Identification and reliability of reinforced concrete structural systems for model updating purposes

摘要

A new strategy for in-situ testing of reinforced concrete structures is presented. Its ultimate goal is to provide real-time information about the structural behavior during the test. Statistical results of structural tests are employed in order to get general information about model statistics so that parameters uncertainties will be represented by a set of mutually consistent probability distributions. The proposed procedure relies upon a probabilistic formulation and takes advantage of the Bayesian Network theory to account for structural and design uncertainties.