Epistemic Uncertainty of Seismic Response Estimates for Reinforced Concrete Bridges

Abstract eng:
Considerable effort has recently been directed toward developing analytical probabilistic seismic response estimates at different earthquake intensities. However, probability distributions of response quantities are sensitive to not only the randomness in the ground motion input, but also the assumptions made in the nonlinear finite element models used in the analyses. There are epistemic uncertainties inherent in selection of model parameters as well as the capabilities of individual analysis platforms themselves. This paper develops quantitative estimates of the epistemic uncertainty, above and beyond the aleatory variability, of typical reinforced concrete overpass bridge structures in California. A series of nonlinear time history analyses of 6 typical existing reinforced concrete bridge structures in California using 60 ground motions is carried out using SAP2000 and OpenSees finite element programs. The response of the bridges in terms of peak displacements at the column top is related to the intensity measure defined as the spectral displacement at the first mode period. A procedure using natural logarithmic regression is applied to the data to estimate the seismic response of the structure and the uncertainty in such estimation. Two methods of obtaining response estimate bias factors between the two programs are illustrated for the 6 bridges under three typical seismic hazard levels (2%-, 5%-, and 10%-in-50 year probabilities of exceedance).

• Export as: BibTeX | MARC | MARCXML | DC | EndNote | NLM | RefWorks
• Add to your basket