Computation tool for assessing the probability characteristics of the stress state of the pipeline part defected by pitting corrosion. (December 2015)

• Record Type: Journal Article
• Authors: Vodka, Oleksii
• Is Part Of: Advances in engineering software. Volume 90(2015)
• Page Start: 159
• Page End: 168
• Publication Date: 2015-12
• Subjects: Computer-aided modeling -- Pitting corrosion -- Pipeline -- Stress–strain state -- Extreme value statistics -- Kernel density estimation; Computer-aided engineering -- Periodicals; Engineering -- Computer programs -- Periodicals; Engineering -- Software -- Periodicals; Periodicals; 620.0028553
• DOI: 10.1016/j.advengsoft.2015.08.012
• Languages: English
• ISSNs: 0965-9978

Abstract

Abstract: One of the factors affecting the life-time of the main pipelines is their susceptibility to different types of corrosion. A common type of pipeline corrosion is pitting corrosion which is a form of localized corrosion – a random process occurring in a certain section of the pipeline, leading to walls thinning under the influence of the external environment. Such a change in the pipeline geometry accompanied by a significant change in the local stress–strain state leads to a large number of emergencies. This work deals with the analysis of probability characteristics of the stress–strain state arising in the areas damaged by pitting corrosion. To determine the parameters of the stress–strain state the finite element method is used. The pipe material is modeled as elastoplastic with a bilinear diagram of plasticity. Pitting corrosion is modeled as a set of randomly positioned half-spheres with the same radius subtracted from the volume of the pipe. Coordinates of pitting defects centers obey to a uniform probability density function (PDF). Using this approach we take into account the mutual influence of defects on the pipe stress–strain state. A special macro was developed to determine the parameters of the stress–strain state of the corroded pipe. It allows performing automatic calculation and processing of the data for pitting corrosion defects. Parameters of the PDF of stress concentration factor depending on the radius of the defect were determined in accordance with the results of simulation. The kernel density estimation and generalized extreme value distribution was chosen as PDF approximation on the basis of the conditions of the Fisher–Tippett–Gnedenko theorem. The change of the PDF shape was investigated for different defect radii. The coefficient of variation, mean and standard deviation of stress concentration factor as function of defect radii were studied as well as reliability function of the pipe part was obtained using the criterion of plastic deformation occurrence.