Accurate Simulation of Non Darcy Flow in Stimulated Fractured Shale Reservoirs
- Barry Rubin (Computer Modelling Group Inc)
- Document ID
- Society of Petroleum Engineers
- SPE Western Regional Meeting, 27-29 May, Anaheim, California, USA
- Publication Date
- Document Type
- Conference Paper
- 2010. Society of Petroleum Engineers
- 5.5.3 Scaling Methods, 1.10.1 Drill string components and drilling tools (tubulars, jars, subs, stabilisers, reamers, etc), 5.1.1 Exploration, Development, Structural Geology, 4.3.4 Scale, 5.8.2 Shale Gas, 3 Production and Well Operations, 4.1.5 Processing Equipment, 2.2.2 Perforating, 5.8.6 Naturally Fractured Reservoir, 5.1.5 Geologic Modeling, 4.1.2 Separation and Treating, 5.5.8 History Matching, 5.3.2 Multiphase Flow, 3.2.3 Hydraulic Fracturing Design, Implementation and Optimisation, 5.5 Reservoir Simulation
- 8 in the last 30 days
- 3,578 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 9.50|
|SPE Non-Member Price:||USD 28.00|
Unconventional shale gas reservoirs require stimulation via hydraulic fracturing of pre-existing fracture networks for practical exploitation, creating a stimulated reservoir volume (SRV). Within the SRV, gas flow from the nano-Darcy shale to the complex stimulated fracture network has been modeled in reservoir simulators using a variety of techniques which upscale/simplify the fracture network. The simulation techniques used in the past were normally not compared with reference solutions.
This work investigates using finely-gridded single well reference solutions (approximately 6-14 million cells) for simulating Darcy and non-Darcy flow within an explicitly modeled SRV complex fracture network, in 2-D, with and without primary hydraulic fractures, as well as scenarios which model stress sensitive permeability and later re-stimulation of a horizontal well. The network fractures use cells which are only 0.001 ft. wide.
The reference solutions are compared with standard dual permeability and MINC (multiple interacting continua) dual continua models as well as novel models which simulate flow inside of the SRV using coarse, logarithmically spaced, locally refined, dual permeability grids, and simulate flow outside of the SRV using unrefined dual permeability grids. These coarse models can be run in minutes on standard hardware, where as the reference models can take days to run on the same hardware.
We will show that excellent matches to the reference solutions are possible using a modest number of refinements to simulate the flow within the SRV when the fracture permeability and the fracture Forchheimer number (for non-Darcy flow) are scaled as described in the paper. These techniques allow the use of 2.0 ft. wide fracture conduits to mimic non-Darcy flow in 0.001 ft. wide fractures.
Good agreement between the reference and coarse models are observed even during the early flow period of the reservoir.
There have been many recent papers discussing the modeling of gas production from unconventional shale gas reservoirs. Work by Mayerhoffer et al.1 and Cipolla2 are two excellent examples. They discuss using numerical simulation of explicit fracture networks created in a stimulated reservoir volume (SRV) to model the physics of flow within a fractured shale reservoir. These papers discussed the simplifications made in order to produce reasonable results and noted(1) the need to model fractures explicitly at their true width (on the order of 0.001 ft.) when accounting for non-Darcy flow. It is this last requirement which can make the simulation of such systems a daunting task. These simulations, with explicit fracture representation, use large detailed grids and require significant execution times due to both the large number of cells used and the small size of fracture cells used.
The ultimate goal of this work is to produce predictive fractured shale gas simulation models which are easy to set up and that run in minutes rather than hours or days. These predictive models would be small enough to be quickly run while actually stimulating the reservoir and/or be used in history matching/optimization software.
It is easy to propose potential models, but in order to ensure the accuracy of the simplified models, it is necessary to compare them with extremely fine grid reference solutions which are capable of modeling fracture flow using cells which are no larger than the width of actual fractures, and flow into the fracture from the matrix using cells small enough to properly capture the very large pressure drops involved.
|File Size||1 MB||Number of Pages||16|