Loading...

A Probabilistic Integrated Framework Compatible with the Reliability Methods for Resilience-Based Design of Structures

Sangaki, Amir Hossein | 2021

0 Viewed
  1. Type of Document: Ph.D. Dissertation
  2. Language: Farsi
  3. Document No: 57171 (09)
  4. University: Sharif University of Technology
  5. Department: Civil Engineering
  6. Advisor(s): Rahimzadeh Rofooei, Fayaz; Vafai, Abolhassan
  7. Abstract:
  8. This thesis presents a probabilistic integrated framework with a set of probabilistic models compatible with the reliability methods for resilience-based design (RBD) of structures. In the present study, resilience has been defined as the ability of a system to maintain its functionality with minimal disruption and to recover the lost functionality quickly. Therefore, the resilience-based design, which considers the recovery process after the earthquake to ensure continued operation (if desired) and liveable conditions immediately after the earthquake, goes beyond the design based on the current building codes and the performance-based design. In the current thesis, The resilience curve (RC) which displays the probability of the resilience index falling below a specific value and can be used in RBD to design a building based on its resilience, has been introduced. The computation of the probability distribution of the resilience index using reliability methods with predictive probabilistic models to generate the resilience curve is one of the innovations of this research. Models for earthquake occurrence, magnitude and location, ground-shaking intensity, response, damage, loss of functionality, recovery time, recovery process, and resilience are all required for the probabilistic seismic resilience estimation of structures, which are proposed in the present study. In the following, the proposed framework and models are implemented in Rt software and their application has been demonstrated by generating a resilience curve for a typical four-story concrete moment-resisting frame building adopted from Haselton and Deierlein, as a case study. This shows that using the proposed framework, it is possible to calculate the probability distribution of the resilience index and consider the effects of an unlimited number of uncertainties. Also, in this research, the acceptable design thresholds for various structures inspired by existing regulations, performance-based design methods, and past studies have been calculated and suggested, if the probability of the low and medium importance structure resilience index falling below 99.95%, 99.996% for high importance structure, 99.998% for very-high importance structure, is 10% or less in the resilience curve (RC); then the structure is resilient and its design based on resilience is acceptable. Finally, in this research, a probabilistic framework compatible with the reliability method for designing structures based on resilience has been presented. The possibility of considering the effects of an unlimited number of uncertainties, the impact of the number of earthquake events that might endanger the resilience of the structure, is one of the unique advantages of the framework presented in this research in comparison with other resilience-based design frameworks. In the final part of the research, a ten-story steel residential apartment has been designed using the presented framework, and it has shown that if the 10-story residential apartment is designed based on existing regulations, it would not possess the necessary resilience. while, if the design of the apartment is done based on the resilience method, although the steel amount utilized in the structure will increase within 7% and as a result, the cost of building the apartment will increase; it is eventually cost-effective and helps to achieve a resilient structure beyond the goals of the current design regulations with minimal additional investment
  9. Keywords:
  10. Resilient Design ; Resilience Index ; Reliability ; Probabilistic Methods ; Resilience-Based Design Criteria ; Resilience Curve (RC)

 Digital Object List

 Bookmark