论文部分内容阅读
将混凝土视为由粗骨料、砂浆质以及两者之间的界面组成的三相复合材料,用细观方法研究混凝土特性,提出了新的随机算法并生成三维细观有限元模型。这种方法基于Mote Carlo随机样本原理,由一个给定的骨料级配随机生成多面体骨料颗粒,按照互不相交原则逐一投放到混凝土模型试件中。将判断骨料是否互不相交转化为判断一个线性规划问题是否有解,能有效地提高模型骨料含量率及模型生成速度。文中提出的所有算法在MATLAB中都易实现,与COMSOL软件相结合能更高效地完成有限元网格剖分。计算实例表明,对于骨料含量高的大体积三四级配混凝土,这种方法能满足数值建模的需要。
The concrete is considered as a three-phase composite material composed of coarse aggregate, mortar and the interface between the two. Concrete is studied by meso-method. A new stochastic algorithm is proposed and a three-dimensional mesoscopic finite element model is generated. Based on the Mote Carlo random sample principle, this method generates randomly generated polyhedral aggregate particles from a given aggregate gradation, which are cast into concrete model specimens one after another according to the principle of disjointness. Will determine whether the aggregate intersects each other to judge whether there is a solution to a linear programming problem, can effectively improve the model aggregate content rate and model generation rate. All the algorithms proposed in this paper are easy to implement in MATLAB, and combined with COMSOL software can finish the finite element meshing more efficiently. The calculation example shows that this method can meet the needs of numerical modeling for the large volume of three or four concrete with high aggregate content.