Electronic Thesis and Dissertation Repository

Thesis Format

Monograph

Degree

Master of Engineering Science

Program

Civil and Environmental Engineering

Supervisor

Sadhu, Ayan

2nd Supervisor

Bitsuamlak, Girma T.

Co-Supervisor

Abstract

Damping estimation is a critical task to perform during the design of slender structures, or for existing structures. This is to ensure the response of the structure is within allowable limits and to determine if additional damping is necessary from auxiliary devices. If a slender structure is experiencing wind loading, a phenomenon known as aerodynamic damping arises, which has the potential to reduce the damping of the structure. The most efficient method to estimate aerodynamic damping is to use a system identification technique, which requires only the input forces and output response of the structure. This thesis describes how to estimate aerodynamic damping ratios of concrete chimneys using a sophisticated output-only system identification technique known as Second-Order Blind Identification. Wind fields generated using drag and lift coefficients and computational fluid dynamics (CFD) are applied to a finite-element concrete chimney model in both along-wind and across-wind directions. The time-series of the wind field is simulated using the power law for the mean wind speed and the von Karman spectrum for the turbulence. Total damping estimates are acquired at various wind speeds and modes in both directions, which are compared to the theoretical values. Aerodynamic damping is acquired by subtracting the structural damping, found using a free vibration test, from the total damping estimate. The aerodynamic damping estimates using drag and lift coefficients are compared with the CFD estimates. It is found that aerodynamic damping in the along-wind direction is always larger compared to the across-wind direction. Also, damping estimates using CFD often exhibit higher values than the wind field simulated using drag and lift coefficients. A general discussion on the results, potential errors, and future work for further research is provided.

Summary for Lay Audience

Structures are affected by natural wind, especially when the wind takes on more chaotic forms such as storms, tornados, and hurricanes. Tall and slender structures are more vulnerable to wind as they exhibit higher deflections and less resistance to motion. The resistance to motion is a critical component when designing a slender structure or when analyzing the performance of an existing structure. Structural parameters such as mass and stiffness determine the resistance of motion, which is easily implemented during the design of a structure. However, this is not usually the case for existing structures, where mass and stiffness may not be known. Therefore, the statistics of a structure’s motion during wind loading can be used to build a mathematical model of its vibration patterns. These vibration patterns contain information about the structure such as the rate at which its motion decreases. A state-of-the-art statistical method is implemented in this paper to analyze the vibration patterns of a slender chimney model and determine the rate at which its motion decreases. The chimney model is subjected to different types of wind loading to examine the effects of wind actions on structures and demonstrate the robustness of the proposed statistical method. The statistical method is efficient at identifying the motion of the chimney model, meaning there is confidence in implementing the method with structural design and monitoring of existing structures.

Download

Share

COinS