Vamvatsikos D., Lignos D.G. (2011). Evaluating the epistemic uncertainty of the seismic demand and capacity for a 9-story steel moment-resisting frame. Proceedings of the 7th Hellenic National Conference on Steel Structures, Volos, Greece
Abstract | The accurate estimation of the seismic performance of steel structures requires reliable information on the effect of our incomplete knowledge of the actual system parameters. Aiming to provide such an outlook we undertake a comprehensive effort to quantify the uncertainty for a single steel moment-resisting frame by bringing together several important advances. Model parameters are described by complete probabilistic distributions including intra-member and inter-member correlation information derived from experimental data from a recently developed database for modeling steel components. Incremental dynamic analysis is employed to accurately assess the seismic performance of the model for any combination of the parameters by performing multiple nonlinear timehistory analyses for a suite of ground motion records. Finally, we use an efficient Monte Carlo simulation algorithm based on incremental Latin Hypercube Sampling to efficiently propagate the uncertainties from the numerous parameters to the actual system demand and capacity. The effect of model parameter uncertainties on the seismic behavior of the 9-story steel moment resisting frame is thus quantified, offering a unique method to assess the actual margin of safety inherent in any steel frame structure.