Hybrid data‐driven hazard‐consistent drift models for SMRF

The seismic design and assessment of steel moment resisting frames (SMRFs) rely heavily on drifts. It is unsurprising, therefore, that several simplified methods have been proposed to predict lateral deformations in SMRFs, ranging from the purely mechanics-based wholly data-driven, which aim alleviate structural engineer's burden conducting detailed nonlinear analyses either as part preliminary iterations or during regional assessments. While many these incorporated codes are commonly used research, they all suffer a lack consideration causal link between hazard level ground-motion suite for their formulation. In this paper, we propose hybrid data-driven models preserve critical relationship hazard-consistency. To end, assemble large database non-linear response history (NRHA) 24 SMRFs different characteristics. These subjected 816 records whose occurrence rates spectral shapes selected ensure consistency our outputs. Two sites with hazards examined enable comparisons under demands. An initial examination resulting drift curves allows us re-visit influence salient modelling assumptions such plastic resistance, geometric configurations joint deterioration modelling. This followed by machine learning (ML)-guided feature selection process considers parameters well key static features, hence nature models. New inter-storey roof displacements then developed. A comparison currently available formulations highlights significant levels overestimation associated previously non-hazard consistent