Deep Neural Networks for Detection and Location of Microseismic Events and Velocity Model Inversion from Microseismic Data Acquired by Distributed Acoustic Sensing Array.

Daniel Wamriew,Dimitri Pissarenko,Roman Pevzner,Evgenii Maltsev

doi:10.3390/s21196627

Abstract

Fiber-optic cables have recently gained popularity for use as Distributed Acoustic Sensing (DAS) arrays for borehole microseismic monitoring due to their physical robustness as well as high spatial and temporal resolutions. As a result, the sensors record large amounts of data, making it very difficult to process in real-/semi-real-time using the conventional processing routines. We present a novel approach, based on deep learning, for handling the large amounts of DAS data in real-/semi-real-time. The proposed neural network was trained on synthetic microseismic data contaminated with real-ambient noise from field data and was validated using field DAS microseismic data obtained from a hydraulic fracturing operation. The results indicate that the trained network is capable of detecting and locating microseismic events from DAS data and simultaneously update the velocity model to a high degree of precision. The mean absolute errors in the event locations and the velocity model parameters are 2.04, 0.72, 2.76, 4.19 and 0.97 percent for distance (x), depth (z), P-wave velocity, S-wave velocity and density, respectively. In addition to automation and computational efficiency, deep learning reduces human expert data handling during processing, thus preserving data integrity leading to more accurate and reproducible results.

Highlights

Fiber-optic cables are increasingly being used as Distributed Acoustic Sensors (DAS)for downhole microseismic monitoring due to their high spatial and temporal resolutions as well as large sensing distances [1] as compared to conventional geophones, which enables them to provide detailed images of the subsurface structure necessary for detection and location of microseismic events as well as velocity model estimation
We demonstrate the feasibility of application of deep learning approach to detect and locate microseismic events and simultaneously estimate the velocity model from DAS-acquired data
The test dataset comprised of 5000 DAS microseismic events from 50 distinct velocity models was used to evaluate the performance of the trained network

Summary

Introduction

Fiber-optic cables are increasingly being used as Distributed Acoustic Sensors (DAS)for downhole microseismic monitoring due to their high spatial and temporal resolutions as well as large sensing distances [1] as compared to conventional geophones, which enables them to provide detailed images of the subsurface structure necessary for detection and location of microseismic events as well as velocity model estimation. DAS technology measures the strain or strain-rate along a fiber-optic cable through. An interrogator sends a laser pulse along the fiber-optic cable and records the backscattered light. The phase difference between backscattered light within a gauge length is calculated to give the signal. An encounter with a seismic wave causes changes in strain/strain-rate, leading to changes in the recorded signal [2,3]. Amongst the many benefits of using fiber-optic cables for downhole microseismic monitoring is that, provided the cable is installed in the well, the system will be able to provide continuous, dense downhole recording while causing no interference with any other activities taking place in or near the well. Limitations of DAS include lack of broadside sensitivity, as discussed in Section 4.1, and gauge-length effects for small gauge-lengths

Methods

Results

Discussion

Conclusion