3D numerical modelling of hydrodynamics and morphodynamics around in-stream structures
Open Access
- Author:
- Xu, Yuncheng
- Graduate Program:
- Civil Engineering
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- February 22, 2019
- Committee Members:
- Xiaofeng Liu, Dissertation Advisor/Co-Advisor
Xiaofeng Liu, Committee Chair/Co-Chair
Peggy Ann Johnson, Committee Member
Chaopeng Shen, Committee Member
Roman Alexander Dibiase, Outside Member - Keywords:
- immersed boundary method
OpenFOAM
CFD
scour modeling
sediment transport
in-stream structure
large woody debris - Abstract:
- In-stream structures include both engineered structures and naturally formed ones. Examples of engineered in-stream structures are bridge piers, abutments, dams, engineered log jams (ELJ), rock vanes, J-hook vanes, among many others. Naturally formed in-stream structures mainly refer to large woody debris (LWD), which consist of fallen trees, logs, stumps, root wads, and piles of branches along the river course. These in-stream structures play a very significant role in flow resistance, sediment transport, invertebrate habitats and other aspects of fluvial ecosystem. For example, LWDs can provide food sources for aquatic insects and create refuge and habitat for fishes. They also create hydraulic diversity and roughness along river banks. Due to the geometrical complexity of these in-stream structures, the surrounding stream flow is extremely complicated and turbulent. The flow around and through complex in-stream structures can also result in local scour, sedimentation, and other morphodynamic changes. Thus, the overarching goal of this thesis research is to model and understand the flow and sediment transport processes associated with in-stream structures. This thesis is organized as three parts: (a) high resolution numerical investigations on the three-dimensional (3D) hydraulics of LWDs, with a focus on the importance of how to represent porosity in computational models, (b) a new immersed boundary (IB) method designed for the accurate prediction of local bed shear stress, which is the driver for sediment motion, and (c) development and application of a coupled hydro-morphodynamics model for complex in-stream structures. The first part tackles the problem of how to represent the porosity of in-stream structures. In many existing literature, the geometry of LWDs are simplified as simple cylinders or solid blocks, which are far from the reality of their complex and irregular shapes. This research tries to understand how much geometric details are needed in the numerical studies of in-stream structures. Three different representations, fully resolved geometry, porosity approximation, and solid barrier simplification, were tested and compared. It is found that the porous media model and the solid barrier model, which are computationally economic, can describe the flow dynamics only to some extent. From the calibration of drag force and wake length, it is found that the equivalent grain size $d_{50}$ in the porosity model should scale as the key element diameter for the simulated ELJ. A wake length scale analysis was performed for the semi-bounded flow around this in-stream structure near the bank. The length estimator in the literature for unbounded vegetation patches can be used with modifications. The results also show that the flow passing through the porous in-stream structure has a significant impact on mean velocity, turbulence kinetic energy, sediment transport capacity and integral wake length. Since geometrically-fully-resolved simulations are not currently feasible for engineering practices, the following suggestions are made based on this study. If the near-field and wake are important for the purpose of the structure, the well-calibrated porosity model seems to perform better than the solid barrier model. However, care needs to be taken when interpreting the results because this work also identified substantial loss of physical information with the porosity model. When the emphasis is the far field away from the structure, both the porosity model and the solid barrier model give comparable results. The second part focuses on the development of a versatile computational fluid dynamics (CFD) code which can be used to track and model the dynamic evolution of the sediment bed as the scour hole develops. The immersed boundary methodology was adopted because it can deal with large and arbitrary bed deformation. More importantly, IB method can easily deal with the interaction between evolving sediment bed and in-stream structures. One technical difficulty with IB method is that in the literature focus was not on the wall shear stress. The use of the IB methods in the literature gave very poor wall shear stress, which is important for sediment transport. The root of the problem is that the original wall functions for turbulent boundary layer flow lack smoothness due to the nonlinearity and discontinuity between the log-law layer and the laminar layer. In IB method, the wall function is enforced through IB cells. However, for complex and evolving surfaces, there is no control on where the IB cells will be located in the boundary layer. The IB cells located in the log-law layer and the laminar layer follow different functions and thus will give non-smooth wall shear distribution. To remedy this, this research introduces a new IB method with a $y^+$-adaptation wall function. The basic idea is that when an IB cell is too close to the immersed boundary, it is automatically replaced by cells in the fluid region further away from the boundary. Thus, all IB cells are in the log-law layer and they use the same function to evaluate turbulent flow quantities. As a result, the wall shear stress is much smoother. In the new IB method, the enforcement of boundary conditions is through IB cells on which the variables are reconstructed (interpolated) from their neighbouring cells with an explicit, iterative scheme. Three interpolation schemes are provided, i.e., quadratic, linear and mixed. Example cases in 1D, 2D, and 3D show the new IB method together with the $y^+$-adaptation wall function produces results compare well with theory and experiments. The third part of the thesis is the utilization of the IB method developed above and the development of a three-dimensional local scour model. The bed is treated as immersed boundary. The major components of the 3D scour model are the CFD part for turbulent flow field and the sediment transport part for updating the bed location. During the simulation, a robust and parallel interpolation scheme between 3D background mesh and 2D immersed boundary mesh is implemented. An edge-center storage method is used to address the divergence calculation problem in the Exner equation. This problem is caused by mesh non-orthogonality. One unique feature of the model that that a diffusion-based sand-slide algorithm is adopted. The relationship between sand slide and the augmented angle of repose is analyzed inside the scour. The model is validated against experimental measurements and its capability is demonstrated with a case where local scour occurs around a bridge pier with complex geometry. The demonstration shows that the model has the capability of simulating the exposure of in-stream structure foundation, which is extremely difficult if other approaches, such as dynamic mesh, are adopted.