THESIS
1997
xiii, 143 leaves : ill. ; 30 cm
Abstract
The reliability and control of structures with uncertainties are investigated in a probabilistic approach....[
Read more ]
The reliability and control of structures with uncertainties are investigated in a probabilistic approach.
The evaluation of reliability integrals commonly encountered in reliability anal-ysis is first investigated. Approximate techniques are studied, including the asymptotic method, perturbation method, Monte Carlo simulation and importance sampling simulation method.
The modeling of random fields with the stochastic finite element method is then applied to the reliability analysis of structures. Numerical studies of a cantilever structure with uncertain material property and loading pattern varying along its length show that the correlation of uncertainties has a significant effect on the response reliability and moments. In particular, assuming a full correlation structure in the material property or loading pattern (or both) of the cantilever results in conservative estimates for reliability and response moments.
Simulation of homogeneous Gaussian random fields is also considered, with applications in the simulation of ground motions modeled by stationary stochastic processes and isotropic homogeneous random fields. Simulation formulas are developed and numerical studies demonstrate the applications in the simulation of earthquake motions.
Finally, the probabilistic framework in the treatment of uncertainties is extended to the design and performance evaluation of tuned mass dampers for structures subjected to structural and loading uncertainties. The performance index takes the uncertainties in the structure and loading into account, and is directly related to the reliability of structure. Parametric studies show that uncertainties have a profound effect on the design and performance of the control device. It is thus recommended that uncertainties be properly accounted for in the designs.
Post a Comment