05

0.0 -I—1—1—1—1—I i..-.-|'___l.-.'-.il—1—1—1—1—I—' ' i'1'ii —|

Period, s

FIGURE 21.20 Original design spectrum, scaled spectrum, and spectra of selected ground motions after scaling.

Component demands from each time-history analysis are recorded, and the maximum values from each earthquake are used to estimate plastic rotation demands. Since seven records were used in the simulation, both FEMA-356 and ATC-40 stipulate that mean values of the peak demands are an adequate measure of the seismic demand. Table 21.16 summarizes the peak and average plastic rotation demands on the columns in the first story for all seven earthquakes. Table 21.16 also shows the maximum demands. This observation is crucial to our understanding of the process governing the choice of earthquake records for an NDP evaluation. If, for example, only three records were used in the evaluation, and Earthquake no. 3 was among the three records, then the seismic demands on the structure would be controlled by the peak values rather than the mean estimates. The maximum demands are considerably different from the mean values, suggesting that engineers should use caution when opting to select mean demands when evaluating performance limits.

In the present case, since all seven data points are used, performance limit states are going to be evaluated on the basis of mean demand estimates. The resulting performance of the building is shown in Figure 21.21. Though only the mean values were used, the NDP results indicate that several columns do not pass IO criteria but do pass LS criteria. Since LS performance is achieved for all elements, the trial design is acceptable.

TABLE 21.16 Peak and Mean Plastic Rotation Demands on First-Story Columns Using NDP
0 0

Post a comment