Reduced-Order Computational Modeling Approach for Precast Concrete Frame Structures under Column Removal Scenarios

This article presents a reduced-order computational modeling approach to quantify the potential for disproportionate collapse in the design of precast concrete framed buildings under column removal scenarios using only the framing geometry and configuration, connection detailing, code provisions, and material specifications. This approach includes the development and validation of mechanics-based moment-rotation curves for novel threaded rod and bolted precast concrete beam-to-column moment connections, which were designed for enhanced robustness under a middle column removal scenario. These moment-rotation curves were then assigned to the beam ends of a 10-story exterior precast concrete moment frame. Nonlinear dynamic analyses were performed to quantify the robustness of each connection under three individual column removal scenarios: middle column, penultimate column, and corner column. Results from an approximate energy-based procedure formulated based on a single nonlinear static analysis are also presented and compared to the nonlinear dynamic analysis results. The results show that, for both connection types, the precast concrete frame was able to arrest collapse for middle and penultimate column removal scenarios with a quantified degree of reserve capacity. The corner column removal scenario, for which the connections were not designed, resulted in collapse, but only by a small margin. A comparison between the results of the nonlinear dynamic analyses and the approximate energy-based procedure yielded close agreement for both connection types; percent differences in their determined ultimate capacity were less than 1% for the middle and penultimate column removal scenarios and less than 4% for the corner column removal scenario.