&&ReWrAp:HEADERFOOTER:0:ReWrAp&&

Date of Award

9-2010

Degree Type

Thesis

Degree Name

Master of Applied Science (MASc)

Department

Civil Engineering

Supervisor

Mike Tait

Language

English

Abstract

Tuned-liquid dampers (TLDs) can be used as vibration absorbers for tall buildings. The ability of TLDs to decrease a building's motions is highly dependent on them being tuned to the building's natural frequency. In the present study natural frequency and damping estimates are determined for a 187 m tall building equipped with TLDs. A properly tuned structure-TLD system acts as a coupled two-degree-of-freedom (2DOF) system. In this study several MATLAB (2009) programs were developed in order to determine the dynamic properties of both single-degree-of-freedom (SDOF) and 2DOF systems. These programs Were based on the statistical maximum likelihood (ML) and least squares (LS) methods. The ML programs are based on earlier work by Montpellier (1997) and the LS programs were developed independently. All of the programs were verified using spectral and time-history data with known dynamic properties. The results of the programs were also compared to results generated by the well-known half-power bandwidth and random decrement methods: The ML and LS programs were found to produce results that were superior to the half-power bandwidth method and comparable to the random decrement program for SDOF systems. The ML and LS programs are superior to the half-power bandwidth and random decrement programs for analyzing coupled 2DOF systems as they are able to determine distinct property estimates for each mass. The natural frequency of the building studied for this project was found to be significantly higher than predicted at the time of its design. Thus the TLDs were not optimally tuned to the building's actual natural frequency and as a result the studied building was behaving as a SDOF system. Therefore the dynamic properties estimated in this study are those of a SDOF system. However, the methods developed herein could be applied to a 2DOF system in the future.

McMaster University Library

Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS