Date of Award

5-2009

Document Type

Thesis

Degree Name

Master of Science (MS)

Legacy Department

Civil Engineering

Advisor

Nielson, Bryant G

Committee Member

Kaye , Nigel B

Committee Member

Pang , WeiChiang

Abstract

Recent hurricane damages have devastated coastal communities and focused national attention on hurricane damage mitigation. Structural damage is one of the most significant impacts of a hurricane. Even small levels of structural damage can result in large economic losses; when gaps open in the roof system, rain water can leak in, ruining the contents of the structure and rendering it uninhabitable although it is still standing. In order to prevent these secondary damages from occurring, a better understanding of the roof system behavior is essential. This research aims to ascertain the behavior of the roof system by determining the influence of variable stiffness in the roof-to-wall connection on system behavior and to develop and propose a method for determining the reliability of a roof system typical to low-rise residential wood construction under wind loads.
Monte Carlo simulations were run on a computer model of the roof system using probability density functions for both structural parameters and load variables. The goal of these simulations was to determine the effect of variable connection stiffness and wind zone discretization on the reliability of the roof system. One significant development this study utilized was an analytical connection model for the roof-to-wall connection, capable of shedding load past a randomly generated capacity value, taken from previous research. Sheathing wind loads were modeled as a lognormal variable and generated within the constraints of a correlation matrix.
Results were obtained utilizing this Monte Carlo simulation. The system reliability was calculated as approximately 0.95 for a wind speed of 100 mph and 0.62 for a wind speed of 130 mph. The study's results suggested that considering the variability in connection stiffness had little effect on the system reliability. The level of correlation between pressures on the roof, however, was shown to have a significant effect on the system reliability.

Share

COinS