Understanding the Mechanisms of Corrosion and their Effects on Abandoned Pipelines

DET NORSKE VERITAS

March 3, 2015

 

Executive Summary

Pipeline abandonment refers to permanently taking a pipeline out of service. It is a subject that requires research to fully understand life cycle implications. Depending on a number of factors, sections of pipeline may be removed or abandoned in place. In the latter case, steel will corrode and the abandoned pipe will lose structural integrity. Ultimately, the corroded pipe will collapse and break into fragments under the load of soil and traffic over it.

This study by Det Norske Veritas (DNVGL) was commissioned by the PTAC Pipeline Abandonment Research Steering Committee (PARSC) to better understand the mechanisms that lead to corrosion and loss of structural integrity over time. Background information about PARSC is provided below. This DNVGL study included a literature review to identify relevant scientific and engineering models for corrosion and structural integrity. Soil data from the United States National Bureau of Standards was accessed. The study used analytical models to predict the time to collapse of abandoned pipelines and determined that time to collapse was dependent on a number of variables, including (i) pipeline diameter, wall thickness and yield strength, (ii) soil type and soil properties and (iii) pipeline depth of cover. Therefore, predictions need to be made on a case-specific basis. The study also estimated soil subsidence after the collapse of a corroded abandoned pipe. In summary, the study provided analytical models useful to estimate the time to collapse and resulting soil subsidence for abandoned pipelines. It is acknowledged that the models developed within this study need further development and refinement.

PARSC requested that DNVGL calculate time to collapse and resulting soil subsidence is circumstances typical to Canadian agricultural soils and the pipelines traversing them. The following tables report calculated results for 3 soil types, 6 pipeline diameters (3 thicknesses each) and 4 depths of cover when bearing the load of a Brent 1082 grain cart loaded with wheat (72,000 lbs or 32,659 kg per axle).

The study report was peer reviewed by Dr. Frank Cheng, Canada Research Chair in Pipeline Engineering, University of Calgary and Dr. Claudio Ruggieri, Professor of Structural Engineering, University of São Paulo, Brazil.

 

Introduction EN

Introduction FR

Final Report – EN

Final Report – FR

Reference Tables Graphs EN

Reference Tables Graphs FR