Lee, Andrew (Massachusetts Institute of Technology), Dahan, Mathieu (Massachusetts Institute of Technology), Amin, Saurabh (Massachusetts Institute of Technology)

Integration of sUAS-Enabled Sensing for Leak Identification with Oil and Gas Pipeline Maintenance Crews

Scheduled for presentation during the "See-and-avoid Systems - II" (ThC2), Thursday, June 15, 2017, 17:00−17:20, Salon AB

2017 International Conference on Unmanned Aircraft Systems, June 13-16, 2017, Miami Marriott Biscayne Bay, Miami, FL,

This information is tentative and subject to change. Compiled on April 24, 2024

Abstract

The U.S. Department of Energy and Transportation considers pipeline security and the timely containment of leaks as a top priority for the oil and natural gas industry. However, despite significant investment in network sensing and maintenance, utilities still incur significant delays (and associated losses) in managing failures. This article focuses on the use of small Unmanned Aerial Systems (sUASs) for the inspection of network components and to facilitate timely repair of failures. Our framework integrates sUAS-enabled sensing with fixed sensing systems and ground-based maintenance crews, and aims to minimize the time to repair multiple network failures. It also reduces human effort in network inspection (e.g. manned reconnaissance, ground patrols, and leak surveys). We consider inspection tasks on a set of failure regions (localization sets) that are generated from fixed sensors installed on the network (e.g. pressure sensors). We focus on the problem of routing a set of available maintenance vehicles at specified yard (base) locations carrying sUASs to optimally identify and repair the network failures. To address this problem, we propose two Mixed- Integer Programming (MIP) formulations: (a) the multi-trip sUAS exploration problem, and (b) maintenance vehicle routing problem. We show that these formulations can be integrated to optimally route the sUASs for identification of multiple network failures that may occur anywhere within the localization sets (MIP-a); and to optimally dispatch maintenance vehicles to repair the identified network failures (MIP-b). We illustrate this approach on a benchmark pipeline network, and demonstrate the inherent tradeoffs between sUAS exploration time and maintenance vehicle travel time in our solution.