Moment matching technique for fast and robust uncertainty quantifications of complex systems

Thumbnail Image
Date
2018-01-01
Authors
Teng, Ya-Lu
Major Professor
Advisor
In Ho Cho
Committee Member
Journal Title
Journal ISSN
Volume Title
Publisher
Authors
Research Projects
Organizational Units
Journal Issue
Is Version Of
Versions
Series
Department
Civil, Construction, and Environmental Engineering
Abstract

Despite the growing popularity of reinforced concrete (RC) core walls, robust uncertainty quantification of their global and local performances remains an intractable challenge. A combination of a random sampling technique in conjunction with a simplified structural analysis has been widely used, but substantial ambiguity remains in the simulation accuracy and the resultant uncertainties. As a fast and robust alternative, this study suggests a moment matching (MM) technique. MM can dramatically reduce required sample points to only a few, which enables a direct use of a high-precision parallel multiscale finite element analysis (PM-FEA) for uncertainty quantification. This study demonstrates how to apply the combination of MM and PM-FEA to quantify uncertainties behind complex RC core walls, notably in terms of global force-resisting capacity and microscopic progressive bar buckling of five U-shaped walls. Results suggest that the combination of MM and PM-FEA will serve as a promising method to improve our understanding of uncertainties behind complex RC structure.

Comments
Description
Keywords
Citation
DOI
Source
Subject Categories
Copyright
Sat Dec 01 00:00:00 UTC 2018