随机凹凸型骨料在混凝土细观数值模型中配置算法研究

Numerical generation and efficient distribution for random shape aggregates in mesoscopic concrete model

混凝土细观结构随机骨料数值模型的建立及其损伤破坏力学分析,可为混凝土宏观强度分析研究提供依据.适于满足各类指定的骨料级配曲线,针对任意形状、大小的骨料颗粒的生成与投放过程提出了一种改进算法,突破了传统凸形骨料的投放限制,使细观数值模型更接近于混凝土试件断面的实际.具体实现过程中,基于计算机图形学多边形重叠判断及区域填充算法,来消除预投放骨料间的交叠现象.此外,利用粒径极值比来控制骨料形状.还提出骨料颗粒投放过程的用户微干预方法,比较传统完全随机法,提高了投放的成功率及效率.算例验证了骨料的实际投放效果.

The meso-mechanics analysis on the basis of numerical random aggregate model can provide convincing theoretical and numerical supports for the macroscopic strength analysis of concrete structure. According to the objective various aggregate grading curves, an improved numerical method is presented for the aggregate numerical generation with random shape and particle size, which breaks the numerical limit of convex aggregates in other common methods. During the aggregate generation process, techniques of distinguishing overlap regions and filling complex regions on the computer graphics are used to remove the regional overlap in the packing process. The definition of dimensional extreme ratio for particle size is proposed to control the figure of random rock aggregates. Also, an effective method under a little artificial interference is adopted in the forming process of concrete meso-mechanics model to improve the successful probability. The efficiency and feasibility of the presented random aggregate generation method is numerically validated by the practical generation results.