Method and system for constructing mesoscopic model for performance prediction of composite materials

Abstract

The invention provides a method and system for constructing a mesoscopic model for composite material performance prediction, which belongs to the technical field of material performance detection. For particles located inside the model space, the size is reduced, and the size range that meets the predetermined gradation is randomly obtained. The compact model of the particles; the particles intersecting the boundary are generated according to the original size, the compact model is not generated, and only the part in the model space is reserved; then the compact model of each particle is expanded to obtain the expansion model of the compact model of each particle; The expansion model of each particle is subjected to area contraction to obtain a mesoscopic model for composite material performance prediction. The invention ensures that there are gaps between the particles, solves the problems that the particles directly contact to form a large number of included angles, the contact points form singular points, and are inconsistent with the actual material structure, and the modeling efficiency is high; it is more in line with the real particle shape in engineering practice , and solves the problem of modeling particles intersecting the model boundary.

Description

technical field [0001] The invention relates to the technical field of material performance detection, in particular to a method and system for constructing a mesoscopic model for performance prediction of composite materials. Background technique [0002] The mechanical and thermal properties of particle-reinforced composites are mainly affected by their mesostructure. Therefore, multi-scale modeling for the structure of such materials and further design and optimization of the mesostructure of such materials are of great significance for analyzing and improving material properties. [0003] Particle reinforced composites are usually complex multiphase discrete systems composed of particles, matrix and the interface between the two. It can also be regarded as a random distribution structure of particles of different scales in a limited space, and the particles are filled with matrix. The material also has defects such as pores and bubbles. In order to more accurately pred...

AI Technical Summary

Problems solved by technology

Since the upper surface of the accumulation body obtained by this method is not flat, it needs to be cut again; in addition, in order to control the gap between particles, it is necessary to divide the surface of the particles into shell units first, and then divide them into solid units again after the particles fall. , and delete the shell element; moreover, when the particles intersect with the model boundary, it is difficult to handle the drop

[0006] The patent document with the application number 201710570950.1 discloses a modeling method for a high-density discrete particle multiphase system based on the principle of expansion. However, the following problems still exist when using this method for modeling: (1) When obtaining the particle compaction model , although eligible random points can be easily generated, but at the same time, due to the reduction of particles, the placement space increases, and because it is too easy to generate qualified random points, resulting in uneven distribution of compressed particles, some particles are close to each other. close, some particles are far away

(2) In the subsequent expansion process, if the particles are divided into solid units, those particles that are close to each other will have direct contact. Each particle is covered with a certain thickness of mortar layer, and the particles are not in direct contact); on the other hand, such direct contact polygonal or polyhedral particles will form a large number of angles with each other, and the contact points will form singular points. , these included angles and singular points will bring serious difficulties to the subsequent matrix meshing

However, when hollow particles formed by surface shell elements intersect with the boundary, the intersecting surface also needs to be divided into shell elements. Since the shell element must have a certain thickness, it is bound to produce an initial penetration with the boundary of the modeling space, so that the contact Definition, dilation calculation and subsequent processing cause trouble

Furthermore, these hollow particles formed by surface shell elements often shrink in volume due to mutual extrusion during the expansion process, so that the final particle model no longer conforms to the predetermined gradation

In addition, these particles still need to be divided into solid elements and delete shell elements in order to be applied in subsequent performance analysis, thereby reducing modeling efficiency

Images