Sub-region retrieval based point-volume-element method for constructing concrete mesostructures with aggregate interference control

This paper proposed a numerical modeling method for constructing concrete model through the point-volume-element aggregate interference discrimination method based on sub-region retrieval. Firstly, the conventional concrete modeling method was enhanced by optimizing the interference discrimination logic between aggregates and the aggregate retrieval approach. Subsequently, a series of concrete models with varied aggregate geometric shapes, interfacial transition zone (ITZ) thicknesses, aggregate volume fractions, and aggregate gradations were generated, validating the reliability of the proposed method. Finally, the efficiency of the proposed method was validated by comparing the number of iterations and modeling time required with those of the conventional method. The results indicate that the proposed method can control the aggregate geometric shape and ITZ thickness by adjusting the number of vertices and the scaling distance. Compared to the conventional method, the proposed method significantly reduced the number of iterations and modeling time, thereby enhancing modeling efficiency. The advantage of the proposed method became more pronounced as the number of aggregates increased. When the number of generated aggregates reaches 100, the required number of iterations is reduced by 83.6%, and the modeling time is reduced by 66.2%.