Secondary flow and longitudinal sediment patterns in open channel flow over a bed of mobile particles

纵脊是水流引起的沉积物形态，在紊流条件下，在明渠中形成的，具有非粘性颗粒的移动的床。这些床的形式，这是不同于横向波纹或沙丘，带来了重要的后果的动量和标量传输，他们发现了广泛的应用，例如在水力学（河流）和其他技术颗粒流系统（化学工程）。它们的形成与垂直于主要平均流动方向的平面中的二次流的存在密切相关，而二次流又与湍流统计中的展向不均匀性有关。为了研究导致沉积脊发展的展向床剪切应力和重力之间的复杂平衡，我们建议进行界面分辨直接数值模拟，再加上颗粒间接触的离散元方法。通过考虑这两种情况，有和没有通道侧壁，我们将试图阐明的影响，先验存在的普朗特的第二类二次流的过程。后者被认为是建设性地干扰不稳定过程的基础上形成的山脊，导致多细胞的二次流模式的情况下，侧壁和移动的沉积物。在这个项目中进行的高保真数值实验将使我们能够提取信息的缩放特性的纵向床的形式和相关的二次运动。这些数据将被进一步利用，以达到改善平板展向扰动稳定性分析的最新水平的目的。可以预期，目前的贡献将提高我们对颗粒流动力学的理解，在一定程度上，它将导致改进的工程型模型，这个问题在未来。

Longitudinal ridges are flow-induced sediment patterns which form under turbulent conditions in open channels with a mobile bed of non-cohesive particles. These bed-forms, which are distinct from transverse ripples or dunes, entail important consequences for momentum and scalar transport, and they find a wide array of applications e.g. in hydraulics (river flow) and other technical particulate flow systems (chemical engineering). Their formation is closely linked to the existence of secondary flow in the plane perpendicular to the primary mean flow direction, which in turn can be related to spanwise inhomogeneities in the turbulent statistics. In order to investigate the intricate balance between the spanwise bed shear stress and gravity that leads to the development of sediment ridges, we propose to perform interface-resolved direct numerical simulation coupled with a discrete element method for the inter-particle contact. By considering both cases, with and without channel side walls, we will attempt to elucidate the influence of a priori existing secondary flow of Prandtl's second kind upon the process. The latter is believed to interfere constructively with the instability process underlying the formation of ridges, leading to a multi-cellular secondary flow pattern in the case with side walls and mobile sediment. The high-fidelity numerical experiments to be carried out in this project will allow us to extract information on the scaling properties of the longitudinal bed-forms and of the associated secondary motion. The data will be further utilized towards the goal of improving the state of the art in stability analysis of a flat bed with respect to spanwise perturbations. It can be expected that the present contribution will enhance our understanding of the particulate flow dynamics to such a degree that it will lead to improved engineering-type models of this problem in the future.