Modeling of Assessment of Seismic Performance of Composite Frames with Reinforced Concrete Columns and Steel Beams

Composite moment frames consisting of steel beams and reinforced concrete columns (so called RCS moment frames) are one of several types of hybrid systems gaining acceptance as cost-effective alternatives to traditional steel or reinforced concrete frames for seismic design. New design standards for composite moment frames have recently been introduced in the United States, and composite RCS frames have been one focus area investigated as part of Phase 5 (Composite and Hybrid Structures) of the US-Japan Cooperative Earthquake Research Program. This research presents an extensive and pioneering analytical study whose focus is on the seismic behavior of composite frames with the objectives to (1) develop and improve existing analytical models and techniques for the nonlinear inelastic static and time history analyses of composite RCS moment frames, (2) propose damage indices and performance criteria to assess seismic performance of such frames, (3) apply accurate nonlinear analysis methods to evaluate building performance under varying seismic hazards, (4) develop and correlate stability limit states to performance levels suggested by modern seismic codes, and (5) investigate response dependency on ground motion parameters so as to reduce the uncertainty in estimating median response. The ultimate goal is to achieve broader acceptance of RCS frames in high seismic regions by demonstrating their reliability through a modern performance-based methodology.

Our approach toward establishing a performance-based design basis for composite RCS frames involves both evaluation of seismic damage indices with test data on member and connection response and comparative behavioral studies between RCS and conventional structural steel moment frames. Trial designs of six- and twelve-story RCS and steel framed buildings are developed to exercise the latest seismic design criteria and standards in the United States including the recently approved International Building Code (IBC 2000) and the 1997 AISC Seismic Provisions. Nonlinear static and time-history analyses are run under two sets of earthquake records (general versus near-fault records with forward directivity) that were selected and scaled to different hazard levels representative of performance levels ranging from immediate occupancy to near collapse. Peak and cumulative performance (i.e., damage) indices are then developed, calculated and compared with structural acceptance criteria established using data from tests and models of structural components. A new methodology is proposed to quantify system stability limit states by integrating the destabilizing effects represented by local damage indices through modified second-order inelastic stability analyses. The proposed method avoids the need for questionable ad-hoc averaging techniques to relate local to global damage indices. Correlation parameters between ground motion intensity measures, such as spectral acceleration, etc., and structural damage are presented, and statistical performance measures of global response are reported.

Supported by test data on structural components, the analyses demonstrate excellent seismic performance of composite framed structures when evaluated both on their own merits and in comparison with steel frames. In particular, by permitting steel beams to run continuous through the reinforced concrete columns, the composite frames avoid the fracture critical details that have caused problems with welded steel moment frames. The design studies do, however, suggest areas for improving current design criteria, in particular, the minimum strength and stiffness requirements for proportioning beams and columns to resist seismic loads. By improving understanding of the seismic response of composite RCS frames this research should lead to their broader utilization for seismic regions and will contribute towards the development of more transparent and reliable performance-based design methodologies.