National Technical University of Athens (NTUA)
Advanced stochastic finite element simulations and reliability analysis
Abstract
dc:descriptionThis thesis presents a series of methodologies that have been implemented in the framework of SFEM and reliability analysis in order to reduce the computational effort involved .The first methodology is a neural network-based subset simulation in which neural networks are trained and then used as robust meta-models in order to increase the efficiency of subset simulation with a minimum additional computational effort. In the second methodology neural networks are used in the framework of MCS for computing the reliability of stochastic structural systems, fields, by providing robust neural network estimates of the structural response. The third methodology consists of constructing an adaptive sparse polynomial chaos (PC) expansion of the response of stochastic systems in the framework of spectral stochastic finite element method (SSFEM). The proposed methodology utilizes the concept of variability response function (VRF) in order to compute an a priori low cost estimation of the spatial distribution of the second-order error of the response as a function of the number of terms used in the truncated Karhunen-Loeve series representation of the random field involved in the problem. Finally, a parametric study of Monte Carlo simulation versus SSFEM in large-scale systems is performed.
Degree
thesis:*- Grantor dc:publisher
- National Technical University of Athens (NTUA)
- Year dc:date
- 2014
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Giovanis, Dimitrios
- Γιοβάνης, Δημήτριος
Subjects
dc:subject × 15- Νευρωνικά δίκτυα
- Μέθοδος υποσυνόλων
- Μέθοδος φασματικών πεπερασμένων στοιχείων
- Εκτίμηση αξιοπιστίας κατασκευών
- Μέθοδος προσομοίωσης Monte Carlo
- Stochastic analysis
- Neural networks
- Subset simulation
- Spectral stochastic finite element method
- Rellability analysis
- Monde Carlo simulation
- Επιστήμες Μηχανικού και Τεχνολογία
- Επιστήμη Πολιτικού Μηχανικού
- Engineering and Technology
- Civil Engineering
Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/38408