ABSTRACT
Hydraulic structures, such as bridges, dikes, engineered logs, rock weirs and vanes, are common in engineering practice. Their existence presents a disturbance to the flow field in rivers and coastal regions. As a result, the sediment transport around these structures is disturbed and consequently results in scour and erosion. This work aims to simulate the sediment transport and bed morphological changes induced by complex structures in real world. Previously we have developed Arbitrary-Lagrangian-Eulerian (ALE) method to track the deformed bed, which is useful for scour around simple geometries. It is difficult and sometime fails to model scour around complex structures. We present a new model which uses an immersed boundary method (IBM) to dynamically track the geometrical deformation of the bed. This method eliminates the need of mesh deformation or remeshing in an ALE approach. The model is developed in the open source computational fluid dynamics platform OpenFOAM.
Local scour is ubiquitous in natural and built environments where flowing water moves sediment. In natural rivers, it is part of the process which creates the landscape. In engineered systems, it is a phenomenon which could endanger the stability of structures. To understand and predict the local scour process is thus of great importance. Computational models have been previously proposed and developed to predict the flow field and scour around hydraulic structures. However, there are still great challenges to accurately predict the evolution of sediment bed around objects with complex shape. Part of the reason is that it is hard to track the surface dynamically during simulation. Previously, mesh deformation method, i.e., the Arbitrary-Lagrangian-Eulerian (ALE) approach, has been used where a body-fitted mesh deforms as the bottom evolves its shape (Liu and Garcia, 2008; Roulund et al, 2005). The ALE method can only be used for simple geometries.
In this work, we present a new model which utilizes the immersed boundary method for the sediment bed. In this method, the background mesh does not change. Instead, the evolving bed is modeled as an immersed boundary and its effect to the flow field is implicitly imposed. The evolution of the immersed bed is governed by the conservation of sediment. Overall, the new model has two parts. One is the hydro-dynamic part and the other is the morphological part. This paper will first briefly introduce the two parts of the scour model and then show some preliminary results.
