Přístupnostní navigace
E-application
Search Search Close
Publication detail
NOVÁK, L. LEHKÝ, D. NOVÁK, D.
Original Title
Sensitivity analysis of Engineering Structures Utilizing Artificial Neural Networks and Polynomial Chaos Expansion
Type
conference paper
Language
English
Original Abstract
This paper is focused on sensitivity analysis of engineering structures using surrogate models. Two different techniques for surrogate modeling are utilized in order to obtain various sensitivity measures of quantity of interest. The artificial neural networks and polynomial chaos expansion are used for efficient sensitivity analysis. Each of the techniques is superior in different areas of uncertainty quantification and thus each of them is used for estimating of different sensitivity measures in two engineering examples – simplified analytical function and complex non-linear finite element model of an existing concrete bridge. On the one hand, artificial neural network is utilized for estimation of sensitivity measures based on Monte Carlo simulation and on the other hand, polynomial chaos expansion is exploited for derivation of global sensitivity measures without additional simulations. It is shown that utilization of both methods leads to efficient and complex sensitivity analysis of engineering structures, and it could be recommended to use combination of both techniques in industrial applications.
Keywords
Artificial neural network; Polynomial chaos expansion; Sensitivity analysis; Uncertainty quantification
Authors
NOVÁK, L.; LEHKÝ, D.; NOVÁK, D.
Released
9. 3. 2023
Publisher
Springer Science and Business Media Deutschland GmbH
Location
Germany
ISBN
9783031255984
Book
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Edition number
13810
Pages from
1
Pages to
15
Pages count
BibTex
@inproceedings{BUT187080, author="Lukáš {Novák} and David {Lehký} and Drahomír {Novák}", title="Sensitivity analysis of Engineering Structures Utilizing Artificial Neural Networks and Polynomial Chaos Expansion", booktitle="Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)", year="2023", number="13810", pages="1--15", publisher="Springer Science and Business Media Deutschland GmbH", address="Germany", doi="10.1007/978-3-031-25599-1\{_}14", isbn="9783031255984" }