Chris Hugenholtz & Thomas Barchyn, University of Calgary
January 2019
Executive Summary
Leak detection and repair (LDAR) programs are a regulatory tool for finding and mitigating fugitive methane emissions from upstream O&G infrastructure. LDAR programs typically rely on close-range techniques outlined in the US EPA’s Method 21 or Alternative Work Practice; however, these techniques are slow, labor-intensive, and costly. Future LDAR programs are likely to incorporate mobile screening technologies (e.g., drones, vehicles, aircraft, and satellites) to achieve mitigation targets more costeffectively. In this Project we developed new hardware and analytics to support vehicle-based LDAR at upstream O&G facilities. The vehicle system measures the advection of methane plumes that cross the vehicle path.
The hardware system we developed consists of an integrated multi-sensor, roof-mounted payload. Data streams from the sensors are fused to provide high-frequency, real-time measurements of the wind vector, vehicle position, and methane concentration. These data are displayed using custom software to support real-time detection and localization of methane plumes at close-range (i.e., on site) and from several kilometers downwind. Post-processing with our software packages (ATLAS and SOLAG) can be used for a more rigorous assessment of detection, localization, and quantification.
Over a 1-year period the hardware/software system completed over 8,000 km of testing in varied terrain, land cover, and weather conditions. To demonstrate the system’s capabilities and data products, we present a case study using a controlled release of methane. The results indicate the system can localize discrete emissions sources to within a few meters in the cross-wind direction and estimate the flux to within 40% of the actual emissions rate. Further testing is required to understand the system’s limitations and operational niches.
Future improvements to the system include the integration of artificial intelligence to improve flux estimates, route planning software tools to optimize plume intersection, automated pad-level source discrimination and localization, and hardware/software upgrades to enable passive sensing – where measurements are collected, stored, and analyzed without any operator input.
# 17-ARPC-05