Buckling Analysis of Steel Tubes With Random Geometric Imperfections


Abstract eng:
The stability analysis of relatively thick steel tubes under external pressure is important, e.g. for the design and construction of offshore pipeline members. The buckling behavior of this type of structures in presence of random imperfections has been studied very scarcely in the literature, in contrast to the case of thin cylindrical shells [1]. In the present paper, the effect of initial (out-ofplane) random geometric imperfections on the buckling load of steel tubes under axial load and lateral pressure is investigated. The geometric imperfections are modeled as a 2D-1V non-homogeneous Gaussian stochastic field simulated using the spectral representation method [2]. The evolutionary power spectrum of the non-homogeneous field is derived from available experimental measurements using the method of separation [3]. For the determination of the limit load variability of the tubes, a stochastic finite element formulation based on Monte Carlo simulation is implemented. It is shown that this variability can be in some cases substantial. A comparison with the results obtained from linearized buckling analyses of tubes with imperfections having the form of the first eigenmode indicates that a realistic description of the imperfections through stochastic fields leads to more conservative results (smaller buckling loads) and thus is preferable.

Contributors:
Publisher:
National Technical University of Athens, 2013
Conference Title:
Conference Title:
COMPDYN 2013 - 4th International Thematic Conference
Conference Venue:
Island of Kos (GR)
Conference Dates:
2013-06-12 / 2013-06-14
Rights:
Text je chráněný podle autorského zákona č. 121/2000 Sb.



Record appears in:



 Record created 2016-11-15, last modified 2016-11-15


Original version of the author's contribution as presented on CD, section: SC-MS 22 THE MULTISCALE STOCHASTIC FINITE ELEMENT METHOD .:
Download fulltext
PDF

Rate this document:

Rate this document:
1
2
3
 
(Not yet reviewed)