一种考虑过滤的短纤维增强复合材料RVE建模方法

代表性体积单元（RVE）法是短纤维增强复合材料性能预测的常用方法，其RVE生成效率是预测效率的重要影响因素。针对现有的随机顺序吸附（RSA）方法生成RVE时，布尔运算次数多、效率低的问题，通过在布尔运算前加入对纤维形心距的判断，过滤掉一部分随机生成的且与已有纤维相交的纤维，以减少布尔运算次数，提出了一种考虑过滤的随机顺序吸附（FRSA）方法。通过将改进后的FRSA方法在不同RVE参数和方法参数下生成RVE所需布尔运算次数和所需时间与基于布尔运算的随机顺序吸附（BORSA）方法进行比较，证明了FRSA方法的先进性。 

A filter based modeling method of RVE for short fiber reinforced composite

The representative volume element (RVE) method is a common numerical modeling method for predicting the material properties of short fiber reinforced composites. The RVE generation efficiency is an important factor affecting the efficiency of the property prediction. For the problem of too many Boolean operation times which cause low efficiency of Boolean operation when the existing random sequential adsorption (RSA) method is used to generate RVE, the centroid distance calculations between the lasted randomly generated fibers and the existing fibers are introduced. Using the centroid distance, some randomly generated fibers, which are intersecting with the existing fibers, are filtered out to reduce the number of Boolean operations. And a filter based random sequential adsorption (FRSA) method for modeling RVE is presented. By comparing the number of Boolean operations and the time cost for generating RVE, the FRSA method has been verified, and the verification is performed with different RVE parameters and method parameters using improved FRSA method and Boolean operation based random sequential adsorption (BORSA).