DEALING WITH NOISY LIMIT STATE FUNCTION IN STRUCTURAL RELIABILITY ASSESSMENT OF A CRASHED THIN-WALLED BEAM


Abstract eng:
Spot welding is the most popular welding technique used in the automobile manufacturing industry for connecting sheet metal car components. However, often due to manufacturing inaccuracies and fatigue failures many spot weld joints may be missing in an operational vehicle. During crash, random spot weld failures may determine failure modes of the entire component. Therefore, in order to properly analyze crashworthiness reliability of such structures the spot weld failures should be considered. Due to a great number of spot weld connections a precise stochastic modeling of their properties with random variables is very inefficient. In this study, spot weld defects are accounted for by deleting a certain number of randomly selected spot weld elements from the finite element model each time the limit state function value is computed. Such an approach introduces a random noise to the limit state function and, in consequence, requires an appropriately adopted technique for reliability analysis. As potentially capable of coping with the noisy failure function a method based on the algorithm proposed by Zou et al. and published in the journal of Reliability Engineering and System Safety in 2002 is investigated. A detailed study on the reliability of thin-walled s-rail subjected to crash is performed. A number of modifications of the original algorithm are suggested.

Contributors:
Publisher:
ASRANet Ltd., 2008
Conference Title:
Conference Title:
4th International ASRANet Colloquium
Conference Venue:
Athens (GR)
Conference Dates:
2008-06-25 / 2008-06-27
Rights:
Text je chráněný podle autorského zákona č. 121/2000 Sb.



Record appears in:



 Record created 2014-11-20, last modified 2014-11-20


Original version of the author's contribution as presented on CD, paper No. 76.:
Download fulltext
PDF

Rate this document:

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