Accurate modeling of sedimentation processes in geotechnical applications requires resolving the complex dynamics that comprise irregularly shaped sand particles interacting with a surrounding fluid. This study provides an integrated numerical model of Computational Fluid Dynamics and the Discrete Element Method, introducing a novel mesh-ready particle morphology generation algorithm to investigate sedimentation behavior in a still fluid domain. The synthetic shape generation approach enables systematic control over elongation, angularity, and surface roughness by combining ellipsoidal transformations with Gaussian surface perturbations. This formulation allows the creation of morphologically varied particle geometries that accurately reflect natural sediments, leading to a detailed investigation of the impact of particle shape on sedimentation behavior. These arbitrarily shaped particles are also converted into clump representations using a voxel-based multi-sphere approximation method, allowing for a direct comparison between the new particle surface-mesh representation and the previously used clump-based models. In contrast to previous studies that rely primarily on idealized shapes, like spheres, or computationally expensive approaches such as the clump models, the proposed method enables efficient, large-scale simulations while preserving key morphological features. The comparative analysis reveals that particle shape has a strong influence on settling velocity, contact force dynamics, and fluid-particle interactions. The simulation results obtained using the new approach show close agreement with validation tests, outperforming spherical models in capturing rotational instabilities, clustering, and non-uniform deposition patterns. The proposed approach enhances the accuracy and quality of sediment transport simulations by integrating advanced particle shape modeling with multiphase numerical methods, offering new perspectives for applications in geotechnical engineering.
Samadi et al. (Mon,) studied this question.