Reliability evaluation of strain-based design for pipelines using probabilistic demand/capacity models

Ground movement can impose excessive deformation violating pertinent pipeline limit states. Currently, the integrity assessment of pipelines subjected to soil movement is generally performed by analyzing the stresses and/or strains in pipelines using various engineering techniques, including finite element analysis (FEA). However, given the wide variability of the pipe and soil engineering properties, using deterministic approaches alone may be inadequate. The desired approach is a semi-probability-based approach using safety factors or a full-probability-based approach, specified in Annex C and Annex O of CSA Z662:2019, respectively. However, probabilistic demand or capacity models are required but missing in Annex C because of the lack of mature and established models for calculating the strain capacity and the demand due to the site-specific nature of ground movement. The objective of this project is to develop probabilistic strain demand and capacity models for reliability analysis of pipelines subjected to geotechnical loads by leveraging FEA and quantitative reliability methods. The results obtained using these desired approaches will be compared. The deliverables of this project will not only help industries, such as Northern Crescent, to improve the pipeline integrity assessment programs with better efficiency and safety but also be useful to update the design codes (CSA Z662).

Faculty Supervisor:

Yong Li;Samer Adeeb

Student:

Sylvester Agbo

Partner:

Northern Crescent

Discipline:

Engineering - civil

Sector:

Professional, scientific and technical services

University:

University of Alberta