AN INNOVATIVE SHAKE TABLE TESTING METHODOLOGY FOR FLOOR ACCELERATION SIMULATION

Over the past three decades, most of the research on earthquake engineering of building structures was on the seismic force resisting system. Yet little research focused on the seismic behavior of the floor diaphragms and their components (such as slab, collectors, and chords). In using conventional shake table testing method to investigate the inertial force load path in floor diaphragms, there is need for multi-story test building so that the higher mode dominated floor acceleration response and the associated inertial force mechanism can be simulated in the laboratory. However, shake table testing on a multi-story test building is challenging to carry out due to the high cost, large amount of time consumed, and limited availability of facilities. In order to enhance the efficiency of the shake table testing for the seismic performance of floor diaphragms, an integrated analytical and experimental research was conducted to develop an innovative floor acceleration simulation test (FAST) methodology, which aims to drive an elastic, reusable single-story test specimen to reproduce the floor acceleration history response from any floor in a prototype multi-story building subjected to inelasticity during an earthquake. This research develops a transfer function approach to compute the required input table motion for FAST. Analytical studies show that, with a special arrangement on the test specimen configuration, the magnitude of specimen deformation can be tuned to a desired level while the FAST is subjecting a specimen to the target floor acceleration history response. In addition, the experimental verification of the FAST methodology was conducted on a half-scale, single-story steel building containing a composite floor slab by using the NHERI@UCSD large high performance outdoor shake table (LHPOST) facility. The analytical and test results confirmed the effectiveness of the proposed FAST methodology. Although the measured floor acceleration in the specimen was higher than the target response due to the overshooting of the table input motion, the expected frequency content which reflected the higher mode effects of the multistory prototype building was well reproduced.