Back to search

National Technical University of Athens (NTUA)

Stochastic analysis and optimum design of structures subjected to fracture

Abstract

dc:description

The present thesis deals with a unified framework for stochastic analysis and optimum design of structures subjected to fracture developed in the ambit of modern numerical techniques for crack growth simulation mainly based on enriched finite elements methods. This has been done within the context of the stochastic finite element method as well as within a modern optimization environment implementing metaheuristic search algorithms.In the first part of this thesis, the stochastic finite element method is presented within the framework of the sequentially linear analysis (SLA) scheme, providing solutions to stochastic nonlinear static problems for structures made with softening materials whose properties are randomly distributed in the structure, but also giving specific information on the probability of failure. The uncertainty characterizing the material properties, is quantified by using the theory of stochastic functions (processes/fields). The response variability of the structures is computed by means of the direct Monte Carlo simulation. Furthermore, the influence of the variation of each random parameter, the probability distribution, coefficient of variation and correlation length of the stochastic fields are examined. The analysis of two benchmark structures has shown that the load-displacement curves and the probability of failure are strongly affected by the statistical characteristics of the stochastic fields. This extension of SLA to the stochastic framework offers an efficient means to perform parametric investigations of the fracture behavior of structures in the case of variable material properties.In the second part of this thesis, the extended finite element method which consists an appropriate framework for the simulation of the fracture process in structural members caused by fatigue, is incorporated into a shape optimization environment. A reliability analysis combined with a structural shape design optimization formulation is proposed where probabilistic constraints are considered in the formulation of the design optimization problem. Shape design optimization problems are formulated aiming at investigating the relation between structural geometry and service life in the design process. Randomness on the crack initialization along with the uncertainty on the material properties are also examined. A sensitivity analysis of four optimization algorithms based on evolution process is conducted in order to identify the best algorithm for solving the structural shape optimization problem. The applicability and potential of the formulations presented are demonstrated with two characteristic numerical examples. It is shown that with proper shape changes, the service life of structural members subjected to fatigue loads can be enhanced significantly. Comparisons with optimized shapes found for targeted service life are also addressed, while the choice of initial imperfection position and orientation was found to have a significant effect on the optimal shapes.

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
  • Georgioudakis, Emmanouil
  • Γεωργιουδάκης, Εμμανουήλ

Subjects

dc:subject × 23

Rights

dc:rights
Statement dc:rights
  • BY
Language dc:language
eng

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/38802

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Georgioudakis, Emmanouil; Γεωργιουδάκης, Εμμανουήλ. Stochastic analysis and optimum design of structures subjected to fracture. National Technical University of Athens (NTUA), 2014. http://hdl.handle.net/10442/hedi/38802