Development of failure frequency, shelter and escape models for dense phase carbon dioxide pipelines

Abstract

Carbon Capture and Storage (CCS) is recognised as one of a suite of solutions required to reduce carbon dioxide (CO2) emissions into the atmosphere and prevent catastrophic global climate change. In CCS schemes, CO2 is captured from large scale industrial emitters and transported, predominantly by pipeline, to geological sites, such as depleted oil or gas fields or saline aquifers, where it is injected into the rock formation for storage. The requirement to develop a robust Quantitative Risk Assessment (QRA) methodology for high pressure CO2 pipelines has been recognised as critical to the implementation of CCS. Consequently, failure frequency and consequence models are required that are appropriate for high pressure CO2 pipelines. This thesis addresses key components from both the failure frequency and consequence parts of the QRA methodology development. On the failure frequency side, a predictive model to estimate the failure frequency of a high pressure CO2 pipeline due to third party external interference has been developed. The model has been validated for the design requirements of high pressure CO2 pipelines by showing that it is applicable to thick wall linepipe. Additional validation has been provided through comparison between model predictions, historical data and the existing industry standard failure frequency model, FFREQ. On the consequences side, models have been developed to describe the impact of CO2 on people sheltering inside buildings and those attempting to escape on foot, during a pipeline release event. The models have been coupled to the results of a dispersion analysis from a pipeline release under different environmental conditions to demonstrate how the consequence data required for input into the QRA can be determined. In each model both constant and changing external concentrations of CO2 have been considered and the toxic effects on people predicted. It has been shown that the models can be used to calculate safe distances in the event of a CO2 pipeline release.