Events Calendar

Oral Dissertation Defense

Speaker: Mehrdad Bozorgnia, CE Ph.D, Candidate

Talk Title: Computational Fluid Dynamics Analysis of Highway Bridges Exposed to Violent Waves

Abstract:
The vulnerability of coastal bridges to damage during Hurricane induced storm surge has been illustrated along the U.S. Gulf Coast in several hurricane events. The total losses during Hurricane Katrina, considering all direct and indirect losses (e.g., job losses), are estimated to exceed 100 billion dollar. Bridges revealed to be the most vulnerable critical components of the transportation system, suffering damage during hurricane induced storm surges and wave loads, and costing over 1 billion dollar to repair or replace TCLEE (2006).

The objective of this study was to calculate the hydrodynamic forces on bridge superstructure via Computational Fluid Dynamic (CFD). Three dimensional numerical wave-load model based on two-phase Navier-Stokes equations is used to evaluate dynamic wave forces exerted on the bridge deck. In order to accurately capture the complex interaction of waves with bridge deck, several millions of mesh cells are used in the simulation domain and simulations are ran on High Performance Computing and Communication (HPCC) cluster at University of Southern California.

First, CFD software was validated by simulating interaction of a solitary wave with a flat plate. The simulation results for pressure under the plate and velocities inside water were in good agreement with experimental data available from French (1970).

Second, Validated numerical model was applied to a 1:5 scale Escambia Bay bridge which was heavily damaged during Hurricane Ivan. Compared to simple flat plate problem, Highway bridge superstructures pose a unique challenge due to their complex geometries, bluff profile, and their relatively large width-to-wavelength ratio. When the added complexities of trapped air, turbulence, and structural response are incorporated, analytical solutions become impractical and the available empirical solutions based on small-scale experiments may be biased by scale effects. Simulation results were compared to experimental data available from the O.H. Hinsdale Wave Research Laboratory at Oregon State University. Influence of modeling (2D vs 3D), time step and grid refinement have been investigated. It has been determined that the two phase Navier Stokes equations are very sensitive to the choice of mesh size and time step. Some guidelines based on simulation results are developed for optimum choice of mesh size and time step for similar wave structure interaction problems.

Third, In order to evaluate scale effects in the wave bridge interaction problem, a bridge prototype with exact Escambia Bay Bridge dimensions is setup. Equivalent wave heights and period are calculated using Froude similitude laws from the wave heights and periods used in model simulations. The forces obtained from CFD simulations for prototype bridge are compared to forces calculated using Froude similitude law from model bridge simulations.

Forth, CFD simulation results for model and prototype bridge was compared with recently published AASHTO guidelines. Recently published AASHTO guidelines for coastal bridges vulnerable to storms provide a series of equations to estimate maximum quasi-steady and slamming forces due to wave impact. These guidelines are developed based on experiments conducted at University of Florida on 1:8 scale Escambia Bay bridge.

Fifth, since air entrapped between bridge girders and diaphragm was determined to be the main reason behind highway bridge failures during recent Hurricanes and
Tsunamis, two retrofitting options are evaluated in terms of their efficacy in reducing hydrodynamics forces applied to bridge superstructure. These two options were using airvents in bridge deck and using airvents in bridge diaphragms.

The ability of CFD to model a complex flow such as described in this dissertation would provide a powerful tool to predict the hydrodynamic forces under various conditions and furthermore to devise effective disaster prevention plan against bridge failure.

Advisor: Prof. J.J. Lee

Location: Kaprielian Hall (KAP) - 460

Audiences: Everyone Is Invited

Contact: Evangeline Reyes