<p>The construction of three-dimensional complex shapes presents a challenge in the application of meshless particle methods, wherein particle distribution significantly affects simulation results. Here, we propose a spherical symmetric smoothing method coupled with pseudo hydrostatic flow for obtaining body-fitted and regular particle distributions before and during simulations. The smoothing method employs a unified geometric form applicable to surface and three-dimensional conditions wherein it conducts regularizations for each virtual sphere or circle, and perform suturing operations using a regularity metric defined to quantify the quality of local particle distribution within a virtual sphere or circle. This new method efficiently transforms irregular particle distributions, such as sawtooth, chaotic and overflowed patterns, into body-fitted and relatively regular configurations that meet the spatial requirements of complex geometries. Importantly, our method ensures no merged or splitted of particles, nor overflow of boundary particles during the smoothing process. By applying this smoothing method in solid and fluid SPH simulations, some novel insights are discovered after increasing simulation accuracy and stability.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

A spherical symmetric smoothing method for regularizing particle distributions in meshless particle methods

  • Haobin Xu,
  • Shuli Sun

摘要

The construction of three-dimensional complex shapes presents a challenge in the application of meshless particle methods, wherein particle distribution significantly affects simulation results. Here, we propose a spherical symmetric smoothing method coupled with pseudo hydrostatic flow for obtaining body-fitted and regular particle distributions before and during simulations. The smoothing method employs a unified geometric form applicable to surface and three-dimensional conditions wherein it conducts regularizations for each virtual sphere or circle, and perform suturing operations using a regularity metric defined to quantify the quality of local particle distribution within a virtual sphere or circle. This new method efficiently transforms irregular particle distributions, such as sawtooth, chaotic and overflowed patterns, into body-fitted and relatively regular configurations that meet the spatial requirements of complex geometries. Importantly, our method ensures no merged or splitted of particles, nor overflow of boundary particles during the smoothing process. By applying this smoothing method in solid and fluid SPH simulations, some novel insights are discovered after increasing simulation accuracy and stability.