Stochastic prediction of collapse of building structures under seismic excitations

Deniz, Derya

Permalink

https://hdl.handle.net/2142/73033 Copy

Description

Title

Stochastic prediction of collapse of building structures under seismic excitations

Author(s)

Deniz, Derya

Issue Date

2015-01-21

Director of Research (if dissertation) or Advisor (if thesis)

• Song, Junho
• Hajjar, Jerome

Doctoral Committee Chair(s)

Song, Junho

Committee Member(s)

• Hajjar, Jerome
• Andrawes, Bassem
• Fahnestock, Larry A.

Department of Study

Civil & Environmental Eng

Discipline

Civil Engineering

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• Structural Collapse
• Collapse Experiments
• Collapse Prediction
• Collapse Probability
• Fragility Curves
• Uncertainty
• Incremental Dynamic Analysis
• Seismic Analysis
• Earthquake
• Performance–Based Earthquake Engineering
• Damage Measure
• Intensity Measure
• Ground Motions
• Energy Analysis
• Collapse Criterion
• Collapse Limit-State
• Gravity Energy
• Seismic Energy

Abstract

Modern seismic design provisions help enhance life safety of building occupants during a strong earthquake-shaking event by ensuring acceptably small likelihood of structural collapse. Therefore, accurate estimate of collapse likelihood of buildings under seismic excitations has recently become critical in efforts to promote hazard-resilience of the society, especially in developing national building codes, regional emergency response plans, and risk management strategies. Despite recent advances in static and dynamic nonlinear constitutive modeling of such structures, accurate prediction of structural collapse with systematic incorporation of uncertainty still remains a question, especially for structural evaluation and design of actual structures. The most commonly used approach to assess the collapse capacity of structures under extreme earthquakes is based on the concept of incremental dynamic analysis (IDA; Vamvatsikos and Cornell, 2002). Uncertainties in structural properties and applied ground motions can be integrated into probabilistic description of structural collapse performance by adopting the probabilistic basis of performance–based earthquake engineering (PBEE) framework together with IDA. The maximum inter-storey drift ratio (IDR) is often selected as the measure to represent the global behavior of structural system in the PBEE framework (Cornell et al., 2002). Likewise, assumed threshold values based on IDR or on slope of IDA curve between IDR and elastic spectral acceleration are most commonly used limit-states to identify structural collapse capacity. However, collapse assessment approaches based on IDR may not accurately represent the overall collapse behavior of structural systems due to redistribution and variation of damage within the structure. Moreover, collapse prediction is found to be sensitive to such subjective collapse limit-states based on the assumed threshold values. Characterization of overall cumulative (i.e., load-path dependent) collapse performance of structures considering aforementioned uncertainties is needed for accurate and reliable collapse risk assessment. Since energy parameters at system-level are aggregated quantities considering redistribution and variation of each individual component-damage within the structural system, they can be excellent indicators to represent total severe structural damage history due to cyclic-loading just before collapse. This paper therefore focuses on energy-based collapse analysis of structures to assess seismic collapse risk of structures. A new energy-based collapse limit-state is first defined to predict collapse in terms of dynamic instability due to loss of structural resistance against the gravity loads, instead of the behavior of the IDA curves. Using the new collapse limit-state, key descriptors that govern collapse capacity are identified for more effective risk assessment. Moreover, a probabilistic approach in collapse assessment is presented for systematic treatment of uncertainties in the ground motion time histories and integration with performance-based earthquake engineering (PBEE) framework. First, nonlinear dynamic analyses are performed for experimental case studies reported in the literature (Kanvinde, 2003; Rodgers and Mahin, 2004; Lignos et al., 2008) by use of OpenSees, an object-oriented software framework developed by Pacific Earthquake Engineering Center (PEER). Using OpenSees computational models validated by corresponding experimental results, new dynamic-instability-based collapse limit-state is developed in terms of energy from the input ground motions and the gravity loads. The selected case studies are then used to test the new collapse limit-state and to identify key parameters that govern the collapse of a structural system. Next, the most effective collapse descriptor representative of structural global behavior history is developed as an equivalent velocity ratio of the system’s dissipated energy to input seismic energy. Using the developed collapse limit-state and new velocity-ratio collapse descriptor, a new method is established to construct collapse fragility models for reliable probabilistic evaluation of structural collapse, considering the uncertainties in both global demand and capacity of the structural system. Finally, the effect of earthquake characteristics and structural parameters on the collapse capacity is investigated for the purpose of estimating and improving structural reliability against collapse.

Graduation Semester

2014-12

Permalink

http://hdl.handle.net/2142/73033

Copyright and License Information

Copyright 2014 Derya Deniz

Owning Collections