1. Introduction
This paper describes three-dimensional FE analysis for a six-story reinforced concrete(RC) wall frame building specimen in a shaking table test. Pretest and posttest analyses were conducted here to investigate analytical accuracy of FE analysis for nonlinear dynamic response of RC buildings. Furthermore, the structural performance of each primary seismic resistance member was investigated using the analytical results.
2. Outline of analyses
Two kinds of analysis model were used to evaluate influence of modeling methods to analysis accuracy. One was a solid model which reproduce configuration of specimen in detail by hexahedral elements, as shown in Figure 3(a). The other one was a shell model by multilayer shell elements which assumed plane stress problem in each layer, as shown in Figure 3(b).
3. Analysis results
Remarkable differences between analysis results of the shell model and that of the solid model were not found as shown in Figure 4 and Figure 8. The pretest analyses satisfactorily predicted the failure mode and nonlinear response of the specimen at an excitation, in which the specimen reached its maximum capacity, as shown in Figure 4 and Figure 6. However, for the excitations performed before the specimen had reached its maximum capacity, the pretest analyses tended to overestimate the displacement responses. Posttest analyses were conducted with the input conditions and material properties measured in the test. The posttest analyses reproduced the test behavior from elastic point up to failure with improved accuracy as shown in Figure 8 and Figure 9.
4. Study of structural performance of primary seismic resistance wall
Shear span ratio, maximum capacity, and skeleton curves of each primary seismic resistance walls were investigated using the analytical results. The findings of this study are as follows.
1) It was found by investigating elemental stresses that shear span ratios of the walls were less than 1.0.
2) Although shear capacities of the walls by FE analysis corresponded well with those calculated by standard minimum evaluation formula, were less than standard mean evaluation formula.
3) Although sliding shear capacities of walls calculated by evaluation method of previous studies corresponded well with maximum shear capacities by FE analysis for 1st story walls, were far higher than those for 2nd story walls. As the reason for that, it was guessed that boundary conditions of 1st story walls were different from those of 2nd story walls. As the results, the boundary condition should be considered on evaluating the sliding failure of walls.
4) Skelton curves of relationships of shear force and drift angle of the walls, which was obtained by FE analysis, could be evaluated well by previous evaluation method.
