詳細

橋本　知久 （ハシモト　トモヒサ）

  • 理工学部 機械工学科　講師
Last Updated :2024/04/25

コミュニケーション情報 byコメンテータガイド

  • コメント

    空気や水などの流れのコンピュータシミュレーションに関する研究を行っています。プラスチック射出成形の設計に必要となる金型キャビティ内の樹脂の流れおよび金型の温度分布を解析している。

研究者情報

学位

  • 博士（工学）

ホームページURL

J-Global ID

現在の研究分野（キーワード）

    空気や水などの流れのコンピュータシミュレーションに関する研究を行っています。プラスチック射出成形の設計に必要となる金型キャビティ内の樹脂の流れおよび金型の温度分布を解析している。

研究分野

  • ものづくり技術（機械・電気電子・化学工学） / 流体工学

研究活動情報

論文

  • Artificial compressibility method using bulk viscosity term for an unsteady incompressible flow simulation
    T. Yasuda; I. Tanno; T. Hashimoto; K. Morinishi; N. Satofuka
    Computers and Fluids 258 105885 2023年03月 [査読有り]
  • Optimized finite difference method with artificial dissipation for under-resolved unsteady incompressible flow computations using kinetically reduced local Navier-Stokes equations
    T. Hashimoto; I. Tanno; T. Yasuda; Y. Tanaka; K. Morinishi; N. Satofuka
    Computers and Fluids 184 21 - 28 2019年04月 [査読有り][招待有り]
  • T. Hashimoto; T. Yasuda; I. Tanno; Y. Tanaka; K. Morinishi; N. Satofuka
    Computers and Fluids 167 215 - 220 2018年05月 [査読有り][招待有り]
     
    Numerical simulations of 2D doubly periodic shear layers and 3D decaying homogeneous isotropic turbulence are presented using Kinetically Reduced Local Navier–Stokes (KRLNS) equations that is applicable to the unsteady incompressible flows without the need for sub-iterations and is capable of capturing the correct transient behavior. To achieve high accuracy, the KRLNS equations is discretized with higher order central difference approximations and 4-stage Runge–Kutta method. The results are compared with the solutions obtained by Lattice Boltzmann method (LBM) and pseudo-spectral method (PSM), which is the standard approach for this problem. Parallel computations are carried out on multiple GPUs (Tesla K40), maximum 4 GPUs available, based on the domain decomposition method and the speedup of the KRLNS equations is investigated. It is found that all three methods can capture the transient flow fields of unsteady incompressible flow and a large speedup for the KRLNS equations is obtained.
  • T. Yasuda; T. Hashimoto; I. Tanno; Y. Tanaka; H. Minagawa; K. Morinishi; N. Satofuka
    INTERNATIONAL JOURNAL OF COMPUTATIONAL FLUID DYNAMICS 31 4-5 258 - 268 2017年 [査読有り][招待有り]
     
    Various collision and velocity models of the lattice Boltzmann model (LBM) were compared to determine their effects on the efficiency of a three-dimensional homogeneous isotropic decaying turbulent flow simulation. We determined that a decrease in the number of velocities, in particular, 13-velocities, which can be used in the quasi-equilibrium lattice Boltzmann and in the multiple-relaxation time models (MRT), could considerably decrease the computational effort. However, decreasing the number of velocities deteriorates the stability and the accuracy of the results. By comparing the collision models, we also determined that the stability of the entropic lattice Boltzmann model (ELBM), and 19- and 27- velocity MRT is much higher than in other models. However, the numerical viscosity introduced by the ELBM underestimates the enstrophy, and the computational effort increases because of the calculation overhead required to solve the additional equations if special care is not given to the calculation.
  • T. Hashimoto; I. Tanno; T. Yasuda; Y. Tanaka; K. Morinishi; N. Satofuka
    COMPUTERS & FLUIDS 110 108 - 113 2015年03月 [査読有り][招待有り]
     
    Higher order approach of Kinetically Reduced Local Navier Stokes (KRLNS) equations are applied for two-dimensional (2-D) simulations of Womersley problem and doubly periodic shear layers in order to demonstrate the accuracy, efficiency and the capability to capture the correct transient behavior for unsteady incompressible viscous flows. The numerical results obtained by the KRLNS equations using higher order difference approximations are in excellent agreement with those obtained by the Lattice Boltzmann method (LBM) and the pseudo-spectral method (PSM), which is a standard approach to incompressible viscous flows. It is confirmed that the KRLNS method can capture the correct transient behavior without use of sub-iterations due to a smoothing effect introduced by using the Grand potential in the continuity equation, and keep the fluctuation of velocity divergence at small level by taking sufficiently low Mach number. Parallel computations are carried out on a GPU based on NVIDIA Tesla C1060 system and the provided CUDA library. High values of speedup are obtained for three methods, the KRLNS equations, PSM and LBM. (C) 2014 Elsevier Ltd. All rights reserved.
  • Nobuyuki Satofuka; Koji Morinishi; Itaru Tanno; Tomohisa Hashimoto; Takahiro Yasuda; Yoshihiro Tanaka
    Computational Methods in Applied Sciences 34 211 - 223 2014年 [査読有り][招待有り]
     
    Numerical simulations of 2D decaying homogeneous isotropic turbulence are presented using three local computational approaches, ACM, KRLNS equations and LBM, in order to investigate accuracy, efficiency and the capability to capture the correct transient behavior of the incompressible flows and the results are compared with the solutions obtained by the Pseudo-Spectral Method (PSM), which is the standard method for this problem. Parallel computations are carried out on a GPU by using the CUDA library provided by the NVIDIA and the speedup is investigated. It is found that all three methods can capture the transient flow fields and LBM is the most accurate and efficient approach for this kind of unsteady incompressible flow.
  • I. Tanno; T. Hashimoto; T. Yasuda; Y. Tanaka; K. Morinishi; N. Satofuka
    COMPUTERS & FLUIDS 88 822 - 825 2013年12月 [査読有り][招待有り]
     
    The authors implemented artificial compressibility method (ACM) with virtual flux method (VFM) on a GPU to simulate incompressible unsteady state flow fields around a cylinder, and lattice Boltzmann method (LBM) on a GPU to simulate incompressible turbulent flows. In this study, unsteady flow fields around a circular cylinder were calculated by LBM with VFM. The history of the length of twin vortex behind a circular cylinder and the Strouhal number were compared with that of the other researchers; also, the speedup obtained by a GPU is shown. (C) 2013 Elsevier Ltd. All rights reserved.
  • Tomohisa Hashimoto; Itaru Tanno; Yoshihiro Tanaka; Koji Morinishi; Nobuyuki Satofuka
    COMPUTERS & FLUIDS 88 715 - 718 2013年12月 [査読有り][招待有り]
     
    Kinetically Reduced Local Navier-Stokes (KRLNS) equations are applied for two-dimensional (2-D) simulation of doubly periodic shear layers in order to demonstrate the accuracy, efficiency and the capability to capture the correct transient behavior. The numerical results obtained by the KRLNS equations are compared with those obtained by the artificial compressibility method (ACM) and the lattice Boltzmann method (LBM). The velocity divergence as a function of time in the KRLNS method is compared with that of the ACM. Parallel computations are carried out on a GPU by using the CUDA library provided by the NVIDIA. (C) 2013 Elsevier Ltd. All rights reserved.
  • Takahiro Yasduda; Tomohisa Hashimoto; Hisato Minagawa; Koji Morinishi; Nobuyuki Satofuka
    Procedia Engineering 61 173 - 178 2013年 [査読有り][招待有り]
     
    Lattice Boltzmann Model is suitable for parallel computation. But it is known that a solution using LBGK model becomes unstable in the case using coarse mesh in high Reynolds number region. In order to overcome this shortcoming, Entropic Lattice Boltzmann model and Quasi-equilibrium Lattice Boltzmann model have been developed. however, comparison between the results obtained by these three method for turbulence flow has not been done yet. In this study, we applied LBGK, ELBM and QELBM to 2-dimensional, incompressible, homogenous, isotropic turbulence and compare the stability, accuracy and computational effort. As the result, enhancement of stability is confirmed by using ELBM and QELBM. Especially, It was found in the case of simulation using coarse grid, ELBM is more stable, and in the case of simulation at high Ma region, QELBM is more stable. © 2013 The Authors.
  • I. Tanno; T. Hashimoto; T. Yasuda; Y. Tanaka; K. Morinishi; N. Satofuka
    Computers and Fluids 80 1 453 - 458 2013年 [査読有り][招待有り]
     
    In this study, the lattice Boltzmann method, pseudospectral method, and artificial compressibility method were implemented on both CPU and GPU machines. Homogeneous isotropic turbulent flows were calculated using these three methods with the C language and CUDA library. The computational results show that the flow field obtained by the lattice Boltzmann method was almost the same as that obtained by the pseudospectral method. Among these three methods, the computational time of the lattice Boltzmann method on a GPU was the shortest of all calculations. Thus, the lattice Boltzmann method was well accelerated by GPUs. These results proved that the lattice Boltzmann method on a GPU has advantages of accuracy and computational speed. © 2012 Elsevier Ltd.
  • T. Hashimoto; I. Tanno; Y. Tanaka; K. Morinishi; N. Satofuka
    Procedia Engineering 61 63 - 69 2013年 [査読有り][招待有り]
     
    Kinetically Reduced Local Navier-Stokes (KRLNS) equations are applied for two-dimensional (2-D) simulations of Womersley problem and doubly periodic shear layers in order to demonstrate the accuracy, efficiency and the capability to capture the correct transient behavior. The numerical results obtained by the KRLNS equations are compared with those obtained by the pseudo-spectral method (PSM), which is a standard approach to incompressible viscous flows for low Mach number, as well as Lattice Boltzmann Method (LBM). It is confirmed that the KRLNS method can capture the correct transient behavior without use of sub-iterations due to a smoothing effect introduced by using the Grand potential in the continuity equation, and keep the fluctuation of velocity divergence at smaller level by giving an appropriately low Mach number. Parallel computation is carried out on a GPU based on NVIDIA Tesla CI060 system and the provided CUDA library. © 2013 The Authors.
  • Tomohisa Hashimoto; Keiichi Saito; Koji Morinishi; Nobuyuki Satofuka
    Lecture Notes in Computational Science and Engineering 74 145 - 152 2011年 [査読有り][招待有り]
     
    In flip-chip packaging technology, the underfill encapsulation is one of the important processes to obtain a significant improvement in fatigue lifetime for the solder joints between IC chip and substrate. The advanced design of electronic devices aiming at the enhancement of the performance involves the increase of the number of solder bumps, smaller size of the IC chip and smaller gap height between IC chip and substrate. That leads to various problems caused by the flow behavior, such as voids in underfill and mis-placed IC chip. The numerical analysis is more and more strongly required for simulating the underfill flow behavior, including the condition of dispensing the underfill material on the substrate. In fact, it is desirable to predict the filling time, the final fillet shape formed around IC chip and the occurrence of air trap especially around the solder bump in the underfill process, considering the effect of contact angle, viscosity and surface tension of the underfill material for increasing the reliability of flip-chip packaging. © 2010 Springer.
  • Tomohisa Hashimoto; Koji Morinishi; Nobuyuki Satofuka
    COMPUTERS & FLUIDS 38 7 1361 - 1368 2009年08月 [査読有り][招待有り]
     
    This paper describes numerical simulations for the shape and its drag of an elastic body deforming under the fluid dynamic force. The simulations were carried out by coupling the Navier-Stokes equations and the equations of motion of the elastic body. The equations of motion are formulated for an elastic shell model which is composed of material particles connected with elastic springs and dampers. The relation between deforming elastic body shape in response to the fluid dynamic force and its drag force was investigated under the constraint of constant volume and fixed center of gravity of the elastic body for incompressible and compressible supersonic flows. In these simulations, an initial shape of the elastic body is a circular cylinder and starts deformation under the fluid dynamic force. (C) 2008 Elsevier Ltd. All rights reserved.
  • Tomohisa Hashimoto; Koji Morinishi; Nobuyuki Satofuka
    COMPUTERS & FLUIDS 38 7 1361 - 1368 2009年08月 [査読有り][招待有り]
     
    This paper describes numerical simulations for the shape and its drag of an elastic body deforming under the fluid dynamic force. The simulations were carried out by coupling the Navier-Stokes equations and the equations of motion of the elastic body. The equations of motion are formulated for an elastic shell model which is composed of material particles connected with elastic springs and dampers. The relation between deforming elastic body shape in response to the fluid dynamic force and its drag force was investigated under the constraint of constant volume and fixed center of gravity of the elastic body for incompressible and compressible supersonic flows. In these simulations, an initial shape of the elastic body is a circular cylinder and starts deformation under the fluid dynamic force. (C) 2008 Elsevier Ltd. All rights reserved.
  • Tomohisa Hashimoto; Koji Morinishi; Nobuyuki Satofuka
    Computational Fluid Dynamics 2006 - Proceedings of the Fourth International Conference on Computational Fluid Dynamics, ICCFD 2006 855 - 860 2009年 [査読有り][招待有り]
  • Tomohisa Hashimoto; Keiichi Saito; Koji Morinishi; Nobuyuki Satofuka
    Computational Fluid Dynamics 2008 573 - 578 2009年01月 [査読有り][招待有り]
  • Tomohisa Hashimoto; Tanifuji Shin-Ichiro; Koji Morinishi; Nobuyuki Satofuka
    COMPUTERS & FLUIDS 37 5 520 - 523 2008年06月 [査読有り][招待有り]
     
    We developed a numerical method for simulating the underfill flow of conventional capillary flow and no-flow types in flip-chip packaging. The analytical models for the two types of underfill encapsulation processes are proposed. In the capillary flow type, the underfill material is driven into the cavity with solder bump by the surface tension with an effect of contact angle as the capillary action. In the no-flow type, the movement of IC chip during the reflow attachment is controlled by an appropriate loading to get the proper interconnect between IC chip and substrate. In both types, the flow behavior and filling time of underfill material in underfilling encapsulation process are investigated, taking the fluid dynamic force acting on the solder bump into account. It is found that the proposed analytical models have a considerable potential for predicting the underfill flow. (C) 2007 Elsevier Ltd. All rights reserved.
  • 橋本 知久; 森西晃嗣; 里深信行
    日本機械学会論文集(B編) 72 713 9 - 16 日本機械学会 2006年01月 [査読有り][招待有り]
     
    一様流中で弾性体が体積一定で重心が固定された拘束条件下で流体力を受けて変形するように、新たに弾性体モデルの運動方程式に人為的な強制力を付加して数値計算を行い、その形状変化が抵抗や流れに及ぼす影響について調べた。交互に渦の放出を伴う非定常流れでは渦の周期に合わせて弾性体後方の形状が振動することがわかった。
  • Tomohisa Hashimoto; Koji Morinishi; Nobuyuki Satofuka
    COMPUTATIONAL FLUID DYNAMICS 2004, PROCEEDINGS 375 - + 2006年 [査読有り][招待有り]
  • Fluid-largely Deformable Elastic Body Interaction Problem Using Overset Grid
    T. Hashimoto; K. Morinishi; N. Satofuka
    CFD Journal 13 2 181 - 190 2004年07月 [査読有り][招待有り]
  • 橋本 知久; 森西晃嗣; 里深信行
    日本機械学会論文集(B編) 70 698 2499 - 2506 日本機械学会 2004年 [査読有り][招待有り]
     
    弾性体モデルの運動方程式と非圧縮性NS方程式を連立させて数値計算を行い、初期に円形で有った物体が一様流中で変形しながら吹き流され物体速度が一様流速度に達した時点で形状が初期の円形に再帰する過程を再現することができた。
  • T Hashimoto; K Morinishi; N Satofuka
    PARALLEL COMPUTATIONAL FLUID DYNAMICS: NEW FRONTIERS AND MULTI-DISCIPLINARY APPLICATIONS, PROCEEDINGS 403 - 410 2003年 [査読有り][招待有り]
     
    A parallel computational algorithm for overset grid using the data strucure of one-dimensional listed type arrangement [1] has been developed for analyzing the flow field of complicated geometries. The parallel code is extended to dynamic problems of fluid-structure interaction. The procedure of parallel computation for moving overset grid according to the motions of structure is proposed and its performance is investigated for simulations of vortex-induced vibration of a circular cylinder. The present numerical results are compared with experimental data and other numerical results.
  • 橋本 知久; 森西 晃嗣; 里深 信行
    日本機械学会論文集 B編 67 663 2649 - 2654 The Japan Society of Mechanical Engineers 2001年 [査読有り][招待有り]
     
    The purpose of the study is to investigate practical advantages, a potential for low computer time and fast convergence as well as accuracy to a steady state solution, of a lower upper symmetric Gauss-Seidel implicit relaxation method using overset grid technique and develop an efficient parallel computational algorithm with a data structure by one dimensional array. This method is applied to viscous turbulent flows around multi element airfoils. Almost linear performance of the speedup ratio in parallel computing is attained up to 16PEs on the Hitachi SR2201 computer. Comparison of the numerical result with available experimental data shows good agreement.

講演・口頭発表等

  • Unsteady incompressible flow simulations by an optimized centered finite difference scheme on Multi-GPU  [通常講演]
    橋本 知久; 安田 孝宏; 丹野 格; 田中 嘉宏; 森西 晃嗣; 里深 信行
    並列計算流体力学国際会議 2018年05月 口頭発表（一般）
  • 最適化高次精度中心差分近似法を適用した非圧縮性流れの数値計算  [通常講演]
    橋本 知久; 丹野 格; 安田 孝宏; 田中 嘉宏; 森西 晃嗣; 里深 信行
    第31回数値流体力学シンポジウム 2017年12月 口頭発表（一般）
  • EDACによる物体まわりの流れ場の計算  [通常講演]
    丹野 格; 橋本 知久; 安田 孝宏; 田中 嘉宏; 森西 晃嗣; 里深 信行
    第31回数値流体力学シンポジウム 2017年12月 口頭発表（一般）
  • Multi-GPU parallel computation of unsteady incompressible flows using kinetically reduced local Navier-Stokes equations  [通常講演]
    橋本 知久; 安田 孝宏; 丹野 格; 田中 嘉宏; 森西 晃嗣; 里深 信行
    並列計算流体力学国際会議 2017年05月 口頭発表（一般）
  • Comparison of artificial compressibility method with and without subiteration for unsteady flow  [通常講演]
    Itaru Tanno; Tomohisa Hashimoto; Takahiro Yasuda; Yoshihiro Tanaka; Koji Morinishi; Nobuyuki Satofuka
    ICCFD9 2016年07月 口頭発表（一般）
  • 山本 恵司; 橋本 知久; 国宗 範彰; 谷川 吉貞; 本間 竹哉; 高見 聡
    年次大会 : Mechanical Engineering Congress, Japan 2015年09月 
    Currently, in the plastic injection molding industry, the manufacturing of a wide variety of products in small quantities increases. High quality of products is strongly needed, reducing the cost and being shorter delivery time. However, many small-to-medium sized companies have difficulties in investing the expansion of equipment, and making the production planning according to the order status in an appropriate manner due to suffering from the shortage of workers. In this research, we addressed the improvement of the work efficiency, and are constructing the production management system that is adequate to the production planning in the plastic injection molding company. The dynamic scheduling system is constituted by the combination of a production scheduler available with free and production management software packages, to solve the problem that initial introduction costs are too much, although maintenance costs are necessary. Our approach is considered to be a promising case of the improvement of the work efficiency in a small-to-medium sized company.
  • 橋本 知久; 丹野 格; 安田 孝宏; 田中 嘉宏; 森西 晃嗣; 里深 信行
    年次大会 : Mechanical Engineering Congress, Japan 2013年09月 
    CAE is an important tool for designing the mold in injection molding and has been used extensively. The filling, packing and cooling stages in injection molding, as well as the warpage after ejection can be simulated. The purpose of this study is to develop a numerical method for solving the incompressible Navier-Stokes equations and the heat conduction equation simultaneously on a Cartesian grid. For computing unsteady incompressible viscous flows, we confirmed that kinetically reduced local Navier-Stokes (KRLNS) equations is a promising method in terms of accuracy, efficiency and the capability to capture the correct transient behavior without sub-iterations. In this paper, the numerical simulations of unsteady heat conduction were carried out by using virtual flux method (VFM), which can treat arbitrary geometries on a Cartesian grid. The two types of boundary conditions were specified. The numerical solutions were compared with those obtained theoretically. The good agreement was obtained.
  • 直交座標系格子を用いた任意形状に対する非定常熱伝導解析  [通常講演]
    橋本知久
    2013年度機械学会年次大会 2013年09月
  • 安田 孝宏; 南川 久人; 橋本 知久; 丹野 格; 田中 嘉宏; 森西 晃嗣; 里深 信行
    流体工学部門講演会講演論文集 2012年11月 
    Quasi-equilibrium Lattice Boltzmann model (QELBM) has been developed in order to get stable solution at high Reynolds number flow simulation by tuning viscosity ratio, namely, the ratio of bulk viscosity and kinematic viscosity. QELBM uses only two relaxation times and does not need to solve non-linear equation at each grid point to enhance stability, thus the algorithm is simpler than Multi-Relaxation Time model and Entropic Lattice Boltzmann model. However, there is degree of freedom for determining the viscosity ratio and the effect of viscosity ratio on the results has not been known yet. In this study, we investigate the region of viscosity ratio where the calculation becomes stable and the effect of viscosity ratio on flow field in various Reynolds numbers and Mach numbers.
  • 橋本 知久; 丹野 格; 安田 孝宏; 田中 嘉宏; 森西 晃嗣; 里深 信行
    年次大会 : Mechanical Engineering Congress, Japan 2012年09月 
    In our previous study, kinetically reduced local Navier-Stokes (KRLNS) equations was applied for simulation of two-dimensional (2-D) unsteady incompressible flow problems in order to demonstrate it's capability to capture the correct transient behavior. The numerical results obtained by the KRLNS equations were compared with those obtained by the artificial compressibility method (ACM), the lattice Boltzmann method (LBM) and the pseudo-spectral method (PSM). The divergence as a function of time in the KRLNS method was compared with that of the ACM. It was confirmed that the KRLNS method can capture the correct transient behavior without use of sub-iterations. In this paper, three-dimensional (3-D) numerical simulations of flows in a circular pipe are carried out by using the KRLNS equations and virtual flux method (VFM) proposed to treat the body with complicated geometries on a Cartesian grid. The solutions and computational time are compared with those obtained by the ACM using sub-iterations.
  • 非圧縮性局所ナビエ・ストークス方程式と仮想流束法を用いた流路内流れの数値計算  [通常講演]
    橋本知久
    2012年度機械学会年次大会 2012年07月 口頭発表（一般）
  • Simulation of Decaying Two-dimentional Turbulence Using Kinetically Reduced Local Navier-Stokes Equations  [通常講演]
    橋本知久
    ICCFD7 2012年07月 口頭発表（一般）
  • 橋本 知久; 齋藤 圭一; 森西 晃嗣; 里深 信行
    日本流体力学会年会講演論文集 2009年 
    The objective of this research is to develop a computational fluid dynamics (CFD) solver for simulating underfill flow in flip-chip packaging, especially for designing the most optimum condition of solder joint performance. Two types of processes for applying the underfill encapsulant to the gap between IC chip and substrate are presented. It is found that the proposed analytical models have a considerable potential for predicting the underfill flow.
  • 橋本 知久; 森西 晃嗣; 里深 信行
    計算力学講演会講演論文集 2003年11月 
    Numerical simulations of a pair of flexible wing flapping symmetrically in two-dimesional space including the effect of wing deformations elastically in response to the fluid-dynamic forces are carried out in order to investigate its contributions to the aerodynamic force production in flapping flight. A flexible wing of insect is formulated as an elastic membrane model using mass-spring with damping effect.

MISC

その他のリンク

researchmap


Copyright © MEDIA FUSION Co.,Ltd. All rights reserved.