Date of Graduation

Spring 2015

Document Type

Honors Research Project

Degree Name

Bachelor of Science

Major

Chemical Engineering

Research Sponsor

Dr. Homero Castaneda Lopez

First Reader

Dr. Qixin Zhou

Second Reader

Dr. Chelsea Monty

Abstract

The goal of this project was to develop a model that could predict the probability of pipeline failure due to internal corrosion in order to prevent pipeline outages/ruptures. The intended audience of this model is oil pipeline owners. The model used the ASME Modified B31G method to determine the pipeline failure pressure. Inputs to the model include corrosion defect length, depth, corrosion rate, yield strength, and pipeline diameter/thickness which are collected using an in-line inspection (ILI) tool. The model outputs include failure probability as a function of time, the time discounted economic risk due to failure, and the recommended year of repair. The failure probability was evaluated using a Monte Carlo simulation which used MATLAB generated distribution parameters. At each corrosion anomaly site, the cost of repair was compared to the discounted economic risk due to failure. For this project, ILI data of a 100 kilometer pipeline was provided by a confidential industry source. The model recommendations include making over 2,600 repairs over a 15 year time span based on a failure cost of $1,000,000 and a repair cost of $10,000. This led to avoiding $13.5 million worth of corrosion induced risk.

Share

COinS