Analysing flow characteristic of breaching embankment using linear hydrodynamic porous model
The study of the overtopping flow associated with breaching embankments is an essential part of water management, particularly for emergency planning. One of the mechanisms that triggers embankment collapse is overtopping. Therefore, it is crucial to identify the zones at risk where the overtopping...
Saved in:
Main Author: | |
---|---|
Format: | Thesis |
Language: | English |
Published: |
2021
|
Subjects: | |
Online Access: | http://eprints.utm.my/id/eprint/101751/1/ZainabMohamedYusofPSKA2021.pdf |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | The study of the overtopping flow associated with breaching embankments is an essential part of water management, particularly for emergency planning. One of the mechanisms that triggers embankment collapse is overtopping. Therefore, it is crucial to identify the zones at risk where the overtopping failure is likely to occur and where the breach might form. The nature of the failure would significantly impact the breach discharge, the variation of reservoir water levels, and the resulting water levels in the downstream valley or floodplain. This thesis presents the characteristics of flow due to an embankment breaching caused by flow overtopping. Laboratory works were carried out to observe the embankment failure, how the erosion is triggered, and factors contributing to the failure. A dimensional analysis was performed to identify the variables involved to analyse the mechanism of the embankment failure. The development of an embankment breach model using Computational Fluid Dynamics (CFD) was carried out to model the failure patterns of a breaching embankment. This required specification of the breach formation and breach widening, and prediction of the resulting breach hydrograph. In this study, the embankment was modelled as a porous medium governed by a generalised form of Darcy’s Law. The erosion is prescribed by systematically decreasing the porous embankment resistance in those areas where erosion is likely to occur linearly. Model validations were performed by comparing CFD simulations with measured data from experimental work in the laboratory for a 2D model. The Eroding models developed were conducted in 2D and 3D, using the Realizable model and the Volume of Fluid (VOF) multiphase model to identify the free elevation surface. The 2D model results have shown good agreement with experimental data for free water surface and velocity profiles over a rigid embankment. For a porous embankment, the profiles displayed reasonable accuracy with that of a Rigid Model. The validations on the 2D porous embankment models gave reasonably good agreement on temporal breach patterns and free surface flow over the breached embankment. The results showed that the overflow volume predicted was close to the theoretical value. The percentage difference was around 13%. The study considered the mesh adaption technique using a grid refinement method. The results indicated that a 10% rule of refining and coarsening produced a difference of 6% (in peak flow of the hydrograph) compared to 10% rule of refining only technique. The 3D Eroding Models allow for the inclusion of lateral breach formation to predict flow features over a breached embankment and predict a breach discharge hydrograph. Three breach shape cases were simulated, namely the side-, trapezoidal, and triangle breach shapes. As a result, parameters such as velocity vectors at the breach area, free water surface profiles, and embankment volume lost during the breaching event were produced. The Eroding Model predicted that the initially triangular shaped beach produced 24% higher peak breach discharge compared with the trapezoidal shape. Comparisons of a maximum velocity at the breached area between the 3D Eroding Models and FLOW-3D simulation ranged from 11% to 52%. Meanwhile, the FLOW-3D simulation predicted more volume lost and peak discharge compared with observed data (Case Study E1) with a percentage difference of 42.7% and 30.2%, respectively. |
---|