ABSTRACTAccurate and highly efficient construction of numerical concrete models with different gradations is important to analyze the acoustic properties of ultrasonic waves in concrete. The random aggregates‐based method is a classic approach to meso‐structural modeling of concrete and is particularly suitable for generating models with low‐density aggregates. However, when facing the simulation requirements of concrete models with large volumes, small particle sizes, and high aggregate content, traditional random aggregates‐based methods encounter challenges in satisfying aggregate packing density and modeling efficiency. In order to improve the efficiency of aggregate placement in modeling concrete meso‐models, we proposed a hybrid‐strategy point cloud placement algorithm that combines the re‐placement and sedimentation strategies. Then, the algorithm's effectiveness in improving the modeling efficiency is verified by reconstruction experiments on random aggregate models with different aggregate grades. Finally, the model's reliability and validity is further confirmed by simulation analysis of the ultrasonic attenuation characteristics of the model. The results show that the algorithm can rapidly generate more than 75% aggregate packing density concrete specimens and significantly outperforms similar algorithms regarding aggregate placement efficiency and model generation time. In addition, the model generated by the proposed algorithm can accurately simulate the attenuation characteristics of ultrasonic waves in concrete.
Read full abstract