Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Mesh-Free Methods with Special Focus on SPH
This chapter presents the principle and basic algorithm of the “smoothed particle method (SPH)” for fluid that belongs to the family of “mesh-free methods (MFMs)” or “mesh-free particle methods (MPMs)”. As in EFGM, the trial functions are no longer created by the basis belong to standard PU, too. Instead, the fluid domain is represented by a set of arbitrarily distributed nodes without regular connectivity. Termed as “kernel approximation”, at the representative node, the field function value in the governing PDEs (strong form) is replaced by the integrated function value with the help of its neighborhood field nodes in the “support domain” similar to that in EFGM. Then the representative node is further approximated by the representative particle assigned with mass, which is termed as the “particle approximation” meaning that the particle is actually the live physical objects. These two steps produce, instead of original PDEs, a set of ODEs that may be solved by routine finite difference schemes w.r.t. time-marching steps to obtain the time history of all the field variables for all the particles. The use of Lagrangian description allows SPH to be adaptive for particle distribution that evolves with the on-going of time. Consequently, it possesses high ability to handle violent “fluid–structure interaction (FSI)” problems with extremely free surface distortion.
Mesh-Free Methods with Special Focus on SPH
This chapter presents the principle and basic algorithm of the “smoothed particle method (SPH)” for fluid that belongs to the family of “mesh-free methods (MFMs)” or “mesh-free particle methods (MPMs)”. As in EFGM, the trial functions are no longer created by the basis belong to standard PU, too. Instead, the fluid domain is represented by a set of arbitrarily distributed nodes without regular connectivity. Termed as “kernel approximation”, at the representative node, the field function value in the governing PDEs (strong form) is replaced by the integrated function value with the help of its neighborhood field nodes in the “support domain” similar to that in EFGM. Then the representative node is further approximated by the representative particle assigned with mass, which is termed as the “particle approximation” meaning that the particle is actually the live physical objects. These two steps produce, instead of original PDEs, a set of ODEs that may be solved by routine finite difference schemes w.r.t. time-marching steps to obtain the time history of all the field variables for all the particles. The use of Lagrangian description allows SPH to be adaptive for particle distribution that evolves with the on-going of time. Consequently, it possesses high ability to handle violent “fluid–structure interaction (FSI)” problems with extremely free surface distortion.
Mesh-Free Methods with Special Focus on SPH
Springer Tracts in Civil Engineering
Chen, Shenghong (Autor:in)
01.01.2023
56 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Mesh-Free Methods with Special Focus on EFGM
| Springer Verlag | 2023
Discrete Element Methods with Special Focus on DDA
| Springer Verlag | 2023
Discrete Element Methods with Special Focus on DEM
| Springer Verlag | 2023
Hybrid Methods with Special Focus on DEM-SPH
| Springer Verlag | 2023
General Finite Element Methods with Special Focus on NMM
| Springer Verlag | 2023