As the demand for oil and gas resources increases pipeline construction pushes further into the geologically unstable Arctic and sub-Arctic regions. Consequently, these buried pipelines suffer much harsh environmental and complex loading conditions. In addition, higher strength and larger size pipes with higher operation pressure are used gradually. These severe and unknown conditions increase the risk of pipeline failure, especially, local buckling (wrinkling) failure. The wrinkling failure and sequential pipe fracture can cause enormous cost loss as well as high risk in safety and environmental impact. In the past, to prevent the buried pipelines from buckling failure, the pipeline maintenance was processed by periodical inspections and excavations in the field. The whole procedure is expansive and time consuming, and has no active warning system for possible failures between the inspection periods. Therefore, to overcome these problems, an automatic warning system for monitoring pipeline buckling is developed. A damage detection model (DDM) with artificial neural network (ANN) is a kern of the warning system and discussed in this paper. The proposed DDM will allow engineers to diagnose the pipe condition reliably and continuously without interrupt the normal operation of buried pipelines. The proposed DDM successfully identifies the distributed strain patterns in local characteristics as well as global trend. Some significant findings in the ANN model working with distributed strain patterns of the pipes are discussed, and a guideline of applying the DDM to the field pipe is also presented in this paper.
Skip Nav Destination
2010 8th International Pipeline Conference
September 27–October 1, 2010
Calgary, Alberta, Canada
Conference Sponsors:
- International Petroleum Technology Institute and the Pipeline Division
ISBN:
978-0-7918-4423-6
PROCEEDINGS PAPER
Prediction of Pipe Wrinkling Using Artificial Neural Network
Z. L. Chou,
Z. L. Chou
University of Alberta, Edmonton, AB, Canada
Search for other works by this author on:
J. J. R. Cheng,
J. J. R. Cheng
University of Alberta, Edmonton, AB, Canada
Search for other works by this author on:
Joe Zhou
Joe Zhou
TransCanada Pipelines Ltd., Calgary, AB, Canada
Search for other works by this author on:
Z. L. Chou
University of Alberta, Edmonton, AB, Canada
J. J. R. Cheng
University of Alberta, Edmonton, AB, Canada
Joe Zhou
TransCanada Pipelines Ltd., Calgary, AB, Canada
Paper No:
IPC2010-31165, pp. 49-58; 10 pages
Published Online:
April 4, 2011
Citation
Chou, ZL, Cheng, JJR, & Zhou, J. "Prediction of Pipe Wrinkling Using Artificial Neural Network." Proceedings of the 2010 8th International Pipeline Conference. 2010 8th International Pipeline Conference, Volume 4. Calgary, Alberta, Canada. September 27–October 1, 2010. pp. 49-58. ASME. https://doi.org/10.1115/IPC2010-31165
Download citation file:
15
Views
Related Proceedings Papers
Related Articles
Profiles of Two JOMAE Associate Editors (A Continuing Series)
J. Offshore Mech. Arct. Eng (October,2021)
Risk-Based Considerations in Developing Strategies to Ensure Pipeline
Integrity—Part I: Theory
J. Pressure Vessel Technol (August,1994)
Risk-Based Inspection Analysis for High-Pressure Hydrogenation Cracking Unit
J. Pressure Vessel Technol (April,2009)
Related Chapters
Use of PSA in Lisencing of EPR 1600 in Finland (PSAM-0160)
Proceedings of the Eighth International Conference on Probabilistic Safety Assessment & Management (PSAM)
Introduction
Computer Vision for Structural Dynamics and Health Monitoring
Section VIII: Division 2–Alternative Rules
Companion Guide to the ASME Boiler & Pressure Vessel Codes, Volume 2, Sixth Edition