Probability analysis of RC member deformation limits for different performance levels and reliability of their deterministic calculations

Y Lu, X M Gu

Research output: Contribution to journalArticlepeer-review

Abstract

The quantified definition of performance levels of RC members, in terms of limit sectional curvature and member drift, is necessary for the implementation of the performance-based design of building structures. Because of various uncertainties involved, the curvature and drift limits satisfying prescribed performance criteria are probabilistic in nature; therefore, it is necessary to understand their probability distributions before any reliable deterministic values can be decided. In this paper, the probability distributions of RC member behavioural parameters, namely the strength and deformation capacities at yield and ultimate stages, and the deformation limits at two intermediate performance levels ("Functional" and "Damage Control"), are established by means of Monte Carlo simulation in conjunction with the RC section analysis. Subsequently, the reliability of employing the deterministic predictions for the deformation limits is assessed. Results show that the limit curvature and drifts at different performance levels generally follow the normal distribution, with a coefficient of variation on the order of 10%, 15%, 35%, and 40% for the yield, Functional, Damage Control, and Ultimate performance levels, respectively. In most cases, the probability of actual curvature and drift capacities exceeding the respective limit values predicted by deterministic calculations ranges from 50% to 70%. However, for heavily confined members and under a relatively low axial force, the above probability could reduce to only about 30%. (C) 2004 Elsevier Ltd. All rights reserved.

Original languageEnglish
Pages (from-to)367-389
Number of pages23
JournalStructural Safety
Volume26
Issue number4
DOIs
Publication statusPublished - 2004

Fingerprint Dive into the research topics of 'Probability analysis of RC member deformation limits for different performance levels and reliability of their deterministic calculations'. Together they form a unique fingerprint.

Cite this