Post-earthquake repairability-based methodology for enhancing steel MRFs

Self-centering beam-to-column connections (SBCs) were widely investigated to enhance conventional steel moment-resisting frame’s (MRF’s) post-earthquake repairability by reducing residual displacements. The peak displacement-targeted design methodology was adopted to develop the design guidelines for enhancing MRFs with the emerging self-centering members, where the desired post-earthquake repairability of the enhanced MRF needs to be achieved with inevitable iterative dynamic analyses. This paper intends to develop a novel post-earthquake repairability-based methodology for enhancing MRFs with SBCs, where residual and peak displacements are both targets. To this end, artificial neural network (ANN) models were developed to estimate the peak and residual displacements of the enhanced MRFs with SBCs under given seismic intensities. Peak and residual displacement-based design (PRBD) procedures were developed based on the proposed post-earthquake repairability-based methodology and established ANN models. Four design cases were designed through the developed PRBD procedure and analyzed. The design and analysis outcomes demonstrate that the improved MRFs can attain the intended post-earthquake repairability and meet the peak and residual displacement objectives without necessitating iterative design and dynamic analyses. This confirms the effectiveness and precision of the proposed post-earthquake repairability-based approach and PRBD procedure. Moreover, the enhanced MRFs with partially self-centering behavior achieved nearly zero residual displacements (lower than 0.2%) under maximum considered earthquakes. This finding shows that the partially self-centering MRF is a more cost-efficient solution for developing seismic resilient MRF than fully self-centering MRF to achieve excellent post-earthquake repairability with a lower requirement of self-centering capacities.