Institution: Queen's University
Department: Civil Engineering
Year: 2015
Keywords: pipe trajectory; horizontal; full scale; buried pipe; olivine sand; mobilization distance; lateral; DIC; PIV; shear strain; digital image correlation; large deformation; pipe-soil interaction; lateral soil force
Record ID: 2058439
Full text PDF: http://qspace.library.queensu.ca/bitstream/1974/12722/1/Burnett_Alexander_J_201501_MASC.pdf


Oil and gas pipelines buried off the east coast of Canada are in a complex and dynamic environment with many inherent risks, including scour events by large icebergs in shallow water, which present a significant hazard to their safety and structural integrity. A thorough understanding of pipe soil interaction under large deformations imposed either by iceberg scour or other means, is critical to the design of oil and gas pipelines and to the current understanding of pipeline-soil interaction. Iceberg keel gouging events are a large strain numerical problem and experimental data evaluating shear strain behaviour of the test material, strain localization of the soil shear bands, and material flow into the void left behind the displaced pipe are needed in order to model the pipe-soil interaction behaviour involved in these processes. A full scale laboratory test configuration was developed to performed experimental work on horizontal pipe-soil interaction, modeling a large deformation event similar to iceberg scour. Lateral pipe displacement tests were conducted with industry grade pipelines up to 0.61 m in diameter, tested at two burial depths in both loose and dense synthetic olivine sand. Transparent sidewalls were incorporated into the test apparatus to enable the use of digital image correlation (DIC) to determine soil displacements. DIC provides a new source of high quality experimental data which can capture the complex large strain soil behaviour in full scale pipe-soil interaction tests. DIC was used to observe the shear strain behaviour of the test soil, strain localization into soil shear bands, and material flow into the void left behind the displaced pipe, and pipe features including mobilization distance, uplift and trajectory. The influence of soil density, burial depth ratio, and pipe diameter has been evaluated on all of these concepts. It was found that significant soil densification occurs in a compression zone in front of the test pipes, causing the soil and pipe behaviour in loose sand to reflect the results of tests conducted in dense sand.