Performance prediction and evaluation of low ductility steel moment frames for seismic loads

Recent earthquakes in Northridge, California and Kobe, Japan have resulted in billions of dollars of damage to buildings, bridges and other structures and the loss of thousands of lives. The suffering of people displaced from their homes and businesses wonder why. Lessons learned from the Michoacan Earthquake in Mexico City in 1985, and from the San Fernando and Loma Prieta Earthquakes that rocked parts of California have been replayed in these recent earthquakes.; A reliability-based, performance-oriented approach has been adopted herein for design and evaluation. This approach was taken in order to account for uncertainties and randomness in seismic demand and capacities in a consistent manner and to satisfy identifiable performance objectives corresponding to various occupancies, damage states and seismic hazard. A new aspect of this procedure is that the confidence that the building will satisfy the performance objective may be calculated.; Guidelines for design and analysis of new ordinary moment frames with weak-column strong-beam configuration are presented. Detailed and realistic modeling of the joint using DRAIN-2DX program is carried out. Evaluation and performance prediction of the system is presented based on the findings from the reliability-based performance evaluation method.; Ordinary moment frames with Weak-Column Strong-Beam configuration were designed, analyzed and evaluated using the evaluation format adopted earlier. Other ordinary moment frame systems that consist of partially restrained connections such as T-stub connections, end-plate connections, and clip angle connections are addressed. Ordinary moment frames with different kinds of partially restrained connections are analyzed and compared for their ability to satisfy the performance objectives.; According to the study, the 20-story building with WCSB configuration in Seismic Design Category C collapsed due to story mechanism and high P-Δ moments. Therefore, it is recommended that WCSB design be restricted to 120 feet tall. All of the buildings with T-stub connections performed well except for the 3-story building in SDC D which exceeded the permanent residual drift limit. As was seen from the measured moment-rotation behavior of the connection, the end-plate connection behaves very similar to the fully restrained connections. Therefore, the connection is expected to perform very well based on the SMF studies performed by Lee and Foutch. The 9-story building with clip angles performed well except the permanent residual drift exceeded the target value.