Microseismic analysis is the primary tool available for fracture characterization in unconventional reservoirs. As DAS fibers are installed in the target reservoir and are thus close to the microseismic events, they hold vast potential for their high-resolution analysis. However, accurately detecting microseismic signals in continuous data is challenging and time-consuming. DAS acquisitions generate substantial data volumes, and microseismic events have a low signal-to-noise ratio in individual DAS channels. Herein we design, train, and deploy a deep learning model to automatically detect thousands of microseismic events in DAS data acquired inside a shale reservoir. The stimulation of two offset wells generates the microseismic activity. The deep learning model achieves an accuracy of over 98% on our benchmark dataset of manually-picked events and even detects low-amplitude events missed during manual picking.
|Number of pages||5|
|Journal||SEG Technical Program Expanded Abstracts|
|State||Published - 2021|
|Event||1st International Meeting for Applied Geoscience and Energy - Denver, United States|
Duration: 26 Sep 2021 → 1 Oct 2021