This paper presents the results of a comparison study performed between the numerically predicted and experimentally observed (through shaking table tests) dynamic behaviors of two scaled models of steel frame buildings structures: one symmetric and one asymmetric in plan. The shaking table tests were carried out at the Earthquake Engineering Center of the University of Bristol. The models were designed and built to be representative of steel buildings designed according to the Eurocodes 3 and 8 (EC3 and EC8). Both 1/5 scale models have a rectangular layout of 2000mm (x-direction) by 1500mm (y-direction). The models consist of two three-storey frames arranged lengthways along the plan. Interstorey height is 700mm. Additional masses (made up of lead bars) were placed on each storey in order to simulate the appropriate mass distributions, to obtain the desired symmetric and asymmetric models. The eccentricity between the center of mass and center of stiffness is about 10% of the side length along the longitudinal (x-) direction. The models were tested using as base inputs the EW and NS components of the El Centro 1940 earthquake ground motion simultaneously applied along the x- and y- directions respectively. These seismic excitations were scaled at various levels in order to first test the model in its linear elastic behavior and then to bring it to its non linear behavior. The results obtained experimentally are compared with their numerical counterparts, obtained through a three-dimensional linear and non-linear modeling, performed using 3-node quadratic spatial beam elements for the columns, 2-node linear spatial beam elements for the girders, and 4-node doubly curved shell elements for the floor slabs. The numerical and experimental comparison indicates that a careful numerical modeling of the structure (especially of the connections between columns and floor slabs) is necessary to correctly simulate and understand the experimental data.
Shaking Table Testing of Symmetric and Asymmetric Three-Storey Steel Frame Structures
TROMBETTI, TOMASO;GASPARINI, GIADA;SILVESTRI, STEFANO;
2004
Abstract
This paper presents the results of a comparison study performed between the numerically predicted and experimentally observed (through shaking table tests) dynamic behaviors of two scaled models of steel frame buildings structures: one symmetric and one asymmetric in plan. The shaking table tests were carried out at the Earthquake Engineering Center of the University of Bristol. The models were designed and built to be representative of steel buildings designed according to the Eurocodes 3 and 8 (EC3 and EC8). Both 1/5 scale models have a rectangular layout of 2000mm (x-direction) by 1500mm (y-direction). The models consist of two three-storey frames arranged lengthways along the plan. Interstorey height is 700mm. Additional masses (made up of lead bars) were placed on each storey in order to simulate the appropriate mass distributions, to obtain the desired symmetric and asymmetric models. The eccentricity between the center of mass and center of stiffness is about 10% of the side length along the longitudinal (x-) direction. The models were tested using as base inputs the EW and NS components of the El Centro 1940 earthquake ground motion simultaneously applied along the x- and y- directions respectively. These seismic excitations were scaled at various levels in order to first test the model in its linear elastic behavior and then to bring it to its non linear behavior. The results obtained experimentally are compared with their numerical counterparts, obtained through a three-dimensional linear and non-linear modeling, performed using 3-node quadratic spatial beam elements for the columns, 2-node linear spatial beam elements for the girders, and 4-node doubly curved shell elements for the floor slabs. The numerical and experimental comparison indicates that a careful numerical modeling of the structure (especially of the connections between columns and floor slabs) is necessary to correctly simulate and understand the experimental data.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
