Integrity Analysis of Dented Pipelines using Artificial Neural Networks

##plugins.themes.bootstrap3.article.main##

##plugins.themes.bootstrap3.article.sidebar##

Published Dec 31, 2019
Janine Woo Muntaseer Kainat Chike Okoloekwe Sherif Hassanien Samer Adeeb

Abstract

The repair of dents in oil or gas pipelines is mandated based on depth and interaction with stress risers, according to pipeline regulations in Canada and the United States. However, there have been cases where dents that did not meet the regulatory repair criteria have ended up failing, leading to operator need for an accurate assessment method for dents in order to maintain safety. While there is no agreed-upon method currently available in industry, conservative techniques employed by operators have led to poor dig efficiency. Recent research in industry has focused on strain- and fatigue-based techniques to assess the severity of dents and prioritize them for excavation and repair. Finite element analysis has been highlighted as an accurate method to evaluate strains and stresses within dented regions of pipe, although the significant computational time required for this method makes it inefficient for system-wide analysis. In this paper, the results from hundreds of finite element analysis models are used to train artificial neural networks. Subsequently, the artificial neural networks output accurate stresses and strains, that would be obtained using finite element analysis, when presented with input dent and pipe information. As a result, the artificial neural networks harness the accurate results that can be obtained from finite element analysis while results can be obtained efficiently for applicability to a pipeline system.

How to Cite

1.
Woo J, Kainat M, Okoloekwe C, Hassanien S, Adeeb S. Integrity Analysis of Dented Pipelines using Artificial Neural Networks. PST [Internet]. 2019Dec.31 [cited 2020Feb.25];3(2):92-104. Available from: https://pipeline-science.com/index.php/PST/article/view/105

Downloads

Download data is not yet available.
Abstract 53 | PDF file Downloads 109

##plugins.themes.bootstrap3.article.details##

Keywords

Pipelines, dents, finite element analysis, artificial neural networks, pipeline integrity, deformation.

References
[1] Kiefner J, Leewis K. Pipeline Defect Assessment – A Review & Comparison of Commonly Used Methods. Worthington, Ohio, USA: Kiefner and Associates, Inc.; 2011. 238 p. Report No.: PR-218- 05405.
[2] Hassanien S, Langer D. A Semi-Quantitative Analysis of Dents Associated with Corrosion: A Reliability-Based Approach. Paper presented at: NACE CORROSION 2018 Annual Conference; 2018 Apr 15-19; Phoenix, AZ, USA.
[3] Canadian Standards Association (CSA) Group. Oil and gas pipeline systems. Toronto, ON, Canada: Canadian Standards Association; 2015. 868 p. Report No: CSA Z662-15.
[4] BMT Fleet Technology. Dent Fatigue Life Assessment. Washington, DC, USA: U.S. Department of Transportation; 2012. 122 p. Report No.: 432.
[5] National Energy Board. Fatigue Crack Failure Associated with Shallow Dents on Pipelines. Calgary, AB, Canada: Transportation Safety Board of Canada; 2010. 3 p. Report No.: NEB SA 2010-01.
[6] Dawson SJ, Russell A, Patterson A. Emerging Techniques for Enhanced Assessment and Analysis of Dents. Paper presented at: 6th International Pipeline Conference; 2006 Sep 25-29; Calgary, AB, Canada.
[7] Noronha D, Martins R, Jacob B, Souza E. Procedures for the strain based assessment of pipeline dents. Int J Pres Ves Pip. 2010 May; 87(5), 254-265.
[8] Okoloekwe C. A Novel Approach to the Strain Based Analysis of Dented Pipelines [dissertation]. Edmonton, AB, Canada: University of Alberta; 2017. 125 p.
[9] Rosenfeld MJ. Guidelines for the Assessment of Dents on Welds. Houston, TX, USA: Technical Toolboxes, Inc.; 1999. 85 p. Report No.: PR-218-9822.
[10] Tiku S, Semiga V, Dinovitzer A, Vignal G. Full Scale Cyclic Fatigue Testing of Dented Pipelines and Development of a Validated Dented Pipe Finite Element Model. Paper presented at: 9th International Pipeline Conference; 2012 Sep 24-28; Calgary, AB, Canada.
[11] Kainat M, Woo J, Langer D, Krausert T, Cheng JJ, Hassanien S, Adeeb S. Effects of loading sequences on the remaining life of plain dents. J Pipeline Syst Eng. 2019 May; 10(2): 1-20.
[12] The American Society of Mechanical Engineers. Gas Transmission and Distribution Piping Systems. New York, NY, USA: The American Society of Mechanical Engineers; 2010. 208 p. Report No.: B31.8-2010.
[13] Arumugam U, Gao M, Krishnamurthy R, Wang R, Kania R. Study of a Plastic Strain Limit Damage Criterion for Pipeline Mechanical Damage using FEA and Full-Scale Denting Tests. Paper presented at: 11th International Pipeline Conference; 2016 Sep 26-30; Calgary, AB, Canada.
[14] Hassanien S, Kainat M, Adeeb S, Langer D. On the Use of Surrogate Models in Reliability-Based Analysis of Dented Pipes. Paper presented at: 11th International Pipeline Conference; 2016 Sep 26-30; Calgary, AB, Canada.
[15] Turnquist M, Smith I. A Life-Cycle Approach to the Assessment of Pipeline Dents. Paper presented at: 11th International Pipeline Conference; 2016 Sep 26-30; Calgary, AB, Canada.
[16] Layouni M, Hamdi M, Tahar S. Detection and sizing of metal-loss defects in oil and gas pipelines using pattern-adapted wavelets and machine learning. Appl Soft Comput. 2017 Mar; 52(2017), 247-261.
[17] Xu W, Chun B, Joonmo C, Lee J. Corroded pipeline failure analysis using artificial neural network scheme. Adv Eng Softw. 2017 Oct; 112(2017), 255-266.
[18] Woo J, Kainat M, Adeeb S. Development of a Profile Matching Criteria to Model Dents in Pipelines using Finite Element Analysis. Paper presented at: ASME 2017 Pressure Vessels & Piping Conference; 2017 Jul 16-20; Waikoloa, Hawaii, USA.
[19] Hyde TH, Luo R, Becker AA. Predictions of threedimensional stress variations in indented pipes due to internal pressure fluctuations. J Strain Anal Eng. 2011 Jun 27; 46(6): 510-522.
[20] Lin M. Characterization of Tensile and Fracture Properties of X52 Steel Pipes and Their Girth Welds [dissertation]. Edmonton, AB, Canada: University of Alberta; 2015. 197 p.
[21] The MathWorks, Inc. Documentation: Deep Learning Toolbox (R2018a) [Internet]. Natick, MA, USA: The MathWorks, Inc. [updated 2018; cited 2019 May 31]. Available from: http://www.mathworks.com/help/deeplearning/index.html.
Section
Original Work