cavitation on the suction surface extends into the blade passage. At a partial flow rate close to the design one, the simulated cavity patterns with the inception and moderate states of cavitation agree with the observation, but they are not well reproduced perhaps due to the insufficient mesh resolution at the deep partial flow rate where the most of cavitation occurs in the vortical structure, typically in the tip separation vortex.

The interaction between bubbles and tissues (bone and fat) in ultrasound and the mechanism of drug transport accompanied by bubble collapse were investigated numerically by using the ghost fluid methods with zonal grids. A parameter defined by the ratio of t₀ to t_s which is proportional to the ratio of bubble radius to wavelength, where t₀ is the bubble collapse time and t_s is the period of an incident pressure wave, was used to characterize the relationship between the incident pressure waveform and the bubble collapse. It was found that the bubble collapse near a tissue surface was classified into three types by considering the penetration of liquid jet and bubble blobs into the tissues. The transport process of matter (imaginary drug) distributed around the bubble surface was calculated with the bubble collapse. The results showed that the mechanism of matter transport due to bubble collapse was summarized as follows. First, both the shock wave by the liquid jet impact on the bubble surface and the rebound shock wave lead to the depression of the tissue wall, and the matter is transported with the vortex flow due to liquid jet formation. Then, the bubble penetrates into the depression, increasing the amount of matter transported into the tissue. The bubble collapse and the matter transport in the bone wall case were compared with those in the fat wall case. Although the maximum wall pressure was higher in the bone case, the amount of matter transported inside the tissue was higher in the fat case. This was caused by the difference in the acoustic impedance between both tissues: the large acoustic impedance of bone induced the reflection of the incident wave, leading to the superposition of the incident wave and the reflection wave that contributed to the acceleration of bubble collapse. Since the stiffer property of bone suppressed the development of the vortex flow and the bubble growth inside the tissue, a larger circulation existed inside the fat, which contributed to the advection of the matter, resulting in a higher concentration of matter inside the fat than inside the bone.

For simulating the behavior of air bubbles injected into the turbulent boundary layer developed beneath the bottom of a ship, we propose a novel experimental system for a different driving force on flow from the bulk pressure gradient. In the proposed facility, the motion of a horizontal belt on the water surface as a moving wall is driven by a motor, and a turbulent boundary layer is developed beneath the belt. To show the usability of this belt-driven system for research on the bubble behavior in turbulent boundary layer, we investigated the bubble clustering beneath the moving belt. Bubble clusters are formed for sufficiently high values of the belt speed and the air injection rate. To quantify this bubble clustering, we calculated the frequency spectrum of the local projected void fraction to estimate the passing frequency of the bubble clusters at two fixed downstream positions and found that it has a peak in the frequency range which is identical to the void wave reported by previous towing model-ship experiments. In addition, our system has the advantage that the wall drag modification can be directly estimated from the torque measurement. The bubble clustering occurs in a parameter regime where the torque reduction ratio is relatively high, which is also consistent with the results of the previous towing experiments.

As a new perspective on the influences of the electrically charged air and electric field on the air flow, the natural air flow on a negatively charged forward-inclined step is analyzed. The electric field formed by the charged step pushes away the negative charge and attracts the positive charge in natural air, so that only the positive charge density is analyzed. A two-dimensional model of flow and electric field is set and solved on COMSOL Multiphysics FEM solver coupling three physics, that is, flow, electrostatics, and charge-transport. The pressure power spectra at the point around the vortex shedding region are calculated in the case of neutral air / uncharged step and charged air / charged step combinations. Whereas the former has large power below 1Hz, the latter has small power below 1Hz but with a larger peak at 30Hz, that is the vortex shedding frequency. In order to analyze the cause of the difference, the electric force effects are evaluated. The result shows that the upper portion of the vortex has forces in the same direction to the mainstream, and it degrades to the lower portion, which accelerates the vorticity of the vortex. Also, the vortex region has a strong downward force. Adding that, the curl of the electric force on the charged air is calculated to show that the upstream half of the vortex has the same direction curl as the vortex rotation, and the downstream half has an opposite curl because of the gradient charge density. These forces strengthen the vortices and raise their negative pressures and are estimated to suppress the fluctuation of the pressures so that the pressure’s spread spectrum reduces. In addition, the results of an experiment using a desktop wind tunnel are given as extra information.

The growth and collapse of a laser-induced bubble near an agarose gel were observed with a high-speed video camera in which 0.7% (Young's modulus E = 13 kPa), 1.5% (E = 55 kPa), 3.0% (E = 200 kPa), and 5.0% (E = 570 kPa) agarose gels were used as tissue-mimicking phantoms. The effects of Young's modulus of agarose and the dimensionless bubble-boundary distance γ on the dynamics of laser-induced bubbles, i.e., bubble oscillation time, bubble migration, bubble shape, jetting, and penetration of the liquid jet into the elastic wall were investigated. It was shown that as Young’s modulus of the wall increased, the amount of migration of the bubble centroid toward the wall increased. The results also showed that as γ decreased, the bubble shape in the late collapse stage changed from a spherical shape, a cone shape, and to a mushroom shape when E = 13, 55, and 200 kPa, while, when E = 570 kPa, it changed from a cone shape, a mushroom shape, and to a volcano shape: the threshold values of γ, where the bubble shape changed from a cone shape to a mushroom shape and from a mushroom shape to a volcanic shape, increased as E increased. These bubble shapes in the late collapse stage were determined from the bubble shapes at the maximum expansion. In the case of mushroom-shaped bubbles, the jet velocity was higher than volcano-shaped bubbles. The value of γ at the onset of the penetration of the liquid jet into the elastic wall increased with an increase in E. The penetration area of bubbles into the elastic wall took the maximum value when the bubble shape in the late collapse stage changed from a mushroom shape to a volcano shape. The difference between the bubble behavior near the agarose wall and that near the PAA wall was also discussed.

Flow driven by co-rotating disks, such as hard disk drives (HDD) have been studied in previous research using axisymmetric models. In our work, a unique non-axisymmetric 3.5" transparent HDD model for information storage was used to investigate the effect of the arm insertion angle on these flow-fields. After introducing the shroud opening region in our model, we focused and illustrated the nature of flow-fields at a parameter α , defined as the arm insertion angle. Comparison was made between α = 20° closer to the shroud opening and a deeper angle of α = 44° closer to the hub. We revealed the flow pattern at the disk-to-disk middle plane (mid-plane) as well as between the arm and the upper disk boundary layer (arm-upper disk plane) across the disk region from the edge of the hub to the edge of the disk, using both flow visualisation and particle image velocimetry (PIV). The visualisation results revealed an accelerated flow around the hub downstream of the arm insertion, with a higher magnitude at α = 44° compared to α = 20°. We observed a rigid body rotation around the hub for both insertion angles.

The objective of this study is to reveal the frequency characteristics caused by Von Karman excitation observed at the stationary field of a Francis turbine. The multiple equidistant frequency (frequency band) of the integer multiple of rotation frequency spacing peaks occur when sound pressure caused by Von Karman excitation is measured at the stationary field. The principles of the generation of multiple equidistant frequencies observed at the stationary field caused by Von Karman excitation have not been clearly revealed. This study revealed the cause of the principles by theoretical analysis. The interesting phenomenon cannot be explained by the blade vibration frequency caused by Von Karman excitation alone, so attention was focused on periodic fluctuations in the relative position between the sound source and observer accompanied by runner rotation i.e. Doppler effect. The equation derived with the Doppler effect implies that the frequency caused by a single vibration by Von Karman excitation is observed by an observer located at the stationary field as frequency band with the span of the rotation frequency. This study yielded the following four findings. Frequency modulation of the integer multiple of rotation frequency occurs at the stationary field. Multiple blades vibrate at the same frequency. The runner is excited as a whole not as individual blade. Since the radius of the excitation location in the runner can be identified, the location to be modified to suppress Von Karman excitation can be estimated.

We aimed to measure the two-dimensional wall shear stress using oil film interferometry (OFI) for the flow around the suction surface of a two-dimensional airfoil. In a previous study, we measured the direction of wall shear stress by applying particle image velocimetry analysis method to Fizeau fringes images. In this study, we developed and compared oil application methods and OFI analysis methods to obtain the wall shear stress magnitude around the reattachment point. When we used the existing oil application methods, the fast Fourier transform (FFT)-based method could not perform the measurement due to the low quality of the Fizeau fringes around the reattachment point. However, narrowing the oil application width improved the quality of the Fizeau fringes. Next, we used an image-analysis-based algorithm to calculate the magnitude of wall shear stress. The old and new analysis methods showed qualitatively similar results for x/c = 0.08 to 0.9. We found that the image-analysis-based algorithm outperformed the FFT-based algorithm in terms of the measurement success rate. Finally, the local skin friction coefficient distribution could be measured using an image-analysis-based method and images of both the new and the old oil application methods around the reattachment point. However, the local skin friction coefficient around the reattachment point was higher for the 1 mm wide oil application than for the 15 mm wide oil application. Therefore, we consider that accounting for the upstream oil spread is important for OFI measurements around the reattachment point of a two-dimensional airfoil.

Heat-driven thermoacoustic coolers (HDTACs) are promising technologies for reusing waste heat. Although numerous studies have been conducted on HDTACs, they have predominantly focused on small units at the laboratory scale, with limited work on larger units for factory installations. This study explored a loop-type HDTAC with multiple thermoacoustic cores connected in series with the cold side of the cooler adjacent to the ambient side of the engine using linear thermoacoustic theory. For a large 1000-core diameter HDTAC designed for factory installations and a small 40-core diameter HDTAC designed for laboratory use, the cold heat exchanger temperature (T_C), number of engine and cooler core pairs (n = 3-5), and flow path diameter of the regenerator were varied. The engine hot heat exchanger temperature (T_H) relative to the cooler cold heat exchanger temperature and the Carnot-specific efficiency coefficient of performance (COP/COP_Carnot) of the entire device are calculated, then design parameters of the device to achieve low temperature operation and high efficiency were indicated. The results show that, for small HDTACs, both low-T_H operation and high-COP/COP_Carnot were achieved at n = 5 (e.g., the maximum value of COP/COP_Carnot for T_C = −50°C was 0.31 at T_H = 248°C). For large HDTACs, the maximum COP/COP_Carnot was achieved with n = 3 (e.g., the maximum value of COP/COP_Carnot for T_C = −50°C was 0.49 at T_H = 178°C).

In order to adopt convolutional neural networks (CNN) for practical use in estimating the source of scalar dispersion in turbulent flows, such as gas leaks in industrial plants, the inference accuracy was verified using scalar concentration distributions in various turbulent environmental conditions. Training and test data were obtained through quasi direct numerical simulations on a flow system with a scalar-source-attached cylinder downstream of a turbulent grid, at two Schmidt numbers. An inference accuracy of at least 90% was confirmed under trained flow conditions, provided that the observation window was large enough to capture the integral scale in scalar fluctuations in turbulence. When the flow conditions (in terms of the turbulence intensity and the Schmidt number) differed between the training and testing phases, the accuracy was significantly reduced. However, the learner supervised with images under all different conditions showed high generalization performance. Singular value decomposition was used to discuss the features essential for training and prediction. In our CNN inference, we concluded that the source estimation was achieved by extracting from the integral-scale scalar patch distribution the features related to each of the turbulent and molecular diffusion progressions and their correlations. The results provide insights into the potential solutions for accurately predicting the sources of material dispersion in turbulent environments.

In recent years, to predict the performance and the aerodynamic sound of box fans with high accuracy, numerical analysis has been necessary for the transitional boundary layer on the blades. In this study, two types of box fans with a baseline fan and a high-load fan, where the boundary layer that develops on the suction surface of the rotor blade transitions to turbulent flow, were targeted for large eddy simulation (LES) analysis and acoustic analysis using different grid resolutions to predict the far-field sound. The box fan in this study has an impeller of 180 mm outer diameter and five blades, and the Reynolds number based on the diameter and the tip velocity is 3.38 × 10⁵. The decoupled method with LES for incompressible flow and computational aeroacoustics (CAA) based on a dipole sound source was performed. The computational grid resolutions had 25 million grid points (Case 1), 52–56 million grid points (Case 2 and Case 4), and 420–450 million grid points (Case 3 and Case 5). The performance of the fans and the sound pressure level (SPL) were compared with each case and experiment. The performance was in good agreement with the experiment in higher grid resolution cases. Boundary layer transition on the blade surface was predicted as the main source of aerodynamic sound generated from the box fan. Increasing the grid resolution improved the prediction accuracy of the SPL in the frequency range corresponding to the scale of the turbulent eddies captured by the grid resolution.