An efficient energy-based methodology for seismic collapse assessment of steel moment frame buildings

Abstract

PurposeNonlinear dynamic analyses are employed for seismic collapse risk evaluation of existing steel moment frame buildings. The standards, such as ASCE 41-17, often define collapse thresholds based on plastic deformations; however, the collapse process involves several factors, and plastic deformation is only one of them. An energy-based approach employs deformation and resistance responses simultaneously, so it can consider various factors such as excessive deformation, stiffness and resistance degradation, and low-cycle fatigue as cumulative damage for seismic assessment. In this paper, an efficient energy-based methodology is proposed to estimate the collapse threshold responses of steel moment frame buildings.Design/methodology/approachThis methodology uses a new criterion based on the energy balance concept and computes the structural responses for different seismic hazard levels. Meanwhile, a pre-processing phase is introduced to find the records that lead to the collapse of buildings. Furthermore, the proposed methodology can detect failure-prone hinges with a straightforward probability-based definition.FindingsThe findings show that the proposed methodology can estimate reasonably accurate responses against the results of the past experiment on the collapse threshold. Based on past studies, ASCE 41-17 results differ from experimental results and are even overly conservative in some cases. The authors believe that the proposed methodology can improve it. In addition, the failure-prone hinges detected by the proposed methodology are similar to the predicted collapse mechanism of three mid-rise steel moment frame buildings.Originality/valueIn the proposed methodology, new definitions based on energy and probability are employed to find out the structural collapse threshold and failure-prone hinges. Also, comparing the proposed methodology results against the experimental outcomes shows that this methodology efficiently predicts the collapse threshold responses.