SEISMIC PERFORMANCE OF BASE ISOLATED BUILDINGS AND STRATEGIES TO MITIGATE VERTICAL ACCELERATION DEMANDS

Abstract

Elastomeric bearing constructed of rubber layers bonded to intermediate steel shim plates is a seismic isolation device used for protecting a structure against earthquake ground shaking. The rubber layers provide the low lateral stiffness required to shift the period whereas the close spacing of the intermediate steel shim plates provides a vertical stiffness that is several thousand times larger than a horizontal stiffness. However, the low vertical isolation frequency can align with the dominant frequency content of the vertical spectrum leading to significant amplification of accelerations. An analytical study was conducted to investigate the influence of the vertical component of ground shaking on the performance of nonstructural systems within multi-story seismically isolated buildings. The research focuses on two dimensional analytical models of 3 and 9-story frames. Three model configurations are explored namely: frames with conventional base isolation, frames with base isolation and viscous dampers, and lastly frames with base isolation and column isolators. Fragility curves are used to assess damage to ceiling systems in both frames. Results from analytical analyses suggests that damage to ceiling systems might be mitigated in low-rise structure using moderate shape factor elastomeric bearings and supplemental damping, and in middle rise structure using high shape factor base bearings and column isolators.

iii

Table of Contents

1. Introduction ………………………………………………………………………………………………………………….1
   • Background and Motivation ……………………………………………………………………………………1
   • Objective ………………………………………………………………………………………………………………3
   • Research significance ……………………………………………………………………………………………..4
   • Scope of Research ………………………………………………………………………………………………….4
   • Research Plan ………………………………………………………………………………………………………..4
   • Organization of thesis proposal ……………………………………………………………………………….5
2. Background information and literature review ………………………………………………………………….6
   • General …………………………………………………………………………………………………………………6
   • Background theory …………………………………………………………………………………………………6
   • Elastomeric bearings ………………………………………………………………………………………………7
   • Effect of the vertical component on structural and non-structural systems …………………..11
   • Past effort to achieve three-dimensional isolation …………………………………………………….13
   • Seismic Fragility Data of Nonstructural System ………………………………………………………14
   • Summary and Research Justification ………………………………………………………………………17
3. Research approach ………………………………………………………………………………………………………18
   • General ……………………………………………………………………………………………………………….18
   • ……………………………………………………………………………………………………………..18
   • Prototype superstructure ……………………………………………………………………………………….20
   • Analytical model ………………………………………………………………………………………………….22
   • Earthquake ground motion …………………………………………………………………………………….28
   • Design of protective system …………………………………………………………………………………..29
   • Parametric study…………………………………………………………………………………………………..31
   • Formulation of fragility curves ………………………………………………………………………………32
   • Estimate probability of failure for ceiling systems ……………………………………………………35
4. Analytical results ………………………………………………………………………………………………………..37
   • General ……………………………………………………………………………………………………………….37
   • Results from parameter study of the base isolated frames ………………………………………….37
   • Results for base isolated frames with supplemental damping …………………………………….46
   • Results for base isolated frames with column isolators ……………………………………………..49
   • Summary …………………………………………………………………………………………………………….54
5. Summary and conclusions ……………………………………………………………………………………………55
   • Summary …………………………………………………………………………………………………………….55
   • Limitations of the study ………………………………………………………………………………………..55
   • Conclusions …………………………………………………………………………………………………………56

References ……………………………………………………………………………………………………………………..57

Chapter 1

Introduction

1.1 Background and Motivation

Seismic isolation is a technique used to protect structures and nonstructural systems from the damaging effect of horizontal earthquake ground motion. This is accomplished by decoupling the superstructure from the substructure using isolation bearings with low lateral stiffness. The low lateral stiffness is required to shift the period of the structure into the long period range, e.g., 2.5s to 4s. This period shift translates into reduced spectral acceleration demands as illustrated by the shaded region on the elastic response spectra shown in Fig. 1.1a (generated using the recorded earthquake ground motions detailed in Chapter 3). The period shift, however, does result in increased displacement demands concentrated at the isolation interface that must be accommodated by the seismic isolation bearings. Seismic isolation bearings fall into two general categories, namely: (1) elastomeric and (2) sliding. Each type has low lateral stiffness to achieve the desired period shift and a large vertical stiffness to support the gravity and vertical forces imposed on during service and seismic loading, respectively.

For elastomeric bearing, the total thickness of rubber provides the low lateral stiffness whereas closely spaced steel shim plates provide a vertical stiffness that is several thousand times larger than the horizontal stiffness. Consequently, the vertical isolation frequencies, typically ranging from 0.03 to 0.15 seconds, can align with the dominant frequency content of the vertical spectrum leading to significant amplification of accelerations. For sliding bearings, the low lateral stiffness is achieved using bearing material such as polytetrafluoroethylene (PTFE) type material mated to polished stainless steel providing a low sliding coefficient of friction ranging from 0.07 to 0.18 (Mohka et al. 1990a, b) with values as low as 0.03 (EPS 2012). However, the materials used to construct sliding bearings produce a large vertical stiffness and thus a low vertical isolation period, typically around 0.03 seconds (EPS 2012).

The short vertical isolation period produced by typical elastomeric and sliding isolation bearing, e.g., 0.03 to 0.15 seconds, can align with dominant frequency content of vertical spectrum as shown by the shaded region plotted in Fig. 1.1b superimposed on spectra generated using the vertical components of recorded earthquake ground motions. Therefore, bearings widely used for seismic isolation provide isolation only in the horizontal direction and are not designed to protect against vertical ground shaking.

 

 

 

Figure 1.1. Elastic, 5% damping, earthquake response spectra from Bin 1: (a) horizontal component; (b) vertical component

 

Observations from reconnaissance following the 1995 Kobe earthquake in Japan and the 1994 Northridge earthquake in the United States indicate that damage to structural and nonstructural components in fixed-base structure were due to demands imposed on these elements from the vertical components of earthquake ground shaking (Papazoglou and Elnashai 1996). Other studies (Kageyama et al. 2004; Yoo et al. 1997) suggest the larger vertical stiffness provided by isolation systems composed of elastomeric bearings could result in acceleration demands that are greater than those in an equivalent non-isolated structure. This suggests structures isolated on bearings with large vertical stiffness could sustain damage to structural and non-structural components due to the vertical component of ground motion.

Recently, two independent test programs on full scale isolated buildings with weights of

5,000-10,000 kN have been conducted at the National Institute for Earth Science and Disaster Prevention (NIED) E-Defense shaking table of Japan (Sato et al. 2010; Ryan et al. 2012). While the influence of vertical excitation has not yet been reported from the Sato et al. tests (2010), Ryan et al. (2012) observed damage to suspended ceilings. Observed damage included ceiling panels falling and in the most extreme cases localized failures of the ceiling grid support system (Soroushian et al. 2012).

1.2 Objective

To date, no comprehensive analytical or experimental study has been conducted to understand the impact of the vertical component of ground shaking on the performance of nonstructural systems within multi-story seismically isolated buildings. Yet, damage to these systems could adversely affect the post event functionality of critical facilities such as hospitals and emergency centers for which seismic isolation is typically employed (Sato 2011).

The first objective of this study is to investigate how the isolation system’s properties, specifically vertical stiffness and vertical damping, affect vertical acceleration demands in the upper levels of the building and the probability of damage to suspended ceiling systems. Suspended ceiling systems were chosen as the nonstructural component because damage to this nonstructural system was observed during recent full scale testing of a base isolated structure (Soroushian et al. 2012). Furthermore, seismic fragility data for suspended ceiling systems is available in the literature (Badillo et al. 2006) though a number of assumptions are required to apply the data to this study as described in subsequent sections.

Secondly, past efforts to achieve three-dimensional isolation have been investigated as will be discussed in the literature review section of this proposal (Tajirian et al. 1990; Huffmann 1984; Kageyama et al 2002 and Takahashi et al. 2008). However, each of these strategies has concentrated on the horizontal and vertical flexibility at the base of the structures. In structures where the center of mass is elevated above the plane of isolation, this vertical flexibility translates into reduced rocking periods of vibration that increases the participation of rocking response, in turn increasing horizontal and vertical acceleration demands in the upper levels of the building. Furthermore, decreasing the vertical stiffness of the isolation device to the point of achieving vertical isolation typically requires using a device with limited stability and is difficult to achieve in practice. Rather than concentrating vertical flexibility at the base to achieve threedimensional isolation, a strategy is proposed in this study to distribute the vertical flexibility up the height of the building. One possibility to achieve this “distributed vertical flexibility” could be through the use of elastomeric laminated bearings that are laterally restrained, similar to pot bearings, installed at various locations up the height of the building. The effect of supplemental vertical damping located at the base of the building will also be explored.

1.3 Research Significance

First, this study offers a systematic analytical parametric study of the influence of bearing shape factor on vertical acceleration demands in multi-story seismically isolated frames. Secondly, this study proposes a use of non-structural fragility data to estimate the likelihood of damage to non-structural components (namely suspended ceiling systems) for evaluation of various systems. Finally, this study provides a development and demonstration of the vertically distributed flexibility concept.

1.4 Scope of Research

This research is limited to analytical studies of two-dimensional, three-story and ninestory models based on the SAC Los Angeles buildings from Gupta and Krawinkler (1999). There are three system configurations considered in this research. The first is structures with conventional base isolation; the second is structures with conventional base isolation and supplemental vertical viscous dampers located at the plane of base isolation. The third configuration is base isolation with vertically distributed flexibility. Finally, fragility curves adapted from Badillo et al. (2006) were used to assess damage due to vertical component of earthquake excitation to ceiling systems and to evaluate the three configurations.

1.5 Research Plan

The objectives of this research will be achieved through the following research plan consisting of four tasks:

1. Perform an analytical parametric study to systematically investigate the influence of the isolation system properties, specifically vertical stiffness, on the vertical acceleration demands in the upper stories of seismically isolated buildings.
2. Assess the probability of damage to suspended ceilings in the upper stories of seismically isolated buildings using seismic fragility data.
3. Explore new systems to minimize damage to suspended ceiling systems in seismically isolated buildings.
4. Using fragility curves to verify the efficacy of the new systems through an analytical study.

1.6 Organization of thesis proposal

This proposal is organized into the following seven sections: (1) Introduction; (2) Background information and literature review; (3) Research approach; (4) Analytical results; (5) Summary and conclusions; (6) References; (7) Appendices.

SEISMIC PERFORMANCE OF BASE ISOLATED BUILDINGS AND STRATEGIES TO MITIGATE VERTICAL ACCELERATION DEMANDS

Sharing is caring!