ABSTRACT Incremental recordwise Latin hypercube sampling (iLHS) and multiple stripe analysis (MSA) are employed to evaluate the seismic performance of low‐rise RC moment‐resisting frames subjected to uncertainties of both the values of the model parameters and the model type itself. The study spans various designs of a given 2‐story frame, from modern seismic‐resistant to older with joint shear deficiencies. It examines fragility curves for global and local behaviors and scrutinizes the selection of intensity measures (IMs). Two levels of fidelity for the beam‐column joint modeling are considered: a high‐fidelity explicit model and a lower‐fidelity implicit model. Findings indicate that model parameter uncertainties have minimal impact for modern frames, while for older frames with joint shear deficiencies, these uncertainties become significant. The study also reveals that no single IM is universally optimal for the ductile or semi‐ductile version of this short‐period frame; the better performing IM fluctuates between average spectral and peak ground acceleration depending on the drift ratio level; for the brittle joint‐deficient one, though, the limited manifestation of plastic behavior makes spectral acceleration at the fundamental period consistently good even close to collapse. The paper concludes with a proposed method for estimating total variance for different limit states, aligning with FEMA P58 guidelines.
Discussion(0)
No comments yet. Be the first to comment.