Seismic risk and post-earthquake recovery of older tall buildings with welded steel moment frames
Abstract/Contents
- Abstract
- Many of the existing tall buildings in the United States use seismically non-conforming welded steel moment frames (WSMFs) of the kind that experienced premature brittle fractures during the 1994 Northridge earthquake. Despite their known vulnerabilities, the vast majority of existing tall WSMF buildings have neither been evaluated nor retrofitted. To support the development of risk management policies for these structures, this study tackles the main impediments to robust quantification of risk and post-earthquake consequences in tall WSMFs by developing (1) a high-fidelity model to improve the nonlinear simulation of fracture-critical connections, (2) a detailed database of a realistic portfolio of tall WSMF buildings to increase the resolution of seismic risk assessments, (3) a screening method for pre-earthquake evaluation of tall WSMF buildings, (4) a damage evaluation method to inform post-earthquake re-occupancy and safety cordon decisions, and (5) a state-of-the-art assessment of earthquake damage, repair cost, and recovery time of tall WSMF buildings. This research presents a new computational model capable of reliably simulating fracture and post-fracture response of welded beam-to-column connections. This model uses a fiber-section element that simulates the welded flanges with single fibers and the shear tab bolted connection with multiple fibers. The flange fibers use a uniaxial stress-strain model that characterizes fracture resistance using a low-cycle fatigue damage rule based on the weld toughness—measured in terms of Charpy-V-notch (CVN) fracture energy—and weld defect size (a0). This stress-strain model also enables a realistic simulation of the connection behavior after flange fracture by capturing crack opening and closure. The stress-strain model is implemented in the OpenSees software as a uniaxial material called, SteelFractureDI. The parameters of SteelFractureDI are calibrated using a comprehensive database of 100 large-scale beam-column connection tests that were digitized and compiled as part of this study. SteelFractureDI is capable of predicting the instant of flange fracture within one half-cycle from the observed instant for 80/100 tests, compared to the 48/100 obtained with state-of-practice rotation limits (ASCE/SEI 2017). Moreover, the proposed connection model provides a more realistic distribution of flange fractures than conventional plastic hinge models due to the proper simulation of post-fracture behavior that avoids fictitious concentrations of drift. The resolution of the proposed connection model is best exploited in structural models that consider the unique features that dominate the seismic behavior of buildings. Thus, we collected a database of the structural characteristic of 89 tall steel moment frame buildings in San Francisco, which are representative of the construction practices in the western U.S. from the 1960's through early 1990's. The information in this database facilitates the creation of high-fidelity nonlinear structural models for several buildings in the database. The seismic response and collapse safety of these building are evaluated using nonlinear response history analyses (NLRHA) under suites of earthquake ground motions at various intensities. This study shows that the collapse safety varies significantly across the portfolio, from collapse risks that are comparable to that expected for current-code buildings to risks that are up to 35 times higher. Furthermore, the estimated performance of WSMFs is especially sensitive to the toughness and quality of the welded connections. These findings motivate broader coordination between structural engineers and public building departments to collect and synthesize data on weld metal toughness. The rigorous collapse metrics from this study enable the development of a rapid screening method to classify the risk of a tall WSMF building as moderate, high, or exceptionally high. This screening method empowers engineers to estimate the relative risk of a tall WSMF building using only basic information on the framing configuration and properties available from structural drawings and linear structural analyses, avoiding time-consuming nonlinear simulations. The screening method is intended to have sufficient specificity to identify the unique building features that influence collapse safety, while being simple enough to expedite the structural evaluation of tall buildings. This screening method can inform preventive decisions, such as whether a building should be prioritized for seismic retrofitting. In high-seismic regions, an earthquake may occur before buildings are retrofitted. Thus, this study also addresses the key post-earthquake issue of determining whether damaged buildings are safe to reoccupy, prior to and during repairs. Using the results of sequential NLRHA on eight WSMF models, we quantify the relationship between an "unsafe" designation using current evaluation guidelines (FEMA 352) and the damaged building's collapse safety. The results show that the FEMA 352 guidelines have two main shortcomings. First, the PGA=0.25g trigger for structural inspection is dangerously misleading for four of the eight frames in this study because more than 30% of the simulations that have a PGA < 0.25g also suffered serious damages; and second, the damage indicator proposed in FEMA 352 fails to effectively identify damage instances that reduced the building's collapse capacity by more than 20% in six from the eight WSMFs models. Recognizing these shortcomings, we propose a set of decision trees with successive single evaluations of damage indicators as a practical post-earthquake tool for tall WMSFs. The proposed decision trees consider the dominant building vulnerabilities to increase the accuracy of the tagging decision from for all the frames considered. The final portion of this study goes beyond building safety to address questions related to post-earthquake building recovery. To address this question, the FEMA P58 (2018a) performance assessment methodology is employed to quantify damage, economic losses, and functional recovery times. The results of expected annual economic loss have a narrow range from 0.4% to 1.22% of building replacement cost across the portfolio. The results of FEMA P58 simulations are further processed to estimate functional recovery time using two recently proposed frameworks for recovery simulations: TREADS (Molina Hutt et al. 2021b) and ATC-138 (ATC 2021). The results show that the average downtime of tall WSMF buildings after a 225-year earthquake is more than one year, grossly exceeding the San Francisco target of four months (SPUR 2009). The assessment methods and data developed in this study are intended to contribute to more reliable estimations of the earthquake risk posed by large inventories of existing steel frame buildings. This risk is understood as the threat to human life and the disruption to societal and economic community functions taking place in WSMF buildings. This information is further intended to support the engineering aspect of a larger and multi-disciplinary disaster risk management conversation that considers societal, economic, and ethical aspects to effectively achieve the goal of community resilience
Description
| Type of resource | text |
|---|---|
| Form | electronic resource; remote; computer; online resource |
| Extent | 1 online resource |
| Place | California |
| Place | [Stanford, California] |
| Publisher | [Stanford University] |
| Copyright date | 2022; ©2022 |
| Publication date | 2022; 2022 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Author | Galvis López, Francisco Alfonso |
|---|---|
| Degree supervisor | Deierlein, Gregory G. (Gregory Gerard), 1959- |
| Thesis advisor | Deierlein, Gregory G. (Gregory Gerard), 1959- |
| Thesis advisor | Baker, Jack W |
| Thesis advisor | Molina Hutt, Carlos |
| Degree committee member | Baker, Jack W |
| Degree committee member | Molina Hutt, Carlos |
| Associated with | Stanford University, Civil & Environmental Engineering Department |
Subjects
| Genre | Theses |
|---|---|
| Genre | Text |
Bibliographic information
| Statement of responsibility | Francisco Alfonso Galvis López |
|---|---|
| Note | Submitted to the Civil & Environmental Engineering Department |
| Thesis | Thesis Ph.D. Stanford University 2022 |
| Location | https://purl.stanford.edu/jb783fz6358 |
Access conditions
- Copyright
- © 2022 by Francisco Alfonso Galvis Lopez
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...