3D fluid modelling of the O.O. Madsen Bridge during flood

Knipe, Russell Alexander (2014) 3D fluid modelling of the O.O. Madsen Bridge during flood. [USQ Project]

[img]
Preview
Text
Knipe_2014.pdf

Download (4Mb) | Preview

Abstract

The O.O. Madsen Bridge in Warwick experiences severe flood debris blockage in the guard rails which the public believes is leading to increased flood depths upstream when the bridge becomes overtopped by flood water. The effects of the debris blocked guard rails were investigated in a 2D flood model of the Condamine River. The study concluded that depths immediately upstream of the bridge decreased in the order of 0.10-0.20m with the removal of the debris blocked guardrails. Additionally, a head loss of 0.5m was experienced over the bridge in a 100 year Average Recurrence Interval (ARI) flood. In order to limit computational times 2D flood models are often made as coarse as they can be while remaining accurate over a large area. While they reflect large scale flow patterns accurately they may not be accurate for smaller objects like the O.O. Madsen Bridge. In order to verify the findings of the 2D flood model a 3D Computational Fluid Dynamics (CFD) model of the bridge was created. ANSYS Fluent was used to model a section of the bridge the width of the centre to centre distance between the piers. A 100 year ARI flood with a depth of 7m and a velocity of 1.5m/s was chosen as the input. The Open Channel settings in the Volume of Fluid method were used to solve the two phase flow and model the surface of the water. The simulation was run twice; once with 100% debris blocked guard rails that allowed no flow to pass through, and once with no guard rails, in a similar fashion to the 2D flood model. The model found there was a 0.08m increase in depth upstream and a 0.09m decrease in depth downstream with the debris blocked guard rails, and no change in depth when no guard rails were present. The data shows there is certainly a change in depth, however it was under the ±0.13m limit of confidence for the model due to the size of the mesh. There were several other limitations to the model which include a lack of validating data, the behaviour of the bridge as a whole and the effect of the boundary and initialisation conditions not being tested. Plots of shear stress on the bed of the river found that the debris blocked guard rails have an impact on the degree of erosion experienced around the bridge pier, increasing shear stress by up to 25%. Although the model was not able to accurately predict the change in depth it serves as a good starting point in understanding the effect of the guard rails on flows around the O.O. Madsen during flood.


Statistics for USQ ePrint 27275
Statistics for this ePrint Item
Item Type: USQ Project
Item Status: Live Archive
Additional Information: Bachelor of Engineering (Civil) project.
Faculty / Department / School: Current - Faculty of Health, Engineering and Sciences - School of Civil Engineering and Surveying
Supervisors: Wandel, Andrew
Date Deposited: 09 Sep 2015 04:57
Last Modified: 10 Nov 2016 02:40
Uncontrolled Keywords: computational fluid dynamics; CFD; flood models; bridge; Warwick; O. O Madsen bridge; flood debris; flood waters; Condamine River; 100 year average recurrence interval; ARI; 2D flood models; ANSYS Fluent
Fields of Research : 09 Engineering > 0905 Civil Engineering > 090502 Construction Engineering
09 Engineering > 0905 Civil Engineering > 090506 Structural Engineering
12 Built Environment and Design > 1204 Engineering Design > 120404 Engineering Systems Design
URI: http://eprints.usq.edu.au/id/eprint/27275

Actions (login required)

View Item Archive Repository Staff Only