Abstract

High-resolution sensors provide essential information to the oil and gas industry when planning for corridor right-of-way, land restoration, damage prevention, risk management and emergency response/damage prevention. The world's growing need for energy has put pressure on the remote sensing community to create even higher resolution and more accurate sensors than ever before. The location of the area of interest where remote sensing is needed is often based on the location of potential petroleum reserves. Political, terrain and weather conditions often determine the choice of technology used to collect the necessary data. Continuous pipeline monitoring is becoming increasingly critical for the oil and gas industry, which faces new challenges due to higher demands for energy. This paper presents the design, development and testing of a smart, wireless sensor network for early leak detection in oil pipelines, based on sensor measurements. Today's low power sensors are capable of measuring pressure, temperature and flow changes due to leaks. We describe a system model for determining decision-making strategies based upon the ability to perform data mining and pattern discovery by utilizing sensor information to detect leaks or abnormal situations. We discuss the development of a method for determining actionable information using game theory. We focus on applying probabilistic models to leak detection and geospatial tools for assisting emergency response and pipeline repairs. Probabilistic predictions are critical in practice on many decision-making applications. However, applicability to big data is complicated by the difficulties of inference in complex probabilistic models, and by computational constraints. Copyright 2015, Society of Petroleum Engineers

You can access this article if you purchase or spend a download.