Time-variant hybrid stochastic interval uncertainty analysis of concrete-filled steel tubular arch structures

Abstract

The long-term structural behaviour of concrete-filled steel tubular arch structures are much related to the inherent viscoelastic effects of creep and shrinkage in concrete core under sustained loading. Such creep and shrinkage effects may lead to excessive deformation during the service period of the structure and consequently impose the risk of failure. Furthermore, uncertainties are inevitable in structural parameters. The failure to accommodate the impact of uncertainties can lead to impractical structural design and even cause significant safety issues.

This dissertation aims to efficiently integrate hybrid uncertainty analysis into the long-term behaviour of concrete-filled steel tubular arch structures by developing a serious of novel computational methods. More specifically, three major uncertainty analysis frameworks have been developed. Firstly, time-dependent linear static analysis of single-spanned concrete-filled steel tubular arch structure with mixed random interval uncertainties is investigated. Secondly, probabilistic interval nonlinear static analysis of concrete-filled steel tubular arch structure considering creep and shrinkage effects is performed with the study object further extended to 3D structures. Thirdly, time-variant random interval dynamic analysis of both 2D and 3D concrete-filled steel tubular arch structures is studied.

The developed uncertainty analysis models bypass the traditional simulation method and are capable of efficiently providing reliable and trustworthy structural outputs. The proposed schemes can offer an applicable way towards the hybrid uncertainty analysis of structures with concrete creep and shrinkage effects at any particular loading age. Finally, the designed uncertainty analysis methodologies can be applied in either pure probabilistic or pure interval problems.