scholarly journals The Use of 3D Convolutional Autoencoder in Fault and Fracture Network Characterization

Geofluids ◽  
2021 ◽  
Vol 2021 ◽  
pp. 1-11
Author(s):  
Feng Xu ◽  
Zhiyong Li ◽  
Bo Wen ◽  
Youhui Huang ◽  
Yaojun Wang
Keyword(s):  
Pattern Recognition ◽  
Spatial Structure ◽  
Seismic Data ◽  
Fracture Network ◽  
Spatial Dimension ◽  
3D Seismic ◽  
Structure Information ◽  
Convolutional Autoencoder ◽  
Network Characterization ◽  
3D Seismic Data

Conventional pattern recognition methods directly use 1D poststack data or 2D prestack data for the statistical pattern recognition of fault and fracture network, thereby ignoring the spatial structure information in 3D seismic data. As a result, the generated fault and fracture network is not distinguishable and has poor continuity. In this paper, a fault and fracture network characterization method based on 3D convolutional autoencoder is proposed. First, in the autoencoder training frame, 3D prestack data are used as input, and the 3D convolution operation is used to mine the spatial structure information to the maximum and gradually reduce the spatial dimension of the input. Then, the residual network is used to recover the input’s details and the corresponding spatial dimension. Lastly, the hidden features extracted by the encoders are recognized via k -means, SOM, and two-step clustering analysis. The validity of the method is verified by testing the seismic simulation data and applying real seismic data. The 3D convolution can directly process the seismic data and maximize the prestack texture attributes and spatial structure information provided by 3D seismic data without dimensionality reduction and other preprocessing operations. The interleaving convolution layer and residual block overcome low learning and accuracy rates due to the deepening of networks.

Advanced Modeling of Interwell-Fracturing Interference: An Eagle Ford Shale-Oil Study

SPE Journal ◽  
2016 ◽  
Vol 21 (05) ◽  
pp. 1567-1582 ◽  
Author(s):  
Matteo Marongiu-Porcu ◽  
Donald Lee ◽  
Dan Shan ◽  
Adrian Morales
Keyword(s):  
Seismic Data ◽  
Pressure Field ◽  
State Of The Art ◽  
Fracture Network ◽  
3D Seismic ◽  
Eagle Ford Shale ◽  
Eagle Ford ◽  
In Situ Stresses ◽  
3D Seismic Data

Summary To investigate interwell interference in shale plays, a state-of-the-art modeling workflow was applied to a synthetic case on the basis of known Eagle Ford shale geophysics and completion/development practices. A multidisciplinary approach was successfully rationalized and implemented to capture 3D formation properties, hydraulic-fracture propagation and interaction with a discrete-fracture network (DFN), reservoir production/depletion, and evolution of magnitude and azimuth of in-situ stresses by use of a 3D finite-element model (FEM). The integrated workflow begins with a geocellular model constructed by use of 3D seismic data, publicly available stratigraphic correlations from offset-vertical-pilot wells, and openhole-well-log data. The 3D seismic data were also used to characterize the spatial variability of natural-fracture intensity and orientation to build the DFN model. A recently developed complex fracture model was used to simulate the hydraulic-fracture network created with typical Eagle Ford pumping schedules. The initial production/depletion of the primary well was simulated by use of a state-of-the-art unstructured grid reservoir simulator and known Eagle Ford shale pressure/volume/temperature (PVT) data, relative permeability curves, and pressure-dependent fracture conductivity. The simulated 3D reservoir pressure field was then imported into a geomechanical FEM to determine the spatial/temporal evolution of magnitude and azimuth of the in-situ stresses. Importing the simulated pressure field into the geomechanical model proved to be a critical step that revealed a significant coupling between the simulated depletion caused by the primary well and the morphology of the simulated fractures within the adjacent infill well. The modeling workflow can be used to assess the effect of interwell interferences that may occur in a shale field development, such as fracture hits on adjacent wells, sudden productivity losses, and dramatic pressure/rate declines. The workflow addresses the complex challenges in field-scale development of shale prospects, including infilling and refracturing programs. The fundamental importance of this work is the ability to model pressure depletion and associated stress properties with respect to time (time between production of the primary well and fracturing of the infill well). The complex interaction between stress reduction, stress anisotropy, and stress reorientation with the DFN will determine whether newly created fractures propagate toward the parent well or deflect away. The technique should be implemented in general development strategies, including the optimization of infilling and refracturing programs, child well lateral spacing, and control of fracture propagation to minimize undesired fracture hits or other interferences.

scholarly journals Phase-Based Image Analysis of 3D Seismic Data

2012 ◽  
Vol 2012 (1) ◽  
pp. 1-4 ◽  
Author(s):  
Peter Kovesi ◽  
Ben Richardson ◽  
Eun-Jung Holden ◽  
Jeffrey Shragge
Keyword(s):  
Image Analysis ◽  
Seismic Data ◽  
3D Seismic ◽  
3D Seismic Data
Sign in / Sign up

Export Citation Format

Share Document