CCEER-16-10: Slab Vibration and Horizontal-Vertical Coupling in the Seismic Response of Irregular Base-Isolated and Conventional Buildings
Loading...
Authors
Pujols, Jean C. G.
Ryan, Keri L.
Issue Date
2016-12-01
Type
Report
Language
Keywords
CCEER
Alternative Title
Abstract
Recent developments in seismic performance objectives that allow superior structural designs have highlighted the importance of providing continued functionality to nonstructural components and systems (NCSs). Modern isolation techniques have been successfully implemented in practice, as well as tested at both component and system levels, but testing of large scale isolated building models subjected to combined horizontal and vertical motions has been limited. Considerable evidence of slab vibration amplification has been reported in both fixed and isolated buildings, and studies have concluded that the response of floor slabs is sensitive to the slab vibration properties rather than the type of configuration (i.e. isolated or fixed-base). Field evidence that supports this claim is inconclusive, as strong vertical excitation has not been reported in instrumented seismically-isolated buildings. The experiments considered in this study were conducted on a full-scale, five-story, steel moment-frame building subjected to a number of 2D (horizontal only) and 3D (combined horizontal and vertical) strong earthquake records using the world's largest shake table at E-Defense. The types of building configurations tested included (1) triple friction pendulum bearings (TPB), (2) a hybrid combination of lead-rubber bearings and cross-linear bearings (LRB/CLB), and (3) conventional (fixed at the base). The datasets corresponding to each building configuration have been permanently archived as separate standalone experiments under NEES TIPS/Project No. 571, and are currently publicly accessible through the DesignSafe-CI (cyberinfrastructure) as part of the Natural Hazards Engineering Research Infrastructure (NHERI). The archiving, organization and documentation of this comprehensive dataset (about 211 GB), are described in detail. These unique, publicly accessible datasets are the backbone of this study. Follow up investigations focused on the specimen isolated with a hybrid combination of lead-rubber and cross-linear bearings (LRB/CLB), and fixed at the base. The seismic response of the buildings was investigated with particular emphasis on slab vibration amplifications in response to vertical excitation and a horizontal-vertical (H-V) coupling effect observed in both buildings. The H-V coupling appeared as a significant amplification of horizontal floor accelerations observed during 3D shaking compared to 2D, and it was partially attributed to the strong asymmetry of the building that was enhanced by supplemental mass placed at the roof to represent equipment or a roof penthouse. Median peak slab vibrations were amplified �" relative to the peak vertical shake table accelerations �" by factors ranging from 2 at the 2nd floor to 7 at the roof. The experimentally observed slab accelerations and the H-V coupling effect were accurately simulated through a 3D model of the specimen using standard software and modeling assumptions. These assumptions included the use of the insertion point method with end joint offsets to represent composite behavior of the floor system model consisting of frame elements for beams/girders and shell elements for floor slabs, as well as adequate distribution of floor masses through refined discretization. H-V coupled modes were positively identified through modal analysis, and verified with evaluation of floor spectral peaks. A supplemental study focused on the evaluation of factors that may influence slab vibration and/or induce a horizontal-vertical (H-V) coupled response of buildings with ii mass irregularities. Parameters that influence the vertical response of the floor system and subsequent H-V coupling effect were investigated through computational simulations of a 3D numerical model of a hypothetical 3-story building both base isolated with lead-rubber bearings (LRBs) and conventionally configured. The parameters investigated included superimposed mass induced eccentricities, vertical acceleration intensity, vertical stiffness of the isolators, and slab design assumptions (i.e. slab stiffness/mass variations). Induced mass eccentricities were observed to influence the vertical and H-V coupling response, but had an unpredictable influence on slab acceleration amplifications. In addition, the direct implications of vertical slab vibrations and the H-V coupling behavior on the design forces of nonstructural components and systems (NCSs) was evaluated, and design modifications that account for these effects are proposed for consideration.
Description
Report No. CCEER-16-10
Citation
Publisher
License
In Copyright