High-precision geometric model particle discretization method, device and system based on meshless method

Abstract

The invention discloses a meshless method-based high-precision geometric model particle discretization method, device and system. The method comprises the following steps: discretizing a three-dimensional geometric model through a plurality of groups of surface grids, extracting grid nodes and generating a plurality of single-layer wall surface particles; defining a normal smooth angle as a judgment criterion, and constructing a visual discrete error distribution diagram; and according to the error graph, applying a variable-scale particle local nesting encryption technology to obtain a wall surface particle model with high precision and low computing resource consumption in a short period. According to the visual discrete precision evaluation method, the encryption position and the encryption degree can be specifically and effectively guided, and the time period of establishing a high-precision wall surface particle model is shortened by combining a coarse-to-fine discrete scheme; herein, the local nested encryption technology overcomes the difficulty that the traditional encryption technology is difficult to realize grid encryption jump between adjacent grid modules, a local structure can be independently encrypted and directly combined with other structures, and high-precision discretization of a thin-wall structure and accurate calculation of a normal vector are realized.

Description

technical field [0001] The invention belongs to the field of discrete geometric model particles, and relates to a high-precision geometric model particle discrete method, device and system based on a gridless method. Background technique [0002] Based on the concept of fluid particles, the meshless method uses discrete particles without fixed topological relationships to replace grids and nodes under the Lagrangian framework. It is suitable for calculating unsteady large deformation flows. The discrete model and the overall solution to the flow field have great potential in the field of flow field calculation for engineering problems; however, the 3D modeling of complex shapes has always been one of the difficulties that restrict the application of the meshless particle method to practical engineering problems. Most of the geometric models are relatively regular and do not have sharp corners or thin-walled structures. The wall boundary used is to arrange three layers of wal...

AI Technical Summary

Problems solved by technology

[0002] Based on the concept of fluid particles, the meshless method uses discrete particles without fixed topological relationships to replace grids and nodes under the Lagrangian framework. It is suitable for calculating unsteady large deformation flows. The discrete model and the overall solution to the flow field have great potential in the field of flow field calculation for engineering problems; however, the 3D modeling of complex shapes has always been one of the difficulties that restrict the application of the meshless particle method to practical engineering problems. Most of the geometric models are relatively regular and do not have sharp corners or thin-walled structures. The wall boundary used is to arrange three layers of wall particles and make the inner wall particles participate in the pressure calculation. The engineering problems of blades, such as the internal flow simulation of fluid machinery, are difficult to arrange directly by traditional methods, and the lack of particle number density caused by insufficient layout accuracy will also lead to non-physical phenomena such as particle penetration; single-layer wall models based on distance functions are often used in Discrete complex geometric walls, but directly introducing the distance function will easily lead to violent fluctuations in the pressure of fluid particles near the wall, which will affect the stability of the calculation; in addition, it is difficult for the wall function to accurately compensate for the lack of particle number density in the three-dimensional case, and the boundary particles involved in the pressure calculation are easy Increase the amount of calculation

Images