Implementing Genetic Algorithms to Assist Oil and Gas Pipeline Integrity Assessment and Intelligent Risk Optimization

Implementing Genetic Algorithms to Assist Oil and Gas Pipeline Integrity Assessment and Intelligent Risk Optimization

Gustavo Calzada-Orihuela (Morelos State Autonomous University, Cuernavaca, Mexico), Gustavo Urquiza-Beltrán (Morelos State Autonomous University, Cuernavaca, Mexico), Jorge A. Ascencio (Polytechnic University of Quintana Roo, Cancun, Mexico) and Gerardo Reyes-Salgado (Computer Science, National Center for Research and Technological Development (CENIDET), Cuernavaca, Mexico)
DOI: 10.4018/IJOCI.2017100104
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

Oil and gas industry, worldwide, needs to monitor, control and assess the elements that are involved in the general oil transportation and production processes. However, these processes are not risk free. The project proposes an intelligent support system that provides optimized projections for effective risk management. The project focuses on the development of a set of Genetic Algorithms (GAs), a branch of AI systems that assists to optimize the usage and distribution of resources. GAs will reduce the latent risks and potential dangers as much as possible. The main purpose is to minimize the risk levels in a pipeline segment based on their condition and by detecting optimal variable configurations: their Risk of Failure (RoF), Probability of Failure (PoF), Consequence of Failure (CoF), and their sub elements (threats and impacts). The heuristic results generated by this set of GAs show a significant reduction on the risk assessment measures, by finding “optimized” configurations of these variables.
Article Preview

Introduction

Nowadays around the world, pipeline systems are the most common method to transport oil, gas and other related products associated to their own production and extraction processes, mainly due to its proven reliability and effectiveness (Hopkins, 2009). However, the very nature of the oil and gas industry carries a constant potential risk around the processes involved. Pipelines could fail for many reasons, including: fabrication or installation flaws, incorrect operation, leaks caused by pipeline corrosion or theft, etc. Any unmitigated threat or unknown exposure could develop a failure, and it could potentially lead to extreme dangerous consequences for the company, the environment, and even to our human life.

Pipeline owners, and companies are compelled to monitor, assess and manage those associated risk and its components (Muhlbauer, 2006). Last decade, in Mexico alone, there have been more than two thousand incidents related to oil pipelines and facilities (PEMEX, 2011; SSPA, 2008). This is why the decision-making process, key factor in most industries, needs constant, accurate and updated feedback in order to prevent, as much as possible, future hazardous events. The oil transportation and distribution systems are usually assessed from different perspectives aiming to evaluate, timely and accurately, the current pipeline integrity.

Although there are different tools and methods to assist the decision-making process, risk assessment has become fundamental for financial analysis (Chang, 2014). It is important to coordinate efforts to provide improved resources in order to avoid incidents, including damage reduction, human losses and environmental disasters prevention caused by these failures. Consequently, oil industries and other companies closely monitor risks associated with their operations because risks that are not controlled nor assessed could produce costly outcomes. Accurate knowledge about the company’s current operational condition, history and future projections can provide a powerful business advantage.

Furthermore, there are three critical elements to consider when performing the pipeline integrity assessment (Muhlbauer, 2006): Risk of Failure (RoF), Probability of Failure (PoF), also known as Likelihood of Failure (Godoy, 2011), and Consequence of Failure (CoF). Risk of Failure depends on the PoF and CoF and defines the possibility of any hazardous event to happen. It also determines the possible, and extent of impact that it could cause. Hence the importance of risk models that provide solid information about the current condition on both sides of the equation, cause and consequence of a possible pipeline failure.

The oil industry, both Mexican and international, are extremely relevant in the global economy. Governments, universities, public and private organizations, and research institutes are continuously making enormous efforts to create state of the art advances related to the industry. These include methodologies and procedures, real time monitoring systems, data acquisition and management systems, decision support systems, AI based applications, Big Data frameworks, etc.

PEMEX, Mexican Petroleum, which is the leading company on oil extraction and transport in Mexico, regularly enhances and updates its systems, techniques, methodologies and protocols to ensure the safe and correct operation of the Mexican pipeline systems.

This project started from the need to continue those updates, and with the ambition to provide alternative systems to help risk reduction by assisting the company to make smarter choices. The Research Centre for Engineering and Applied Sciences (CIICAp) at the Morelos State Autonomous University (UAEM) in collaboration with the Physical Sciences Institute (ICF) at the National Autonomous University of Mexico (UNAM), and the National Technology and Science Council (CONACyT) supported this particular project.

Complete Article List

Search this Journal:
Reset
Open Access Articles
Volume 7: 4 Issues (2017)
Volume 6: 4 Issues (2016)
Volume 5: 4 Issues (2015)
Volume 4: 4 Issues (2014)
Volume 3: 4 Issues (2012)
Volume 2: 4 Issues (2011)
Volume 1: 4 Issues (2010)
View Complete Journal Contents Listing