Date of Award

2016

Document Type

Open Access Dissertation

Department

Civil and Environmental Engineering

Sub-Department

College Of Engineering and Computing

First Advisor

M. Hanif Chaudhry

Abstract

Despite numerous embankment failures around the world, in-depth assessments as well as reliable, comprehensive sets of data of such failures are limited. Detailed understanding of the earthen embankment failure process and the dominant parameters affecting the failure are needed to predict and model the breach process precisely. Therefore, a series of laboratory experiments on embankment (i.e., dam and levee) overtopping as well as erodibility measurements using the Submerged Jet test are conducted. Also, to study the influence of different model parameters (e.g., turbulence model, bedload formula), one of the dam embankment experiments is simulated. The main goal of this research is to quantitatively determine the failure process of earthen embankments due to overtopping and to investigate the effects of compaction and cohesion. Results of the current research are as follow: 1) For the non-cohesive dam breach tests with different levels of compaction, two dimensionless equations are developed to predict the variation of crest height and embankment bottom length with time as a function of dry density of the embankment material. A simple model assuming a triangular shaped embankment is proposed to predict the development of breach with time for different compaction levels of the embankment. Except during the initial stage of the failure, the match between the proposed model and the observed data is satisfactory. The proposed model for the crest height is also validated against a laboratory test reported in the literature and three dam failures in South Carolina during October 2015. 2) For the numerical simulation of the non-cohesive, non-compacted dam breach test, the bedload transport formula is the main component. A comparison between the simulated and observed breach outflow hydrograph shows that the simulations and observations are in better agreement by using the Meyer-Peter and Muller (M.P.M.) formula than the Ashida and Michiue (A.M.) formula. However, the A.M. formula predicts better the bed and water surface profiles after the initial failure stage. 3) For the Jet tests of cohesive samples with different levels of compaction, erodibility varies over a wide range and the effects of compaction energy and water content on the erodibility coefficients are investigated. 4) For the non- cohesive, non-compacted levee breach tests, breach shape with time is measured using a sliding rod technique developed by the author, and water surface velocity distribution is recorded using particle image velocimetry. Breach outflow is also estimated by using a calibrated sharp-crested weir. Erodibility coefficients in both vertical and horizontal directions are determined from the experimental results and corresponding bed-load transport formulas are proposed. 5) For the laboratory investigation of the effects of hydraulic loads on the failure of non-cohesive, noncompacted levees, non-dimensional relations are proposed for the breach top width and depth. Moreover, a trapezoidal shape model is presented to predict the breach evolution. 6) The effects of compaction and cohesion on the failure of an overtopped levee are studied. From the results of the experiments, three non-dimensional relationships are developed which express the breach widening rate, breach deepening rate, and breach eroded-load as a function of the excess Shields number and compaction energy for noncohesive levees. The proposed models of breach morphology compare satisfactorily with the results of the current laboratory tests. Furthermore, the levee material cohesion and specifically the clay content is found to be the dominant parameter controlling the failure process. Based on the experimental observations, envelope curves are presented to express the effects of compaction and cohesion on the breach characteristics (i.e., breach depth, breach top width, breach eroded volume, breach total area, and breach submerged area).

Rights

© 2016, Ali Asghari Tabrizi

Download

Share

COinS