Probability of strength reduction of aged ship structural profiles due to corrosion wear

Download files
Access & Terms of Use
open access
Abstract
A relatively simple, straight-forward statistical method is presented here to assess the probability of reduction of geometrical properties of aged ship structural profiles. These rolled profiles such as bulb plate, tee bar, flat bar etc. begin to shrink mainly due to corrosion wear following the breakdown of protective coatings. Geometrical properties may include the stiffener cross-sectional area with or without the attached plating, moment of inertia, section modulus etc. among many others. The first set of inputs includes the statistical data about the dimensions or scantlings of the structural profiles usually collected by the manufacturers. For example, the mean and coefficient of variation (c.o.v.) of the web height, web plate thickness etc. The second set of inputs for this analysis is the evenly distributed corrosion wear; the mean wear and the c.o.v. or standard deviation of wear. Based on certain assumptions the mean and variance of the geometrical properties may be derived statistically from these basic inputs. The outputs may be used to address the following issue. The Classification Societies set up Renewal Criteria for the stiffeners/profiles in ship hull structures when they do not meet the requirements which may be, for example, cross-sectional area falling below 90% of the original nominal value. However, the means and variances of the geometrical properties as functions of age have many other useful applications.
Persistent link to this record
Link to Publisher Version
Author(s)
Chowdhury, Mahiuddin
Supervisor(s)
Creator(s)
Editor(s)
Translator(s)
Curator(s)
Designer(s)
Arranger(s)
Composer(s)
Recordist(s)
Conference Proceedings Editor(s)
Other Contributor(s)
Corporate/Industry Contributor(s)
Publication Year
2006
Resource Type
Conference Paper
Degree Type
UNSW Faculty
Files
download Final Manuscript.pdf 119.6 KB Adobe Portable Document Format
Related dataset(s)