Hydraulic fracturing induces a network of fractures to promote the flow of oil and gas in tight shale reservoirs. In this paper, we discuss the detection of fracture hits during stimulation to avoid reaching an observing well. Our aim is to devise an early warning system for cases where fracture interference may adversely affect production. We propose the use of image-based algorithms such as template matching and convolutional neural network (CNN) on lowfrequency signals of distributed acoustic sensing (DAS) data to detect precursor events leading to a fracture hit.

This content is only available via PDF.
You can access this article if you purchase or spend a download.