Oil and Gas Pipeline Risk Assessment Model by Fuzzy Inference Systems and Artificial Neural Network

Date
2015-02
Authors
Wu, Wentao
Journal Title
Journal ISSN
Volume Title
Publisher
Faculty of Graduate Studies and Research, University of Regina
Abstract

Pipeline transportation has been widely used and recognized as the best way to transport oil and gas in energy industry, because of its excellent safety features and significant economic advantages. However, the failure accident of pipelines is one of the most frustrating issues, as its significant adverse impact on people, environment and public safety; it can also cause severe economic loss. Due to pipelines mostly being installed underground, information limitation and data uncertainties make it difficult to predict and assess failure risks by a single methodology based model. Intelligence Systems (IS), in particular Fuzzy Inference System (FIS) and Artificial Neural Networks (ANNs), have been significantly developed in recent years. Besides traditional experts’ knowledge risk assessment methods, the IS based assessment methods have been well established to assess risks in many industries, because of their capabilities of dealing with uncertainty and vagueness. In this thesis two hybrid risk assessment systems have been developed which combine the FIS, ANNs, and expert risk assessment methodology to accomplish risk assessment. The FIS based and the ANNs based model are both introduced to give comparable results, which provides experts with a more confident risk score. The proposed hybrid models have been built and tested by using Fuzzy Logic Toolbox, the Neural Network Toolbox and the GUI (guide) of MATLAB®. Each methodology of the two risk assessment models have been tested and analyzed, which proves that these two Pipeline Risk Assessment models can be utilized in pipeline risk assessment areas.

Description
A Thesis Submitted to the Faculty of Graduate Studies and Research In Partial Fulfillment of the Requirements for the Degree of Master of Applied Science in Industrial Systems Engineering, University of Regina. vii, 148 p.
Keywords
Citation
Collections