arXiv subjects

Hideki Yanaoka

Publications and source records attributed to Hideki Yanaoka.

12 recordsLinked to original sources

arXiv2025

Energy-conserving finite difference scheme for compressible magnetohydrodynamic flow at low Mach numbers using nonconservative Lorentz force

In magnetohydrodynamic (MHD) flows, incompressibility is assumed for low Mach numbers. However, even at low Mach numbers, the Mach number influences flow and magnetic fields. Therefore, it is necessary to develop a method that can stably analyze low Mach number compressible MHD flows without using the incompressible assumption. This study constructs an energy-conserving finite difference method to analyze compressible MHD flows at low Mach numbers with the nonconservative Lorentz force. This analysis method discretizes the Lorentz force so that the transformation between conservative and nonconservative forms holds. This scheme simultaneously relaxes velocity, pressure, density, and internal energy, and stable convergence solutions can be obtained. In this study, we analyze four types of models and verify the accuracy and convergence of this numerical method. In the analyses of two- and three-dimensional ideal periodic inviscid MHD flows, it is clarified that momentum, magnetic flux density, and total energy are conserved discretely. The total energy is conserved even in a nonuniform grid. Even without correction for the magnetic flux density, the divergence-free condition of the magnetic flux density is satisfied discretely. Analysis of a Taylor decaying vortex under a magnetic field clarifies that the present numerical method can be applied to incompressible flows and can accurately predict the trend of energy attenuation. In the Orszag-Tang vortex analysis, an increase in Mach number reduces the magnitude of vorticity and current density. In addition, compression work increases more than expansion work, and the influence of compressibility appears. An increase in Mach number slightly delays the transition to turbulent flow. This numerical method has excellent energy conservation properties and can accurately predict energy conversion.

physics.flu-dyn↗

arXiv2025

Resonance phenomena of thermo-bioconvection generated by chemotactic bacteria under unsteady heat condition

Bioconvection is a phenomenon caused by microorganisms with tactic properties. To effectively utilize bioconvection for industrial purposes, it is necessary to find a way to control it. In this study, we performed a three-dimensional numerical analysis of thermo-bioconvection generated by a suspension of chemotactic bacteria under unsteady heating conditions at the bottom. Under unsteady heating conditions, thermal convection and bioconvection coexist, and unsteady thermo-bioconvection occurs around plumes. When the frequency of the temperature fluctuation is low, thermo-bioconvection follows the temperature fluctuation. However, as the frequency increases, the ability of thermo-bioconvection to follow the temperature fluctuation deteriorates. A resonance phenomenon occurs at the frequency where the instability of the suspension owing to the density difference between the bacteria and water is maintained and where thermo-bioconvection can follow temperature fluctuations. At the resonance frequency, the transport characteristics of bacteria and oxygen throughout the entire region within the suspension improve significantly. As the amplitude of temperature fluctuations and thermal Rayleigh number increase, the interference between thermal convection and bioconvection intensifies, leading to a noticeable improvement in transport characteristics owing to the resonance phenomenon. At this time, the amplitude of temperature fluctuations and thermal Rayleigh number do not almost affect the resonance frequency. This study demonstrated the possibility of thermal control of transport properties in bioconvection.

physics.flu-dyn↗

arXiv2025

A numerical method for low Mach number compressible flows by simultaneous relaxation of dependent variables

Density varies spatiotemporally in low Mach number flows. Hence, incompressibility cannot be assumed, and the density must be accurately solved. Various methods have been proposed to analyze low Mach number flows, but their energy conservation properties have not been investigated in detail. This study proposes a new method for simultaneously relaxing velocity, pressure, density, and internal energy using a conservative finite difference scheme with excellent energy conservation properties to analyze low Mach number flows. In the analysis for sound wave propagation in an inviscid compressible flow, the amplitude amplification ratio and frequency of sound wave obtained by this numerical method agree well with the theoretical values. In the analysis for a three-dimensional periodic inviscid compressible flow, each total amount for the momentum, total energy, and entropy are discretely conserved. When no approximation, such as low Mach number approximations, is applied to the fundamental equations, the excellent conservation properties of momentum, total energy, and entropy are achieved. In decaying compressible isotropic turbulence, this computational method can capture turbulence fluctuations. Analyzing the Taylor-Green decaying vortex, we confirmed the validity of this computational scheme for compressible viscous flows. In the calculation for the natural convection in a cavity, the validity of this numerical method was presented even in incompressible flows considering density variation. In a three-dimensional Taylor decaying vortex problem, it was shown that this numerical method can accurately calculate incompressible flows. We clarified the accuracy and validity of the present numerical method by analyzing various flow models and demonstrated the possibility of applying this method to complex flow fields.

physics.flu-dyn↗

arXiv2024

Numerical method for the magnetic vector potential in incompressible magnetohydrodynamic flows and the conservation properties of magnetic helicity

Analyzing magnetohydrodynamic (MHD) flows requires accurate predictions of the Lorentz force and energy conversion. Total energy, cross-helicity, and magnetic helicity can be used to investigate energy conservation properties in inviscid MHD flows. However, the conservation property of magnetic helicity has not been fully clarified using the magnetic vector potential equation. This study presents a numerical method to simultaneously relax magnetic vector and electric potentials for incompressible MHD flows using a conservative finite difference scheme that discretely conserves total energy. First, it was proven that the transport equations of total energy, cross-helicity, and magnetic helicity can be discretely derived from the equations of momentum, magnetic flux density, and magnetic vector potential, thereby elucidating the conservation properties of these quantities. Subsequently, five models for steady and unsteady problems were analyzed to verify the accuracy and convergence of the proposed numerical method. Additionally, the computational approach involving the magnetic vector and electric potentials was validated. A comparison of the calculated results with exact solutions in the analysis of one- and two-dimensional flow models and Hartmann flow further validated the numerical method. Unsteady analyses of two- and three-dimensional decaying vortices were performed. The ideal periodic inviscid MHD flow exhibited good conservation properties for total energy and cross-helicity. Magnetic helicity was discretely preserved even in three-dimensional flow. Furthermore, in viscous flow, the attenuation trends of total energy, cross-helicity, and magnetic helicity aligned with the exact solution. The numerical method accurately captured the decay trends of energy. Thus, the proposed method can facilitate the investigation of energy conservation and conversion in compressible MHD flows.

physics.flu-dyn↗

arXiv2024

Deformation and breakup of the liquid ligament with various disturbances on the interface in shear flow

This study performed a numerical analysis of the deformation and breakup of a liquid ligament with various disturbances on the interface in shear flow. The shear flow generates a three-dimensional flow and vortices around the liquid ligament. These vortices promote the movement of the liquid inside the liquid ligament. When the velocity difference of shear flow increases, a nonlinear effect becomes strong, and turbulence with higher wavenumber components than the initial disturbance occurs at the interface. This turbulence accelerates the ligament splitting and increases the number of breakup droplets. Then, the droplet diameters become uniform, and the atomization quality improves. As the wavenumber of the disturbance applied to the interface increases, the liquid moving velocity along the central axis of the liquid ligament increases. Furthermore, the breakup time of the liquid ligament becomes short. In addition to the initial reference disturbance with a low wavenumber, when turbulence with twice the wavenumber of the reference disturbance is applied to the interface, the interface deformation and the splitting of the liquid ligament are similar to those with the single reference disturbance. When turbulence with four times the wavenumber is added to the initial disturbance with the low wavenumber at the interface, the deformation of the liquid ligament is accelerated, and the liquid ligament splits faster. Additionally, the number of breakup droplets increases; hence, the total surface area of the liquid increases. The droplet diameter becomes uniform; therefore, the atomization quality of the liquid ligament is improved.

physics.flu-dyn↗

arXiv2023

Numerical simulation for axis switching of pulsating jet issued from rectangular nozzle at low Reynolds number

Axis switching of a jet ejected from a rectangular nozzle affects flow mixing characteristics. To elucidate such a mixing mechanism, the axis switching and vortex structure deformation should be investigated in detail. This study performed a numerical analysis of the axis switching of a pulsating jet ejected from a rectangular nozzle at a low Reynolds number. At all aspect ratios, a rectangular vortex ring similar to the shape of the nozzle cross-section is periodically shed downstream, and the side of the vortex ring deforms into a hairpin shape downstream. A vortex pair is generated inside the vortex ring downstream of the nozzle corner. When the aspect ratio is AR=1.0, the vortex pair consists of symmetrical vortices, while as AR increases, the asymmetry of the vortex pair enlarges. At AR=1.0, regeneration of a vortex ring occurs downstream. For AR=2.0, alternately on the long and short sides of the nozzle, an upstream vortex ring overtakes a downstream vortex ring. Regardless of AR, downstream near the nozzle, as the vortex pair existing inside the vortex ring distorts the vortex ring, the positions of the side and corner of the vortex ring exchange, resulting in a 45-degree axis switching. For AR>1.0, further downstream, the hairpin part of the vortex ring on the long side develops away from the jet center compared to the short side, causing a 90-degree axis switching. As a result, high turbulence occurs over a wide area, strengthening the mixing action. As AR increases, intensive interference between the vortex rings on the upstream and downstream sides diffuses the vortices downstream. Then, as turbulence by the diffused vortices widely occurs, the mixing effect is further strengthened.

physics.flu-dyn↗

arXiv2023

Energy-conserving finite difference scheme based on velocity interpolation applicable to unsteady flows using collocated grids

The collocation method uses the Rhie-Chow scheme to find the cell interface velocity by pressure-weighted interpolation. The accuracy of this interpolation method in unsteady flows has not been fully clarified. This study constructs a finite difference scheme for incompressible fluids using a collocated grid in a general curvilinear coordinate system. The velocity at the cell interface is determined by weighted interpolation based on the pressure difference to prevent pressure oscillations. The Poisson equation for the pressure correction value is solved with the cross-derivative term omitted to improve calculation efficiency. In addition, simultaneous relaxation of velocity and pressure is applied to improve convergence. Even without the cross-derivative term, calculations can be stably performed, and convergent solutions are obtained. In unsteady inviscid flow, the conservation of kinetic energy is excellent even in a non-orthogonal grid, and the calculation result has second-order accuracy to time. In viscous analysis at a high Reynolds number, the error decreases compared with that of the Rhie-Chow interpolation method. The present numerical scheme improves calculation accuracy in unsteady flows. The possibility of applying this computational method to high Reynolds number flows is demonstrated through several analyses.

physics.flu-dyn↗

arXiv2023

Three-dimensional decaying magnetic field belonging to Beltrami flow

This study analysed a three-dimensional Taylor decaying vortex under an applied magnetic field as a benchmark test problem to verify the calculation method of an electromagnetic fluid flow and investigated the validity of the decaying magnetic field model. First, we observed the flow structure of a three-dimensional Taylor decaying vortex without an applied magnetic field. We investigated the changes in the error between the calculation result and the exact solution when the number of grid points and the Reynolds number varied and showed the effectiveness of the benchmark test. Next, we analysed a three-dimensional Taylor decaying vortex under an applied magnetic field and clarified the characteristics of the decaying magnetic field. When a magnetic field is applied, low magnetic pressure regions are connected in a mesh pattern, and the magnetic pressure distribution with a distorted cubic structure occurs to surround a high magnetic pressure region. In a stagnation region, the magnetic energy becomes low, and the magnetic flux line is similar to the streamline of the velocity field. High current densities occur in a grid pattern, and the magnetic flux lines swirl around the high current density region. The magnetic pressure and magnetic energy are high in the high current density region. When the Reynolds number and the magnetic Reynolds number vary, the decay trends of various energies agree well with the exact solution. The transition to turbulent flow occurs at a high Reynolds number, and the kinetic and total energies decrease rapidly. After the dissipation rate of kinetic energy becomes maximum, the vortex structure decays, and the flow field approaches a stationary state without magnetic fields. The three-dimensional Taylor decaying magnetic field belonging to the Beltrami flow is a valuable model for verifying the calculation method of electromagnetic fluid flows.

physics.flu-dyn↗

arXiv2023

Hairpin vortices and heat transfer in the wakes behind two hills with different scales

This study performed a numerical analysis of the hairpin vortex and heat transport generated by the interference of the wakes behind two hills in a laminar boundary layer. In the case of hills with the same scale, the interference between hairpin vortices in the wake is more intensive than in the different-scale hills. When the hills with different scales are installed, hairpin vortices with different scales are periodically shed. Regardless of the scale ratio of the hills, when the hill spacing in the spanwise direction is narrowed, the asymmetry of the hairpin vortex in the wake increases due to the interference between the wakes. At this time, the turbulence caused by the leg and the horn-shaped secondary vortex on the spanwise center side in the hairpin vortex increases, and heat transport around the hairpin vortex becomes active. In addition, the leg approaches the wall surface and removes high-temperature fluid near the wall surface over a wide area, resulting in a high heat transfer coefficient. These tendencies are most remarkable in the same-scale hills. In the case of hills with different scales, the heat transfer coefficient decreases because the leg on the spanwise center side in a small hairpin vortex does not develop downstream.

physics.flu-dyn↗

arXiv2022

Pattern wavelengths and transport characteristics in three-dimensional bioconvection generated by chemotactic bacteria

We conducted a three-dimensional numerical simulation of bioconvection generated by oxygen-reactive chemotactic bacteria. This study investigated the bioconvection patterns, interference between plumes, and the wavelength of bioconvection patterns. In addition, we clarified the transport characteristics of cells and oxygen in the bioconvection. Multiple plumes occur in the suspension and three-dimensional bioconvection is formed around the plumes by the cells with vortex rings arising around the plumes. Even if bioconvection at a high Rayleigh number is disturbed, the bioconvection is strongly stable with respect to disturbances, and the pattern does not change due to disturbances. Bioconvection changes depending on the physical properties of bacteria and oxygen, and, in particular, the rate of oxygen consumption by bacteria significantly affects the strength of bioconvection. Bioconvection patterns with different plume arrangements and shapes are formed for different Rayleigh numbers or initial disturbances of the cell concentration. As a result, the wavelengths of the patterns also vary. As the Rayleigh number increases, interference between plumes is strengthened by the shortening of the pattern wavelength, so the velocities of both upward and downward flows increase. Many cells are located under the plumes and a strong shear flow occurs in these regions. As the pattern wavelength decreases, the cells are affected by high shear stress. Then, the convective transport of the entire suspension strengthens and the transport characteristics of cells and oxygen improve. When the chamber boundary is changed to side walls, bacteria adhere to the wall surface, and plumes are regularly arranged along the side walls.

physics.bio-ph↗

arXiv2022

Interference and heat transfer between hairpin vortices in wakes behind staggered hills

The present study performs a numerical simulation of the interference and heat transfer between hairpin vortices formed in wakes behind staggered hills in a laminar boundary layer. Hairpin vortices are periodically shed in the wake of a row of hills, causing interference between the hairpin vortices. As the spanwise distance between the hills decreases, interference increases and the hairpin vortices become strong. At that time, because the interference between the legs of the hairpin vortex and the Q2 ejection becomes strong, the head of each hairpin vortex rises sharply. When the hill spacing decreases, the turbulence caused by the head and both legs of the hairpin vortex generated from a hill in the second row increases remarkably. In addition, the secondary vortex also generates turbulence. The hairpin vortex and the secondary vortex are attracted to adjacent hairpin vortices, causing widespread high turbulence in the spanwise direction near the wall surface. Regardless of the hill spacing, Q2 ejection and Q4 sweep due to the hairpin vortex occur, and the secondary vortex forms around the hairpin vortex, activating heat transport and increasing the heat transfer coefficient in the wake. When the hill spacing becomes narrower, the interference between the hairpin vortices strengthens the legs of each hairpin vortex and secondary vortex, and heat transport near the wall surface becomes very active. The heat transfer increases over a wide range of the wake because the legs of hairpin vortices flowing downstream are spread in the spanwise direction.

physics.flu-dyn↗

arXiv2022

Large-eddy simulation of turbulent separated and reattached flow in enlarged annular pipe

This study performs a large-eddy simulation of turbulent separated and reattached flow in an enlarged annular pipe. A vortex ring is periodically shed from the sudden expansion part. A longitudinal vortex occurs around the vortex ring, making the flow three-dimensional. As a result, the vortex ring becomes unstable downstream and splits into small vortices. A tubular longitudinal vortex structure occurs downstream of the reattachment point near the wall surface on the inner pipe side. A low-frequency fluctuation occurs at each pipe diameter ratio. The smaller the pipe diameter ratio is, the more downstream the influences of small-scale vortices and low-frequency fluctuation on the flow field appear. The smaller the pipe diameter ratio, the slower the pressure recovery downstream from the reattachment point. The pressure recovery on the inner pipe side is delayed compared to the outer pipe side. Turbulence is maximum upstream of the reattachment point due to the small-scale vortices generated by the collapse of the vortex ring. This maximum value decreases as the pipe diameter ratio decreases. The smaller the pipe diameter ratio, the higher the turbulence downstream from the reattachment point.

physics.flu-dyn↗