Deep Learning for Quantitative Hydraulic Fracture Profiling from Fiber Optic Measurements

2021 ◽  
Author(s):  
Weichang Li ◽  
Han Lu ◽  
Yuchen Jin ◽  
Frode Hveding
Keyword(s):  
Deep Learning ◽  
Hydraulic Fracture ◽  
Fiber Optic

Leveraging coherent wave field analysis and deep learning in fiber-optic seismology

2021 ◽  
Author(s):  
Benjamin Schwarz ◽  
Korbinian Sager ◽  
Philippe Jousset ◽  
Gilda Currenti ◽  
Charlotte Krawczyk ◽  
...  
Keyword(s):  
Deep Learning ◽  
Wave Field ◽  
Real Time ◽  
Learning Strategies ◽  
Data Storage ◽  
Large Scale ◽  
Fiber Optic ◽  
Positive Impact ◽  
Field Analysis ◽  
Coherent Wave

<p><span>Fiber-optic cables form an integral part of modern telecommunications infrastructure and are ubiquitous in particular in regions where dedicated seismic instrumentation is traditionally sparse or lacking entirely. Fiber-optic seismology promises to enable affordable and time-extended observations of earth and environmental processes at an unprecedented temporal and spatial resolution. The method’s unique potential for combined large-N and large-T observations implies intriguing opportunities but also significant challenges in terms of data storage, data handling and computation.</span></p><p><span>Our goal is to enable real-time data enhancement, rapid signal detection and wave field characterization without the need for time-demanding user interaction. We therefore combine coherent wave field analysis, an optics-inspired processing framework developed in controlled-source seismology, with state-of-the-art deep convolutional neural network (CNN) architectures commonly used in visual perception. While conventional deep learning strategies have to rely on manually labeled or purely synthetic training datasets, coherent wave field analysis labels field data based on physical principles and enables large-scale and purely data-driven training of the CNN models. The shear amount of data already recorded in various settings makes artificial data generation by numerical modeling superfluous – a task that is often constrained by incomplete knowledge of the embedding medium and an insufficient description of processes at or close to the surface, which are challenging to capture in integrated simulations.</span></p><p><span>Applications to extensive field datasets acquired with dark-fiber infrastructure at a geothermal field in SW Iceland and in a town at the flank of Mt Etna, Italy, reveal that the suggested framework generalizes well across different observational scales and environments, and sheds new light on the origin of a broad range of physically distinct wave fields that can be sensed with fiber-optic technology. Owing to the real-time applicability with affordable computing infrastructure, our analysis lends itself well to rapid on-the-fly data enhancement, wave field separation and compression strategies, thereby promising to have a positive impact on the full processing chain currently in use in fiber-optic seismology.</span></p>

scholarly journals Neural Turbo Equalization: Deep Learning for Fiber-Optic Nonlinearity Compensation

2020 ◽  
pp. 1-1
Author(s):  
Toshiaki Koike-Akino ◽  
Ye Wang ◽  
David Millar ◽  
Keisuke Kojima ◽  
Kieran Parsons
Keyword(s):  
Deep Learning ◽  
Fiber Optic ◽  
Turbo Equalization ◽  
Nonlinearity Compensation

Fiber-optic distributed acoustic sensing of microseismicity, strain and temperature during hydraulic fracturing

Geophysics ◽  
2019 ◽  
Vol 84 (1) ◽  
pp. D11-D23 ◽  
Author(s):  
Martin Karrenbach ◽  
Steve Cole ◽  
Andrew Ridge ◽  
Kevin Boone ◽  
Dan Kahn ◽  
...  
Keyword(s):  
Hydraulic Fracturing ◽  
Hydraulic Fracture ◽  
Fiber Optic ◽  
Data Set ◽  
New Approach ◽  
Acoustic Sensing ◽  
Fiber Optic Technology ◽  
Physical Effects ◽  
Distributed Acoustic Sensing ◽  
Temperature Strain

Hydraulic fracturing operations in unconventional reservoirs are typically monitored using geophones located either at the surface or in the adjacent wellbores. A new approach to record hydraulic stimulations uses fiber-optic distributed acoustic sensing (DAS). A fiber-optic cable was installed in a treatment well in the Meramec formation to monitor the hydraulic fracture stimulation of an unconventional reservoir. A variety of physical effects, such as temperature, strain, and microseismicity are measured and correlated with the treatment program during hydraulic fracturing of the well containing the fiber and also an adjacent well. The analysis of this DAS data set demonstrates that current fiber-optic technology provides enough sensitivity to detect a considerable number of microseismic events and that these events can be integrated with temperature and strain measurements for comprehensive hydraulic fracture monitoring.

Hydraulic fracture characterization using fiber optic DAS and DTS data

2020 ◽  
Author(s):  
Robert Hull ◽  
Craig Woerpel ◽  
Kirk Trujillo ◽  
Rob Bohn ◽  
Ben Wygal ◽  
...  
Keyword(s):  
Hydraulic Fracture ◽  
Fiber Optic ◽  
Fracture Characterization

Case study of hydraulic fracture monitoring using multiwell integrated analysis based on low-frequency DAS data

2020 ◽  
Vol 39 (11) ◽  
pp. 794-800
Author(s):  
Masaru Ichikawa ◽  
Shinnosuke Uchida ◽  
Masafumi Katou ◽  
Isao Kurosawa ◽  
Kohei Tamura ◽  
...  
Keyword(s):  
Hydraulic Fracture ◽  
Fiber Optic ◽  
Low Frequency ◽  
Fracture Propagation ◽  
Propagation Direction ◽  
Integrated Analysis ◽  
Fiber Direction ◽  
Distributed Temperature Sensing ◽  
Fiber Optic Cable ◽  
Cumulative Strain

Distributed acoustic sensing (DAS) is an effective technique for hydraulic fracture monitoring. It can potentially constrain fracture propagation direction and time while monitoring strain perturbation, such as stress shadowing. In this study, we acquired passive DAS and distributed temperature sensing (DTS) data throughout the entire fracturing operations of adjacent production wells with varying offset lengths from the fiber-optic cable in the Montney tight gas area. We applied data processing techniques to the DAS data to extract low-frequency components (less than 0.5 Hz) and to construct the strain rate and cumulative strain maps for detecting responses related to fracture hits along the fiber-optic cable. We used low-frequency DAS (LF-DAS) results to estimate the fracture hit position and time, and in certain cases, to additionally estimate the fracture connection. By integrating LF-DAS results with DTS results, we detected the temperature changes around the compression response near the fracture hit position and time. Furthermore, we observed that timing of the fracture hit can be constrained more precisely by using high-frequency DAS data (greater than 10 Hz). We estimated the fracture propagation direction and speed from the estimated fracture hit position and time. The fracture propagation direction deviated slightly from a perpendicular line to the fiber direction. In addition, as estimated from the first fracture hit time, the fracture length and fluid injection volume had a proportional relationship. Due to challenges associated with the data, it is important to design data acquisition geometry and fracturing operations on the premise of acquiring LF-DAS data. It is also important to apply an additional noise reduction process to the data.

Near-field strain in DAS-based microseismic observation

Geophysics ◽  
2021 ◽  
pp. 1-49
Author(s):  
Ge Jin ◽  
Frantisek Stanek ◽  
Bin Luo
Keyword(s):  
Hydraulic Fracture ◽  
Fiber Optic ◽  
Moment Tensor ◽  
Near Field ◽  
Source Parameters ◽  
Field Strain ◽  
Far Field ◽  
Normal Strain ◽  
S Wave ◽  
Sign Patterns

Microseismic monitoring with surface or downhole geophone arrays has been commonly used in tracking subsurface deformation and fracture networks during hydraulic fracturing operations. Recently, the use of fiber-optic DAS technology has improved microseismic acquisition to a new level with unprecedentedly high spatial resolution and low cost. Deploying fiber-optic cables in horizontal boreholes allows very close observation of these micro-sized earthquakes and captures their full wavefield details. We show that DAS-based microseismic profiles present a seldomly reported near-field strain signal between the P- and S-wave arrivals. This near-field signal shows monotonically increasing (or decreasing) temporal variation, which resembles the previously reported near-field observations of large earthquakes. To understand the near-field strain behavior, we provide a mathematical expression of the analytic normal strain solution that reveals the near-field, intermediate-near-field, intermediate-far-field, and far-field components. Synthetic DAS strain records of hydraulic-fracture-induced microseismic events can be generated using this analytic solution with the Brune source model. The polarity sign patterns of the near-field and far-field terms in these synthetics are linked to the corresponding source mechanism’s radiation patterns. These polarity sign patterns are demonstrated to be sensitive to the source orientations by rotating the moment tensor in different directions. A field data example is compared to the synthetic result and a qualitative match is shown. The microseismic near-field signals detected by DAS have potential value in hydraulic fracture monitoring by providing a means to better constrain microseismic source parameters that characterize the source magnitude, source orientation, and temporal source evolution, and therefore better reflect the geomechanical response of the hydraulically fractured environment in the unconventional reservoirs.

Thermal Effects on Far-Field Distributed Acoustic Strain-Rate Sensors

2021 ◽  
Author(s):  
Smith Edward Leggett ◽  
Ding Zhu ◽  
Alfred Daniel Hill
Keyword(s):  
Phase Shift ◽  
Temperature Gradient ◽  
Strain Rate ◽  
Hydraulic Fracture ◽  
Thermal Effects ◽  
Temperature Effects ◽  
Fiber Optic ◽  
Displacement Discontinuity ◽  
Observation Well ◽  
Optical Phase

Abstract Fiber-optic cables cemented outside of the casing of an unconventional well measure cross-well strain changes during fracturing of neighboring wells with low-frequency distributed acoustic sensing (LF-DAS). As a hydraulic fracture intersects an observation well instrumented with fiber-optic cables, fracture fluid injected at ambient temperatures can cool a section of the sensing fiber. Often, LF-DAS and distributed temperature sensing (DTS) cables are run in tandem, enabling the detection of such cooling events. The increasing use of LF-DAS for characterizing unconventional hydraulic fracture completions demands an investigation of the effects of temperature on the measured strain response by LF-DAS. Researchers have demonstrated that LF-DAS can be used to extract the temporal derivative of temperature for use as a differential-temperature-gradient sensor. However, differential-temperature-gradient sensing is predicated on the ability to filter strain components out of the optical signal. In this work, beginning with an equation for optical phase shift of LF-DAS signals, a model relating strain, temperature, and optical phase shift is explicitly developed. The formula provides insights into the relative strength of strain and temperature effects on the phase shift. The uncertainty in the strain-rate measurements due to thermal effects is estimated. The relationship can also be used to quantify uncertainties in differential-temperature-gradient sensors due to strain perturbations. Additionally, a workflow is presented to simulate the LF-DAS response accounting for both strain and temperature effects. Hydraulic fracture geometries are generated with a 3D fracture simulator for a multi-stage unconventional completion. The fracture width distributions are imported by a displacement discontinuity method program to compute the strain-rates along an observation well. An analytic model is used to approximate the temperature in the fracture. Using the derived formulae for optical phase shift, the model outputs are then used to compute the LF-DAS response at a fiber-optic cable, enabling the generation of waterfall plots including both strain and thermal effects. The model results suggest that before, during, and immediately following a fracture intersecting a well instrumented with fiber, the strain on the fiber drives the LF-DAS signal. However, at later times, as completion fluid cools the observation well, the temperature component of the LF-DAS signal can equal or exceed the strain component. The modeled results are compared to a published field case in an attempt to enhance interpretation of LF-DAS waterfall plots. Finally, we propose a sensing configuration in order to identify the events when "wet fractures" (fractures with fluids) intersect the observation well.

Sign in / Sign up

Export Citation Format

Share Document