Nearly 23,000 miles of oil-gathering pipelines have been installed in North Dakota in recent years to move large volumes of fluids from oil and gas fields to various processing facilities. This increased infrastructure requires new technology for pipeline safety to achieve the goal of zero spills or leaks.
Technology developed for monitoring pipeline systems needs to handle large amounts of data and provide real-time information to operators. This project, conducted by the University of North Dakota (UND) Energy & Environmental Research Center (EERC)–Research Institute for Autonomous Systems (RIAS), received exploratory funding from the State Energy Research Center (SERC). The project concept was based on the combination of subsurface and surface field data with machine learning to integrate information for monitoring for leaks from buried pipelines efficiently.
In Phase I, magnetic geophysical data and images from sensors attached to an unmanned aircraft system (UAS) platform were used for detecting buried steel pipelines in controlled experiments conducted by EERC and RIAS researchers. Machine learning workflows were developed for real-time pipeline detection.
The state-of-the-art tools developed and interdisciplinary knowledge gained in this project could support application of the technology to future projects funded by the U.S. Department of Transportation Pipeline and Hazardous Materials Safety Administration or presentation to the iPIPE consortium for consideration. Additional phases of research will be needed to explore the technology developed and tested during Phase I.