Year

2018

Season

Summer

Paper Type

Master's Thesis

College

College of Computing, Engineering & Construction

Degree Name

Master of Science in Civil Engineering (MSCE)

Department

Engineering

NACO controlled Corporate Body

University of North Florida. School of Engineering

First Advisor

Dr. Raphael Crowley

Second Advisor

Dr. Adel ElSafty

Third Advisor

Dr. Don Resio

Department Chair

Dr. Murat Tiryakioglu

College Dean

Dr. Mark A. Tumeo

Abstract

In recent years, a number of researchers have applied various computational methods to study wind wave and tsunami forcing on bridge superstructure problems. Usually, these computational analyses rely upon application of computational fluid dynamic (CFD) codes. While CFD models may provide reasonable results, their disadvantage is that they tend to be computationally expensive. During this study, an alternative computational method was explored in which a previously-developed diffraction model was combined with a previously-developed trapped air model under worst-case wave loading conditions (i.e. when the water surface was at the same elevation as the bottom bridge chord elevation). The governing equations were solved using a finite difference algorithm in MATLAB for the case where the bridge was impacted by a single wave in two dimensions. Resultant inertial and drag water forces were computed by integrating water pressure contacting the bridge superstructure in the horizontal and vertical directions, while resultant trapped air forces (high-frequency oscillatory forces or sometimes called “slamming forces” in the literature) were computed by integrating air pressure along the bottom of the bridge deck in the vertical direction. The trapped air model was also used to compute the buoyancy force on the bridge due to trapped air. Results were compared with data from experiments that were conducted at the University of Florida in 2009. Results were in good agreement when a length-scale coefficient associated with the trapped air model was properly calibrated. The computational time associated with the model was only approximately one hour per bridge configuration, which would appear to be a significant improvement when compared with other computational technique

Share

COinS