Session: 06-03-02 Ground Monitoring

Paper Number: 133219

133219 - A Probabilistic Method for Assessing Pipeline Strain Demand Using InSAR Ground Movement Data 

Abstract: 

Ground movement hazards, such as landslides and ground subsidence, pose a major threat to buried pipelines and resulted in $391M USD in property damages in the United States from 2002 to 2021. The dynamic nature of ground movement makes it necessary to actively model and predict pipeline integrity in order to maintain a reliable pipeline network. Conventional stress-based design and assessment methods struggle to predict pipeline failure in the presence of large longitudinal strains that commonly result from ground movement hazards; this has prompted the industry to adopt strain-based design and assessment (SBDA) methods instead. Applying SBDA methods to operational pipelines requires the estimation of strain demand, the strain induced on the pipeline by its operational environment. The easiest way to assess strain demand is by monitoring the displacement of the pipeline over time through the use of inertial measurement units (IMUs) or strain gauges. However, IMUs are expensive to run and strain gauges only measure strain at a single location. Synthetic Aperture Radar (SAR) data, acquired by satellite, can be used to compute ground movement using SAR interferometry (InSAR). This offers an attractive alternative for estimating strain demand; InSAR can cover large areas of interest and provide precise ground displacement measurements at a high spatial and temporal resolution. InSAR data is commonly used for geohazard monitoring but has not yet been used to assess strain demand for buried pipelines experiencing ground movement.  

This paper presents a novel method for predicting pipeline strain demand using InSAR ground movement data. InSAR data are processed and segmented before being sent to a Bayesian network model for strain demand calculation. In the Bayesian network, pipe-soil interaction models from prior research are used to account for environmental effects on pipeline displacement. With the help of several pipeline companies, model outputs were validated using pipeline strain measurements collected during past ground movement events. The high resolution of InSAR data and the use of Bayesian network models enable the probabilistic evaluation of strain demand for multiple pipeline segments while accounting for the effects of data quality on strain demand estimates. The proposed method can empower pipeline companies to perform strain-based assessments with greater ease, promoting a fast and data-informed response to ground movement hazards.

Presenting Author: Colin Schell University of Maryland, College Park

Presenting Author Biography: Colin is a PhD candidate enrolled in the Reliability Engineering program at the University of Maryland. Advised by Dr. Katrina Groth at the Systems Risk and Reliability Analysis lab (SyRRA), his research focuses on using causal models to better understand pipeline risks stemming from mechanical and natural hazard loading conditions, as well as third-party excavation. Colin also received his B.S. in mechanical engineering from the University of Maryland in 2020.

Authors:

Colin Schell University of Maryland, College Park
Mirka Paluchova 3vGeomatics
Ernest Lever GTI Energy
Katrina Groth University of Maryland, College Park