Publication detail

Application of polynomial chaos expansion to reliability analysis of prestressed concrete roof girders

NOVÁK, L. NOVÁK, D. SLOWIK, O.

Original Title

Application of polynomial chaos expansion to reliability analysis of prestressed concrete roof girders

Type

conference paper

Language

English

Original Abstract

The paper is focused on the application of the surrogate model to reliability analysis of real structure. First, the basic theory of polynomial chaos expansion (PCe) is described. Usage of the PCe for moment and sensitivity analysis is then presented, which can be done efficiently by post-processing of explicit function without any additional computational demands. Implementation of theory into new software tool and its application on prestressed concrete roof girder failing in shear is presented in the last part of the paper. As can be seen in the example, PCe for reliability analysis represents very strong and efficient tool, especially for variance based sensitivity analysis and estimation of statistical moments.

Keywords

structural reliability, Polynomial chaos expansion, surrogate model, sensitivity analysis

Authors

NOVÁK, L.; NOVÁK, D.; SLOWIK, O.

Released

14. 5. 2018

Publisher

Institute of Theretical and Applied Mechanics of the Czech Academy of Sciences

Location

Praha

ISBN

978-80-86246-91-8

Book

ENGINEERING MECHANICS 2018 PROCEEDINGS, VOL 24

Edition

First

Pages from

609

Pages to

612

Pages count

4

URL

http://www.engmech.cz/improc/2018/609.pdf

BibTex

@inproceedings{BUT150898,
  author="Lukáš {Novák} and Drahomír {Novák} and Ondřej {Slowik}",
  title="Application of polynomial chaos expansion to reliability analysis of prestressed concrete roof girders",
  booktitle="ENGINEERING MECHANICS 2018 PROCEEDINGS, VOL 24",
  year="2018",
  series="First",
  pages="609--612",
  publisher="Institute of Theretical and Applied Mechanics of the Czech Academy of Sciences",
  address="Praha",
  isbn="978-80-86246-91-8",
  url="http://www.engmech.cz/improc/2018/609.pdf"
}