高级检索
许立群, 王佳炎, 袁海鹏, 陶建新, 唐勇, 赵静. 物质点法流固交互表面细节优化算法[J]. 计算机辅助设计与图形学学报, 2024, 36(2): 303-312. DOI: 10.3724/SP.J.1089.2024.19763
引用本文: 许立群, 王佳炎, 袁海鹏, 陶建新, 唐勇, 赵静. 物质点法流固交互表面细节优化算法[J]. 计算机辅助设计与图形学学报, 2024, 36(2): 303-312. DOI: 10.3724/SP.J.1089.2024.19763
Xu Liqun, Wang Jiayan, Yuan Haipeng, Tao Jianxin, Tang Yong, Zhao Jing. Fluid-Solid Coupling Surface Detail Optimization Algorithm Based on Material Point Method[J]. Journal of Computer-Aided Design & Computer Graphics, 2024, 36(2): 303-312. DOI: 10.3724/SP.J.1089.2024.19763
Citation: Xu Liqun, Wang Jiayan, Yuan Haipeng, Tao Jianxin, Tang Yong, Zhao Jing. Fluid-Solid Coupling Surface Detail Optimization Algorithm Based on Material Point Method[J]. Journal of Computer-Aided Design & Computer Graphics, 2024, 36(2): 303-312. DOI: 10.3724/SP.J.1089.2024.19763

物质点法流固交互表面细节优化算法

Fluid-Solid Coupling Surface Detail Optimization Algorithm Based on Material Point Method

  • 摘要: 为了增强现有物质点法流固交互中表面细节,同时改进大量固体粒子对其表面重建没有贡献的问题,提出一种高效的流固交互自适应物质点法.首先使用一种接触面离散积分点方法解决物质点法中出现的数值黏性伪影问题,实现流固移动接触面上粒子的自由滑动;然后引入自适应广义插值方法,在流体和固体的不同区域自适应地调整网格分辨率,增强表面细节;最后采用窄带仿射粒子网格方法,通过粒子重采样减少固体内部粒子数量,从而在节省内存空间的同时减少计算量.利用斯坦福开源模型进行多组消融实验,结果表明,与均匀背景网格物质点法相比,所提方法流固交互表面细节更丰富,消融实验场景下整体粒子数减少80%,平均模拟效率提高50%.

     

    Abstract: In order to enhance the surface details of the existing material point method in fluid-solid coupling, and improve the problem that a large number of solid particles have no contribution on its surface remeshing, the paper proposes an efficient fluid-solid coupling adaptive material point method. First, an interface quadrature method is used to solve the problem of viscous numerical artifact in the material point method, and realize the free-slip effect on the moving contact interface of the fluid and solid. Then, an adaptive generalized interpolation method is introduced to adjust the grid resolution in different areas of fluid and solid to enhance surface details. Finally, the narrow band affine particle in cell method is adopted to reduce the number of particles inside the solid with the particle resampling method, thereby saving memory space and reducing the amount of calculation. The paper makes several experiments with Stanford open source models. Compared with the uniform background grid material point method, experiments show that the proposed method has more abundant details in the surface of fluid and solid, the overall particle number is reduced by 80%, and the average simulation efficiency is improved by 50% in the ablation experiment scene.

     

/

返回文章
返回