In numerical modelling of scour around riverine and coastal structures using a hydro-morphodynamic model, a discrete sand sliding procedure needs to be implemented to avoid the occurrence of unrealistic bed profiles. In this study, two commonly employed sand sliding techniques, namely, the artificial transport rate method (ATRM) and the geometry-based method (GBM), are implemented in OpenFOAM and their performances are evaluated for five three-dimensional test cases. The test cases are classified into cases with and without sediment transport induced by a flow field. In the first three test cases, in the absence of a flow field, sand heap avalanches for different geometries and bed boundary grid structures are modelled to compare the methods in terms of simulation time and mass continuity. In test cases 4 and 5, in the presence of a flow field, the sensitivity of the sand sliding methods coupled with a hydro-morphodynamic model to different bed mesh structures is evaluated in modelling scour. The results of the analysis demonstrate that modelling of the sand sliding procedure using ATRM requires higher computational time, while its results are highly independent of the bed mesh structure with lower mass continuity error, <0.2% in all test cases, in comparison with GBM.

中文翻译：

---

用于水体动力学建模的滑砂方法的数值研究

在使用水文形态动力学模型对河流和沿海结构进行冲刷的数值模拟中，需要实施离散的滑沙程序，以避免出现不切实际的河床剖面。在这项研究中，两种常用的滑砂技术，即人工传输速率法 (ATRM) 和基于几何的方法 (GBM)，在 OpenFOAM 中实施，并针对五个三维测试案例评估了它们的性能。测试案例分为有和没有由流场引起的沉积物传输的案例。在前三个测试案例中，在没有流场的情况下，对不同几何形状和床边界网格结构的沙堆雪崩进行建模，以比较模拟时间和质量连续性方面的方法。在测试用例 4 和 5 中，在存在流场的情况下，在建模冲刷中评估了砂滑动方法与水体动力学模型对不同床网格结构的敏感性。分析结果表明，与 GBM 相比，使用 ATRM 对滑砂过程进行建模需要更多的计算时间，而其结果高度独立于床网格结构，质量连续性误差较低，在所有测试案例中 <0.2%。