Development of Numerical Algorithm for Multi-Phase Flow by Distributed Computer

分布式计算机多相流数值算法的开发

基本信息

  • 批准号:
    11650166
  • 负责人:
  • 金额:
    $ 2.37万
  • 依托单位:
  • 依托单位国家:
    日本
  • 项目类别:
    Grant-in-Aid for Scientific Research (C)
  • 财政年份:
    1999
  • 资助国家:
    日本
  • 起止时间:
    1999 至 2000
  • 项目状态:
    已结题

项目摘要

We have develped the numerical scheme, CIP method, which is able to solve solid, fluid and gassimultaneously. By this method, the boundary of multiphase flow can be captured with the fixed mesh. Most importantly, gas and fluid that have completely different playsical properties can be calculated simultaneously.Since the performance of computation is limited when only one CPU of computer is used, parallel and distributed computers becomes mainstream at present. In accordance with this trend, we need to look for the scheme that is appropriate for parallel calculation. The purpose of this research is to establish the best numerical scheme for multiphase flow.Since parallel algorithm can be readily applied to the advection phase of the CIP, we constructed PC clusters, and established at first the parallel scheme of only advection phase. Thus, 3 GFLOPS has been realized with 24 PCs. Since one PC has the performance in 100MFLOPS, we were able to achieve a speed in the range of "super linear".Nextly, we considered the parallel scheme to solve pressure equation. As one of them, we tried the SOR method with the Red-Black method. We confirmed that the calculation speed with 24 PCs increases almost in proportion to the number of PC as well. This performance is similar for SSI (Symmetrical Semi Implicit) scheme we planed to use, and the SOR method with the Red-Black method is faster in view of convergence. With this method, we succeeded to calculate three-dimensional multiphase fluid with 24 CPUs.
我们发展了一种能求解固体、流体和气体的数值格式-CIP方法。该方法可以在固定网格下捕捉多相流的边界。最重要的是,具有完全不同物性的气体和流体可以同时进行计算,由于单CPU计算机的计算性能有限,并行和分布式计算机成为目前的主流。根据这一趋势,我们需要寻找适合并行计算的方案。本研究的目的是建立多相流的最佳数值格式,由于并行算法可以很容易地应用于CIP的平流相,我们构造了PC集群,首先建立了对流相的并行格式。因此,用24台PC实现了3 GFLOPS。由于一台PC机的性能达到100 MFLOPS,我们可以实现“超线性”范围内的速度。作为其中之一,我们尝试了SOR方法与Red-Black方法。我们证实,24台PC的计算速度几乎与PC的数量成正比。这种性能与我们计划使用的SSI(对称半隐式)格式相似,而从收敛性来看,采用Red-Black方法的SOR方法更快。利用该方法,在24个CPU上成功地实现了三维多相流的计算。

项目成果

期刊论文数量(28)
专著数量(0)
科研奖励数量(0)
会议论文数量(0)
专利数量(0)
T.Yabe: "A Universal Numerical Solver for Solid, Liquid and Gas-Application to Laser-Induced Melting and Evaporation-"SPIE Proc.On Laser-Assisted Microtechnology. 4157. 1-7 (2000)
T.Yabe:“固体、液体和气体的通用数值解算器 - 在激光诱导熔化和蒸发中的应用 -”SPIE Proc.On 激光辅助微技术。
  • DOI:
  • 发表时间:
  • 期刊:
  • 影响因子:
    0
  • 作者:
  • 通讯作者:
M.Tajima and T.Yabe: "Simulation on Slamming of a Vessel by CIP Method"J.Phys.Soc.Japan. 68. 2576-2584 (1999)
M.Tajima 和 T.Yabe:“通过 CIP 方法模拟船舶撞击”J.Phys.Soc.Japan。
  • DOI:
  • 发表时间:
  • 期刊:
  • 影响因子:
    0
  • 作者:
  • 通讯作者:
R.Tanaka,T.Nakamura,T.Yabe: "Constructing exactly conservative scheme in a non-conservative form"Comput.Phys.Commun. 126. 232-243 (2000)
R.Tanaka、T.Nakamura、T.Yabe:“以非保守形式构建完全保守的方案”Comput.Phys.Commun。
  • DOI:
  • 发表时间:
  • 期刊:
  • 影响因子:
    0
  • 作者:
  • 通讯作者:
Y.Zhang and T.Yabe: "Effect of Ambient Gas on Three-Dimensional Breakup in Coronet Formation"CFD Journal. 8. 378-382 (1999)
Y.Zhang 和 T.Yabe:“环境气体对冠层形成三维破裂的影响”CFD 期刊。
  • DOI:
  • 发表时间:
  • 期刊:
  • 影响因子:
    0
  • 作者:
  • 通讯作者:
{{ item.title }}
{{ item.translation_title }}
  • DOI:
    {{ item.doi }}
  • 发表时间:
    {{ item.publish_year }}
  • 期刊:
  • 影响因子:
    {{ item.factor }}
  • 作者:
    {{ item.authors }}
  • 通讯作者:
    {{ item.author }}

数据更新时间:{{ journalArticles.updateTime }}

{{ item.title }}
  • 作者:
    {{ item.author }}

数据更新时间:{{ monograph.updateTime }}

{{ item.title }}
  • 作者:
    {{ item.author }}

数据更新时间:{{ sciAawards.updateTime }}

{{ item.title }}
  • 作者:
    {{ item.author }}

数据更新时间:{{ conferencePapers.updateTime }}

{{ item.title }}
  • 作者:
    {{ item.author }}

数据更新时间:{{ patent.updateTime }}

YABE Takashi其他文献

YABE Takashi的其他文献

{{ item.title }}
{{ item.translation_title }}
  • DOI:
    {{ item.doi }}
  • 发表时间:
    {{ item.publish_year }}
  • 期刊:
  • 影响因子:
    {{ item.factor }}
  • 作者:
    {{ item.authors }}
  • 通讯作者:
    {{ item.author }}

{{ truncateString('YABE Takashi', 18)}}的其他基金

Mg recycle method using Solar-pumped-laser for Mg fuel cell
镁燃料电池太阳能泵浦激光镁回收方法
  • 批准号:
    20560032
  • 财政年份:
    2008
  • 资助金额:
    $ 2.37万
  • 项目类别:
    Grant-in-Aid for Scientific Research (C)
The Simulations of Laser-Micro-Airplane by the CIP Method and its Application to CO_2 measurement
CIP法模拟激光微型飞机及其在CO_2测量中的应用
  • 批准号:
    15607007
  • 财政年份:
    2003
  • 资助金额:
    $ 2.37万
  • 项目类别:
    Grant-in-Aid for Scientific Research (C)
Development of Numerical Method for Mesoscopic Flow
细观流动数值方法的发展
  • 批准号:
    09680474
  • 财政年份:
    1997
  • 资助金额:
    $ 2.37万
  • 项目类别:
    Grant-in-Aid for Scientific Research (C)

相似海外基金

Data driven approach integrating optics and machine learning to uncover parallel computation mechanism in brain circuits
集成光学和机器学习的数据驱动方法揭示大脑回路中的并行计算机制
  • 批准号:
    21H02801
  • 财政年份:
    2021
  • 资助金额:
    $ 2.37万
  • 项目类别:
    Grant-in-Aid for Scientific Research (B)
Development of stencil calculation and communication model to achieve high scalability on massively parallel computation
开发模板计算和通信模型,以实现大规模并行计算的高可扩展性
  • 批准号:
    18K11336
  • 财政年份:
    2018
  • 资助金额:
    $ 2.37万
  • 项目类别:
    Grant-in-Aid for Scientific Research (C)
Multiphase parallel computation method to predict and understand disasters caused by gas-liquid-solid interactions
预测和理解气-液-固相互作用引起的灾害的多相并行计算方法
  • 批准号:
    18K11337
  • 财政年份:
    2018
  • 资助金额:
    $ 2.37万
  • 项目类别:
    Grant-in-Aid for Scientific Research (C)
Development of a highly integrated circuit using DNA and a demonstration test of the parallel computation using a complementary DNA as an operator
使用DNA的高度集成电路的开发以及使用互补DNA作为算子的并行计算的演示测试
  • 批准号:
    17K06395
  • 财政年份:
    2017
  • 资助金额:
    $ 2.37万
  • 项目类别:
    Grant-in-Aid for Scientific Research (C)
CAREER: Multi-Objective Optimization via Simulation: Theory, Methods, and Parallel Computation
职业:通过仿真进行多目标优化:理论、方法和并行计算
  • 批准号:
    1554144
  • 财政年份:
    2016
  • 资助金额:
    $ 2.37万
  • 项目类别:
    Standard Grant
Parallel Computation in Game AI
游戏AI中的并行计算
  • 批准号:
    1972112
  • 财政年份:
    2016
  • 资助金额:
    $ 2.37万
  • 项目类别:
    Studentship
ParCell: A Parallel Computation Framework for Scalable and Mechanistic Modeling and Simulation of Multicellular Systems
ParCell:用于多细胞系统可扩展和机械建模与仿真的并行计算框架
  • 批准号:
    1609642
  • 财政年份:
    2016
  • 资助金额:
    $ 2.37万
  • 项目类别:
    Standard Grant
Stability Analysis of Large-Scale Nonlinear Systems using Parallel Computation
使用并行计算的大规模非线性系统的稳定性分析
  • 批准号:
    1538374
  • 财政年份:
    2015
  • 资助金额:
    $ 2.37万
  • 项目类别:
    Standard Grant
SHF: Small: Empirical Autotuning of Parallel Computation for Scalable Hybrid Systems
SHF:小型:可扩展混合系统并行计算的经验自动调整
  • 批准号:
    1527706
  • 财政年份:
    2015
  • 资助金额:
    $ 2.37万
  • 项目类别:
    Standard Grant
Development of the method for multibody system analysis based on the canonical theory for constrained system applicable to the parallel computation and the state estimation
基于适用于并行计算和状态估计的约束系统规范理论的多体系统分析方法的发展
  • 批准号:
    15K17994
  • 财政年份:
    2015
  • 资助金额:
    $ 2.37万
  • 项目类别:
    Grant-in-Aid for Young Scientists (B)
{{ showInfoDetail.title }}

作者:{{ showInfoDetail.author }}

知道了