Author : Kimiya Zakikhani
Publisher :
Page : 0 pages
File Size : 36,23 MB
Release : 2021
Category :
ISBN :
Book Description
As the most frequent failure source, external corrosion has led to more than 1700 failures in gas transmission pipelines in US since 1996, causing a property damage of approximately $189M. Such numbers highlight the importance of maintaining gas transmission pipelines in safe conditions to postpone corrosion failures. As the most widely applied method of corrosion monitoring technique, in-line inspection is expensive and time-consuming due to requiring high frequencies. On the other hand, the recent efforts directed towards developing failure prediction or maintenance planning models for oil and gas pipelines seem to have some limitations. As such, most of the failure prediction models are based on limited number of inspection or historical records or are limited in application due to their subjectivity. Furthermore, in the domain of maintenance planning, the current procedures are merely based on considering the associated costs and safety thresholds in the decision-making process. Such methodologies do not address the importance of pipeline availability and continuation of operation as a critical asset in the selection of the maintenance strategy. This research has two main objectives. As the first objective, the proposed research aims to develop historical data-based failure prediction models for gas transmission pipelines by considering geo-environmental features. As the second objective, this research aims to propose a reliability-centered availability-based maintenance planning framework that considers the criticality of pipeline operation. For these objectives, a detailed literature review was carried out on current methodologies to predict failures in oil and gas pipelines and maintain them. As the most important limitations, current failure prediction models do not consider geographical and environmental properties of pipelines to predict failures. On the other hand, in maintenance planning scope, none of current practices highlight importance of pipeline operation and availability in making the proper decision. In addition, these methodologies are often subjective, i.e. they are merely applicable to limited pipelines. To overcome these limitations the mentioned objectives of this research were defined and failure and maintenance data were collected from accessible historical records and reports. The failure prediction models were developed from best-subset and multiple regression analyses on the historical failure data and were then validated. On the other hand, the maintenance planning framework was developed from a coupled cost and availability-based maintenance planning procedure on different maintenance scenarios. For each scenario, a discrete event simulation was carried out through MATLAB programming. Such simulation was performed on the pipeline reliability profile obtained from a Monte Carlo simulation and consideration of improvement in availability per unit cost as the decision criteria. Monte Carlo simulation was carried out to consider wide range of design parameters for development of the reliability profile. The developed failure prediction models were able to satisfactory predict time of corrosion failures in gas transmission pipelines for Great Plains and South East Regions of the U.S. These models were validated with mean absolute error (MAE) and root mean square error (RMSE) of 0.12 and 0.04, for Great Plains, and 0.11 and 0.07, for South East regional classifications, respectively. The proposed maintenance planning framework reveals that for a case study of a 24-inch pipeline, considering an availability-cost indicator, the second maintenance scenario, with interventions at the service life of 30.1 and 40.5 years is more effective. This order is followed by the first scenario with interventions at service life of 33.3 and 42.2 years, and finally the third scenario with intervention at service life of 24.2 years, respectively. The developed failure prediction models can assist decision makers and pipeline operators to predict the expected time of corrosion failure in gas transmission pipelines in the selected regions by considering geo-environmental and pipeline design parameters. In addition, for maintenance planning of oil and gas pipelines, this research proposes a novel methodology that considers oil and gas pipelines as critical assets for which continued operation is of high importance. Such consideration provides a compensation between the costs incurred and pipeline availability to avoid over/under maintenance.