Fully probabilistic analysis of FRP-to-concrete bonded joints considering model uncertainty

Dongming Zhang, Xiang-Lin Gu, Qian-Qian Yu, Hongwei Huang, Baolin Wan, Cheng Jiang

Research output: Contribution to journalArticlepeer-review

99 Citations (Scopus)

Abstract

This work presents a full reliability-based analysis framework for fiber-reinforced polymer (FRP)-to-concrete bonded joints considering model uncertainty. Eight frequently used bond strength models for FRP-to-concrete bonded joints were calibrated by defining a model factor. A total of 641 well-documented tests were considered. Four of the eight models had model factors that correlated with input design parameters and the systematic part of the model factor was removed by a regression equation f. By doing this type of characterization, all eight model factors could be comparatively uniform and described by lognormally distributed random variables. The merit of the uniform model uncertainties after calibration for the eight models was established by the reliability analysis. This study improves the predictability of concrete strengthened with fiber composites and provides useful suggestions on their model uncertainties in engineering practice.
Original languageEnglish
Pages (from-to)786-806
Number of pages21
JournalComposite Structures
Volume185
DOIs
Publication statusPublished - 2018

Fingerprint

Dive into the research topics of 'Fully probabilistic analysis of FRP-to-concrete bonded joints considering model uncertainty'. Together they form a unique fingerprint.

Cite this