Performance prediction and evaluation of steel special moment frames for seismic loads.

摘要

Following the January 17, 1994 Northridge, California Earthquake, investigations showed that more than 150 steel moment resisting buildings experienced beam-to-column fractures. Even though no structural collapse was reported after the Northridge earthquake, classified as a moderate earthquake, the significant economic loss due to the damage and service disruption resulted in a concern of design and construction procedures for steel moment frame buildings. As part of Performance Prediction and Evaluation Team of the FEMA/SAC Steel Project, this study focused on performance prediction and evaluation for new, existing, and damaged steel moment frame buildings.; Several classes of post-Northridge buildings, which are representative of typical buildings designed by the 1997 NEHRP provisions and constructed using pre-qualified ductile connections, were developed considering different heights, locations, and soil types. Then, the frame analysis models were made including the effect of post-Northridge ductile connection, panel zone deformation and interior gravity frames. The 1973 UBC, 1985 UBC and 1994 UBC provisions were selected for the pre-Northridge 3-, 9- and 20-story buildings in Los Angeles and Seattle and designed for wind and seismic loads. A total of 15 pre-Northridge buildings, which are representative of typical buildings based on year of construction and brittle pre-Northridge connections, were designed considering different height, locations and year of construction. Then, the frame analysis models were made including the effect of post- and pre-Northridge connection, panel zone deformation and interior gravity frames. The seismic demands and capacities were calculated using each set of 20 SAC ground motions representing 2/50 and 50/50 hazard levels. Using the drift demands and capacities, the performance prediction and evaluation procedure based on the reliability framework for SMRF buildings is presented. The randomness and uncertainty associated with predicting the capacity and demand are explicitly accounted for this method. Confidence levels that existing buildings will exceed the pre-defined performance level against different hazard levels are calculated based on the Collapse Prevention (CP) and Immediate Occupancy (IO) performance levels.