Seismic Demands for Nondeteriorating Frame Structures and their Dependence on Ground Motions

The objective of this study is to improve the understanding of behavior patterns and the quantification of seismic demands for nondeteriorationg regular frames subjected to ordinary ground motions. In this study, the term ordinary refers to ground motions that are recorded at distances greater than 13 km from the fault rupture, do not exhibit pulse-type characteristics and are recorded on stiff soil sites. Engineering demand parameters (EDPs) of interest include roof and story drifts, local deformations, absolute floor accelerations and velocities, story shears and overturning moments, and energy terms, which are obtained by means of nonlinear time history analyses. Since nondeteriorating frames are used, the EDPs of primary interest are those that correlate best with structural, nonstructural and contents damage at performance levels related to dollar losses and downtime. A relational database management system is used to perform statistical evaluation of EDPs and establish relationships between structural and ground motion parameters. The primary intensity measure, IM, used is the spectral acceleration at the first mode of the structure, Sa(T1).

The emphasis of this study is on quantification of EDPs for performance evaluation, but includes a discussion of issues related to the design of components that need to be protected to avoid brittle failure in the response, e.g., columns in a moment resisting frame. An exploration of probabilistic evaluation of EDPs is summarized, in which EDP hazard curves are developed and based on numerical integration procedures and closed form solutions. The use of global collapse fragility functions (for a frame in which P-Delta causes dynamic instability in the response) along with an IM hazard curve to estimate the mean annual frequency of collapse is explored and illustrated in an example.

This study has provided much insight into the inelastic dynamic response characteristics of moment resisting frames and the statistical properties of important engineering demand parameters. The data from the extensive nonlinear analyses of various frames with variations in their properties is stored in a database management system and can be exploited to obtain comprehensive statistical information on other EDPs of interest in the performance evaluation of frame structures.