采用多边形壁面的MPS粒子分裂算法研究

张静, 熊进标

张静, 熊进标. 采用多边形壁面的MPS粒子分裂算法研究[J]. 原子能科学技术, 2022, 56(6): 1138-1146. DOI: 10.7538/yzk.2021.youxian.0977
引用本文: 张静, 熊进标. 采用多边形壁面的MPS粒子分裂算法研究[J]. 原子能科学技术, 2022, 56(6): 1138-1146. DOI: 10.7538/yzk.2021.youxian.0977
ZHANG Jing, XIONG Jinbiao. Investigation on MPS Particle Splitting Algorithm Based on Polygon Wall Model[J]. Atomic Energy Science and Technology, 2022, 56(6): 1138-1146. DOI: 10.7538/yzk.2021.youxian.0977
Citation: ZHANG Jing, XIONG Jinbiao. Investigation on MPS Particle Splitting Algorithm Based on Polygon Wall Model[J]. Atomic Energy Science and Technology, 2022, 56(6): 1138-1146. DOI: 10.7538/yzk.2021.youxian.0977

采用多边形壁面的MPS粒子分裂算法研究

Investigation on MPS Particle Splitting Algorithm Based on Polygon Wall Model

  • 摘要: 反应堆严重事故模拟中涉及大量含有自由界面或组分界面的多相流动,不具有拓扑结构的粒子法在这类流动的模拟中有其独特的优势。本文在移动粒子半隐式(MPS)法的基础上开发了适用于多边形壁面的粒子分裂模型。针对多粒径模拟及多边形壁面的特殊性,对粒子有效半径、数密度、梯度算子模型、拉普拉斯算子模型和表面粒子识别模型等进行改进,并对分裂过程进行简要阐述。将改进后的MPS法运用于无挡板及有挡板的溃坝实验模拟计算。计算结果表明,采用多边形壁面的粒子分裂模型对复杂自由液面捕捉清晰,且计算时间仅为传统MPS法的18.75%,但监测点的压力波动较大。改进后的模型在保证准确性和精度的同时提高了计算效率,为进一步计算三维相变传热奠定基础。

     

    Abstract: Simulation of severe accident in nuclear reactor involves multiphase flow phenomena containing free surface or component interfaces. The moving particle semi-implicit (MPS) method waiving the topological mesh has been extensively employed for multiphase simulation. However, the uniform-size particle used in the traditional MPS method cannot provide high enough spatial resolution in the region of interest, e.g., near the phase interfaces, since utilization of uniform-size fine particle significantly raises the computation cost. Meanwhile, a large number of wall particles are utilized to model the wall boundary, which remarkably increases the computation resource. In order to improve the computation efficiency and accuracy, the particle splitting algorithm is implemented along with the polygon wall boundary. Utilization of particle splitting in the specified region refines simulation of flow with complex free surfaces. The polygon wall model eliminates the wall particle. In the model, the wall effect is modelled based on a flat wall scenario where the contribution of the wall to the particle number density is estimated as a function of wall distance. The zero-pressure gradient at the wall is realized with the virtual mirror particles about the wall. In order to facilitate the implementation of particle splitting algorithm and polygon wall model, several basic models of the MPS method were modified, including particle effective radius, particle number density, gradient operator model, Laplace operator model, and free surface identification model. For algorithm verification, the static pressure of water column was simulated with the improved MPS method. The discrepancy between the pressure at the monitoring point and the theoretical value is 6.25%. Then the baffle-free dam-break flow was simulated. The simulation is in good agreement with the experiment, which indicates that the particle splitting algorithm has little effect on the flow pattern. In the simulation of dam-break with baffles, the free surface motion can be replicated with the modified MPS method. The modified method predicts appreciably larger pressure fluctuation range than the traditional MPS method, which should be improved in the future. The splitting model can reduce the computation time by about 55%, while the polygon wall can reduce the computation time by around 62.5%. Combining both of the models, the computation time can be reduced by over 80%. Hence, the particle splitting model together with polygon wall boundary can improve local resolution and computational efficiency in the same time, which solidates a foundation for the particle merging calculation, three-dimensional calculation, and phase change heat transfer calculation.

     

  • [1] 李勇霖,陈荣华,田文喜,等. 基于MPS方法的共晶反应研究[C]∥第十五届全国反应堆热工流体学术会议暨中核核反应堆热工水力技术重点实验室学术年会论文集. [出版地不详]:[出版者不详],2017:1028-1036.
    [2] KOSHIZUKA S, OKA Y. Moving-particle semi-implicit method for fragmentation of incompressible fluid[J]. Nuclear Science and Engineering, 1996, 123(3): 421-434.
    [3] 张明昊,陈荣华,郭凯伦,等. 基于粒子法的流量脉动条件下单气泡上升行为研究[J]. 核动力工程,2019,40(S2):15-20.
    ZHANG Minghao, CHEN Ronghua, GUO Kailun, et al. Research on single bubble rising behavior under flow pulsation condition based on particle method[J]. Nuclear Power Engineering, 2019, 40(S2): 15-20(in Chinese).
    [4] ZHU Y, QIU Z, XIONG J, et al. Verification and validation of MPS potential force interface tension model for stratification simulation[J]. Annals of Nuclear Energy, 2020, 148: 107753.
    [5] XIONG J, ZHU Y, ZHANG T, et al. Lagrangian simulation of three-dimensional macro-scale melting based on enthalpy method[J]. Computers & Fluids, 2019, 190: 168-177.
    [6] ZHU Y, XIONG J, YANG Y. MPS eutectic reaction model development for severe accident phenomenon simulation[J]. Nuclear Engineering and Technology, 2020, 53(3): 833-841.
    [7] TANG Z, WAN D, CHEN G, et al. Numerical simulation of 3D violent free-surface flows by multi-resolution MPS method[J]. Journal of Ocean Engineering and Marine Energy, 2016, 2(3): 355-364.
    [8] CHEN X, SUN Z G, LIU L, et al. Improved MPS method with variable-size particles[J]. International Journal for Numerical Methods in Fluids, 2016, 80(6): 358-374.
    [9] SHIBATA K, KOSHIZUKA S, MATSUNAGA T, et al. The overlapping particle technique for multi-resolution simulation of particle methods[J]. Computer Methods in Applied Mechanics & Engineering, 2017, 325: 434-462.
    [10] HARADA T, KOSHIZUKA S, SHIMAZAKI K. Improvement of wall boundary calculation model for MPS method[J]. Transactions of the Japan Society for Computational Engineering and Science, 2008: 20080006.
    [11] ZHANG T, KOSHIZUKA S, MUROTANI K, et al. Improvement of boundary conditions for non-planar boundaries represented by polygons with an initial particle arrangement technique[J]. International Journal of Computational Fluid Dynamics, 2016, 30(2): 1-21.
    [12] ZHANG T, KOSHIZUKA S, MUROTANI K, et al. Improvement of pressure distribution to arbitrary geometry with boundary condition represented by polygons in particle method[J]. International Journal for Numerical Methods in Engineering, 2017, 112(7): 685-710.
    [13] MITSUME N, YAMADA T, YOSHIMURA S. Parallel analysis system for free-surface flow using MPS method with explicitly represented polygon wall boundary model[J]. Computational Particle Mechanics, 2020, 7(2): 279-290.
    [14] KOSHIZUKA S, AND A N, OKA Y. Numerical analysis of breaking waves using the moving particle semi-implicit method[J]. International Journal for Numerical Methods in Fluids, 1998, 26(7): 751-769.
    [15] TANAKA M, CARDOSO R, BAHA H, et al. Multi-resolution MPS method[J]. Journal of Computational Physics, 2018, 359: 106-136.
    [16] MITSUME N, YOSHIMURA S, MUROTANI K, et al. Explicitly represented polygon wall boundary model for the explicit MPS method[J]. Computational Particle Mechanics, 2015, 2(1): 73-89.
    [17] SHIBATA K, MASAIE I, KONDO M, et al. Improved pressure calculation for the moving particle semi-implicit method[J]. Computational Particle Mechanics, 2015, 2(1): 91-108.
    [18] MARTIN J, MOYCE W, MARTIN J, et al. Some gravity wave problems in the motion of perfect liquids, Part Ⅳ: An experimental study of the collapse of liquid columns on a rigid horizontal plane[J]. Philosophical Transactions of the Royal Society of London, 1952, 244(882): 312-324.
    [19] HU C, KASHIWAGI M. A CIP method for numerical simulations of violent free surface flows[J]. Journal of Marine Science & Technology, 2004, 9(4): 143-157.
计量
  • 文章访问数:  147
  • HTML全文浏览量:  0
  • PDF下载量:  184
  • 被引次数: 0
出版历程
  • 刊出日期:  2022-06-19

目录

    /

    返回文章
    返回