Date of Graduation

12-2007

Document Type

Thesis

Degree Name

Bachelor of Science in Civil Engineering

Degree Level

Undergraduate

Department

Civil Engineering

Advisor/Mentor

Selvam, R. Panneer

Committee Member/Reader

Grimmelsman, Kirk A.

Committee Member/Second Reader

Heymsfield, Ernest

Abstract

The flow around suspension bridge decks was traditionally investigated using wind tunnel experiments. Computer modeling, an alternative to wind tunnel testing is now being used to simulate the wind flow around a bridge cross section. This alternative is less expensive and takes less time compared to wind tunnel experiments. The motion of flow around a bridge deck is described by a set of partial differential equations called Navier-Stokes (N-S) equations. Flow features such as velocity, pressure and vorticity are predicted by solving these equations. These flow features are then used to compute flow parameters such as aerodynamic drag, lift and moment coefficient; Cd, Cl and Cm and Strouhal number St. In 2001, Selvam and Govindaswamy created two computer models. These computer models were used to compute flow parameters using quadrilateral finite elements (structured meshes). The main problem with both computer models was creating a proper grid around the bridge section and controlling the error within the solution. To address this problem, automatic generated triangular grids (unstructured grids) are used in the existing models. Adaptive finite element techniques and error estimation algorithms are also used in the model. Error estimation algorithms determine the error in the FEM mesh. Adaptive techniques remesh the grid around the bridge section whenever the calculated error exceeds a specified limit. In this research, the influence of specified permissible errors and grid sizes on the accuracy of flow parameters (drag coefficient and Strouhal number) is studied. It is found that the computed drag coefficient and Strouhal number for Great Belt East Bridge Section are improved using the adaptive finite element model.

Keywords

Civil engineering; Structural engineering

Share

COinS